You are on page 1of 232

Prescribing under Pressure


General Editors Nikolas Coupland Adam Jaworski Cardiff University

Recently Published in the Series:

Talking about Treatment: Recommendations for Breast Cancer Adjuvant Treatment Felicia D. Roberts Language in Time: The Rhythm and Tempo of Spoken Interaction Peter Auer, Elizabeth Couper-Kuhlen, Frank Mller Whales, Candlelight, and Stuff Like That: General Extenders in English Discourse Maryann Overstreet A Place to Stand: Politics and Persuasion in a Working-Class Bar Julie Lindquist Sociolinguistic Variation: Critical Reections Edited by Carmen Fought

Prescribing under Pressure

Parent-Physician Conversations and Antibiotics

Tanya Stivers


Oxford University Press, Inc., publishes works that further Oxford Universitys objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright 2007 by Oxford University Press, Inc.

Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Stivers, Tanya. Prescribing under pressure: patient-physician conversations and antibiotics / Tanya Stivers. p. cm.(Oxford studies in sociolingistics) Includes bibliographical references and index. ISBN 978-0-19-531115-0 1. Antibiotics. 2. AntibioticsEffectiveness. Physician and patient. I. Title RM267.S75 2007 615'329dc22 2006048256

9 8 7 6 5 4 3 2 1 Printed in the United States of America on acid-free paper


lthough only one name appears on the cover of this book, it represents the thoughts, ideas, and contributions of many people. Some I formally acknowledge here, some appear in the list of references, but many who appear in neither place nonetheless helped me along the way by challenging me to consider yet another way of approaching the set of problems that this book brings together. I owe the biggest debt of gratitude to the people who let me study them: the nearly 800 parents and physicians I videotaped over the course of 5 years of eldwork, much of which was used for the present book. Although the research discussed here relies on videotapes of these interactions, many of these individuals (as well as others) shaped my thinking in quite different ways: parent comments in physician waiting rooms, physician comments or questions before or after parents arrived as well as during ethnographic interviews, and medical assistant comments during the course of a typical day all inuenced my thinking about what happens when a parent walks through the door with a sick child. Each of the subjects in my study granted me just a small amount of timemany would not even remember participating; most would surely be shocked to nd out that their few minutes would sum to provide me with a careers worth of data. Still, those minutes add up, and so I thank all of these anonymous individuals for their generosity. My biggest supporter from beginning to end has been John Heritage. He believed in the project from the moment I oated the idea by him, and his encouragement materially, emotionally, and intellectually was invaluable. I shudder to think how many versions of this material he has read over the years in the form of grant proposals, dissertation chapters, articles, and nally book chapters. Besides offering tireless support, he taught me how to study social interaction without losing track of either the structures through which interactants accomplish social actions or the fact that interactants have feelings, concerns, and agendas. He challenged me each step of the



way to be more ambitious, to be more creative in my thinking, and to consider the interplay between the microlevel interaction and the macrolevel social context. I am innitely grateful for the support, education, and contributions that he has provided. Many others to whom I owe a substantial debt of thanks were, in one way or another, connected to John. It was he who introduced me to Rita Mangione-Smith. Rita provided both data and much support for the work after qualitative analyses pointed me toward antibiotics. Rita had independently been investigating parent attitudes toward, and expectations for, antibiotics and had been surveying physicians on their perceptions. Thus, with our work combined, we could, for the rst time, see whether behaviors derived from conversation analysis could be linked with exogenous survey variables, and the merger was a success. As a student of conversation analysis at the University of California at Los Angeles, I was mentored by both John Heritage and Manny Schegloff. Then and since, Manny has consistently challenged me not to forget that whatever people are doing in a medical encounter, they are doing it Turn Constructional Unit by Turn Constructional Unit, turn by turn, sequence by sequence, as in all forms of social interaction. And time and time again, the importance of understanding each level of orderliness has proven critical to the analysis. To Manny I owe a thank-you for teaching and modeling both rigor and enthusiasm in studying interaction. Both were contagious early on, and have not waned. This book is heavily based on my UCLA dissertation in applied linguistics: Negotiating Antibiotic Treatment in Pediatric Care: The Communication of Preferences in Physician-Parent Interaction (2000). Several of the chapters are based on previously published articles. The ideas of chapter 2 previously appeared as Presenting the Problem in Pediatric Encounters: Symptoms Only versus Candidate Diagnosis Presentations, published in Health Communication (2002b). The ideas in chapter 5 previously appeared in Parent Resistance to Physicians Treatment Recommendations: One Resource for Initiating a Negotiation of the Treatment Decision, published in Health Communication (2005c). Many of the ideas in chapter 6 appeared in Participating in Decisions about Treatment: Overt Parent Pressure for Antibiotic Medication in Pediatric Encounters, published in Social Science and Medicine (2002a). Finally, chapter 7 is a compilation of work based partly on Non-Antibiotic Treatment Recommendations: Delivery Formats and Implications for Parent Resistance in Social Science and Medicine (Stivers, 2005b) and partly on Heritage and Stivers (1999), Online Commentary in Acute Medical Visits: A Method of Shaping Patient Expectations in Social Science and Medicine. I acknowledge Lawrence Erlbaum and Elsevier for allowing me to incorporate these articles into the present book. Bringing together the previous ideas in the form of this book was most importantly prompted by a conversation with Steve Clayman. It was his timely prodding and support that convinced me to revisit this work and draw it together. Besides prompting the writing, he generously read drafts of the chapters and provided much needed feedback, consistently pushing me to clarify my writing and, in turn, my thinking. I am also grateful to the Max Planck Institute for Psycholinguistics in Nijmegen, The Netherlands, and especially to Steve Levinson, for making the writing of



this book possible. The Language and Cognition Group, particularly Nick Eneld, has through their own work, their way of approaching problems, and their questions, pushed me to come to terms with disciplines and ways of thinking that I had previously been hopelessly ignorant of. This sentiment wove its way into this book and led me to attempt to situate the problem and discussions of it more broadly than I otherwise would have. Much appreciation is owed to my friends and family, who have been (and will probably continue to be) subjected to ranting over the years about antibiotic overprescribing and the structures of social action. Instead of asking me to stop talking about my work, they have engaged with my ideas and, through their own stories and thoughts, prompted me to consider new analytic angles. Special thanks to Ignasi Clemente, Amy Miller, Rob McClinton, Heidi and Luella Hood, Dr. Valentine, Kathi and Milt Schmutz, Jim and Jean Stivers, and Julian and Riley Scaff. Finally, thank you to Nik Coupland, Adam Jaworski, and Peter Ohlin for their support of this book.

This page intentionally left blank


1. The Miracle Drug: The Context of Modern Antibiotic Usage, 3 2. Foregrounding the Relevance of Antibiotics in the Problem Presentation, 23 3. Alternative Practices for Asking and Answering History-Taking Questions, 51 4. No Problem (No Treatment) Diagnosis Resistance, 77 5. Treatment Resistance, 105 6. Overt Forms of Negotiation, 131 7. Physician Behavior That Inuences Parent Negotiation Practices, 155 8. Conclusion, 185 Appendix: Transcript Symbols, 195 Notes, 199 References, 203 Index, 219

This page intentionally left blank

Prescribing under Pressure

This page intentionally left blank

The Miracle Drug

The Context of Modern Antibiotic Usage

The History of Medicine 2000 BCHere, eat this root. 1000 BCThat root is heathen. Here, say this prayer. 1850 ADThat prayer is superstition. Here, drink this potion. 1920 ADThat potion is snake oil. Here, swallow this pill. 1945 ADThat pill is ineffective. Here, take this penicillin. 1955 ADOops . . . bugs mutated. Here, take this tetracycline. 19601999 AD39 more oops . . . Here, take this more powerful antibiotic. 2000 ADThe bugs have won! Here, eat this root. Anonymous

n 2000, primary care physicians in the United States handed out approximately 126 million prescriptions for antimicrobials (McCaig, Besser, & Hughes, 2003). Basic arithmetic shows this to be approximately 2.5 billion doses consumed by ambulatory care patients alone. Although in some respects rates of antimicrobial drug use have fallen in the recent past, the annual population-based rate of prescribing in the United States remains 461 prescriptions per 1,000 people (McCaig et al., 2003). In pediatrics, the primary care specialty with the highest rate of prescribing, 235 of every 1,000 medical visits result in an antibiotic prescription (McCaig et al., 2003). Unfortunately, many of these prescriptions are for the treatment of viral illnesses. Because antibiotics are ineffective against viruses, such prescriptions are inappropriate, and their prevalence threatens the effectiveness of antibiotics in treating bacteria that cause pneumonia, strep throat, and ear infections (Streptococcus pneumoniae), some of the most common childhood illnesses. This book asks why the problem of inappropriate antibiotic prescribing persists and seeks answers by investigating the


details of interactions between pediatricians and parents in visits for children with symptoms of an upper respiratory tract infection.

History of Antibiotics
The history of antibiotics is relatively short. By most accounts, penicillin was discovered by British scientist Alexander Fleming in 1928, and initial results were published less than a year later (Fleming, 1929). For various reasons, it was not until 1942 that the British and Americans began mass-producing the drug. Clinical use became widespread during World War II, when penicillin was heavily marketed to the public using what we now call direct to consumer advertising, such as Thanks to penicillin, he [showing a picture of an American soldier on the ground] will come home (Levy, 1992: 10). At this time, penicillin was widely heralded as a miracle drug. People were astonished at the ability of antibiotics to cure illnesses overnight that had previously been fatal. But the golden era of antibiotics was to be short-lived. Fleming himself noted early on that the drug required careful dosing and that bacteria mutate quickly in response to exposure. When accepting his 1945 Nobel Prize, Fleming warned of bacterial resistance. Although his concerns were primarily with underdosing (a problem that persists both through noncompliance and, primarily in developing countries, through lack of knowledge or proper antibiotic supplies), his illustration of the threat of bacterial resistance is still relevant today:
Here is a hypothetical illustration. Mr. X has a sore throat. He buys some penicillin and gives himself, not enough to kill the streptococci but enough to educate them to resist penicillin. He then infects his wife. Mrs. X gets pneumonia and is treated with penicillin. As the streptococci are now resistant to penicillin the treatment fails. Mrs. X dies. Who is primarily responsible for Mrs. Xs death? Why Mr. X whose negligent use of penicillin changed the nature of the microbe. (Fleming, 1945)

As it turns out, the resistance problem is actually worse than Fleming envisioned in two ways. It turned out that even restrained use of antibiotics will generate bacterial resistance over time. Moreover, the resistance problem emerged very quickly. As early as 1946, just a few years after mass production began, when penicillin was still available without a prescription in the United States, there were reports of penicillin-resistant bacteria probably due in no small part to the misuse that Fleming had been concerned about (Levy, 1992). But at that point, new antibiotics were quickly coming onto the scene. Fast forward just 30 years to the early 1970s, and antibiotic resistance had already come to be considered a real public health threat. Strains of bacteria that cause meningitis and ear infections in children and a strain that caused gonorrhea once again proved fatal. Both had previously been treated successfully with penicillin or a derivative (Levy, 1992). At present, just over 60 years since the beginning of wide-scale antibiotic use, the growing problem of bacterial resistance to antibiotics is widely recognized as one


of societys greatest health threats (Adam, 2002; Baquero, Baquero-Artigao, Canton, & Garcia-Rey, 2002; Doern & Brown, 2004; Harbarth, Albrich, & Brun-Buisson, 2002; Jacobs, Felmingham, Appelbaum, Grneberg, & Group, 2003; McCaig & Hughes, 1995; Neu, 1992; Reichler et al., 1992; Schwartz, 1999; Smolinski, Hamburg, & Lederberg, 2003; Whitney et al., 2000; Wise et al., 1998). As discussed in a brief introduction to a recent special issue of Emerging Infectious Diseases devoted to antimicrobial resistance, bacterial resistance promises to pose a still larger problem soon because the manufacture of new drugs is at a virtual standstill and has been since 1968 (Weber & Courvalin, 2005). This means that we are on the brink of returning to an era when common illnesses, long thought to have been conquered, may once again prove fatal. At present, illnesses caused by such bacteria are already more difcult to treat (Dagan, 2000; Friedland, 1995; Watanabe et al., 2000), more expensive to treat (Gums, 2002; Holmberg, Solomon, & Blake, 1987), and result in increased mortality (Feikin et al., 2000). For all of these reasons, the problem of bacterial resistance is a paramount public health concern worldwide.

Determinants of Bacterial Resistance

What lies behind the bacterial resistance problem? The answer is far from simple. A 2000 World Health Organization (WHO) report points to a number of issues, including the overuse of antibiotics in livestock (World Health Organization, 2000) and international travel that spreads resistant bacteria (Fidler, 1998; Memish, Venkatesh, & Shibl, 2003). But the biggest single factor across both developing and developed nations appears to be the very problem of misuse that Fleming pointed to in 1945 (Albrich, Monnet, & Harbarth, 2004; Harbarth & Samore, 2005). Prescribing when it is not clinically appropriate is still relatively common (Kaiser et al., 1996; Orr, Scherer, MacDonald, & Moffatt, 1993; Todd, Todd, Damato, & Todd, 1984), and this is a primary contributor to the generation of bacterial resistance (Albrich et al., 2004; Cristino, 1999; Deeks et al., 1999; Gomez et al., 1995; Nava et al., 1994; Watanabe et al., 2000). A second contributing form of misuse involves prescriptions for an inappropriate type of antibiotic (e.g., using a second-line, stronger antibiotic rather than a rst-line one when clinical guidelines support the latter) (Hossain, Glass, & Khan, 1982; Hui, Li, Zeng, Dai, & Foy, 1997), and this, too, has been shown to contribute to bacterial resistance (Kozyrskyj et al., 2004; McCaig et al., 2003). Although there is substantial overlap, misuse takes somewhat different forms in developing versus developed nations. Misuse in Developing Countries In developing countries, the factors that lead to misuse revolve around problems of supply and regulation. For instance, with respect to supply issues, developing countries often have difculties in accessing the right medication or adequate doses of medication (Guyon, Barman, Ahmen, Ahmen, & Alam, 1994; Uppal, Sarkar, Giriyappanavar, & Kacker, 1993). It is also not always possible to gain access to diagnostic tests that would allow health practitioners important insight into the condition(s) that


they are treating (e.g., Bosu & Ofori-Adjei, 1997; Horgerzeil et al., 1993; Mamun, 1991). Because of restricted access to information, developing countries may also suffer from a lack of physician knowledge (Igun, 1994) and a corresponding lack of patient knowledge (Braithwaite & Pechere, 1996). Regulation is also a serious problem in developing countries, where antibiotics are often available directly from pharmacies, health care providers, or roadside stalls without government restriction (Bartoloni et al., 1998). Evidence suggests that when people can self-medicate, misuse is rampant in terms of both usage for inappropriate conditions (Haak, 1988; Radyowijati & Haak, 2003; Vuckovic & Nichter, 1997) and inappropriate dosing (Indalo, 1997; Kunin et al., 1987). To drive down the cost of antibiotics, some patients purchase only one dose at a time, use insufcient dosages, or truncate the course of antibiotics. Both a lack of understanding of how the drug works and the implications of such practices, as well as economic issues, contribute to this problem. These are practices that effectively educate bacteria to resist antibiotics. However, prescribing issues are quite a complex problem in developing countries because in spite of fewer antibiotic regulations, they do not necessarily have higher rates of bacterial resistance. This is probably because many people, particularly in rural areas, still have substantial barriers to access, including transportation, cost, or lack of providers in the area. For instance, India has few regulations over antibiotic prescribing but very low rates of resistance in the bacteria relevant to our discussion here, at least in rural areas (Thomas, 1999). This means that there may actually be less misuse of antibiotics in rural areas, despite the lack of regulations in place (Quagliarello, Parry, Hien, & Farrar, 2003). But as a consequence, people who should appropriately be treated with antibiotics may not receive them either. Although it is difcult to fully assess the situation across developing nations, there is substantial evidence that penicillin and erythromycin resistance is an emerging problem in community-acquired Streptococcus pneumoniae across many regions of the world, even in more rural areas (see Okeke et al., 2005, for a review). Misuse in Developed Countries Developed countries generally have fewer problems with respect to physician knowledge, access to high-quality drugs (barring problems with counterfeit drugs and the like), and adequate amounts of medication. Moreover, developed countries typically regulate access to antibiotics, and many have public health campaigns in place to educate patients about antibiotics. However, existing research shows that inappropriate prescribing of antibiotics for viral infections is nonetheless common in many developed countries. In the United States, researchers and policy makers are strongly advocating more judicious prescribing practices (e.g., Bell, 2002; Belongia et al., 2001). But advocacy alone, even from national and international organizations such as the Centers for Disease Control and the World Health Organization, has not stopped doctors from inappropriately prescribing (Finkelstein et al., 2000; Gonzales, Malone, Maselli, & Sande, 2001; Gonzalez, Steiner, & Sande, 1997; Mainous, Hueston, & Clark, 1996; Mangione-Smith et al., 2004; McCaig, Besser, & Hughes, 2002; Metlay, Shea, Crossette, & Asch, 2002; Pennie, 1998). For viral colds, the


prescribing rate across populations is estimated to be approximately 30%, and for bronchitis and other illnesses typically of viral origin, it is estimated to be as high as 60% (Gonzales et al., 2001). Cultural factors may also play a role. People with different cultural backgrounds may be more or less likely to visit a physician for a particular condition (Pachter, 1994). People from particular cultural backgrounds may be more likely to expect that a visit to a health care provider will result in a prescription for treatment (Radyowijati & Haak, 2003). Such expectations may, in turn, affect prescribing rates among particular ethnic and cultural groups (Froom et al., 2001; Harbarth et al., 2002; Radyowijati & Haak, 2003). Differences in cultural attitudes specically toward antibiotics may lead to another related issue: the transportation and sales of noncontrolled antibiotics into developed countries where prescriptions are required (Mainous et al., 2005). When physicians are asked why they prescribe against clinical evidence and national guidelines, they commonly cite issues such as patient pressure (Avorn & Solomon, 2000; Little et al., 2004; Stevenson, Greeneld, Jones, Nayak, & Bradley, 1999), lack of time (Little et al., 2004), and a concern with avoiding lawsuits over missed bacterial infections (Sargent & Welch, 2001). Whether in developing or developed nations, antibiotic misuse is unlike most types of medical errors in that it is an error that has far greater social impact than individual impact. As Avorn and Solomon observe, Antibiotics are the only drug class whose use inuences not just the patient being treated but the entire ecosystem in which he or she lives, with potentially profound consequences (2000: 128). This was foreseen by Fleming, as noted earlier in the excerpt from his Nobel Prize acceptance speech (1945). The social consequences of misuse may be at the level of the community, as in the Fleming example, because bacteria can easily spread from children to adults within the local community, but they can also be at the regional, national, or even international level (McCormick et al., 2003). As Levy points out, bacteria do not have respect for national borders (Levy, 1992). Thus, it is possible for resistant bacteria to cross the world within 24 hours (Fidler, 1998). And travelers denitely spread bacteria around: Approximately 1,500 of every 100,000 travelers returning from developing countries bring with them an acute febrile respiratory tract infection (Steffen & Lobel, 1996). Misuse that does not result from lack of knowledge or supply problems typically pits the individual against society and thus represents a social dilemma: situations in which individual rationality leads to collective irrationality (Kollock, 1998).1 Antibiotic misuse by individuals is perhaps best understood as the type of social dilemma Hardin made famous in his article in Science in 1968. Hardins example was that of herders who collectively have access to common land for cattle grazing. Individual rationale would have it that each herder should put as many grazing cows as possible onto the land, even though the commons will be damaged as a result. To make the decision that would be collectively best would require all herders to act in a way that is not in the interest of the individual (i.e., putting fewer cattle on the land) (Hardin, 1968). This type of social dilemmausing the land for individual gain even at the cost of the collectiveHardin terms a commons dilemma. Typically, patients and parents of child patients do not view their illnesses as viral or bacterial but as minor or serious. A serious illness is one that they have tried


to wait out or to treat but that has persisted or an illness that stands in the way of some special event for which they do not want to be sick. At least among those who believe that it is not good for human society to use antibiotics often, these cases are much like the grazing situation described by Hardin.2 For parents, the desire to use antibiotics is rational at the individual level because of their belief that antibiotics will help their children get better faster. For physicians, the desire to prescribe antibiotics is rational at the individual level because they believe that it will satisfy the parent, enable the visit to be closed, and allow both parties to move on. The desire to use antibiotics is irrational at the collective level because they expose bacteria to the drug and thereby enable them to mutate and develop resistance to the drug. In general, humans may be inclined to prioritize themselves as individuals or families over the larger community. As noted by Humphrey in his important paper on social intelligence, individuals have evolutionary reasons for prioritizing the survival of their own genes over others and thus to do well for oneself whilst remaining within the terms of the social contract on which the tness of the whole community ultimately depends calls for remarkable reasonableness (Humphrey, 1988). And solutions to such problems are challenging, to say the least. (See Kollock, 1998, for a discussion of strategic solutions to social dilemmas.) We will return to this in chapter 8.

Bacterial Resistance as a Global Problem

Ultimately, the community that stands to suffer because of individual-level decisions is the global one. Even in countries where inappropriate prescribing is relatively low, such as The Netherlands (Melker & Kuyvenhoven, 1994; Otters, van der Wouden, Schellevis, van Suijlekom-Smit, & Koes, 2004), inappropriate prescribing does occur (Otters et al., 2004). Still, national policies clearly do make a difference, as evidenced by the broad range of rates of antimicrobial resistance and generally corresponding rates of inappropriate antibiotic prescribing across countries. Europe is particularly interesting in this respect because of the close proximity of so many countries. According to reports from the Alexander Project 19982000 (a continuing surveillance study that examines the susceptibility of bacteria involved in respiratory tract infections), even countries that share a border can have dramatically different rates of penicillin resistance to bacteria (Jacobs et al., 2003). Whereas Portugal has a rate of 8.2% penicillin resistance, Spain has a rate of 26.4%. Whereas Germany has a rate of 1.9%, Switzerland a rate of 8.6%, and Belgium a rate of 5.7%, France has a staggeringly higher rate of 40.5%. The same goes for nearby countries as well. And although relative to France, Belgiums rate of 5.7% is quite good, relative to its other border country, The Netherlands, it is quite poor. The Netherlands has the lowest rate in the European Union, only 1.1%. Nearby UK has a rate of 10.9%, and growing still worse, neighboring Ireland has a rate of 24.1%. In general, the data show a generally consistent pattern between antimicrobial usage and resistance prevalence, and these patterns also appear generally consistent with outpatient antibiotic sales (Cars, Mlstad, & Melander, 2001) and antibiotic usage (Albrich et al., 2004). France is again the highest, Spain is also quite high, and Germany and The Netherlands are again very low.


The Current Investigation

The problem of inappropriate prescribing is clearly large and amorphous. In this book, we focus on one problem area: inappropriate prescribing of antibiotics for viral upper respiratory tract infections (URTIs) among children in the United States. There are a variety of reasons for thinking this a useful population to study. First, 65 to 70% of URTIs are viral (Wald, Guerra, & Byers, 1991). Second, according to the National Ambulatory Medical Care Survey, American pediatric patients receive two to three times more antibiotic prescriptions than any other patient group, including the elderly (Aronoff, 1996). Third, compared with adult populations, the pediatric population has been particularly resistant to efforts to alter inappropriate prescribing for viral URTIs (Belongia, Knobloch, & Kieke, 2005). Fourth, according to Alexander Project data, the United States has one of the highest rates of Streptococcus pneumoniae resistance to penicillin worldwide. Its rate of 25% surpasses its neighbor Mexico, a developing country with minimal antibiotics regulation, which has a rate of 22%. Only Spain, Japan, Israel, France, and Hong Kong (with a whopping 69.9% rate of bacterial resistance to penicillin) surpass the United States (Jacobs et al., 2003). Thus, U.S. prescribing practices for children with URTI symptoms may yield insight into the larger problems of misuse, both nationally and internationally. This book demonstrates that pediatrician-parent interactions provide a critical window into the macrolevel problem of bacterial resistance and antibiotic misuse in the United States. Close examination of such encounters reveals the impact that microlevel interactional phenomena have on diagnostic and treatment outcomes in URTI visits. This investigation will argue that while the root of misuse in developing countries is more clearly a public health issue, the root of misuse in developed countries like the United States is at least equally a sociological issue.3 Misuse in the United States Earlier we discussed various determinants of misuse in developed countries. If we move more specically to the United States, we can look more closely at this issue. One rather obvious contributor to inappropriate prescribing is whether physicians understand the relationship between viral infections and antibiotics. Because research suggests that 89% to 97% of U.S. physicians do understand this relationship (Schwartz, Freij, Ziai, & Sheridan, 1997; Watson et al., 1999), the question remains as to why physicians continue to overprescribe antibiotics in the face of the antibiotic resistance problem. As mentioned earlier, physicians commonly cite patient and parent pressure as a reason for prescribing (Barden, Dowell, Schwartz, & Lackey, 1998; Palmer & Bauchner, 1997; Schwartz, 1999; Schwartz et al., 1997; Watson et al., 1999). In pediatrics, 50% to 70% of parents visiting report an expectation that their child will be given antibiotics (Hamm, Hicks, & Bemben, 1996; Mangione-Smith et al., 2004; Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; MangioneSmith, McGlynn, Elliott, Krogstad, & Brook, 1999; Sanchez-Menegay & Stalder, 1994). However, parents reports of expectations are not necessarily associated with inappropriate antibiotic prescribing (Mangione-Smith et al., 2006; Mangione-Smith



et al., 1999). And physicians are not accurate predictors of which parents expect antibiotics and which do not (Mangione-Smith et al., 2006; Mangione-Smith et al., 1999). Additionally, researchers in both adult and pediatric contexts have found that doctors perceptions of patients expectations for antibiotics have a signicant effect on whether doctors prescribe antibiotics, even in cases where they judged them to be not indicated (Britten & Ukoumunne, 1997; Cockburn & Pit, 1997; Gani et al., 1991; Hamm et al., 1996; Mangione-Smith et al., 2006; Mangione-Smith et al., 1999; Mangione-Smith, Stivers, Elliott, McDonald, & Heritage, 2003; Vinson & Lutz, 1993). Specically, in one study there was a 25.5% increase in the probability that the physician would prescribe an antibiotic if he or she perceived the parent to expect it, controlling for a range of other issues (Mangione-Smith et al., 2006). Additionally, when physicians thought parents expected antibiotics, they diagnosed middle ear infections and sinusitis more frequently (49% and 38% of the time, respectively) than when they did not think antibiotics were expected (13% and 5%, respectively). These gures are likely to be low because of improved behavior during the study period (Mangione-Smith, Elliott, McDonald, & McGlynn, 2002). This suggests a disconnect between what parents report and how physicians perceive them. Because what physicians perceive appears to inuence their behavior, and they can access parents expectations only through parents behavior, it is this that appears to be most consequential, which raises the issue of what parental behaviors lead physicians to believe that parents are looking for antibiotics. Overt parent requests for antibiotics might be expected to be the culprit, and physicians typically cite and complain about this occurring (Schwartz et al., 1997; Stevenson et al., 1999). Although overt requests and other forms of overt lobbying for antibiotics do occur (discussed in chapter 6 primarily), they are quite rare (Fischer, Fischer, Kochen, & Hummers-Pradier, 2005; Stivers, 2002a). On the other hand, this book will argue that less direct interactional behaviors also communicate pressure for antibiotic prescriptions, even if, at times, unintentionally. This book will argue that in a variety of ways, parents actively participate in the visit in ways that pressure physicians in the direction of bacterial diagnoses and antibiotic prescribing. This book will also argue that parents, even when vying for antibiotics, are oriented to this interactional work as in the physicians domain of expertise, and thus this issue is something parents work to manage. Pressure as Parent Participation Patient participation is currently an important topic of discussion in health services research and health policy circles. Much of the emphasis from local to national levels is to encourage physicians to involve patients or parents in treatment decisions. According to the goals of Healthy People 2010, patients who participate actively in decisions about their health care can have a positive impact on national health (U.S. Department of Health and Human Services, 2000). Researchers assert that patients should, whenever possible, be offered choices in their treatment decisions (Brody, 1980; Butler et al., 2001; Deber, 1994; Emanuel & Emanuel, 1992; Evans, Kiellerup, Stanley, Burrows, & Sweet, 1987; Falloweld, Hall, Maguire, & Baum, 1990;



Kassirer, 1994; Levine, Gafni, Markham, & MacFarlane, 1992). Several American medical associations now recommend that physicians overtly involve patients in their decision making. For instance, the American Cancer Society, the American Urological Association, the American Gastroenterological Association, the American College of Physicians, and the National Institutes of Health (NIH) all recommend shared decision making for decisions surrounding cancer screening (Frosch & Kaplan, 1999). Although there is recognition that not every patient wants to participate in their health care, the primary rationale fueling these recommendations is, rst, that patients generally do have the desire and are entitled to participate in treatment decisions (Blanchard, Labrecque, Ruckdeschel, & Blanchard, 1988; Cassileth, Zupkis, Sutton-Smith, & March, 1980; Emerson, 1983; Ende, Kazis, Ash, & Moskowitz, 1989; Faden, Becker, Lewis, Freeman, & Faden, 1981; Swenson et al., 2004; Thompson, Pitts, & Schwankovsky, 1993) and, second, that patients have improved outcomes when they participate in medical decision making, including satisfaction (Brody, Miller, Lerman, Smith, & Caputo, 1989; Brody, Miller, Lerman, Smith, Lazaro, et al., 1989; Evans et al., 1987), patient health (Brody, 1980; Greeneld, Kaplan, Ware, Yano, & Frank, 1988; Kaplan, Greeneld, & Ware, 1989; Mendonca & Brehm, 1983; Schulman, 1979), and patient mental well-being (Brody, Miller, Lerman, Smith, & Caputo, 1989; Evans et al., 1987; Falloweld et al., 1990; Greeneld et al., 1988). Although the movement toward shared decision making in health care has certainly taken root in the care of chronic conditions, the issues have been far less explored in acute care. But the social factors encouraging partnership in chronic care may nonetheless also be affecting acute care encounters. First, because of the consumerist movement in health care, patients can be seen to be moving away from the guidance-cooperation models depicted as normative by Parsons in the 1950s (1951) and documented empirically in the 1970s (Byrne & Long, 1976). Specically, as Haug and Lavin (1983) suggest, the consumer model refocuses the balance of power on the patients rights (as purchaser) and on the physicians obligations (as seller) rather than on the physicians rights (to direct) and patient obligations (to follow directions) (p. 213). As summarized by Roter and Hall (1992), patients, particularly younger and more highly educated ones, are becoming more likely to exhibit consumerist behaviors (Ende et al., 1989; Hibbard & Weeks, 1985; Reeder, 1972), which include having sophisticated medical knowledge, seeking information through reading, and exercising independent judgment in following physicians recommendations. Kravitz, Bell, and Franz (1999: 873) suggest that patients are more than the passive recipients of doctors actions; they inuence the clinical encounter through use of their own linguistic resources. Second, as of 1985, drug companies have been allowed to market prescription medication directly to consumers (DTC). Promotional strategies encourage patients to act as consumers both in terms of product knowledge and with explicit suggestions to ask your doctor if X is right for you. As summarized by Pinto, Pinto, and Barber (1998), this shift was brought about because of pressure by drug companies who felt that managed care companies restricted their access to physicians and to patients in the sense that often only particular drugs would be covered by a given



company. Competitiveness in the drug market also added pressure. Additionally, though, patients with the activist mindset of many . . . baby boomers (p. 93) also gave the movement a needed push. Once in place, the DTC move fueled still more patient activism: patients are active decision makers for their own health needs (or the health needs of their babies) and . . . physicians are not the exclusive inuencers in brand selection (pp. 9192). Research suggests that advertising is encouraging patients to ask about medication (Peyrot, Alperstein, van Doren, & Poli, 1998) and that physicians are likely to prescribe or consider prescribing a drug requested by a patient (Borzo, 1997). With respect to patients asking, Sleath, Svarstad, and Roter (1997) examine data involving patients receiving care for chronic conditions and who are prescribed a psychotropic drug. They claim that patient-initiated talk showing that they wanted a new prescription was associated with prescribing 20% of the time. More recently, Kravitz and his colleagues have shown that requests do affect prescribing rates (Kravitz et al., 2005). Moreover, there is additional evidence that patients are becoming more consumer oriented: In a survey asking how people are likely to react to being denied a requested prescription, Bell, Wilkes, and Kravitz (1999) found that 46% of respondents reported being likely to be disappointed by a denied prescription request. Additionally, 25% of all respondents reported being likely to pressure physicians. Finally, 24% of all respondents who were denied a prescription said they would be likely to seek a prescription from another physician. Related to DTC advertising is a third factor that may affect not only chronic but also acute care: A vast amount of medical information is now readily accessible to the public through the Internet (du Pr, 2000). Thus, patients are not only being inuenced to be more proactive in their own health care but also being given resources with which to become more knowledgeable. Kravitz and colleagues (1999) found that the most common information requests involved questions about medications and that the most common action request was for medications. This has largely been studied with respect to chronic or serious conditions, but these factors also affect acute care. As du Pr points out, It is no longer enough (if it ever was) to simply tell patients what to do. Empowered patients want information and the right to make their own decisions (2000: 15). Thus, even in situations such as acute care, where physicians may feel that a more medical or disease-based approach is appropriate, there may be pressure from both patients and policy makers for patient participation in the visit. Treatment for an acute illness is typically conceptualized as something that the patient is directed to do by the physician or that the physician recommends or advises (e.g., Byrne & Long, 1976). And some research suggests that in the primary care context, doctors are much less likely to involve patients or parents in treatment decision making (Braddock, Edwards, Hasenberg, Laidley, & Levinson, 1999; Elwyn, Edwards, & Kinnersley, 1999; Tuckett, Boulton, Olson, & Williams, 1985). Generally, the treatment phase of an acute medical encounter is thought of as doctor driven (at least in the sense that doctors make recommendations and provide advice for treating the patients medical problem), particularly in contexts where there is a view that only one course of action is correct and thus decision making is basically straightforward (Coulter, 1997; Szasz & Hollender, 1956). Additionally, patients



with acute problems are generally seeking a physicians treatment recommendation or advice, and the physicians recommended treatment is generally grounded in his or her medical knowledge. Therefore, patients in the acute contexts might be assumed to interact within the guidance modelnot disagreeing or querying the physicians treatment recommendation (Szasz & Hollender, 1956). And if patients were to inuence acute visits at all, we might expect it to be done subtly, given the general orientation of these visits. For these reasons, only a method that examines interactions at a detailed level would be able to identify such practices. This is the method adopted in this book. Here what I show is that when we look carefully at the details of parent-physician interaction, we see that parents and physicians frequently go through a subtle but very much observable negotiation of the childs illness. Negotiation occurs at virtually every stage in the visit, from the opening statements to the doctor to the reception of the physicians treatment recommendation and a variety of places in between. This book will step through each of the phases of the medical visit in order to show a variety of different practices (mostly covert) that parents initiate and that can be seen to affect the diagnostic and treatment outcome of their childs medical visit. Thus, contrary to what would be deduced from the existing literature, this book shows that even in rather doctor-centered visits, where the physician has made no real effort to explicitly involve the parent in the treatment process, parents affect the diagnostic and treatment outcomes of the visits through their interactions with the physician. Methodology Historically, there has been relatively little connection between large-scale social or public health problems and microlevel studies. Rather, these problems have been most typically investigated through large-scale surveys of either medical records or of physicians and parents. [(As one example of a survey that has generated studies relevant to this domain, see the National Ambulatory Medical Care Survey (CDC, 2004), which is a key survey in primary care for the Centers for Disease Control.)] When provider-patient interactions are examined, this most commonly involves process analysis methodology: a coding of the interaction, followed by an analysis only of the coding rather than the interaction. Debra Roter has been the leading gure in the development of this work since the pioneering research of Barbara Korsch in the 1960s (e.g., Korsch, Gozzi, & Francis, 1968). Interactions in this and similar types of approaches are coded on the basis of analyst- or literature-driven constructs (a top-down approach), whether for issues of patient participation or physician behavior, rather than codes that have emerged from an understanding of how interactions in the context of interest work. (See Roter & Hall, 1992, for a review of one of the major coding schemes in interaction studies in public health, and Roter, 2002, for a full bibliography of studies using the Roter Interaction Analysis System coding scheme.) Microlevel analytic methods such as discourse and conversation analysis have (but to a far lesser extent) taken root in the realm of health communication. These studies have illuminated important dimensions of medical interaction but, with rare exceptions (e.g., Waitzkin, 1991), the results are not generalized or are not generaliz-



able. For instance, using a microanalytic approach to discourse, Mishler shows that physicians and patients commonly pursue very different (and, at times, conicting) agendas during medical visits: the medical agenda and the lifeworld agenda (Mishler, 1984). West (1984) published the rst conversation analytic book on medical interaction. She wanted to understand the role language played in structuring both the social and the power relationships between physicians and patients. Todd and Fisher published a collection of discourse analytic papers dealing with the organization of medical communication (Todd & Fisher, 1993), and Heritage and Maynards recent collection of conversation analytic studies examines each phase of the medical visit and its organization (Heritage & Maynard, 2006). Both discourse analysis (DA) and conversation analysis (CA) take the perspective that medical interaction is, at its most basic level, still basic social interaction that is occurring in an institutional context (Drew & Heritage, 1992a). These methodologies treat social interaction as a highly structured domain where the structural underpinnings, like the structural underpinnings of a molecule, can be examined and understood. For conversation analysts in particular, social interactants accomplish social actions through language (See Heritage, 1984b, for summaries; Levinson, 1983). Thus, interactants greet each other, request things, and complain or invite through language, and the doing of these social actions is itself highly structured. One of the hallmarks of conversation analysis is that in analyzing any bit of social interaction, analysts must validate their understandings of participants social actions through an examination of interactants responses. This virtually necessitates that analysts look at interaction through the lens of the sequence (e.g., an initiating turn and a response) rather than restricting themselves to individual words, phrases, or sentences, as linguists have historically done. This methodology was particularly valuable in the present study because of the problem of understanding what parent behaviors physicians were understanding as communicating pressure to prescribe, regardless of parent intent. In examining social interaction in sequence structural terms, CA looks for patterns in the interaction that form evidence of systematic usage such that a particular turn design, for instance, can be identied as a practice through which people accomplish a particular social action either vocally or visibly. For example, from ordinary interaction contexts, we see practices for opening telephone conversations (Schegloff, 1968, 1972b, 1977), competing for epistemic rights over a claim (Heritage, 1998; Heritage & Raymond, 2005), or inviting another interactant to complete ones turn at talk (Lerner, 1996). To be identied as a practice, a particular communication behavior must be seen to be recurrent and to be routinely treated by a recipient in a particular way such that it can be discriminated from related or similar practices. The signicance of these practices can be understood in terms of (1) the immediate sequences in which they occur, (2) the larger activities in which they are embedded (Heritage & Sorjonen, 1994), and (3) the overall organization of the phases in the interaction. The latter two levels of organization are particularly signicant when CA is used to analyze interaction in institutional contexts, such as medical visits, because of the general goal orientation of participants in these interactions (Drew & Heritage, 1992b).



A CA approach is highly structural in orientation. To this end, CA researchers in medicine have been interested in practices of social interaction that reveal structure at different levels: the visit (i.e., overall structure), the activity (e.g., the treatment activity/phase), the sequence (e.g., the opening question and its response), or the turn (e.g., turn constructional practices). Some studies clearly t into one level, such as studies of the visits overall structure (for relevant work on phases, see Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991) or studies of turn design within a phase, such as Heritage and Stiverss examination of online comments offered during the physical examination (Heritage & Stivers, 1999). Other studies cross multiple levels. For instance, Perkyl examines alternative ways of designing a diagnosis delivery. This then is relevant both at the activity and the turn levels (Perkyl, 1998). Very few conversation analytic studies have attempted to connect interactional practices to large-scale exogenous issues, whether they are relational, socioeconomic, demographic, or public health issues (but see Boyd, 1998; Clayman, Elliott, Heritage, & McDonald, 2006; Clayman, Heritage, Elliott, & McDonald, in press; Heritage, Boyd, & Kleinman, 2001; Kleinman, Boyd, & Heritage, 1997). This book represents a conversation analytic investigation of how parents and physicians communicate about children with routine upper respiratory tract infection symptoms, and it demonstrates that conversation analytic ndings can offer results that bear on the large-scale public health problem of inappropriate antibiotic prescribing. Data This book draws on three data sets. First, there is a corpus of 65 videotaped encounters involving 6 pediatricians from 5 practices collected as pilot data that I will refer to as the Hillside data set. Then there is a corpus of 295 audiotaped encounters involving 10 pediatricians from 2 practices that I will refer to as the Seaside data set. Finally, there is a corpus of 522 videotaped encounters from 38 physicians in 27 practices that I will refer to as the Metro data set. All data were collected between September 1996 and June 2001 in Southern California. Children ranged in age from newborn through 16 years old, and all were accompanied by parents. Most visits involve children from 6 months to 10 years of age. All visits involve children who were being seen for routine illnesses, and approximately 98% of children had routine upper respiratory tract infection symptoms. Informed written consent was obtained from all participating parents and physicians in all samples. Only parents who could conduct the visit in English were admitted into the studies. For purposes of anonymity, in all transcripts pseudonyms replace any use of a subjects name or other identifying information (e.g., school names). Demographic information was not collected for the Hillside data set, though I expect it represents a midpoint (socioeconomically) between the Metro and Seaside data. The Metro data involved parents who were 34 years old, on average, with a median household income of $40,000. Most of the caregivers were mothers: 86% were female. Fifty-three percent of the parents were Latino, with 28% white, 12% African American, and 7% Asian. Sixty percent of parents were high school graduates but did not have a college degree. Twenty-four percent had at least an undergrad-



uate degree, whereas 16% had less than a high school degree. The Seaside data were, on balance, somewhat wealthier, more likely to be white, slightly older, and more likely to have a college education than the more diverse Metro sample. The demographic backgrounds of the Seaside and Metro data sets are described elsewhere in greater detail (Mangione-Smith et al., 2006; Mangione-Smith et al., 1999). Statistical information that is mentioned throughout the book refers to either the Seaside or Metro data. All data shown here were transcribed by the author according to the conventions originally developed by Gail Jefferson (see Appendix for conventions). The Role of Children in the Pediatric Visit In earlier work (Stivers, 2001), I aligned myself with other researchers of pediatric interactions against studies that fail to take into account the role of the child in the interaction (e.g., Pantell, Steward, Dias, Wells, & Ross, 1982). Here, I appear vulnerable to this very criticism. However, in my analyses of these interactions, what became clear is that children play quite a small role in the domain of treatment negotiation. Interestingly, even when children are active participants in the visit, it is rare that they perform the behaviors outlined in chapters 2 through 5 (but for a rare case, see Extract 6.1). They typically do orient to the doctor as someone who can tell them what is wrong and give medicine (see Extract 2.1 for an example). For this reason, they can occasionally, without intention, make it difcult for a parent to perform a behavior that might have pushed for a bacterial diagnosis or antibiotic treatment. For instance, if a child presents his or her own problem, parents are, at least in that sequential location, blocked from presenting it in their own words: words that, as we will see, can communicate a very particular stance toward the outcome of the visit. For this reason, there will be relatively little discussion of the childs contributions, despite the fact that in general interaction, their role is certainly important. The Social Context of These Visits Research on physician-patient interaction has been growing steadily, most signicantly since Barbara Korschs groundbreaking pediatric communication studies (e.g., Korsch et al., 1968). I will not attempt to review that literature here (but see Heritage & Maynard, 2006, for a comprehensive review). Instead, I will focus on the issues that are most relevant to understanding what pediatricians and parents are dealing with in their interactions involving children with routine childhood illnesses, and specically with upper respiratory illnesses. Many American readers who have attempted to reach their primary care physician during their lunch hour, after hours, or even when the phone lines are busy are probably familiar with a common recording that physicians place on their voicemail that instructs patients to call 911 emergency or to seek help in an emergency room if their problem is life threatening. Virtually anything about which parents or patients seek assistance from their primary care doctor is something that they do not perceive as life threatening. But of course, many adults and children have health problems that they live with or manage by themselves (Dunnell & Cartwright, 1972). In the case of the visits with which we are dealing, it is likely that every child has had a similar



illness in the past, and it is virtually certain that every parent has experienced an illness similar to the one the child has several times, if not on an annual or semiannual basis. Thus, parents in these visits have made a decision to seek medical care at this time. Physicians, too, are aware that each visit was not a virtual certainty (as might be expected with acute pain that results in a trip to the emergency room). Rather, visits are understood to be the result of an active consideration of alternatives. In general, this book will argue that parents seeking medical help for these routine illnesses feel they have gone beyond the point where their own expertise is sufcient. Some parents may be coming specically to get antibiotics; some may be coming because they are getting no sleep, and their child is cranky, disturbing the household, and they do not know what to do; others may want reassurance that what they have been doing is right and that there is no more to be done. In all cases, though, they have a problem that they no longer feel comfortable handling on their own. Two issues seem to inhabit these interactions: (1) the legitimacy of the visit and (2) the treatability of the child. Legitimacy In the way that adult patients present their reason for visiting the physician, they often include statements that work to show they have not rushed to the doctor at the rst sign of a problem but have waited a reasonable length of time, have come for good reason, and have attempted to manage their troubles prior to seeking medical assistance (Heritage & Robinson, 2006a). Similarly, patients work to show that they have not been overly attentive to their bodiesnoticing the slightest or most minimal changebut rather are coming to the physician only with rather unusual noticings or problems (Halkowski, 2006). Heritage and Robinson (2006a) argue that there are three basic ways that patients in acute primary care encounters display their orientations to their conditions as doctorable or worthy of evaluation as a potentially signicant medical condition, and worthy of counseling and, where necessary, medical treatment (p. 58): 1. Patients routinely include in their problem presentations attributions to third parties in order to give support to their decision to seek medical assistance (i.e., physicians, spouses, friends, or acquaintances said they should see a doctor). 2. Patients routinely display troubles resistance both (a) in the report of their decision to visit the doctor (e.g., stating that they have waited some length of time, or that they tried over-the-counter medications) and (b) in their description of their condition (e.g., they offer objective rather than subjective evidence of their difculty as severe enough to warrant the visit). For example, patients with shoulder pain will state that they cannot latch a seat belt by way of indirectly indexing the severity and doctor worthiness of their complaint, rather than describe the degree of their pain. 3. Patients rarely offer any diagnosis of their condition and furthermore orient to this as a behavior to be avoided (Gill, 1998). In this way, they defer to the physicians knowledge for solving their medical problem.



The ways parents communicate about their childrens conditions appear to be somewhat different. Whereas adult patients and parents share an orientation to the doctorability of the medical conditions, the pediatric context appears to have characteristics that are at variance with adult acute visits. First, parents more rarely formulate the reason for their childs visit as based on a third partys recommendation. Rather, this is typically reserved as a justication for a concern either offered subsequent to the presentation or following indications by the physician that the child may not have a problem (e.g., see Extract 7.24). Second, parents typically report their decision to visit the doctor more straightforwardly and in less troubles-resistant ways than adults. For example, parents appear more willing to go to the doctor quickly on behalf of their child than on their own behalf, and they provide less justication for this behavior. Although as Parsons (1951) pointed out, adults are normally obliged to resist the sick role and to make light of their troubles, the sufferings of little children are another matter (Strong, 1979: 204). While troubles resistance may be invoked during pediatric encounters in the form of showing, for example, that they did not rush to the doctor, the data in my corpus also support Strongs suggestion that concerns to justify a visit to the doctor may be somewhat relaxed in the pediatric context. Third, parents orientations to bodily attentiveness appear to be markedly different when they are acting as caregivers rather than patients. For instance, according to Halkowski (2006), adults who are overly concerned about themselves risk being thought to be seeing the doctor in a motivated way. Halkowski further suggests that adult patients regularly show a balance between attentiveness and inattentiveness to their bodies and emergent symptoms. By contrast, parents acting as an advocate for their child appear to be more attentive. In this context, the balance normally seen in adult primary care may be recalibrated. Concern over a childs well-being is generally seen as the sign of a good, if slightly overanxious, parent. This is supported further by data involving British health visitors and rst-time mothers (Heritage & Lindstrm, 1998; Heritage & Se, 1992). For example, in this data extract, the health visitor is illustrating the types of noticings and the level of detail at which noticings should be made. The parents are instructed to notice when she smi:les (line 9), when . . . shes holding her head up better (lines 910), and when she can see (lines 1112).
(1.1) Extract from Heritage & Lindstrm, 1998: 404 1 2 3 4 5 6 7 8 9 10 HV: F: HV: .hh These uhm (1.0) are the notes that I carry arou(t) with me:, Mm hm, And I (0.2) I uh record your babys progress on he:re. (0.2) .hhh [So that uhm (.) I want to know when shes [(Oh) doing new things when she smi:les and when she (.) .hh uh:m you know shes holding her head up

HV: M: HV:



11 12 13

better: .hh I want you to notice if she: (.) .hh can see:_ ((datum continues with father volunteering information about babys sight).

In contrast to an avoidance of overly self-attentive behavior in adults, with children we see an orientation to both the acceptability and desirability of close bodily monitoring. Related to this argument, parents may be more sensitive to potential perceptions of negligence when they are acting as their childs caregiver than when they are acting on their own behalf. For example, a parent may hear a doctors questions about her childs health as testing her capabilities as a mother (Bates, Bickley, & Hoekelman, 1995; Heritage & Se, 1992). Parents have some reason to be concerned about their pediatricians perceptions of them. Sheridan (1994) surveyed pediatricians and family practitioners about their perceptions of the accuracy of parents reports of their childrens symptoms. She showed that while only 1% of parents were perceived as actually falsely reporting or inducing their childs symptoms, 23% were perceived to be in some way misrepresenting their childs symptoms (e.g., exaggerating them) (Sheridan, 1994). Fourth, parents appear markedly more likely to offer possible diagnoses in the pediatric context. I will analyze this practice in detail in chapter 2, but for this discussion it is important to recognize that this behavior may indicate that parents feel more entitled to have expertise over, and to participate in, the diagnosis and treatment of their child than in their own care. Specically, a parent who is knowledgeable about childhood illnesses, symptoms, remedies, and the like is displaying good parenting. By contrast, an adult patient who is knowledgeable about acute illnesses may be viewed, and treated, as an uncooperative or bossy patient (Papper, 1970). Within the pediatric context, we have observed that parents, on the whole, are less oriented to (1) diffusing responsibility for seeking medical care and (2) exhibiting troubles resistance and are more willing to (3) diagnose their childs condition and (4) be attentive to their childs body. That said, we observed earlier that with routine illnesses parents do make a decision that this illness at this point in time requires medical attention, whereas similar illnesses at other times have not. So, although in many ways parents are under fewer constraints to legitimize their visit, there nonetheless appears to be some underlying concern that their visit be legitimate, or at least that its legitimacy be validated, and in various behaviors a concern with legitimating their visit can be observed. However, to some extent it appears that the concern is less to establish legitimacy than to have it validated. Whereas adults appear to frequently treat their visits legitimacy as questionable from the outset, parents appear more concerned that the physician not undermine the inherent legitimacy of their visit. This will be most clearly visible in chapter 3 on history-taking questions and chapters 4 and 5 on diagnosis delivery and treatment recommendations. One reason for this might be that parents may demand and expect more doctoring when they are acting as an advocate for their child rather than as a patient. Roter and Hall (1992) suggest that parents are more likely to be assertive on anothers behalf, especially a childs (p. 17). They provide an example of a father who, after coming to an emergency room, announced that he would only see a doctor who



was a parent (p. 17). Researchers have also found that parents are more willing to seek out information, ask questions, and voice concerns when speaking on behalf of their child (Korsch et al., 1968). And parents may have a stronger concern than adult patients about the cause of their childs illness (Korsch et al., 1968). Treatability These issues come together in a marked way with respect to treatment decisions. Here, several key issues converge: First of all, parents have familiarity with these illnesses both in their own experience and in observing their children. Therefore, they may often feel that they know what the problem is and how it may need to be treated. Just as parents and physicians orient to greater latitude across a range of behaviors in the pediatric encounter, in contrast to the adult encounter, this may also affect parents willingness to explicitly or implicitly seek out treatment for their child. That is, in contrast to an adult patient, a parent may be more willing to (1) ask about treatment, (2) pressure physicians for treatment, or (3) display expertise about treatment options for their child. Second, in contrast with adult care, parents may feel additional pressures to cure their child. One pressure parents face in terms of sick children is the need for a quick solution to their problem. As caregivers, parents are responsible for both properly caring for a sick child in terms of keeping them at home and taking them to a doctor if needed and also in terms of getting medicine for them. In todays society, the pressure to accomplish these things quickly has increased. As a pediatrician quoted in a newspaper article put it, Years ago, parents might keep a child at home and just sit out an infection. Now most dont have that luxury (Warren, 1998). In a society where two working parents are increasingly common, a sick child poses a problem for the family in terms of both nighttime sleep and daytime care. The pressure parents feel to get their children well may understandably translate to pressure on the doctor to make them well. Thus, for parents the pressures they feel may translate to more latitude in terms of offering accounts of what they think is bothering their child, pressing the physician to treat the child, and offering opinions on how to treat the child. Third, as was mentioned earlier, in the pediatric context, physicians may feel more social pressure to cure a child patient than an adult patient. Insofar as children are considered more of the societys responsibility than adults are (e.g., see Strong, 1979), pediatricians may feel self-imposed pressure to do what they can to help the child get well quickly. Furthermore, they may feel parental pressure to make the child well. In this way, the physician may feel a greater obligation to proactively address the condition of a sick child than a sick adult. In sum, from both the parents perspective and the doctors perspective, there is a pressure within the pediatric context to cure the child of an illness. As was mentioned earlier, for a variety of reasons parents come to the medical visit seeking a solution for their childs problem. Throughout the visit, physicians must balance two issues that are, at times, competing: maintaining the legitimacy of the visit and communicating whether the child has a treatable problem or not. If the childs problem is treatable, this is unproblematic: The legitimacy of the visit is quite



easily upheld through the prescription of medication. But if the childs illness is not treatable, as in the case of most viral colds or other infections, then this becomes a struggle. For parents, the issues are inverted: If a physician threatens the legitimacy of the visit by indicating that the condition is either not treatable or not problematic, they may do work to reestablish the legitimacy of their visit and/or advocate for the problematic and treatable nature of the childs problem. Interestingly, what parents consider treatable is rooted in their folk model of illness. Helman notes that prior to World War II and the mass production (and availability) of antibiotics, the typical illness model viewed colds as something triggered by external sorts of causes such as being in the cold air, drafts, or getting chilled when not wearing shoes (Helman, 1978). With this model came home remedies, so few physician visits were made. By contrast, postantibiotic patients suddenly were more likely to visit physicians for colds. Arguably, this was because all forms of illness were suddenly lumped together as caused by germs and therefore were considered treatable (Helman, 1978). Although we are nearly 30 years past the time when Helman wrote about his suburban general practice, the basic problem is the same. A concern with treatability is different from a concern with legitimacy. A parent can orient to the visit as legitimate and to their childs illness as treatable. Parents can respond to a physicians recommendation against treatment as problematic either because it delegitimates the visit (e.g., Yeah, it was my wife who called; I gured there wasnt much you could do) or because they feel that their child needs treatment (e.g., Can I at least have thuh prescription an Ill decide whether or not to ll it in a couple days?). However, it may not be entirely clear, to an analyst or to the physician, whether a given physicians action is problematic because it deals a blow to the legitimacy of the visit or to the treatability of the illness. Even with respect to treatability, parents can want treatment without desiring antibiotics specically, as will be discussed further in chapter 6. Britten and her colleagues have shown that adult patients do not always or simply want antibiotics (Britten, 1994; Britten, Jenkins, Barber Bradley, & Stevenson, 2003; Britten, Stevenson, Barry, Barber, & Bradley, 2000). Moreover, other studies show that patients and parents alike are not, across the board, less satised if they fail to get an antibiotic (Himmel, LippertUrbanke, & Kochen, 1997; Mangione-Smith et al., 2001). Much of this book will be concerned with interactional practices that are taken by the physician to be solely concerned with treatability and, even more specically, taken to be directly indexing antibiotics. It is precisely because of this perception by physicians that we will see that parents accomplish negotiation for antibiotics.

Overview of the Book

This book will examine the parent-pediatrician negotiation of legitimacy, treatability, and antibiotics in particular, as these issues emerge throughout the visit. The book is generally laid out in the order of the acute care medical visit (drawing on Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991): (1) opening; (2) establishing the reason for the visit; (3) history taking; (4) physical examination; (5) diagnosis; (6)



treatment; and (7) closing. Chapter 2 will examine the beginning of the visit proper, when parents are frequently offered an opportunity to explain their reason for seeking medical help. This chapter will suggest that parents display rather different stances toward their childrens illness and thus toward their preferences for the visit outcome through their formulation of their reason for visiting. Chapter 3 focuses on taking the childs illness history, during which physicians primarily ask questions and parents and children answer questions. Through the ways that physicians design their questions, they reveal their current diagnostic and treatment trajectories. We will see that although in one sense parents are put in a relatively constrained and arguably powerless position in this phase, they are nonetheless quite capable of working within these sequential and structural constraints to encourage physicians away from one diagnostic trajectory and/or toward an alternative diagnostic and treatment trajectory. Chapter 4 examines parent resistance as a response to no-problem diagnoses. This chapter argues that through the use of three different types of responses and because of the structural organization of the diagnosis, parents can take issue with the physicians diagnosis and, at times, lead the physician to alter the diagnostic and/ or treatment trajectory from no problem and/or nontreatable to problematic and/or treatable. Chapter 5 examines parent resistance in a second environment: following a treatment recommendation. Taken together with chapter 4, this chapter shows that resistance can take quite different forms, depending on the normative sequential organization of the action it is responding to. Still, as with diagnosis resistance, treatment resistance can be observed to be a powerful tool to negotiate in favor of antibiotic treatment. Chapter 6 examines the relatively rare behavior of parents overtly lobbying for antibiotics. This chapter shows that this type of behavior can take several particular formats that can be more or less direct (and more or less coercive). It also shows that the practices, though more frequent during the treatment recommendation phase of the visit, can be offered throughout the visit. Chapter 7 shifts from the parent to the physician to explore several ways in which physicians can, through the actions they perform and their design, shape whether parents perform problematic behaviors such as diagnosis and treatment resistance. This chapter focuses on three behaviors: online commentary, formulation of the diagnosis, and formulation of the treatment recommendation. Finally, chapter 8 concludes by speculating about the issues underlying the problems of inappropriate antibiotic prescribing, particularly in developed countries but also as it contributes to the global problem of bacterial resistance to antibiotics.

Foregrounding the Relevance of Antibiotics in the Problem Presentation

s discussed in chapter 1, medical visits are generally conducted in a way that proceeds rather systematically through a series of activities, beginning with an opening and progressing through to treatment discussion and closing. To understand the variety of ways that parents inuence the outcome of the visit, we will, in the course of this book, look at a number of these activities in detail. This chapter is concerned with an activity that generally occurs very early in the visit: when physicians and parents establish why the child is visiting the physician. Depending on the scheme of the medical encounters structural organization, this activity may be treated as the beginning of the history taking or as initiating an activity in its own right. I follow Byrne and Long (1976) in treating it as shifting to establish the reason for the visit. The question that physicians generally ask parents (e.g., What can I do for you today?) offers parents an opportunity to shape the course of the visit by describing their childs problem in their own words and thus emphasize particular dimensions of the illness and de-emphasize others. With this question, physicians also provide parents with an opportunity to formulate their worries (or not), to project and tell a story about the problem (or not), and/or to offer their own speculations about the problems cause, all in the course of their response. When physicians solicit the problem, this represents the rst and sometimes only sequentially provided for opportunity that parents have to shape the physicians view of the problem and directly inuence the treatment decision.1 Thus, this is the obvious starting point for our study of parentphysician negotiation of treatment.



Parents generally respond to physicians questions about the reason for their visit by using one of two problem presentation formats. The rst is simply a description of the childs symptoms (e.g., He has a rash all over his body), and for this reason we term it a symptoms only presentation. The characteristic feature of the second format is that it includes the mention of a possible diagnosis (e.g., We were thinking she has an ear infection because shes been having pain), and for this reason we term it a candidate diagnosis presentation. In distinguishing between these two primary forms of problem presentation, this chapter describes alternative responses by physicians that display different analyses of the parents stance toward the childs illness. Specically, in cases in which the childs problem is presented with a symptoms-only description, parents are treated as having adopted the stance that they are primarily seeking a medical evaluation of the child. By contrast, in cases where the childs problem is presented with a candidate diagnosis, parents are treated as having adopted the stance that they are seeking conrmation of their diagnosis and seeking treatment for the illness condition. Each of these patterns will be discussed in turn in the following sections.

Establishing the reason for the visit is an activity that is present a large majority of the time in acute care medical visits. Even when physicians skip the activity and begin by starting the childs history taking (e.g., by beginning with a history-taking question such as So how long has the cough been going on?), parents nonetheless commonly offer their reason for visiting. Current research suggests that around 85% of acute encounters involve the patient or parent presenting the problem (Heritage & Robinson, 2006a; Stivers, 2001, 2002b). Establishing the problem is important because it generally plays an important role in structuring the rest of the visit: Both what the problem is and how it is presented set the agenda for the visit. But the problem presentation is important for other reasons as well: A physician needs an accurate and thorough description of the patients problem to provide a correct diagnosis (Ong, de-Haes, Hoos, & Lammes, 1995; Pendleton, 1983). Related to this, the problem presentation allows patients to formulate their problem or concern in their own words and allows the inclusion of both biomedical and life world dimensions of the problem and its impact on the patient (Fisher, 1991; Frankel, 1984; Heritage & Robinson, 2006b; Mishler, 1984). Ruusuvuori (2000) showed that the problem presentation is internally structured. She examined several key aspects of the problem presentation, including how patients begin and end their problem presentations and their vocal and visible resources for holding the oor during their presentation. Establishing the reason for the medical visit constitutes the shift to medical business or the rst topic (Byrne & Long, 1976; Heath, 1981; Robinson, 1998). Previous interaction research has established that openings typically affect later activities in institutional interactions, including the way the issue is addressed or the remedy that is suggested. For example, the way a problem is presented to 911 emergency call takers can affect whether they agree to dispatch help immediately



following the problem presentation (e.g., see Whalen & Zimmerman, 1987; Whalen, Zimmerman, & Whalen, 1988). Boyd (1998) has shown that the way interactions are opened can not only have interactional consequences but also affect whether the request being made is granted. She explored medical peer review telephone calls in which physician-reviewers representing a national utilization review rm call physicians who have proposed the surgical insertion of tympanostomy tubes for the management of recurrent ear infections. The reviewers, at the end of the phone call, approve or decline the surgery on behalf of the patients insurance company. Boyd found that the formulation the reviewer employed in moving to the business of the call was signicantly associated with whether the surgery was approved. Additionally, she found (1997) that in cases where the reviewers decision was negative, certain initiating formulations were associated with less interactional conict. Although this research involves relating the same speakers actions (i.e., the speakers openings and decisions), it shows the importance of the opening as an activity in these contexts.

Symptoms-Only Problem Presentations

The most common way that childrens problems are presented is with a symptomsonly presentation. Across the Seaside and Metro data sets, this format was used about 55% of the time (see Stivers, 2002b, regarding the Seaside data set results). The symptoms-only term underscores the fact that the problem presentation offers only a description of the problems the child is experiencing and does not attempt to identify the illness condition. If children present their own problem, it is virtually always with this format. For example, see Extract 2.1. Here a boy who is about 10 years old and his father are visiting the physician because the boy has, as he says, red spots.
(2.1) 202 1 2 3 4 5 6 7 DOC: O:kay: Robert. (0.5) DOC: Whats up.=h BOY: -> Uhm I have these little red s:pots all over my body. (0.5) BOY: -> An:- we dont know what they are: (really)

In this case, the boy rst describes his primary symptom (lines 45). As a response to Whats up. (line 3), the telling of his primary symptom displays his orientation to that symptom as being the reason for their visit. Then, after a bit of silence, he adds a second turn constructional unit (TCU) (Sacks, Schegloff, & Jefferson, 1974) that emphasizes his and his fathers (and perhaps his familys) concern for a diagnosis of this symptom (line 7). With this second unit An: - we dont know what they are: (really) the boy focuses on the evaluation of the spots as the reason for his visit. By contrast, the question of whether the spots are treatable (i.e., treatable with



prescription medication) is not raised and is thus understandably left contingent on the evaluation. Here, then, the matter of diagnosis is the focus of the problem. This case is unusual because the boy explicitly indexes his and his fathers and/or familys desire for a diagnosis of the illness. It is more common that the request for evaluation be left implicit but nonetheless to be the underlying reason for visiting. This can be seen in Extract 2.2.
(2.2) 1188 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 DOC: MOM: -> -> -> DOC: MOM: -> DOC: MOM: DOC: And so: do- Whats been bothering her. (0.4) Uh:m shes had a cou:gh?, and stufng- stuffy no:se, and then yesterday in the afternoo:n she started tuh get #really goopy eye:[s, and every= [Mm hm, =few minutes [she was [(having tuh-). [.hh [Okay so she haso when she woke [up this morning were her eyes= [( ) =all stuck shut,

In line 1, the physician asks for the reason for the childs visit with an open solicitation. The mother describes several symptoms in response. In lines 35/7 she lists a cough, a stuffy nose, and goopy eyes. As was the case with the symptoms offered in Extract 2.1, here, too, the mother makes no inference about the cause of the problem but simply states the symptoms as the basis for the visit. Whether the mother believes that the childs condition is treatable is not disclosed in her problem presentation. Rather, the presentation offers only symptoms for evaluation and leaves it to the physician to determine whether and how the condition will be treated. In offering only the symptoms of the childs illness, the parents communicate an orientation to the childs problem as in need of evaluation but as only possibly treatable. We can see this again in Extract 2.3.
(2.3) 2058 (Dr. 5) 1 2 3 4 5 6 7 8 9 DOC: BOY: MOM: -> DOC: MOM: -> MOM: -> -> And whats going on with you:, (2.0) (Well-) (0.4) .tlkh He ha:s uh: rash all over his body, Uh [huh:, [Like head to toe, (0.6) An:d uh:m he ha:s uh #fever#,=es ((kid making noise)) uh hundred n one today,



10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

DOC: MOM: MOM: -> -> DOC: DOC: MOM: -> DOC: MOM: -> DOC: MOM: -> -> DOC: MOM:

Mm hm:?,= =Stop it- Stop that. (Zack. Stop it.) ((to child)) (0.8) Hes had uh fever for two day:s, Hes had [uh persistent cough=for uh few weeks, [Mm hm, Uh hu[h:?, [But it w=(h)asnt been bad enough to bring him in, Uh huh? And hes (complai:ned) for- uhm- (0.3) (.ml[h) [two days about uh stomach:=ache_ uh: (.)stomach cramping. (1.0) .Tlkh n- n- uhm: for two days? #Yeah:. (an it started yesterday.)

In this interaction, the physician solicits the problem with an open question about the boys medical problem. His mother, in response, offers several symptoms. She mentions a rash (line 4), a fever (lines 8 and 13), a cough (line 14), and a stomachache (lines 19/2122)). As in the other interactions shown thus far, the mother does not offer any theory of what is causing these problems but only details the symptoms. In this way, she treats the symptoms as problematic and as the reason for seeking medical help. For example, the mother also calls the cough persistent. At a point where the physician might have begun history taking, she does not, instead offering a continuer (Schegloff, 1982) (line 16). Then, with it w=hasnt been bad enough to bring him in, (line 17), the mother emphasizes the gravity and doctorability of the childs condition (Heritage & Robinson, 2006). The self-repair (Schegloff, Jefferson, & Sacks, 1977) from what was probably wasnt bad enough in the simple past tense to hasnt been bad enough using the present perfect also suggests a progression of his condition to the current state where he is in need of an evaluation (Bybee, Perkins, & Pagliuca, 1994). She here suggests that because of the accumulation of symptoms, the mother now feels that her son does need a medical evaluation. Although part of this evaluation may include treatment, the parent effectively remains silent on this topic, thus embodying an agnostic stance on the treatability of the childs condition. In this section, we looked at one communication practice for outlining the reason for visiting: a symptoms-only problem presentation. When parents make use of this communication practice, they convey that their reason for visiting is to have a medical evaluation of their childs condition and to seek advice for the management of that condition. They remain agnostic about the exact nature of the illness and its treatability. As noted earlier, this type of problem presentation was most common and is typically treated as unmarked or canonical: Both parents and physicians treat this type of presentation as doing nothing special. This will be discussed in more detail shortly.



Candidate Diagnosis Problem Presentations

The second way that parents regularly present their childrens problems involves mentioning a candidate diagnosis. Although many illnesses could, in principle, be presented with a candidate diagnosis, in these data this format is overwhelmingly reserved for bacterial diagnoses (e.g., ear infections, strep throat, sinusitis) or for illnesses that parents might assume to be bacterial (e.g., pneumonia, bronchitis) rather than viral (e.g., cold or u). In contrast, with symptoms-only presentations, candidate diagnoses were much less frequent: They represented only 16% of the problem presentations in a subset of the data (Stivers, 2002b).2 But despite being less frequent, this is still a relatively common type of problem presentation. As mentioned in chapter 1, this is at odds with existing research in the adult context that suggests that patients very rarely offer explicit or implied diagnoses (Gill, 1998; Heritage & Robinson, 2006a; Ruusuvuori, 2000). In a sample of 300 acute care adult encounters, candidate diagnoses were estimated to be present only 8% of the time (J. Heritage, personal communication, February 12, 2006). Reasons for the relative scarcity of this behavior may include patients orientations to the physicians expertise, as well a reluctance to voice more serious diagnostic possibilities. Heritage and Robinson (2006a) suggest that patients may introduce diagnostic claims in support of the doctorability of their problem in cases where a condition has been previously diagnosed or in cases where a rather benign explanation is possible. Additionally, Gill (1998) notes that when patients offer their own theories of causation, they frame them as delicate actions, either by downgrading the certainty of their theory or by offering them speculatively. Ruusuvuori (2000) suggests that such tentative framing of a diagnostic suggestion suggests patients orientations to the action as stepping into medical territory (p. 165). Although candidate diagnoses appear to be more frequent in these data than in the adult context data, as we will see, parents in these data still orient to the action of offering a candidate diagnosis as delicate. In contrast to symptoms-only presentations, candidate diagnoses can be heard to convey a stance that the nature of the childs medical problem is already known and thus the reason for the medical visit is primarily to seek treatment for a known condition. We can see an illustration of this in Extract 2.4. Here, in response to a problem solicitation (lines 12), the mother offers a candidate diagnosis (lines 45) and then offers the childs symptoms as evidence for the diagnostic conclusion (lines 8 and 11).
(2.4) 305 1 2 3 4 5 6 7 DOC: Al:ri:ght, well what can I do [for you today. MOM: [(hm=hm=hm=hm.) MOM: -> .hhh Uhm (.) Uh- Were- thinking she might -> have an ear infection? [in thuh left ear? DOC: [Okay, DOC: Oka:y,



8 9 10 11


Uh:m because=uh: shes had some pain_ (.) [Alrighty? [over thuh weekend:(.)/(_) .h[h

As her reason for the visit, the mother offers her inference that her daughter has an identiable and treatable problem (an ear infection). The claim is epistemically downgraded (e.g., with thinking and might, as well as with the strong questioning intonation). Additionally, the diagnostic claim is offered with supporting evidence. That turn begins with because, suggesting that what will follow is evidence for the prior inference. The observation that is provided is that the girl has had ear pain. In itself, this observation could have been offered as the reason for the visit but, placed as it is here, it is offered as an account for her candidate diagnosis. Despite the mitigation and the account, which both treat the action as delicate, the mothers turn in lines 45 nonetheless asserts the existence of a known and treatable conditionan ear infection. Because this diagnosis suggests a treatable condition, it looks forward to a specic treatment recommendationa prescription for antibiotics. A similar situation can be seen in the next example, shown in Extract 2.5. Here the physicians question in line 1 is an initial history-taking question but in an environment where no problem presentation was solicited. The mother responds with a full problem presentation, including a candidate diagnosis.
(2.5) 615 1 2 3 4 5 6 7 8 9 10 11 12 DOC: .hh So how long has she been sick. (1.2) MOM: Jus:t (.) I came down with it last Wednesday, so shes probably had it (0.2) DOC: Uh huh_ MOM: (Like) over- four days? (1.0) MOM: An shes been complaining of headaches. (.) MOM: -> So I was thinking she had like uh sinus -> in[fection er something. DOC: [.hhh

With her TCU initial So (line 10), the mother formulates her candidate diagnosis, similar to that of the mother in Extract 2.4, as an inference based on her childs symptoms. Also similar to the mother in that extract, this mother downgrades the epistemic certainty of the diagnosis with I was thinking, like, and er something. In this case, the symptom of headaches precedes the conclusion offered by the mother as a candidate diagnosis: that the headaches are a symptom of an underlying sinus infection. Another way that parents can work to mitigate explicit self-diagnosis is to further downgrade the epistemic authority embodied in their formulation. A candidate



diagnosis can, for example, be offered speculatively. An example of this is shown in Extract 2.6. Here, although the presenting concern involves the recurrence of similar symptoms, this is not a follow-up visit; rather, the child was treated for a condition previously, and the mother has scheduled a new appointment for a new problem condition, albeit similar to the last illness.
(2.6) 316 1 2 3 4 5 6 7 8 9 10 11 12 13 DOC: MOM: DOC: PAT: MOM: Alrighty? Well- Here:=we go:! Howre you do^ing. Fine howre you. Im hanging in there:?, Well hi Matthew howre you[:. [Fine, (.) .hh I brought im back because is- .hh He tu- we took all thuh medication but hes been complaining of uh sore throat off n o[n fer like uh week, [O:kay? .hh An I [didnt (know) [(Youll hafta) refresh my=uh: myHe [had strep. [horrible memory, ((12 lines reviewing medical history not shown)) 26 27 28 29 30 31 32 33 34 35 36 37 38 39 MOM: =But fer like thuh la:st week. Off n on he- he tells me. (Not even just but) going hell go Mom my throat is hurting again. An I noticed it was I- (0.5) DOC: [Huh huh huhMOM: -> [(I-) I thought (0.5) maybe I better just- >I dont -> know if ya know strep has secondary er anything like -> that I wasnt sure.But he hasnt had thuh fever er thuh nausea er anything that he[s had before. DOC: [O:kay:, DOC: .hh [(Goo:d?,) MOM: [But I thought since tdays Veterans Day n theyre off school itd be easier fer me tuh bring im tday than-


Here, the mother states her worry as a generalized possibility: I dont know if ya know strep has secondary er anything like that (lines 3132). This boy had a diagnosed strep throat infection several weeks prior to this encounter and was prescribed antibiotics. Here, the mother suggests that strep has secondary. Secondary infections are bacterial infections that occur following a viral infection. A secondary throat or ear infection, for example, would normally be treatable with antibiotics. The mothers use of secondary here may be the result of her confusing secondary



infections with relapses. Relapses normally involve the return of an infection after it has appeared to go away. Regardless, what appears quite clear is that the mother is concerned that her son has a throat infection. The mother initiates the move to business by offering her reason for the visit as her sons complaint of a sore throat, despite his having completed a full course of medication (lines 68). After an intervening sequence about the history of the prior illness (omitted lines), the mother reasserts her sons symptoms by animating her sons complaint in direct reported speech (lines 2728). Subsequent to this, she continues with her own observation of his symptoms (line 28) and her diagnostic inference (3133) that her son has a type of secondary infection from strep throat. Again, this claim is mitigatedthis time with I dont know, er anything, and I wasnt sure. However, even with the mitigation, the claim of infection makes treatment for that infection relevant. In this case, the mother has taken a small step away from a direct statement of a diagnostic theory by formulating it as a speculation. Mitigation of a candidate diagnosis can also be accomplished with indirection. When a candidate diagnosis is offered indirectly as, for example, a statement about past illnesses, parents regularly formulate it without mitigation (see Gill, 1998, on indirection). An example of this is shown in Extract 2.7.
(2.7) P201 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 An:- An what didju bring her in: for today? She had uh fever this morning, Mm hm?, .h An:d shes complai:ned of: uh pai:n in her left ca:lf?, (.) DOC: Mm hm:?, DAD: -> And we have ha:d: (1.0) some experience in -> thuh pa:st with s:inus::=sinusitis? DOC: Mm hm? DAD: .hh A:nd it was: (.) uh lo:ng ti:me being diagnosed_=We had tuh go t thee emergency room, DOC: Mm hm::?, DAD: Uh::m a:nd nally thuh doctor the:re (a[t- could nd it.) This was- (1.0) ve months ago= DOC: [Mm hm. DAD: or so so-.hh So she has had something in the past, <Shes had her- .hh I think her knee was hurting in thuh pa:st. DOC: O:kay. DAD: (An) they did x-rays ( ) (0.2) DOC: Okay so she had fever: just today: it started, (.) DAD: Yeah. DOC: DAD: DOC: DAD:



Here, the father mentions that they have some experience in thuh pa:st with s:inus::=sinusitis? (lines 89). Although he does not directly state, I think she has a sinus infection, the father nonetheless communicates his belief that his daughters condition may be sinusitis. Whether direct or indirect, in each of these examples the parent does two primary things: (1) describes one or more symptoms that the child is experiencing and 2) offers an inference about the underlying diagnosis that is producing the symptom(s). As discussed earlier, when parents formulate their reason for the visit with a symptoms-only problem presentation, they make no claims about the treatability of their childs symptoms. They formulate their childs medical problem as, in the rst instance, in need of a physicians evaluation. However, when parents formulate their childs problem with a candidate diagnosis, they adopt a stance that their childs condition is medically problematic and in need of prescription treatment overwhelmingly antibiotics.

Implied Candidate Diagnoses

The candidate diagnoses discussed thus far offer examples that are clearly articulated. However, parents and physicians alike orient to implied candidate diagnoses in very similar ways. Thus, ultimately I will argue that candidate diagnoses include both explicit and implied varieties. In this section, though, we will discuss implied candidate diagnoses separately. Implied candidate diagnoses were less frequent than the explicit candidate diagnoses, only 10% of the cases in a subset of the data (Stivers, 2002b). This type of problem presentation represents a hybrid of the two practices outlined thus far. On the one hand, it involves the presentation of symptoms only; however, the symptoms are highly specic. Their specicity appears to imply a particular bacterial condition. When compared with many formulations of symptoms, these diagnosis implicative symptoms involve a ner level of detail and specicity, or what Schegloff terms granularity, than is typical (Schegloff, 1972a, 2000). The rst feature suggests that these problem presentations could be classied with other symptoms-only presentations because they involve no actual articulated diagnosis. But the second feature suggests that when a parent offers diagnosis implicative symptoms they are displaying their stance that the child has the implied condition and is thus in need of treatment for that condition. In particular, they canonically employ a level of technical specicity that is relevant for a health professional rather than for an ordinary recipient (e.g., mentions of the color of nasal discharge or the color of spots in the throat) and is for that reason understood to imply a particular diagnosis. For example, a parent might mention that her child has a barky cough to index croup, she might mention green nasal discharge to index sinusitis, or she might mention white or yellow spots on the childs throat to index strep throat. When parents mention symptoms such as these, the stance they take is similar to that taken up with a direct candidate diagnosis: They treat the symptoms as both medically problematic and treatable. One type of evidence that offering these specic sorts of symptoms works to index particular infections is that physicians treat them as conveying a concern about



a particular diagnosis. As an example, we can look at Extract 2.8, in which the mother asserts that she saw yellow spots on her daughters throat (lines 10/12), which regularly index strep throat. Although the physician does not reject strep throat specically, he displays his orientation to the parent as having implied a diagnosis in his formulation of lines 1718/2021. He treats a diagnosis of blisters or cold sores as a position that contrasts with that of the parent. This is accomplished particularly with his mention of actually as a preface to his identication of the spot as blisters (line 17). Actually marks the nding as counter to what was previously offered (Clift, 2001; Schegloff, 1996).
(2.8) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 MOM: DOC: MOM: DOC: DOC: MOM: DOC: MOM: DOC: GIR: DOC: DOC: MOM: DOC: MOM: DOC: And I- s- she was complaining about her #throa:t.# Nkay:, an she had uh fever last night?, (.) Uh::- (.) uh little bit. so I- I kept plying her with Tylenol just to help [#her throat pai:n.# [>Okay,< Sure. (0.2) -> And then uh- I looked down her throat yesterday-> last ni:ght, an I could see thuh yellow:_ ^Okay. -> #spo:[t so:. ((trails off)) [.hh Well open up rea::l big. lets take uh look an (say-) say #Ah:::[:::.=hh [Ah::::=hh .hh (0.5) Yeah:. You know actually what those a:re pr=h .hh are primarily blisters back there. Yea:h? Its almost like shes got cold sores in thuh back of er throa:t. (Oh:[::.)/(Aw:::.) [And u:sually thatll go along with this just being viral. (.) [Really.= [#er-# =Y:eah. .hh => One v thuh teachers told me it might be stre:p => so:[:_ [.mlk Yeah we are starting to see some strep so Im gonna culture just in case .hh shes got both going on at the same ti:me but- .hh when you


35 36 37 38 39



see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it[s ya know like shes got blisters n [Oh yeah:_ Oh:::.

That the parents original diagnosis implicative symptom was designed to imply a bacterial candidate diagnosis is made explicit in line 30. Here, the mother identies the diagnosis of concern as strep and further asserts an account for this concern: One v thuh teachers told me it might be stre:p. Her turn nal so retroactively casts her suspicion of the spots as having been related to this teacher-offered candidate diagnosis (Raymond, 2004). Typically the implied diagnosis is not brought to the surface of the interaction, but in this case we have clear evidence that the parents mention of spots earlier in the encounter was an indirect way of conveying her worry about strep. The candidate diagnosis that was implied is attributed to a third party, distancing the mother from the diagnosis that was previously discounted (Clayman & Heritage, 2002). Another illustration is offered in Extract 2.9. Following some detailing of symptoms (earlier in data not shown and here in lines 811/13), the physician, in overlap, shifts into a joking examination of the girls stomach. The mother returns to the symptoms and the problem presentation as the physicians joking talk is reaching completion.
(2.9) 1050 (Dr. 1) ((just following some joking about responses to DOCs initial inquiry Whats up. . . . the sky)) 1 2 3 4 5 6 7 8 9 10 11 12 13 DOC: MOM: And what else. (2.2) Tell thuh doctor what did you told me this morning.= When I was brushing=uh (.) your hair. (0.5) What=do you have. (.) <Tu:mmya::che.> Uh tummyache.=[h [.h Shes had (uh) fever for three [days shes had a cold off an= [Lemme feel y- ((move to examine girls stomach)) =o:n for about (three) days.


((14 lines not shown DOC begins exam feeling childs stomach - joking)) 28 29 30 31 32 MOM: -> I thought I saw the little white (.) dot[s, DOC: [.h There was one little sp:o:t_ but it didnt look -too ba:d. MOM: -> Because sh- theres strep throat goin around -> [in her class an:- an I cant seem to get rid of=



33 34 35 36


[Yeah: w=this (.) co:ld an_ .h shes beenTurn your hea:d, really high fevers.

Here the mother offers another symptomthis one a diagnosis implicative symptom (line 28). When the mother says that she saw the little white (.) dots, with the denite article the, she conveys that these dots are specic and have a previously established meaning. Additionally, this symptom hearably indexes a diagnosis of strep throat. In response, the physician rejects the implicit claim that the dots are problematic (lines 2930). Subsequently, the mother makes explicit that the diagnosis she was alluding to with the diagnosis implicative symptom was strep throat (line 31). But this more overt stance toward her daughters illness as treatable is not displayed until the physician rejects the parents less direct conveyance. The candidate diagnosis offered in lines 3132 hearably accounts for her prior statement. The physician also treats this as an account by accepting the turn at rst possible completion with Yeah. However, both statements (lines 28 and 3132, respectively) appear to convey the same diagnostic theory. A nal example is shown in Extract 2.10. Here, as part of the narrative presentation of her daughters problem, the mother states that she started with uh little clear ui:d on: uh:m h tlk Saturday. . . . And then- by yesterday it turned- gree:n, (lines 67/15/17). With her use of started, she projects that there has been some change. Additionally, clear suggests that the change may be in terms of color because clear uid is nonproblematic, and typically a problematic formulation would be simply a runny nose or a lot of drainage rather than the naming of a color.
(2.10) 1046 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 DOC: MOM: Oka::y, so:, lets see whats doin he:re?=hh We:ll, Erin:, thuh rst up to bat here, (0.2) she: uhm (.) ^shes been ac[ting prettyDOC: [.hh DOC: Come clo[se to me (Er,) MOM: [pretty happy but- .hh she started with uh little clear ui:d on: uh:m h tlk Saturday. (.) MOM: running out of her no:se_ DOC: (Who[o hoo) ((whistled)) MOM: [and draining into er throat_ DOC: I think theres uh bird in er ear:.= MOM: =#huh hu[h# ((throat clear)) DOC: [Did=you hear tha:t? MOM: -> And then- by yesterday it turnedGI?: #Hu::h hu[h# ((cough)) MOM: -> [gree:n, DOC: Kay her ears look perfect.

19 20 21 22 23 24 25



Okay:, .h And=uh .h its mostly at night when- it drains dow:n [its:DOC: [Yeah_ MOM: -> An Ive had uh sinus infec[tion, DOC: [Okay. Open up your mouth real wide-

As the mother continues her narrative, she does, in fact, assert that the color changed to green. With this level of technical specicity, the characterization indexes a diagnosis of a sinus infection. That this is the mothers design is made explicit when, in her next turn, the mother states that she has had a sinus infection. The implication is that the mother has experienced similar symptoms and thus believes that green discharge can be a symptom of sinusitis. That the mothers turn in 23 is designed to be connected to her earlier diagnosis implicative symptom is partly carried by the An, which connects it back to what was said previously. This helps the turn to be heard for its ramications for the daughter rather than as a discrete unrelated announcement. Implied candidate diagnoses can be seen to have a resemblance to both symptoms-only presentations and articulated candidate diagnosis. However, in this section we have seen that parents appear to use them to index particular diagnoses rather than suggesting them outright or explicitly with articulated candidate diagnoses. In this way, they appear to be displaying the stance that their child has a given condition and that they are seeking treatment for that condition. Combined, then, in over a quarter of cases, parents are identifying their reason for visiting in a way that physicians treat as seeking antibiotic treatment. Primary evidence for this claim is found in the way physicians respond to the two different types of problem presentations.

Responses to Problem Presentations

When parents offer a symptoms-only presentation, physicians treat them as embodying a stance that their child has a doctorable condition for which they are seeking medical evaluation. By contrast, when parents offer a candidate diagnosis, physicians treat them as embodying a stance that their child has not only a doctorable condition but also a treatable one and, further, that parents are lobbying for antibiotic treatment. So by using a candidate diagnosis rather than a symptoms-only problem presentation, parents initiate a negotiation of their childs treatment in favor of antibiotics. Responding to Symptoms-Only Presentations As noted earlier, symptoms-only presentations are the most common type of problem presentation and are oriented to as the unmarked format for presenting the problem.3 This section focuses on two ways in which physicians respond to symptoms-only presentations preparatory to a contrast with physicians responses to candidate diagnoses. First, physicians typically move from a symptoms-only presentation directly



into an investigation of the childs problem, which, as has been noted, is the most common trajectory for a medical visit (Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991). This may mean a move directly to physical examination (as seen in Extract 2.1) or (most commonly) a move into history taking (as shown in Extracts 2.2 and 2.3). But what is critical is that the physician does not in any case in these data take issue with parents about the symptoms they describe. This suggests that physicians treat symptoms-only presentations as making an investigation of the patients problem the most immediately relevant next activity. Moreover, physicians typically formulate their subsequent diagnoses as direct, positively formulated announcements. That is, they offer the diagnosis without an orientation to a previously implied or articulated diagnosis, thereby treating an explanation of the problem as the primary task set by the parents problem presentation. Both of these features are illustrated in Extract 2.11.
(2.11) 1188 (Dr. 3); [problem presentation shown previously in 2.2] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DOC: And so: do- Whats been bothering her. (0.4) MOM: Uh:m shes had a cou:gh?, and stufng- stuffy no:se, and then yesterday in the afternoo:n she started tuh get #really goopy eye:[s, and every= DOC: [Mm hm, MOM: =few minutes [she was [(having tuh-). DOC: [.hh [Okay so she haso when she woke [up this morning were her eyes= MOM: [( ) DOC: =all stuck shut, MOM: Yeah but- Well actually during thuh middle of the ni:ght [she woke u[:p_ and they we[re stuck shut n_ DOC: [Okay, [Okay_ [Okay_ 1-> An how about fever. Any fever at all? ((33 lines of history taking/examination not shown)) 49 50 51 52 53 54 55 56 57 58 58 60 61 DOC: 2-> 2-> 2-> MOM: DOC: DOC: 2-> 2-> Basically shes mov- i- shes: :yknow> kinda: developed the co:ld an respiratory thing thats goin arou:nd. [Uh huh, [.hh An its moved into her eyes, so shes got like #uh:# pink eye or conjunctivitis. .hh and so thuh: cou:gh, and the stufness I would treat symptomatically with uh cough an cold medicine like Pediaca:re, Dimetapp, whatever:. DOC: .hh And then Im gonna give you some eyedrops to put in her eyes_ MOM: Okay?, ((DOC continues on to detail dosage))



Here, we rst see that at arrow 1 the physician moves from establishing the reason for the childs visit directly to taking the patients history. Second, at arrows 2, when the physician delivers his diagnosis, it is simply asserted rather than framed as rejecting an alternative, denying the parents theory, or conrming it. It states that the condition is a cold and pink eye. In lines 5560, the physician outlines his treatment recommendation for the two conditions. This, too, is formulated straightforwardly as a proposal. Like the problem presentation, the diagnosis and treatment recommendation are offered in an unmarked way, suggesting that they are providing only an evaluation and advice on treatment. This is further supported by the example shown as Extract 2.12. As with Extract 2.11, here, too, the physician moves directly from establishing the reason for the visit into history taking (arrow 1).
(2.12) 2058 (Dr. 5); [problem presentation shown previously in 2.3] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 DOC: And whats going on with you:, (2.0) BOY: (Well-) (0.4) MOM: .tlkh He ha:s uh: rash all over his body, DOC: Uh [huh:, MOM: [Like head to toe, (0.6) MOM: An:d uh:m he ha:s uh #fever#,=es ((kid begins noise)) uh hundred n one today, DOC: Mm hm:?,= MOM: =Stop it- Stop that. (Zack. Stop it.) ((to child)) (0.8) MOM: Hes had uh fever for two day:s, Hes had [uh persistent cough=for uh few weeks, DOC: [Mm hm, DOC: Uh hu[h:?, MOM: [But it w=(h)asnt been bad enough to bring him in, DOC: Uh huh? MOM: And hes (complai:ned) for- uhm- (0.3) DOC: (.ml[h) MOM: [two days about uh stomach:=ache_ uh: (.) stomach cramping. (1.0) DOC: 1-> .Tlkh n- n- uhm: for two days? MOM: #Yeah:. (an it started yesterday.) ((48 lines of history taking and exam not shown)) 74 75 76 77 78 DOC: MOM: MOM: DOC: DOC: 2-> .Tlkhh You want to [know what you ha:ve? [HisHis chest and his genital:s are the reddest, #Yeah:.#=h Hes got scarlet #fever:#.



After the history taking and physical examination (data not shown), the physician moves to offer her diagnosis (shown in line 78). Similar to Extract 2.11, here, too, the physician formulates the diagnosis afrmatively and straightforwardly. This sequence begins quite early in the physical examination. The physicians turn in line 74 is hearably a preannouncement (Terasaki, 2004) addressed to the boy with you. This may indicate that the forthcoming news is delicate or unusual. However, the mother does not orient to the physicians turn as initiating a pre-sequence. Rather, she does some additional work to assert the problematic nature of her childs condition by offering an additional problematic symptom (lines 7576). In this way, the mother may be treating the pre-announcement as preceding the full investigation of the boy. In line 77, the physician offers minimal agreement with the mothers turn before moving directly to her diagnostic assertion that the boy has got scarlet #fever:#. It is also notable that the physician has now shifted from addressing the boy to addressing his mother (evidenced by her reference to the boy using the thirdperson pronoun he). In looking at these cases, what we can see is that when parents use a symptomsonly problem presentation formulation, doctors treat them as taking a stance toward their visit as legitimate and to their childs condition as doctorable, but doctors do not treat parents as making any claims about the childs treatability. Parents in these encounters specically orient to the diagnosis, leaving treatment to the physician. There is no explicit orientation to whether the condition is in need of treatment. And physicians in these encounters routinely treat these parents as primarily seeking an evaluation of their childs illness. In these cases, then, parents are not observably vying for any type of treatment, nor are they treated by physicians as doing so. Earlier in this section, I suggested that the symptoms-only presentation may be the unmarked way of presenting a childs problem. In addition to it being the most common format, evidence supporting this is that physicians respond to these presentations with an unmarked diagnosis delivery: Physicians routinely move from establishing the reason for the childs visit into an investigation of the problem, and they routinely offer their diagnoses and treatment recommendations as simple, straightforward announcements (i.e., not apparently responsive to, in the sense of conrming or disconrming, any particular previous diagnostic theory). In the straightforwardness of their formulation, these diagnosis announcements appear to be doing nothing special. Responding to Candidate Diagnosis Presentations If symptoms-only presentations are the way parents show themselves to be doing nothing special, candidate diagnoses convey the reverse: that something special is being done. Physicians display this analysis in the way that they respond to candidate diagnoseswhether suggested or impliedby (1) treating conrmation or disconrmation of the parents proposed diagnosis as relevant and (2) orienting to the relevance of antibiotic treatment. The two most common locations where physicians address parents candidate diagnoses are immediately after the presentationparticularly with disconrmations or challengesand during the diagnosis phase. We will look at each of these contexts.



Candidate Diagnosis Uptake: Just Following the Presentation As was just discussed, following symptoms-only problem presentations, physicians typically move directly into history taking or physical examination. In these data, there are no cases of physicians challenging the existence of a parent-reported symptom such as ear or throat pain, congestion, or a runny nose. By contrast, if a parent presents a candidate diagnosis, the physician may counter that diagnosis then and there. In the Seaside data set, this occurred 19% of the time (Stivers, 2002b). Thus, physicians treat the two types of presentations quite differently. As an illustration, we can return to the case shown earlier as Extract 2.5. After the mother presents her daughters problem and offers a candidate diagnosis of a sinus infection, the physician moves to counter the proposed diagnosis.
(2.13) 615 [shown earlier in Extract 2.5] 10 11 12 13 14 15 16 MOM: -> So I was thinking she had like uh sinus in[fection= DOC: [.hhh MOM: -> =er something.= DOC: => =Not necessarily:, Thuh basic uh: this is uh virus basically:, an=uh: .hh (.) thuh headache seems tuh be:=uh (0.5) pretty prominent: part of it at r:st uh: (0.2) .hh

The physicians turn in lines 1316 is clearly responsive to the mothers candidate diagnosis at lines 10/12. In the rst TCU of line 13, although slightly mitigated, the physician rejects the mothers assertion as unlikely. The forcefulness of the counter is partly carried by its being latched to the mothers turn in line 12. Although the physicians rst TCU does not completely rule out a sinus infection, in the second TCU he asserts that this is uh virus basically:,. This offers an alternative diagnosis unequivocally and thus strongly rejects the mothers candidate diagnosis. The third TCU suggests that the headache is part of this viral condition, accounting for one of the symptoms that the mother stated had led her to her own candidate diagnosis, thus rejecting not only the mothers conclusion but also her logic. Unlike the way physicians respond to symptoms-only presentations, after candidate diagnoses, physicians are more likely to take up candidate diagnoses, whether to contest or support them. A similar example can be seen in Extract 2.14. In this case, the mother presented her candidate diagnosis as: >I dont know if ya know strep has secondary er anything like that I wasnt sure. But he hasnt had thuh fever er thuh nausea er anything that hes had before (shown earlier in 2.6). Following the completion of a somewhat extensive joking sequence (data not shown), the physician moves to address the mothers diagnosis.
(2.14) 316 [full presentation shown in 2.6] 31 32 MOM: -> [(I-) I thought (0.5) maybe I better just- >I dont -> know if ya know strep has secondary er anything like



33 34 35 36

-> that I wasnt sure.But he hasnt had thuh fever er thuh nausea er anything that he[s had before. DOC: [O:kay:, DOC: .hh [(Goo:d?,) ((16 lines not shown including joking about BOY having day off but not MOM))

53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

DOC: DOC: => => => DOC: => => MOM: DOC:


O:kay:, .hh Well:, (.) o:ne good thing is: that- uhm (0.5) strep infections:- respond really well tuh amoxicillin. .hh so wh:ile he may not have strep any more (.) he could still have- uh viral process going on, he could still have just residual sore throa:t, .h[h dry weather kind of things, .hhh [(Okay.) Uhm: besides having an actual infection so we can always look at those issues, .hh an then if you want we can also just retest his throat. (.) An make sure theres no more strep there too. (.) Well (you it) kinda depends on what you- what you [think. [Mkay,

The mother framed her candidate diagnosis in a way that allowed both agreement and disagreement. That is, on the one hand, she suggested evidence that her son no longer has strep (hasnt had thuh fever er thuh nausea in lines 3334). But she still speculated about strep as a diagnostic possibility (I dont know if ya know strep has secondary er anything). With this possible diagnosis on the interactional table, the physician addresses both dimensions of the mothers presentation. In line 54, he takes an inbreath, prefaces his turn with Well stretches the Well:, and then delays the turn further with a micropause. All of these features are common in dispreferred turn formats (Heritage, 1984b; Pomerantz, 1984). This turn design projects that the physician is headed toward disafliation with the mothers candidate diagnosis. But the physician also frames his response as in agreement with the mother through his wh:ile he may not have strep any more (line 56). This works to maximize the appearance of agreement with the mothers problem presentation. The physician also goes on to validate the mothers reason for visiting: he could still have- uh viral process going on, he could still have just residual sore throa:t, .hh dry weather kind of things, (line 58). In this way, the physician both counters the mothers candidate diagnosis of strep and validates her reason for coming (by referencing alternative causes of the sore throat). Finally, in lines 6162, the physician offers another sort of response; he offers to retest the boys throat to make sure that he no longer has any strep. When a candidate diagnosis is implied, physicians also routinely act responsively. Here, we can see an example both of a disconrmation and of an orientation to the physicians orientation that the candidate diagnosis was looking forward to antibiotic treatment. In Extract 2.15, the mother presents her childs symptom of



nasal discharge by specically mentioning the color as problematic. She says, its gotten- it was green (line 4).
(2.15) P110 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 DOC: GIR: MOM: -> DOC: MOM: -> -> DOC: MOM: -> DOC: => => MOM: DOC: => MOM: DOC: => MOM: DOC: => => DOC: => MOM: Youre sick_ well whats u:p. (1.1) I dont kno[: [w, [B[etween yesterday and toda:y, she-= [How- hh =.hh ya know its (this-)/(js-) nasal crap an its gotten it was gree[:n.=it was [uh= [[Nkay:, =really uh beauti[ful color (yesterdange ) [.hh Okay well just because its green [it doesnt [(doesnt mean [an-) (I kn-) [ma- mean its bacterial. (Right right right right.) Theres a [who:le new thing about: uh: .h - sinus [I know. an everybodys saying wed so weve been trying very hard not tuh put kids on antibiotics if we can avoid it, Ri:ght. right.

We can see evidence that the physician hears the mothers mention of the green color as indexing a bacterial sinus infection if we look at his virtual rejection of the diagnosis with well just because its green it doesnt ma- mean its bacterial. (lines 1011/13). Although no direct candidate diagnosis is offered here, the physician treats the particular formulation of symptoms as clearly implying one, and in that way he is enabled to reject it in a way that does not emerge in response to symptoms-only formulations. Moreover, following the rejection of the mothers implied candidate diagnosis, the physician goes on to foreshadow his own unwillingness to prescribe antibiotics (lines 14/1619). With this, he displays his understanding that the mothers candidate diagnosis was not only seeking conrmation or disconrmation but also working to advocate for, and thus to initiate a negotiation in favor of, antibiotic treatment. In this section, we have examined immediate responses to candidate diagnosis formulations. Although no instances of a conrmation or disconrmation of a symptoms-only presentation occurred in these data, such responses immediately following a candidate diagnosis do occur. By responding to candidate diagnoses in this way, physicians treat these presentations as involving diagnostic and treatment implications that are not implied by symptoms-only presentations. This case is made stronger when we observe that such responses occur not just immediately but are even



more commonly addressed later in the encounter, when there is a candidate diagnosis problem presentation. Candidate Diagnosis Uptake: During the Counseling Phases The second primary area where physicians can be seen to directly address parents candidate diagnosis presentations is in the counseling phases, when they offer their nal diagnosis and treatment recommendation. In these locations, physicians work to show that the diagnosis and treatment recommendations they are providing are being offered in light of the earlier candidate diagnosis. This occurred 71% of the time in a subset of these data (Stivers, 2002b). As an example, see Extract 2.16. In this case, the mother earlier offered a candidate diagnosis that her daughter has a sinus infection. In line 42, the physician is completing his examination of the girl and offering his diagnosis.
(2.16) 615 [shown earlier in 2.5 and 2.13] 10 11 12 MOM: -> So I was thinking she had like uh sinus in[fection= DOC: [.hhh MOM: -> =er something.= ((29 lines of history taking and examination not shown)) 42 43 44 45 46 47 DOC: => .hh Uh: (1.4) Lets see (now ) (1.1) I think uh: I => dont think she h:as: uh: sinus infection,Have you noticed uh lot of (0.2) heavy drainage:?, (0.2) MOM: Yeah shes been:: (.) When she does cough she coughs up (the-)/(tht) (.) gree:n, (1.0) that mucus stuff?

When the physician begins his diagnosis, he appears to be headed for a diagnostic assertion with I think but, initiating repair on this turn beginning, changes tack and instead designs a diagnosis that is responsive to the mothers problem presentation. He does this by disconrming her candidate diagnosis with I dont think she h:as: uh: sinus infection (lines 4243). Additionally, the formulation the physician uses to disconrm the diagnosis is the very one the mother used: sinus infection. This may seem inconsequential, but in the next case (Extract 2.17), we can see that the father uses a different expression for the same diagnosis (this time sinusitis) in his candidate diagnosis in lines 89. When the physician reaches her own diagnosis, it is this formulation that she returns to.
(2.17) P201 (Dr. 7) [full problem presentation shown in 2.7] 8 9 10 DAD: DOC: <And we have ha:d> (1.0) some experience in thuh pa:st with s:inus::=sinusitis? Mm hm?

((254 lines of history taking and examination not shown))

265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280



Ah::[:=hhh [Just a teeny teeny teeny bit. See in thuh back [there, DAD: [Uh huh, (.) DAD: [Uh huh, DOC: [But nothing too #ba:d,# so that might be ea:rly or- .hh uh:m:=but ^otherwise :her ears look great.= =Shes not< having uh lot uh mucus or stuff. You usually get- i- bad ear infectio:ns_ .hh a know -after you get uh lot of co:ld, => I=dont=know if thats been=er history in thuh => pa:st but- .hh uh lotta times youll get sinusitis => or ear infections after a lot of mucus up here, => and right now shes - pretty clear_ it seems like its mostly the fever,

At this point in the visit, the physician has completed most of her examination and is, across lines 265270, inspecting the girls throat and perhaps inviting the father to look with her at the girl (with See in thuh back there,). The physician mentions both sinusitis and ear infections as conditions that can occur after a cold but asserts that right now shes - pretty clear_ as well. The initial assertion (line 273) that Shes not> having uh lot uh mucus or stuff. appears to be mentioned as support for her claim that there is no sinus infection. Here, as in Extract 2.16, the physician uses the characterization of the illness that was used by the parent to disconrm it as a diagnosis. We can see another illustration of a diagnosis that is delivered as a disconrmation subsequent to an earlier candidate diagnosis in Extract 2.18. Here, the physician returns to the childs ears and presents his diagnosis as regretfully disconrming (lines 6566/68).
(2.18) 1017 (Dr. 1) ((Simplied; BRO is the child patients brother)) 1 2 3 4 5 6 7 8 MOM: BRO: BRO: DOC: BRO: DOC: MOM: He- no[::. hes thuh- hes got the ear infection. [HesHe- he (los-) Hes SICK. You ^think so? Yeah. Hes sick. [#Oh.# Well I can see hes not smi:[ling, C[(Hm_) ((laugh)) [Kind of listless.

((53 lines of problem presentation, history taking and examination not shown)) 62 63 64 DOC: .hh Yeah #Say Ah:- <Lemme look (in)/(at) those ears agai:n. (23.5)



65 66 67 68 69 70

DOC: => hh=Wish (we) could s(h)ay h(h)e h(h)ad an e(h)ar => i(h)nfection butMOM: ( [I dont know what_ Yeah:.) DOC: => [I dont see: it. (.) MOM: Go^od.

The physicians diagnosis is clearly retrieving the mothers prior candidate diagnosis of an ear infection by repeating it here as something he wishes he could say. Additionally, in formulating his diagnosis in the negative I dont see: it. (line 68), he treats his nal diagnosis as one that is, by design, disconrming the mothers candidate diagnosis in second position rather than asserting a diagnosis in rst position.4 In the counseling phases, as earlier in the visit, physicians treat direct and implied candidate diagnoses as being basically functionally equivalent. So, just like direct candidate diagnoses, implied candidate diagnoses are also routinely conrmed or disconrmed during the physicians nal diagnosis. In the following case (Extract 2.19), the father implies a diagnosis of swimmers ear when he mentions Hes been swimming a lot, (line 6) after having mentioned ear pain. (See Gill, 1998, for a discussion of patients diagnostic explanations in this format.) Although this case differs slightly in that this is not the symptom, note that the symptom of eara:che is here being explained by a fact of swimming, although the diagnosis is not here stated. This is quite similar to stating a noticing of green discharge or pus on the tonsils without asserting a diagnosis of sinusitis or strep throat.
( 2.19) 1189 (Dr. 2) 1 2 3 4 5 6 7 8 DOC: DAD: Well Charlies got an eara:che. #A[w::.# [Well- YeahHis- ba- its bothering him la- lot of swimming. (0.5) Hes [been swimming a lot, an then he went= [Okay. =to thuh snow.


((30 lines not shown; history taking and physical examination)) 39 40 41 42 43 44 DOC: DOC: DOC: Lets peek at=your ea:r. (3.0) .hh so=what:=h=ow:. (0.5) Well he does not have a swim ear: but he does have a middle ear infection,

In this case, after the physician has examined the child, he says, Well he does not have a swim ear: but he does have a middle ear infection, (lines 4344). With this formulation, the physician treats the parent as having implied a diagnosis of swim-



mers ear to explain his sons ear pain. Thus, the physician does not simply assert the ear infection but rst rejects the implied diagnosis. Here, the disagreement is evidenced by the Well preface (Pomerantz, 1984) and the rejection component does not have a swim ear:. The physician only then provides the positive diagnosis. This is marked as contrastive both with the negative diagnosis and with the but. Additionally, with the second does the physician can be heard to support the fathers claim that his son has ear pain. The stress marks the second component as contrastive with the negative and suggests that although the parent was wrong on one count (with the implied diagnosis), he was right in his recognition of a medical problem. There is also evidence that when parents present their childrens problems with a candidate diagnosis, conrmation-disconrmation is treated as relevant in the course of the visit. For instance, in Extract 2.20, when a physician fails to address the parents candidate diagnosis, the parent reinvokes it as a question (in lines 7576).
(2.20) 1141 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 73 74 75 76 77 DOC: So: whats goin on he:re. (1.2) MOM: Hes got uh:- (0.2) tlk (Theyre kinda-) He stayed out of school on Monday:, DOC: [Uh huh, MOM: [(w-) MOM: With kind of #uh-# low grade fever an- (.) uh crummy no:se_ an now hes complaining about -> ears_<Hes [very susceptible [to thuh infections, DOC: [Uh oh. [Is (e-) DOC: Uh huh, DOC: Has he had uh fever at all?, ((60 lines not shown)) DOC: Yeah I think hes got thuh bug thats goin around right now: [anMOM: => [(Oh) you dont- He doesnt have uh => infection? DOC: I dont think so. Not yet.

In line 9, the mother offers a candidate diagnosis of an infection. This is responded to during the physical examination (data not shown), but in the nal diagnosis, the physician neither conrms nor disconrms it until the parent requests that conrmation (lines 7576). Here, the actual diagnosis is given with I think hes got thuh bug thats goin around right now: (lines 7374). In responsein fact at rst possible completionthe parent treats her primary concern as remaining unaddressed and pursues conrmation of that diagnosis (lines 7576). Although the physicians turn was not apparently designed to be complete at that point (both intonationally and with the presence of a cutoff following an), the mother still begins her turn, coming in immediately upon possible grammatical completion, to ask about whether the child has uh infection? (Sacks et al., 1974). Here, she uses the same formula-



tion she used previously, which helps display her action as requesting conrmation of a diagnosis she had offered earlier in her problem presentation rather than asking about something new. It is only at this point that the physician actually disconrms the diagnostic possibility. This example suggests that parents actively monitor physicians diagnoses for the way they address previously suggested diagnoses. Interestingly, this appears to be the case even when physicians display attending to the parents concerns earlier in the visit. Here, the physician had attended to the parents concern of infection during the physical examination, but the mother did not treat this as sufcient and still pursued the physicians conrmation-disconrmation in the ofcial diagnosis phase. I have claimed that the use of a candidate diagnosis represents one of the earliest forms of parent behavior that can systematically inuence the treatment outcome by conveying pressure for antibiotic treatment. Earlier, we saw that the relevance of antibiotics following a candidate diagnosis could be observed immediately following the problem presentation (Extract 2.15). Physicians also display an orientation to the relevance of antibiotics following a candidate diagnosis later on in the visit, such as in the diagnosis or treatment phase. For example, see Extract 2.21.
(2.21) 305 ((shown earlier in Extract 2.4; Extract 2.21 begins just following the physical examination)) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 DOC: .hh So: it would loo:k hh like she is:=uhm (.) probly ghting some (.) viral: upper respiratory kinda stuff, DOC: .hh More on thuh left than on thuh right, which c[an account for some pain maybe, MOM: [Okay. DOC: .hh Uhm:=hh Ears are- hh I mean .hh theres not even uh lot of wax in her ears. Her ears are prett[y clea:n. MOM: [( ) DOC: I mean [they look s- sec- exceptional_ .hh (ya know.) MOM: [(Great.) MOM: Yeah: [uhm DOC: [For uh kid [her age. MOM: [huh huh huh .h[hh She loves tuh have= DOC: [Good job. MOM: =her ears cleaned. hzuh hu[h huh .hhh huh huh huh DOC: [Well- (.) fantastic (cuz) DOC: (theyve be-) you guys are doing uh great jo:b, .hhh MOM: (M[m.) DOC: [Uh:- I would tell you though I dont hhh (.) I -> dont see anything that requires like antibio:tics er anythi:ng, but certainly sympto[matic treatment might= MOM: [Mm. DOC: =be in order, DOC: .hh MOM: Okay.



Similar to Extracts 2.162.19, here the physician formulates his diagnosis as responsive to the mothers candidate diagnosis: First, he conrms that the girl has an illness. This is carried with the full form is . . . ghting and the additional stress on is that works to conrm. (See Stivers, 2005a, for a discussion of this practice in immediately subsequent position.) Second, he accounts for the pain that the mother mentioned previously as evidence for her candidate diagnosis, noting that there may be more infection on the left than the right, which can account for some pain maybe, (lines 34). Here, the physicians use of the word pain ties back to the mothers own use of pain in her candidate diagnosis. Third, the physician disconrms the candidate diagnosis in that he specically targets the ears to note that there is not even wax, which suggests that there was something else being searched for, and neither the searched-for item (i.e., infection) nor the more minimal wax could be found. Finally, when the physician begins his treatment recommendation in line 19, he formulates this as responsive also. Although antibiotics had not been explicitly raised previously in this visit, the physician frames this recommendation as responsive. This is accomplished in part by his raising them at all. Other potentially relevant medications are not ruled out, so the raising of this treatment is signicant and displays an orientation to their signicance for the visit. Additionally, the use of I would tell you though suggests that this is part of his response to the mother. He has provided a conrming response in that the child is ghting an illness. Here, he is providing the counterpart, disconrming a need for antibiotics. The though carries much of the weight in establishing the utterance as contrastive with the position taken by the mother in the earlier problem presentation. Delivering the treatment recommendation in these ways suggests the physicians understanding that the parent was oriented to antibiotics as the appropriate treatment for the illness she said she believed the child hadan ear infectioncommunicated through her use of a candidate diagnosis early in the visit. Quantitative evidence also supports the claim that physicians perceive candidate diagnoses as treatment implicative. For the Seaside data, when parents used a candidate diagnosis, physicians were 5.23 times more likely to report having perceived the parent as expecting antibiotics (p >.05) (Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). The same pattern emerged in the replication of the rst study with the larger and more ethnically and socioeconomically diverse Metro data (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). But parents may not always be using candidate diagnoses with the motivation to advocate for antibiotics. Parent use of candidate diagnoses was not signicantly associated with parents reporting an expectation for antibiotic treatment (Mangione-Smith et al., 2006; Stivers et al., 2003). We have seen that physicians treat symptoms-only problem presentations as making relevant only an investigation and evaluation of the childs problem. By contrast, we have seen that physicians treat candidate diagnoses as inviting conrmation or disconrmation and that physicians often respond directly to parents candidate diagnoses. I have shown that by offering a candidate diagnosis in the problem presentation, parents adopt a stance that their childs illness is medically problematic and treatable. Physicians responses support this analysis. Specically, the responses to candidate diagnoses are generally conrmatory or disconrmatory and often men-



tion antibiotic treatment. Physicians treat parents as in search of conrmation that their own diagnosis was correct and in terms of treatment for that condition. As we saw in Extract 2.18, physicians may even be somewhat apologetic if they are unable to conrm a candidate diagnosis. By contrast, in cases where parents present their child with a symptoms-only problem presentation, physicians routinely offer their diagnoses as straight afrmative statements.

This chapter has outlined the two primary ways that parents present their childrens problems during the reason for the visit phase of the encounter. We saw that these practices convey parents alternative stances toward the childs problem in terms of its doctorability and treatability. Symptoms-only problem presentations are treated as the unmarked type of presentation and display a stance that parents are, rst and foremost, seeking an evaluation of their child. By virtue of having come to the physician, they claim a need for medical assistance but make no claims about diagnosis or treatment. In contrast to the actions of presenting symptoms only, presenting a candidate diagnosis pushes forward across the physicians medical judgment by anticipating this judgmentwhether straightforwardly or more obliquelythereby making treatment directly relevant. The fundamental claim of this chapter is that through the way they present their childrens problems, parents effectively initiate a negotiation of the treatment their child will receive at the visits end. When parents put a candidate diagnosis onto the interactional table, physicians must then contend with it. When that diagnosis is for a bacterial infection, then by extension they must also deal with the relevant treatment: antibiotics. This behavior is associated with physicians perceiving parents as expecting antibiotics. As mentioned in chapter 1, physicians who perceive a parent to expect them to prescribe antibiotics for an illness are more likely to inappropriately prescribe antibiotics (Mangione-Smith et al., 2006; Mangione-Smith, McGlynn, Elliott, Krogstad, & Brook, 1999). So, the use of this behavior is extremely consequential both for the interaction and for the treatment decision, even if parents use of it may be motivated by a range of different reasons. In this chapter, we observed the interactional consequentiality of the different presentation formats. We saw that physicians deal with candidate diagnosis presentations as entering into their domain of expertise. This is true both when physicians conrm and when they disconrm the parents candidate diagnosis. When the parent conrms, the physician claims, from second position, to have been the person to make such a determination. Conversely, and when the parent was incorrect, physicians commonly state that antibiotics were not necessary. Parents typically did not bring up antibiotics in these contexts, but physicians were nonetheless treating other parent behaviors as indexing antibiotics and a parent preference for them. At the same time parents, even while using behaviors that apply pressure for antibiotics, show their own orientation to and respect for medical authority and the physicians primary rights over that domain. In part, this is evidenced through the infrequency of candidate diagnoses and the contrasting prevalence of symptoms-only problem



presentations. But it is also evidenced by the accounts provided and the downgrading used when parents do begin to tread into the medical domain. This chapter offers strong evidence that parents affect prescribing decisions in ways that might be quite unexpected. Traditionally, parent and patient participation has been thought to (1) require physician invitations, (2) occur during the treatment phase of the visit, and (3) involve some direct statement of preferences. With problem presentation formats, none of these elements is present, and yet parents can nonetheless be observed to shape the outcome of the prescribing decision by shaping how the physician sees the childs illness and by suggesting what sort of stance they have toward the illness. As mentioned earlier, these data suggest that parents may not always (or necessarily) intend their problem presentation format to have the impact that it does. Whereas to a physician a bacterial diagnosis indicates antibiotics, this is only one thing that it may be associated with for a parent. Parents may be more concerned to underscore the legitimacy of their visit, and one resource for accomplishing this might be suggesting that the reason for the visit is because the parent thinks the illness is treatable with medicine. Thus, it is not the medicine that they want per se but rather a cure. Another possibility is that parents may be concerned that their child is seriously ill. Although physicians may understand that viral and bacterial illnesses can involve similar discomfort and severity of illness in terms of the illness experience, parents seem to generally believe that viral illnesses are minor whereas bacterial illnesses are serious. Their model of illness is somewhat at odds with physicians in this respect. Thus, mentioning a concern about a bacterial illness could be a practice through which parents underscore their concern for their children. In all cases, parents seem to view their use of candidate diagnoses as marked, delicate, and rather restricted in use. But even if delicate, rarely used, and differentially motivated, when used, candidate diagnoses represent the behavior that most commonly lls the rst sequentially provided-for opportunity to inuence the diagnostic and treatment outcome of the medical visit.

Alternative Practices for Asking and Answering History-Taking Questions

fter a reason for the childs visit has been established, physicians typically begin taking the childs illness history. During this activity physicians generally ask questions about what sorts of symptoms the child has had, and prior treatment. In contrast to the reason for the visit phase, the history-taking phase is generally driven by very particular physician questions that make relevant quite constrained answers from parents. The structure of the history-taking activity does provide parents with an opportunity to participate, but their participation is arguably rather constrained because questions are generally quite focused. This is different from the reason for the visit phase, during which parents were answering a very unconstrained question about why they were there that allows them the interactional space to begin a brief narrative about the illness, detail the symptoms, or report on their concerns. Given all of this, parents might be thought to have few if any resources for participating in this activity at all, let alone shaping the diagnostic and treatment outcomes of the visit. This chapter shows that substantial negotiation over the childs condition does occur during this phase of the visit. Through the ways physicians design their questions, they display their stance toward the childs illness as problematic or not and, by implication, forecast whether it will be treatable. Similarly, parents have resources both for displaying their own stance toward the illness and for pushing physicians toward or away from particular diagnostic and treatment trajectories, as understood through physicians questions, in the way that they respond to questions.




In this chapter, I rst discuss the principles that underlie the way physicians design their questions and then offer evidence that parents hear history-taking questions as implicating diagnosis and treatment. I then discuss two resources parents have for shaping the diagnostic-treatment trajectory during this phase of the visit. Finally, I demonstrate that physicians treat these practices as ways of negotiating the diagnostic and treatment outcome of the visit.

History taking is generally characterized as a series of doctor-initiated questionanswer sequences (Beckman & Frankel, 1984; Boyd & Heritage, 2006; Byrne & Long, 1976; Mishler, 1984). However, these sequences are best understood as part of larger courses of action concerned with both gathering information and initiating a differential diagnosis of the patients condition (Athreya & Silverman, 1985). Medical students are taught to begin considering diagnostic possibilities as soon as they read about the patients symptoms in a chart or hear about them during the encounter (Athreya & Silverman, 1985; Cohen-Cole, 1991; Greenberger & Hinthorn, 1993). Textbooks that offer prescriptive techniques for taking patients histories typically suggest questions to ask in the face of given symptoms, with the idea being that each question should rule out a possible diagnosis or move a step closer to ruling one out. For example, if a child is having abdominal pains, one textbook suggests inquiring into factors that either aggravate or relieve the symptoms. Answers to such questions do not simply provide more information about the condition; they also indicate whether the problem is inammatory or spastic in nature (Athreya & Silverman, 1985). So, each answer to a history-taking question furthers the physicians progress toward a particular diagnosis. When we examine interactional data, we see that it is not only doctors who treat questions as part of a diagnostic and evaluative process. Drew (2006) showed that in calls to a British on call doctor after hours, callers treat doctors history-taking questions as a time during which the doctor is determining the urgency of the patients needs and whether a home visit is necessary. Often, callers treat history-taking questions as an opportunity to embellish their initial accounts, in order to convince the doctor of the seriousness or urgency of the condition (p. 423). Additionally, he found that callers often volunteer further details of new or unrelated patient symptoms. These are often presented following a no-problem answer to a question about symptoms. Through the design of history-taking questions, physicians reveal much about their presuppositions and, in turn, about their stances toward patients illnesses. As Clayman and Heritage note with respect to news interviews, there are not really any truly neutral questions (2002). Researchers in the medical context have noted that the design of history-taking questions is consequential. At the most basic level, researchers of provider-patient communication have discriminated between closed-ended and open-ended questions (Byrne & Long, 1976; Mishler, 1984) and examined the relative merits of these alternative designs (Beckman, Markakis, Suchman, & Frankel, 1994; Mishler, 1984; Roter & Hall, 1992). In particular, the narrower the range



of appropriate responses, the more a question limits the patients participation (Roter & Hall, 1992). This may include the way that a question can exclude the patients life world perspective and personal experience (Mishler, 1984). Boyd and Heritage (2006: 163) observe that questions (arguably, particularly yes-no questions) unavoidably establish agendas, and embody presuppositions, and preferences concerning patient responses (See also Heritage, 1984a, 2002; Koshik, 2002; Lindstrm, 1995; Pomerantz, 1984). They further argue that, in the context of comprehensive history taking (i.e., the kind of history taking that occurs in annual checkups as opposed to acute care visits) (Bates, Bickley, & Hoekelman, 1995), the design of doctors questions is guided by two principles: optimization and recipient design. Optimization involves designing questions to prefer or to facilitate no-problem or prosocial responses. (See Heritage, 1984; Pomerantz, 1984, for further information about structural preference.) For example, even in the case of a patient in late middle age, a history-taking question is structured as Is your father alive? rather than Is your father dead? (Boyd & Heritage, 2006: 165). This principle of optimization is departed from, they argue, only for cause. So, a question that is not optimized in its design is marked and conveys that something out of the ordinary is being done with the question. The second principle, the principle of recipient design, means that questions should be tted to some matter the patient has raised where an optimized question would be inappropriate. (For discussions of the broad principle of recipient design, see Sacks & Schegloff, 1979; Sacks, Schegloff, & Jefferson, 1974; Schegloff, 1972a.) Thus, although asking if patients parents are alive reects the principle of optimization, if patients are relatively old or have said things that suggest that their parents are not living, then physicians normally reect this information in their question design. Failure to do so will raise questions about how attentive they have been to the patient. A major piece of evidence for these claims is that recipients of questions respond differently, depending on their design. It is only a small step from here to the observation that through the design of recipient responses, recipients do substantial interactive work to align or disalign with the questioners stance toward the question. For instance, Heritage (1998) has shown that people who respond to a question with an Oh-prefaced response convey that they held a position on the topic prior to the inquiry being made, and Raymond (2003) has shown that when recipients of yes-no questions provide non type conforming answers, they resist a questions agenda or the terms (e.g., the presuppositions) of the question. With respect to the medical context, Stivers and Heritage (2001) have shown that in comprehensive history taking, patients may not be as conned by question agendas as was previously thought. Rather, they may implement various forms of sequence expansion, including narratives, to offer unsolicited information. In these ways, we see that patients have multiple resources for maneuvering even in a seemingly narrow answering space. This maneuverability is particularly important for conveying the patients own stance toward the illnessparticularly its doctorability and treatability. Thus, during the history-taking activity of interest to us, doctors and parents alike have a variety of resources for displaying their relative stances toward the childs problem and thus toward each other. We will see that through question asking



and responding, parents and physicians once again negotiate what kind of problem the child has: both whether it is doctorable and whether and how it is treatable.

Problem Presumption: Principles of Question Design in Acute Care

The principles underlying question design in the acute care context are, of course, related to the principles underlying question design in the comprehensive or routine care context (Boyd & Heritage, 2006). By virtue of structural preference, yes-no history-taking questions are inevitably tilted toward either a problem or a no-problem response (Boyd & Heritage, 2006). For instance, in Extract 3.1, the question Are you eating? prefers a yes answer through its use of an unmarked interrogative construction (Pomerantz, 1988), whereas the formulation Any vomiting or diarrhea? prefers a no answer because of the addition of an unstressed negative polarity item any (Boyd & Heritage, 2006; Horn, 1989; Koshik, 2002). Negative polarity items reverse the structurally preferred answer from an afrmative (e.g., following Vomiting?) to a negative.1 But both of these questions can be said to presume no problem in these health domains because the preferred answer to the question is a no-problem answer.
(3.1) 206 [girl presents with a sore throat] 1 2 3 4 5 6 7 8 9 10 11 12 DOC: 1-> Are you eating? (0.2) PAT: Y:eah:, (0.5) DOC: ( ate, (0.2) Okay.) DOC: 2-> .hh Any vomiting er diarrhea? (.) PAT: Mm[:_ [((PAT looks to MOM)) MOM: That wasnt diarrhea honey, DOC: No, Okay. So no vomiting, no diarrhea, MOM: (Mm mm.)

In acute care, the principle of optimization may, at times, conict with concerns of both visit legitimacy (Heritage & Robinson, 2006a) and condition treatability because questions that are optimized are built for responses indicating that a symptom is not problematic. When a question is designed to presume wellness, parents may understand the physician as displaying a stance that the childs condition is neither serious nor in need of treatment. But a physician who designed all history-taking questions to display a no-problem presumption might be heard as questioning the likely existence of symptoms and hence the legitimacy of the childs visit altogether. Conversely, a physician who designed routine questions about the childs general health to prefer problem answers could be heard to convey a stance that the childs



illness was quite serious. Questions about unmentioned problems or routine background questions are generally designed to prefer a no-problem response. Evidence for this can be seen both in Extract 3.1 and again in Extract 3.2.
(3.2) 308 [girl presents with stuffy nose, sore throat, and a blister in her mouth] 1 2 3 4 5 6 7 8 9 DOC: -> .hh Any fever at home? at all? MOM: No^:.= DOC: =Okay, MOM: [NoDOC: -> [No vomiti:ng, er[: MOM: [No:_ DOC: Any other problems like that huh, (1.5) MOM: N:o.

The doctors question at line 1 prefers a no (and a no-problem) answer through the use of the polarity marker any (Boyd & Heritage, 2006; Quirk, Greenbaum, Leech, & Svartvik, 1972). Similarly, the doctors request for conrmation in line 5 (continuing into 7) prefers a no answer in that it solicits a conrmation of a negatively formulated assertion. This suggests that similar to comprehensive history taking, as discussed by Boyd and Heritage (2006), acute care history-taking questions are guided by two partially competing principles. Like comprehensive history taking, there is a similar orientation to the principle of optimization: Physicians display a presumption that if the child had a fever, vomiting, or other problems like that, they would have been mentioned. Questions about these symptoms are generally designed to embody a presupposition that they do not exist because they were not mentioned and are generally optimized to prefer a no-problem response. However, unlike the comprehensive history-taking context, in the acute care context, physicians are aware that there is at least one problem that prompted the visit. It is not simply an issue of recipient design because this underlies optimization in this context as well. Rather, physicians also typically design some of their questions with an eye toward the principle of problem attentiveness. This principle asserts that physicians should design questions about dimensions of the childs condition that have been mentioned or implied to be problems in such a way as to prefer a problem response in order to display that they do, in fact, regard these issues as problematic. These two guiding principles are very much in line with the general conversational principles Levinson covers in his theory of generalized conversational implicature (2000). The design of physicians history-taking questions in the acute care context juxtaposes the Q (quantity) and I (informativeness) principles. The Q principle states both that speakers should provide the strongest possible statement of knowledge that they can and that, most relevant for us here, recipients will assume that what is stated by their interlocutor is the strongest possible statement or description.2 The I principle states that speakers should say as little as necessary to achieve communicative ends. The corresponding recipients corollary assumes this and therefore



allows recipients to assume the richest description possible, consistent with what is taken for granted. The practices described here are very much an interactional version of the dilemma described by Levinson because doctors are provided only a very small amount of information about the child, and yet this information is clearly utilized in terms of the design of history-taking questions from that point forward. And, as mentioned earlier, doctors appear oriented to the assumption that if the parent did not mention particular symptoms, they are not likely to exist (Q principle), and if particular symptoms were mentioned, then questions broadly in line with those symptoms should be designed to presuppose a problem (I principle). Moving from the theoretical principles to interaction principles, these issues broadly correspond to the interactional constraints surrounding the two principles described earlier: optimization and problem attentiveness. A clear illustration of this can be seen in Extract 3.3, where a toddler-age girl has come in for a runny nose. The questions beginning in line 3 reect the principles discussed by Levinson because in their design the physician displays her own orientation to the parents not having mentioned symptoms as indicating that the child has no other symptoms: nothing else, no fever, no vomiting, no diarrhea.
(3.3) 211 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 DOC: DAD: DOC: MOM: DAD: DOC: DOC: MOM: DAD: DOC: DOC: DAD: MOM: DOC: DOC: She has uh runny nose, #Ye:[:s.=h# [Okay anything el:se? That[s (it.) [N::ah.=h No:?, (.) No fever? [No:. [No. No fever, no [coughing so far, [Okay. No cough:, Okay:?, No vomiting, no diarrhea?, N:o. No. (nothing.) Okay.

When physicians design questions to be problem attentive, they underscore the legitimacy of the childs visit by displaying themselves to assume a problem. We can see this in the design of the two questions in Extract 3.4.
(3.4) 206 1 2 3 DOC: -> Okay:? .hhh Uh:=uhm Are you coughing? (0.2) PAT: Yeah:.



4 5 6 7 8 9 10


Mkay:, (0.5) DOC: -> Uh runny nose? (1.0) PAT: #m# Like every once in uh while itll start running. (8.0) DOC: O::kay.

Here, both inquiries are designed to prefer a yes answer that conrms the existence of problems. Both use the positively formatted question design, with the second question acting as a second in that series, thereby also utilizing that structure (Heritage & Sorjonen, 1994). Whereas the questions in this example have to do with identifying the existence of particular symptoms, we can also observe the same principle in questions about the quality, quantity, or duration of a particular symptom. An illustration of the rst two is shown in the questions in Extract 3.5.
(3.5) 505 [presented with worsening cold and lethargy] 1 2 3 4 5 6 7 8 9 10 11 DOC: -> Has he been coughing uh lot? (0.2) MOM: .hh Not uh lot.=h[h DOC: [Mkay:?, MOM: But it- it <sound:s:> deep. (1.0) MOM: An with everything we heard on tee v(h)ee=hhhh we got sca:re. DOC: -> Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm.

The questions asked in lines 1 and 9 are both designed to prefer yes answers and thus to conrm the existence of a problem. In the rst case, the question is afrmatively designed. In the second, the request for conrmation uses a similarly afrmative assertion with rising intonation (as noted with ?) and stress on deep to indicate what is being questioned. The design of history-taking questions in these acute care visits can be seen as guided by the principles of optimization and problem attentiveness. This is an important component to our argument because through an orientation to these two principles, parents are able to ascertain the physicians stance toward their childs illness and can therefore negotiate treatment. The way a question is designed unavoidably conveys the physicians stance toward the childs condition (or some aspect of it) as problematic or nonproblematic. When questions are built as part of a larger activity of history taking, the presence of a trajectory toward a problem or a no-problem diagnosis is even more visible. Thus, understanding how questions are designed is important for our larger argument.



The Diagnosis Implicativeness of History-Taking Questions

One primary mechanism through which physicians reveal their stances toward a childs illness is their question design. A major reason is that questions are understood to be directly linked to or in the service of the activity of diagnosing. Earlier, we reviewed medical textbook literature stating that physicians should use history taking to begin differential diagnosis. But not only in theory is the activity of history taking in the service of diagnosis; physicians and parents treat history-taking questions as beginning the diagnosis process. This becomes visible both in the way physicians design their questions and in the way parents respond to them. Although parents and children are not likely to recognize all of the diagnostic implications of any given question, they nonetheless commonly display an orientation to history-taking questions as being part of a differential diagnosis. For example, in Extract 3.6 a grandmother attending the visit with her son and grandchildren responds to a physicians history-taking question and then negates the diagnosis that would have been implicated, had the question been answered afrmatively and which the physicians trajectory of questions was likely to have been pursuing.
(3.6) 506 1 2 3 4 5 6 DOC: Does he have uh history of wheezing? (0.5) GMA: No. (.) GMA: -> He doesnt have asthma. DOC: (Okay,)

The grandmother rst answers only the question that was asked (line 3), but then, after a micropause (line 4), she asserts that her grandson does not have asthma, thereby treating the prior question as part of a differential diagnosis for asthma. The doctor receipts this assertion with (Okay,) which acknowledges the grandmothers understanding of the implicit link between the question and the diagnosis without overtly conrming it. Besides recognizing the link between questions and diagnoses, insofar as the question at line 1 could initiate a series of questions to investigate whether the child has asthma, the grandmothers He doesnt have asthma. works to shut down that diagnostic trajectory. Even young children can observe that questions are being asked in the service of diagnosing. We can see this in Extract 3.7, where an 8-year-old girl orients to the diagnosis-implicative nature of history taking by relating a possible diagnosis teasingly offered by her parents to the doctor.
(3.7) 517 1 2 DOC: Does it itch? ((begins to dry hands with paper towel)) (2.1)



3 4 5 6 7 8 9 10 11 12 13 14 15 16 17



[Sort of. (0.2) DOC: Mka:y, (.) DOC: An whered it start off. (0.5) DAD: (In thee=uhm) (.) thuh ba:ck? DOC: Mka:y?, (12.5) PAT: -> My mah- .hh my mommy an daddy said Im allergic to chocolate. (.) DAD: -hhh ((laugh)) DOC: (hhh) ((laugh))

That the girl performs this action at this juncture during history taking displays her orientation to diagnosis as a process being accomplished partially through history taking, even though the possible diagnosis is neither reported nor received in a serious way.

The Treatment Implicativeness of History-Taking Questions

Even treatment can be implicated through history-taking questions. Although the relevance of treatment is generally introduced through an orientation to diagnosis, evidence of the treatment implicativeness of history-taking questions is also visible in the ways parents and physicians deal with history taking. And parents hear the stance revealed by doctors questions as relevant to whether the child will ultimately be treated. Although we will be exploring the practices through which parents negotiate with physicians in the next section, here I want only to observe that treatment is oriented to by both parents and physicians as linked to the problem versus no-problem nature of the condition, and the stance physicians and parents take to the childs condition begins early. We can see this quite clearly in Extract 3.8. Here, the mother provides a problematic response (line 2) to a question designed to prefer a no-problem answer and then upgrades this response with quite uh little bit. Interestingly, this upgrade is revised to a downgrade in response to a repair initiation at line 5 (see line 6). It is following the physicians receipt of this that the parent goes on to assert, rst, that she is a relatively troubles-resistant parent (Heritage & Robinson, 2006a; Jefferson, 1988) (lines 910/12/1416) and, most relevant for this analysis, that were doing everything that I think were supposed to be doing and it is still not getting better.
(3.8) 161307 1 DOC: Is she uh (0.2) eating okay?

2 3 4 5 6, 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23





No, I would say her appetites down. (.) [quite uh little bi[Pardon, I would say her appetites down a little. Yeah. (0.3) She just- You know I dont like to bring her- iI dont typically bring my kids in (.) you [know [Yeah. drop of a [hat_ .hh but its been goin on about= [(Right.) = ten days now and its: (.) I mean Im- were doing everything I think were supposed to [be doing and its not getting any better. [Yeah. (Yeah ) (7.5) Actually now that Im thinking about it she was sick before Thanksgiving_ (0.8) So: yeah ^its probably been about ten days; (I always ) Is the cough worse at night?

The mothers response here appears defensive in the sense that she appears to hear the question as indexing the doctors stance toward her daughters condition as not terribly problematic. In this position, she invokes their failure to resolve or manage the girls condition through home remedies (lines 1416). So history-taking questions are heard as consequential for both diagnosis and treatment outcomes. Parents also sometimes use history-taking questions to bring up very specic treatment options. In Extract 3.9, the parent brings up antibiotics. Earlier in this encounter, the parent presented the childs condition as a cold with congestion and headaches. Just prior to this extract, the physician asked about previous sinus infections, and the parent conrmed that the child has suffered from this in the past. Here, they are in the middle of extensive history taking. The parent re-raises sinus infections as part of her response to a question about when they had previously met this physician, who is not their regular pediatrician. The way that she refers to the prior infection with last .h sinus infection. treats this illness as a sinus infection, though this is very indirect because it is embedded in the reference to the prior. But last is relevant only with respect to next, which she here conveys is her stance.
(3.9) 2069 (Dr. 9) 1 DOC: ^Oh I- I knew Id seen you guys before_ I didnt 2 realize it was- ((sniff))/(0.2) 3 MOM: Yeah, its [(been uh while) 4 DOC: [>was< back in No[vember. 5 MOM: [( ) [Ye:s. I think it=



6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27


[Yeah:.=h =was -his last .h sinus in[fection. huh huh [huh [Yeah. [Yeah, .hh (2.0) So- November. So its four months uh, Yieah:. (8.0) Did we get an x ray then? (.) No [(there was no_ Hm mm. [Or did we(6.5) It=did clear up though. (.) With=that- (1.0) I think it was like twenty days or whate[ver it was. [Uh huh,=hh of antibiotics it cleared up. (3.2) And thuh headaches went away? Yeah.

Moreover, in response to a question that pursues information about a previous sinus infection (line 14), the parent mentions that the condition resolved with antibiotics (lines 19/2122/24). Although the physicians question Did we get an x ray then? (line 14) suggests a pursuit of how the condition was diagnosed rather than how it was treated, the parent, after a no answer to this question (line 16) and after a long silence, instead goes on to provide information about the treatment that was given rather than the diagnosis. With this turn, the mother suggests that the antibiotics worked to cure a condition similar to the one her daughter has now. Thus, with this turn, the parent implies that they are seeking similar treatment. Important for this point in our analysis, parents orient to the relevance of information about treatment during history taking because it may well shape the physicians stance toward the childs condition as problematic and treatable when they reach that phase of the visit. Physicians also treat the way parents design their responses as having direct treatment implications. This is illustrated in Extract 3.10. After the parent has taken a stance that the childs condition is getting worse and is therefore problematic (line 4), the doctor, in her response, addresses the concern as lobbying for antibiotics (lines 1112).
(3.10) 119 1 2 3 DOC: MOM: DOC: .hh And this has been going on for about uh week?, .hh Seven day:s, Okay:, [(yeah.)

4 5 6 7 8 9 10 11 12 13 14 15



[But thuh cough is (wearing) wor:se. Getting worse. (.) DOC: Okay. Well well take uh- I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on -> for uh while. Well look an see whether or not she -> needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an then thuh cough sorta kinda hangs o:n.

Although the parent does conrm the physicians question about the illnesss duration (line 2), she does it with a nonconforming answer: She uses an alternative formulation rather than the more minimal alternative yes (Raymond, 2003). Additionally, in her expansion (line 4), the physician hears her to be treating her daughters illness as more problematic than the physician had treated it. But rather than dealing with this in terms of diagnosis, she deals with it in terms of treatment. Finally, even if antibiotics are not overtly broached at this stage in the visit, treatment can nonetheless be seen to be relevant in relation to the questions. We saw this in the previous example with the physicians comment that Well look an see whether or not she needs something that needs an antibiotic. We have seen that the diagnostic and treatment import of history-taking questions and responses are understood by both parents and physicians. In terms of negotiation, this has important implications for what parents do in their responses to history-taking questions. A physician who takes a stance toward the condition as nonproblematic is not likely to recommend prescription treatment and, in particular, not antibiotic treatment. This stance is potentially problematic for parents because, at a minimum, it threatens their own judgment that the child was ill and therefore threatens the legitimacy of their visit. Parents have resources that are deployed systematically to deal with these issues and through which they negotiate their childs treatment and, by extension, their diagnosis and visit legitimacy.

Parent Resources for Negotiating Physicians Diagnosis and Treatment Trajectories

Although physicians history-taking questions are quite constraining in terms of the relevant next action, as we have already begun to see, parents can still do interactional work to break out of those constraints and, for instance, interactionally push physicians away from the stance they conveyed through a question and/or even toward an alternative stance. As we have already seen, these stances have clear implications for diagnosis and treatment. This section examines two interactional resources for accomplishing this and thus for accomplishing visit outcome negotiation: mentioning additional symptoms and mentioning possible diagnoses.



Mentioning Additional Symptoms Parents are frequently confronted with a dilemma in responding to physician questions: They must respond in a way that supports a no-treatment outcome despite this being potentially contrary to their interests and beliefs. One way that parents manage this interactional dilemma is to steer physicians away from a no-problem diagnosis by mentioning additional symptoms as part of their question response. This is not a common practice. In the Seaside data, it was observed to occur in the history-taking context 9% of the time. Although not terribly frequent, the behavior still represents yet another interactional resource parents have for initiating a negotiation of the visit outcome with the physician. We can see an example of a parent mentioning an additional symptom in Extract 3.11 (shown earlier as 3.5). In line 1, the physician asks a question that, contrary to the other questions just prior (lines 111), prefers a problem response. But the parent provides a dispreferred, nonconforming (Raymond, 2003), and no-problem answer when she disconrms that it is not uh lot. However, following that, she offers a dimension of the illness that is problematic: the quality of the cough: But it- it <sound:s:> deep. This works to shape the trajectory of the physicians diagnosis in two ways: First, it pushes the physician away from her present direction by introducing a new dimension of the illness. By providing problem-implicative information at this point in the interview, the parent provides the physician with an alternative diagnostic path. Second, the practice invites pursuit and sequence expansion (see Schegloff, 2006) that would move the trajectory in a different direction. Evidence for this is provided in the physicians reaction: Although she had proposed closure of the sequence following Not uh lot. with Mkay:?, (line 16), following the additional symptom presentation, she initiates a line of further questioning. (More on physician uptake of additional symptoms will be discussed at the end of this chapter.)
(3.11) 505 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 DOC: BOY: DOC: DOC: DOC: DOC: DOC: MOM: DOC: Anything hurt? (1.5) No:_ (.) Your ears dont hurt? (0.3) ((BOY shakes head laterally)) Your throat doesnt hurt? (0.6) ((BOY shakes head laterally)) Your tummy doesnt hurt? ((extra rise at the end)) (0.5) ((BOY shakes head laterally)) No:? (0.2) Has he been coughing uh lot? (0.2) .hh Not uh lot.=h[h [Mkay:?,

17 18 19 20 21 22 23


MOM: -> But it- it <sound:s:> deep. (1.0) MOM: An with everything we (heard) on tee v(h)ee=hhhh we got sca:re. DOC: Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm.

Mentions of additional problematic symptoms can be found in both answers to individual questions and in environments where a series of questions (and their responses) may be taken into account. For instance, in Extract 3.12, the mother provides a series of no-problem responses to questions. The rst two answers are to optimized questions. The third question is hearable as beginning to search for a possible problem, but once again, the parent provides a no-problem answer. And in fact, the mother goes on to further support the physicians no-problem trajectory with an additional no-problem statement (arrow 4). However, immediately after (line 24), she shifts position, stating, So thats why it didnt concern me too much and which overtly displays her orientation to the prior sequences as aligned with a no-problem trajectory. In line 25, the mother continues her turn by moving toward the event that brought her to the medical visit, this time providing problem-implicative informationthat she saw something in her daughters throat. This, like the previous instance, is prefaced by the conjunction but, which treats the action to follow as contrastive with the prior.
(3.12) 308 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 DOC: 1-> .hh Any fever at home? at all? MOM: No^:.= DOC: =Okay, MOM: [NoDOC: 2-> [No vomiti:ng, er[: MOM: [No:_ DOC: Any other problems like that huh, (1.5) MOM: N:o. MOM: Thank g(h)oodn(h)ess [(huh huh,) DOC: [N:kay:, DOC: 3-> .hh What abou:t uhm as far as her appetite, eating wi:se, MOM: Still has uh good appeti:te? DOC: Thuh appetite is [goo:d? MOM: [(Yeah:,???) DOC: hh=nkay, so thisMOM: 4-> [An she still (have) energy tuh pla:y, DOC: [(Thuh) sore throat-



20 21 22 23 24 25 26 27 28 29

MOM: DOC: DOC: DOC: MOM: -> => => DOC: MOM: => DOC:

[ T u h r i d e h e r bicy]cle: n[Still has energy tuh play:?,] Nkay, .hh Uhm: So thats why it didnt concern me too much but I:- I looked at it this morning: an I thought I saw (.) Something i[n there. [something in there. [Yeah:. [Mkay,

Similar to Extract 3.11, the mother pushes the physician away from an entirely noproblem trajectory toward an investigation of a possibly problematic dimension of the childs illness: in this case, something visible inside the childs throat. Yet another type of environment where we see parents using this practice is exemplied in Extract 3.13. Here the pattern is similar to the examples in Extracts 3.11 and 3.12, but it follows not a question to which a no-problem answer must be given but after the doctor has indicated a shift away from illness-specic history and to more general questions about whether other family members are sick. The mother perhaps hears this question as indicating that the physician has completed an investigation of the illness and with her action redirects the physician back toward a problem trajectory.
(3.13) 2053 (Dr 8) 1 2 3 4 5 6 7 8 9 10 11 12 13 DOC: -> Mlk Uh:m: how bout this #morning.# Has he thrown up? MOM: No. Not since uh:- i- it must have been uh combination of the medicine and the [(milk). DOC: [#an thuh mi(h)lk,=hh huh huh huh huh (h)o(h)kay, DOC: -> .hhh O:kay. Is anyone else at home sick?=hh= MOM: =No. (0.5) MOM: => But thuh temperatu:re- w- went up every time: #uh:# (0.7) after taking the Tylenol [after a while it goes up. DOC: [Uh huh,

In this case, after providing a no-problem answer at lines 3 and 9, the mother then goes on to offer further problem-implicative informationthat her sons temperature can be kept down only with the use of Tylenol. In their own responses, physicians show that they hear parents to be pushing them away from no-problem trajectories when they respond to their questions with the presentation of additional symptoms. We can see evidence of this in Extract 3.14.



(3.14) 119 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DOC: .hh And this has been going on for about uh week?, MOM: .hh Seven day:s, DOC: Okay:, [(yeah.) MOM: -> [But thuh cough is (wearing) wor:se. DOC: Getting worse. (.) DOC: Okay. Well well take uh- I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on for uh while. Well look an see whether or not she needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an then thuh cough sorta kinda hangs o:n.

Here, the physician has requested conrmation that the duration of the girls condition is about uh week?, a characterization that does not treat this as either a long or a short period of time. However, the mothers response Seven day:s, is hearably an upgrade. Although technically Seven days is merely an alternative formulation of uh week, this formulations precision works to emphasize the conditions long duration and therefore treats the illnesss duration as problematic. Moreover, this answer fails to conform with the terms of the question (Raymond, 2003). The physician nonetheless proposes closure of the sequence with okay:, after which the mother offers the additional and problematic symptom that the cough is getting worse. In this case, it is not that the physician goes on to do a further inquiry, but she addresses the import of the mothers action in several ways. First, the doctor assures the mother that she actually heard it (that is, the cough), so she knows whether to be concerned about it or not (lines 78). Second, she addresses the mothers assertion that the cough is a problem by suggesting that, in fact, it may not be: sometimes coughs certainly can hang on for uh while. (lines 1112). Third, the doctor states that she will look an see (line 12). Finally, returning to whether the cough should, in fact, be treated as a problem, she suggests that it may or may not be something that needs an antibiotic (lines 1213). Her account is just following: very often (yuh) get uh col:d an then thuh cough sorta kinda hangs o:n. (lines 1415). Thus, across this turn, the physician addresses a range of issues embedded in the mothers turn and treats the mother as having conveyed a stance that the childs cough is problematic. We have looked at one resource parents have for pushing physicians away from a no-problem diagnostic and treatment trajectory during history taking: the mentioning of additional symptoms. It works because it offers a symptom as problematic at a point where only something unproblematic has been raised. This effectively encourages physicians to consider an alternative diagnostic trajectory, and it invites



sequence expansion, which, if done, does place physicians on an alternative diagnostic trajectory. Mentioning Alternative Possible Diagnoses Another communication practice that parents use following a question-answer sequence involving a no-problem answer is to mention an alternative possible diagnosis. This, too, is not frequent, occurring 12% of the time in the Seaside data. These diagnoses are not typically offered in quite the same way that they are offered in the problem presentation position (i.e., as candidate diagnoses as discussed in chapter 2). They are only rarely overtly stated as the parents diagnostic theory. Instead, they are more typically offered as justications for their childs condition being problematic when their question response is otherwise aligned with a no-problem stance toward the childs illness. What they do, though, is to propose an alternative diagnosis to be considered, and this is quite similar to candidate diagnoses. How overtly this is done certainly varies, as we will see. As an initial example, we can look at Extract 3.15.
(3.15) 313 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 DOC: 1-> How has he been today.Has he [eaten anything tday? DAD: [(uh t-) DAD: Uhp- uh:=l:little bit_ (he di:d.??) (0.2) DOC: 2-> Did he throw it up today? DAD: Uh: no he hasnt. (0.2) DOC: 3-> Ok:ay.=Has he had fever at home today? DAD: Uh:m, (0.2) No No. DOC: (2.0) DAD: We give him some Tylenol for thuh fever, DOC: When was tha:t. DAD: Uh:m, last ni:ght, DOC: 4-> But today hes had no fever? DAD: t- Today (we h-) no:. (.) DOC: O[kay. DAD: [Yeah no fever. (2.1) DOC: 5-> So no fever today. So he seems okay today. (.) DAD: Yeah seems tuh be okay. DOC: (h[Mkay.??) DAD: => [Its just that: I wanna make sure that (1.0) (might not have uh) i[nfection er something ( ) DOC: [O:ka:y,



The doctor asks a series of history-taking questions that receive no-problem answers (questions 15). At line 11, the father offers information that might problematize the previous answer. If the boy did not have a fever because it had been brought down with Tylenol, then he could potentially have misanswered the doctors question. However, because the Tylenol was given the previous evening, this is ruled out, and the line of questioning (at line 14) underscores the childs present no-problem status. This is still further underscored when, at arrow 5, the doctor requests conrmation that he seems okay today. prefaced with a summative marker So (Raymond, 2004). Here, the father rst conrms (line 22) but then continues his turn with a contrastive Its just that: and an alternative possible diagnosis: that he has an infection. The alternative diagnosis works to push the physician toward a conceptualization of the illness that is at odds with the prior line of questioning and indicates the parents own stance toward the child as not as healthy as he might seem. This alternative diagnosis defends the fathers decision to seek medical care. Alternative possible diagnoses are sometimes delivered quite explicitly. In Extract 3.16, the grandmothers mention of a diagnosis is specic: throat infection and ear infection. But both are still offered as general possibilities and not as strong theories, and there are two possibilities. The mention of more than one diagnostic possibility underscores that it is not a single theory being advocated, but rather it is a class of problematic diagnoses that is being pushed by the grandmother.
(3.16) 506 1 2 3 4 5 6 7 8 9 10 11 12 13 14 DOC: GMA: GMA: -> -> GMA: -> GMA: -> GMA: -> -> -> Has he ever needed uh breathing treatment? (.) No. (7.0) No but what he gets is like throat infections an ear infections. (0.4) Frequently. (0.4) An he (gets with thuh high fever.) (0.5) So thats why I=uhm (1.5) we decided to bring im in because (0.4) with thuh temperature. (.) (I know ) somethings gonna be wrong.

The grandmother rst answers a history-taking question with a no-problem answer. After 7 seconds of silence, the grandmother again offers a No and then goes on to contrast this with the type of illnesses her grandson has had trouble with in the past. This conveys the grandmothers stance that the physician may be pursuing a less productive diagnostic trajectory by suggesting an alternative possible diagnosis (i.e., ear/throat infection) to the one being pursued by the physician (i.e., allergies/ asthma). The grandmothers reuse of No and the connective but work to tie her



turn in lines 56 back across the lengthy silence to her turn in line 3. In this way, her concerns are voiced as an expansion of her prior answer and hence as part of a history-taking question-answer sequence rather than as an assertion of a candidate diagnosis. Single, specic diagnoses are, occasionally, offered by parents during history taking. We can see an example of this in Extract 3.17. Here the mother responds to a history-taking question about the duration of her sons fever (line 1) with an answer that might suggest she rushed to the doctor too quickly (line 3). The question embodies a presupposition that the child has a fever and orients to this as a problematic symptom. However, the mothers response treats the answer she gives as inadequate to evidence a problematic illness, and perhaps as potentially undermining the legitimacy of her medical visit (Heritage & Robinson, 2006a).
(3.17) 2067 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11 DOC: BOY: MOM: DOC: How long has he had uh fever for? Ah [bah buh[Just since last ni:ght_ Mm=hm::?, (0.2) MOM: -> But his brother an his sister have ear infection. DOC: Mm:, okay. MOM: So:, (.) DOC: .h MOM: Im uh li[ttle ( -) DOC: [Uhm has he been sick at all:?

The mother displays her orientation to 1 day as a minimal and relatively insignicant length of time with her use of just. But following the doctors receipt of this (line 4), the mother adds a further component to her turn with But his brother an his sister have ear infection. This offers a possible diagnosis as an account for seeking medical care within what is normally a relatively unproblematic duration of time (less than 24 hours). This addition also indexes her primary concernthat her son may also have an ear infection, and infection stands as a proxy for taking antibiotics as well. Extract 3.17 and the following Extract 3.18 both show cases where the diagnostic theory approaches that shown in the reason for the visit phase. Although alternative diagnoses are rarely presented as overtly during history taking as they are during the problem presentation, in these last two cases we can see that parents do approach this degree of directness. Here, in response to a question about coughing, the mother (following an exchange with her daughter in lines 34) offers a no-problem response (line 5): No cough.
(3.18) 2074 (Dr. 9) 1 2 DOC: An:d coughing too?, (0.8)

3 4 5 6 7 8


MOM: GIR: MOM: DOC: MOM: -> ->

Uh little bit or no coughing. #No:?# No [cough. [No cough. [Okay. [.hh <Her b::rother tested positive> for strep, Thuh results came in last Tues:day.

Subsequent to the physicians receipt of the answer (line 6), the mother offers a possible diagnosis (strep throat) that pushes the physician away from a line of inquiry regarding coughing and toward one about a throat infection. Considering both mentions of additional symptoms and mentions of alternative possible diagnoses, we can see that they do slightly different but related interactional work. Whereas mentions of additional symptoms push doctors away from a particular trajectory of questioning, mentions of alternative possible diagnoses in this same sequential environment actively instruct physicians as to which trajectory of inquiry they should pursue in the service of conrming or disconrming this diagnosis. Thus the latter pushes them toward an alternative diagnosis (and, by extension, treatment) trajectory. This is precisely how these behaviors can be understood as negotiation tools. The best evidence for this is the way that physicians tend to respond to these actions. They consistently treat these behaviors as pressuring them to consider an alternative diagnosis and treatment and often as specically indexing a desire for antibiotics.

Physician Responses to Parent Actions

When physicians ask questions, in addition to conveying particular presuppositions, they also often display that the questions were asked either as part of a relatively standard anticipatable series of questions or, alternatively, as a follow-up to some prior comment by the patient. Heritage and Sorjonen (1994) discuss the difference between two similar types of questions in the context of health visitorparent interaction: what they term routine or agenda-based questions and contingent inquiries. They observe that agenda-based questions typically embody a forward movement within a larger sequence, and the use of an agenda question following a previous question-answer sequence treats the prior response as unproblematic (p. 11). By contrast, contingent questions emerge in environments in which there is some unexpected or problematic response to the prior question, where the inquiry sustains the topical focus of the preceding question-answer sequence, and where it is recognizably produced as . . . contingent in character, rather than as anticipated (p. 11). Similarly, Byrne and Long (1976) discussed questions that relate a given symptom to a particular time, place, or activity as supplementary questions (p. 34). They discuss the importance of these questions in making a diagnosis. The contingent or supplementary quality of a question may also be suggested by its placement within a given sequence. Whereas the previous question types show, in Heritage and Sorjonens terms, a forward movement through an agenda of questions (1994: 11),



doctors also sometimes expand a question-answer sequence by asking further questions about a particular symptom as a postexpansion sequence (Schegloff, 2006). When they do this, they fail to progress in the same way that a new agenda-based question would. Halts in progressivity are important here because they represent one way that physicians treat a symptom as problematic. We can see an example of this type of question and response in Extract 3.19 (shown earlier as 3.11). The doctor asks a question about the amount of coughing (line 13). In response, the parent offers an additional symptom about the quality of the cough in line 17, as we discussed earlier.
(3.19) 505 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 DOC: 1-> Has he been coughing uh lot? (0.2) MOM: .hh Not uh lot.=h[h DOC: [Mkay:?, MOM: But it- it <sound:s:> deep. (1.0) MOM: An with everything we heard on tee v(h)ee=hhhh we got sca:re. DOC: a-> Kay. (An fer i-) It sounds deep? (.) MOM: Mm hm. DOC: b-> Like uh barky cough? MOM: .hh (1.1) Uhhhm=hhh It sounds very:=uhm (.) (I dont know:=wwlike:) (0.2) It sounds- (2.5) Tlk .hh Tlk Not like that like: DOC: [Not (barky.) MOM: [Like when someone has bronchitis that it sounds ( ) DOC: Okay. DOC: c-> Does he sound like uh dog er uh seal barking? MOM: No. DOC: Okay. (0.5) DOC: .hh Not bark(y).=hh (0.9) DOC: Okay.

In lines 1920, the mother explains that this quality to the cough caused her (and presumably her husband or family) concern because of what had been on television lately. At line 21, the doctor further expands the sequence by asking another question, this time one that builds off the second component of the mothers turn. In this way, the physician initiates postexpansion of the sequence and pursues the mothers additional symptom, rather than progressing to a next agenda item. This shows that



the mothers effort to move the doctor away from her existing trajectory and toward an alternative trajectory was successful. This sequence is pursued until that line of inquiry is exhausted (lines 3438). The effectiveness of additional symptom mentions is also evidenced by cases where a physical examination is generated by the additional symptom mention. We can see this in Extract 3.20 (shown earlier as 3.12). Here, the mention of an additional symptom is in lines 2528. This immediately generates a new sort of inquiry (lines 3032) and just after this, a physical examination (line 36).
(3.20) 308 24 25 26 27 28 29 30 31 32 33 34 45 36 MOM: So thats why it didnt concern me too much but I:- I looked at it this morning: an I thought I saw (.) Something i[n there. [something in there. [Yeah:. [Mkay, .hh Uhm tl but it hasnt hurt her enough that she has had trouble swallowing for example. She complain:ed of that: today. Is that right? Yeah. => Lets see. Sit up (straight,)



What these cases show is that in the question-answer sequence, parents and physicians can push and pull to negotiate the doctors orientation toward the illness. Beyond this, we can observe that what is truly being negotiated is whether the child will be treated. Physicians understand parents who are using these practices to push them away from particular inquiry or diagnostic trajectories and/or toward alternative trajectories to be negotiating the diagnostic and treatment outcome of the visit. We can see evidence for this in Extract 3.21 (shown earlier as 3.16). The grandmother offers an alternative diagnosis (or rather two: throat and ear infections). The physicians initial response is to claim that a regular cold can also be accompanied by a fever. By mentioning regular cold in this environment, the physician does several things: (1) she shows that this was, in fact, the trajectory she was going down because she defends the diagnosis; (2) she treats the grandmother as proposing a counterdiagnosis to the one she was pursuing through her inquiries; and (3) when she mentions its just uh virus. this is a shift from cold to a category of diagnosis that is directly related to whether antibiotics are prescribed. This suggests that the doctor hears antibiotics as having been made relevant by the grandmother.
(3.21) 506 5 6 GMA: No but what he gets is like throat infections an ear infections.



7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27





(0.4) Frequently. (0.4) An he (gets with thuh high fever.) (0.5) So thats why I=uhm (1.5) we decided to bring im in because (0.4) with thuh temperature. (.) (I know ) somet[hings gonna be wrong. [Daddy. [Stop that. [( ) -> Well actually uh regular cold can give you uh high -> fever for: three day:s. (0.4) -> As long as they act okay then actually its- it: -> (0.8) may go away by itself, (its) just uh virus. (2.0) => Uh:m, an you said hes had uh lotta ear infections? (.) In thuh pa:st yes. Okay. An when was thuh last time he had one?

Finally, note, too, that at the double arrow the physician returns to questioning and does so in pursuit of one of the issues mentioned by the grandmother. Thus, there is evidence that the behaviors are both successful in terms of moving the physician to a pursuit of alternative trajectories of inquiry and are understood as working toward this end. This case shows the physician explicitly orienting to the diagnosis as under negotiation. In Extract 3.22, we see that the physician treats antibiotics as being negotiated (and pushed for) by the parent.
(3.22) 119 1 2 3 4 5 6 7 8 9 10 11 12 13 14 DOC: MOM: DOC: MOM: DOC: .hh And this has been going on for about uh week?, .hh Seven day:s, Okay:, [(yeah.) [But thuh cough is (wearing) wor:se. Getting worse. (.) DOC: Okay. Well well take uh- I hear- actually heard it out there. (.) DOC: .hh (Yep-)/(eh-) .h I heard your cou^gh. DOC: .hh Uh:m sometimes coughs certainly can hang on -> for uh while. Well look an see whether or not she -> needs something that needs an antibiotic. DOC: .hh Uh:m (.) very often (yuh) get uh col:d an



Here, following the parents mention of an additional problematic symptom (i.e., the worsening cough), the physician goes on to offer a no-problem explanation for this (lines 1112) and explicitly treats the parent as lobbying for antibiotic treatment (lines 1213). Physicians respond very similarly when the parents action has been to offer an alternative possible diagnosis. We can see an example of this in Extract 3.23.
(3.23) 2074 (Dr. 9) 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 MOM: .hh <Her b::rother tested positive> for strep, Thuh results came in last Tues:[day. DOC: [(^Oo.) Okay. MOM: So: (0.5) Uhm: h(h) .h(h)h Just giving you that piece of information. DOC: Yeah. (.) DOC: -> It could very possibly.<I mean .hh u:sually you dont -> see strep with uh runny nose. DOC: h- Uhm_ .lkh ^Bu:t (.) if youve been exp^osed tuh strep, an exposed to uh co:ld I suppose you could get them both at thuh same: -ya know at thuh same ti:me, DOC: .h[h MOM: [And my son didnt have a fe[ver ei[ther. DOC: [.mlk [Really, MOM: So:_ DOC: #Okay.#

In line 7, the mother announces an alternative possible diagnosis relative to the prior questioning trajectory. This is receipted as news in line 9 but is not pursued at that point. The mother pursues a fuller response from the physician with Just giving you that piece of information.. With this turn, the mother orients to her previous possible diagnosis as a departure from the norms of doctor-parent interaction. She does this through her smile voice (noted in the transcript with the ). As Haakana described, patients often mark their actions as delicate through the use of such resources (Haakana, 2001). With this action, though, she also creates a context for the doctor to elaborate on his responsea post response pursuit of response (Jefferson, 1981). The physician initially receipts the information with Yeah. but then does go on to address the diagnostic implications (It could very possibly.). She draws on her previous line of questioning to defend not going toward a strep throat diagnosis. (The mother had previously said the child had a runny nose, and this would be counter to a standard symptom list for strep throat.) What this shows is that physicians can resist a parents pressure to move toward a particular diagnostic trajectory, but they still feel the pressure and reect this either by resisting or by succumbing to it. It is interesting, too, that later on in the treatment phase of the medical visit there is a statistical association between parent mentions of additional symptoms and



the way physicians talk about their treatment recommendation. Following a statement of additional symptoms, physicians were more likely to offer parents delayed prescriptions (i.e., prescriptions that they start if the child fails to improve) or contingency plans (i.e., offers to prescribe following a phone call if particular circumstances follow, such as if the fever reaches a particular level) (Mangione-Smith et al., 2001). In the Seaside data, if a statement of additional symptoms was present in the visit, physicians exhibited these types of concessionary treatment recommendation behaviors 24% of the time. By contrast, in the absence of any mention of additional problematic symptoms, these concessionary behaviors were present only 9% of the time (chi2 (1) = 7.67; p = 0.006). Although unlike candidate diagnoses, mentions of additional problematic symptoms were not signicantly associated with physicians perceiving parents to expect antibiotics, there is still both qualitative and quantitative evidence that the behavior is understood as applying pressure to the physician to treat the child with antibiotics.

In this chapter, we examined how physicians design their history-taking questions and how parents respond. We observed that through the details of how these question-answer sequences are produced, physicians and parents negotiate whether the childs illness (or specic symptoms) is problematic. Although it is rare that the history-taking activity involves any overt discussion of whether antibiotics should be prescribed for a child (but see Extract 6.9), it is quite common for physicians and parents to take part in a relatively covert verbal push-pull over their respective stances toward the childs illness and its treatability. And through a negotiation of whether some aspect of an illness is problematic, the patients diagnosis and treatment are also being negotiated. History taking may seem an unlikely environment for negotiation to occur, but physicians and parents treat history-taking questions as part of a diagnostic trajectory. Whether or not parents are always able to determine what trajectory a physicians questions may be part of could be seen as less important than that they understand physicians to be on a trajectory. If a parent understands the physician to be pursuing a diagnosis, even if they do not know which diagnosis, they may still be able to assess whether it is a problem-implicative one that will ultimately yield treatment for their child. In cases where a no-problem diagnostic or treatment trajectory appears to be projected by the questioning trajectory, parents make use of interactional practices to push physicians away from this trajectory and/or toward an alternative one. We examined two practices: mentioning additional symptoms and mentioning alternative possible diagnoses. Both of these practices are solutions to an interactional dilemma: Parents must offer a no-problem answer to a history-taking question, which potentially undermines their own position that the child has a problem. Thus, at this juncture, when parents extend their no-problem response in order to mention either the additional symptom or a possible diagnosis, we observed that this pressures physicians to adjust their stance toward the childs illness by inviting sequence expansion regarding the new information (the symptom or the diagnosis).



The two practices are slightly different in terms of the action that they accomplish. Whereas the additional symptoms pressure physicians by pushing them away from their prior questioning trajectory (usually a no-problem trajectory), alternative possible diagnoses afrmatively push physicians toward another trajectory (usually one that involves a bacterial diagnosis with corresponding antibiotic treatment). Thus, although both practices are a form of lobbying, the latter practice can be seen to be somewhat stronger. Finally, we observed that physicians treated these behaviors as lobbying for a problem trajectory even when the parent did not offer an alternative. That is, even when only additional problematic symptoms are offered, physicians treat parents as lobbying for a problematic diagnosis and often for antibiotics in particular (recall 3.22, for example). As in earlier work, here I argue against the idea that in the history-taking activity the patient is imprisoned within courses of action that are overwhelmingly undertaken at the doctors initiative (Stivers & Heritage, 2001: 178). Although this portrayal is common (Byrne & Long, 1976; Fisher & Todd, 1983; Mishler, 1984; Waitzkin, 1991; West, 1984), particularly with respect to yes-no types of questions (Mishler, 1984; Roter & Hall, 1992), social interaction is much more exible than it is often credited to be. Parents, as shown in this chapter, hear the consequences of their responses relative to the question trajectory environment and exploit the sequential opportunity to add to their turn, and thus they do work that can, and frequently does, affect not only subsequent actions interactionally but also diagnostic and treatment outcomes later in the visit. This chapter also illustrates a second opportunity that parents and physicians utilize in covertly negotiating the treatment outcome of the visit. Whereas the primary way that the candidate diagnosis problem presentation works to negotiate is by asserting the parents stance that the childs condition is a problem that can and should be treated, during history taking, parents react to physicians stances by either afliating with or resisting physicians stances toward the childs condition. And further, they make use of turn constructional resources to manage the direction physicians proceed with the visit. History taking leads to physical examination and quite often substantially overlaps it. Both activities represent an investigation of the childs problem. However, during the physical examination, two actions relevant to our discussion are common: (1) additional inquiries that investigate the problem and (2) statements about what the physician is observing. We do not deal with the physical examination in its own right here, but we can readily understand that the rst type of action will provide parents with a similar response environment to the environment discussed here. The second is similar to the diagnosis environment. I will therefore look at both online and nal diagnoses in the next chapter. Online comments will be discussed as a distinct physician behavior in chapter 7.

No Problem (No Treatment) Diagnosis Resistance

ntil the diagnosis phase, covert forms of parent pressure typically work to encourage the physician in a particular diagnostic direction on the basis of parents own inferences about where the physician is heading. A diagnosis of a childs condition transforms the interactional context because negotiations of what the diagnostic or treatment outcome will be must now contend with a diagnostic result and its implications. This chapter will examine parent resources for dealing with a no-problem diagnosis and its corresponding no-treatment outcome. In particular, we will examine parent resistance to the physicians diagnosis as a form of pressure. Diagnosis resistance, like the other covert behaviors we have examined, is not terribly frequent. In the Seaside data, it was observed 17% of the time (Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). But as we will see, it is nonetheless another consequential interactional resource in the negotiation of diagnostic and treatment outcomes. In this chapter, we will discuss what constitutes resistance and how it works. Diagnoses of the type typically given in the context of upper respiratory tract infection symptoms are rather transparently antibiotics (or nonantibiotics) implicative. But because the treatment recommendation has typically not yet actually been made, physicians are still subject to pressure for a particular outcome. Furthermore, even the diagnosis can be delivered as more or less nal. For instance, some diagnoses are delivered following a complete physical examination, whereas others are offered online during an otherwise in-progress physical examination activity (Heritage & Stivers, 1999; Stivers, 1998). Although most of my comments about diagnoses will be for diagnoses that follow a physical examination, the claims appear to hold for



both online and nal diagnoses; in either situation, a treatment recommendation is pending and will be forthcoming, even if what may be immediately subsequent is more physical examination.

Heath (1992) and Perkyl (1998) have shown that patients rarely respond to diagnoses at all or tend to offer only minimal acknowledgments (e.g., mm hm or uh huh). One analysis of why this would be the case is that diagnoses fall within the physicians domain of expertise and authority. As Heritage argues, Diagnostic reasoning is an activity based on special knowledge possessed and controlled by the profession of medicine (2006: 85). Because the physician and layperson are so far apart in their relative degree of insight into the diagnosis, when physicians deliver a diagnosis, they are at the point in the visit where they are most able to deploy their cultural authority to dene the nature of the patients problem (Heritage, 2006: 85). Because of this discrepancy of insight, patients are at risk of being only marginal participants in the diagnostic process. Additionally, patients may not generally respond to diagnoses because of the implications of the diagnosis for treatment. Patients exhibit a general orientation to diagnoses as preliminary to treatment recommendations (Perkyl, 1998; Robinson, 2003). Perkyl found that patients are more likely to respond to physicians diagnoses when physicians explain their diagnostic reasoning. Both Heath and Perkyl identify alternative ways of formulating the diagnosis that signicantly affect the patients degree of responsiveness, and both also suggest that patients are more likely to resist no-problem diagnostic evaluations. Across these studies, there is general agreement that patient and parent responses to diagnoses are interactionally marked. This means that patients and parents who do respond are treated by physicians as doing something special. Despite this, the diagnosis delivery still offers a further locus for the negotiation of what kind of problem the child has and whether it is treatable. Physicians communicate their stance directly through the diagnosis, and parents have that announcement as an opportunity to communicate their own stance: that their childs condition is problematic. Although they may be marginal participants in the clinical dimension of viewing signs, listening to symptoms, and triangulating the diagnosis, they are not marginal participants in terms of inuencing the diagnosis and the treatment. When physicians offer their diagnoses, the category of treatment that will be suggested is usually projectable either as prescription treatment (most commonly antibiotics) or symptomatic treatment (overwhelmingly over-the-counter medication). Thus, in this chapter I argue that parent resistance to no-problem diagnostic evaluations is a communication behaviorlike candidate diagnoses, mentioning additional symptoms, and mentioning alternative possible diagnosesthat displays parents stances toward their childrens conditions as problematic and, in many cases, as in need of treatment. One issue that underlies this is the acceptability of the diagnosis for the parent. As Helman (1978: 125) points out, No diagnosis would be acceptable to patients, it



appears, unless it was to a large extent consonant with their world view, and particularly with their interpretation of illness. What we may further see is that a diagnosis that does not correspond with a treatment that is consonant with the patients model of their childs illness is equally problematic.

A Preference to Progress to Treatment Recommendation

In considering how diagnoses are responded to, we should keep in mind that parents bring children to the physician as the expert: the one with the cultural authority to give advice. Parents do not position themselves as experts, so when the physician reaches the diagnosis phase of the visit, parents are hardly in a position to corroborate the diagnosis. These issues support an interactional structure where the physicians diagnosis receives minimal if any response. On top of these issues of cultural authority and expertise, there is the treatment dimension. One of the most common reasons parents seek help from physicians is that they feel they cannot successfully self-manage their childs illness. Thus, they are seeking a solution. This then provides yet another sort of pressure to keep response to the diagnosis at a minimum in order to facilitate progress to the treatment phase. This pressure for progressivity is, of course, very consequential for parent involvement in this phase of the visit, including making it difcult for them to participate. This is true, rst, because there is no structurally provided opportunity for parents to respond to the diagnosis: No action is conditionally relevant (Sacks, Schegloff, & Jefferson, 1974; Schegloff, 1968), and there is no other form of pressure for response. Conversely, there is pressure against responding because diagnoses are actions that can be understood as part of a larger structural organization for which treatment is the next activity (Byrne & Long, 1976; Robinson, 2003); any action that responds to a diagnosis (and particularly any action that itself makes relevant a responsive action) impedes the progressivity of the physician to the treatment recommendation and thus may be seen as a dispreferred action in this context. Empirical support for the preference for progressivity in this context is offered in two ways: First, neither physicians nor parents treat nonresponse or minimal response (e.g., acknowledgments) as impediments to progressing to the treatment recommendation. When physicians deliver diagnostic comments during the physical examination or offer ofcial diagnoses following the completion of the examination, they display no orientation to seeking acceptance or even acknowledgment from parents. That is, physicians do not wait for parents. Rather, they move directly into treatment recommendation (in the case of ofcial diagnoses) or to the next stage of the physical examination (in the case of online diagnoses). Similarly, parents typically offer no, or only minimal, acknowledgment of diagnoses (Heath, 1992; Heritage & Stivers, 1999; Perkyl, 1998). Conversely, virtually any action that initiates a sequence in this environment is treated as resistant to the diagnosis, even if the action does not appear, on its face, to be treating the diagnosis as problematic. Thus, what appears to be problematic about such actions is, in large part, that they delay the physicians progress to the treatment recommendation.



Physicians typically progress immediately from delivering the diagnosis into treatment without waiting for the parent. This can be seen in Extract 4.1, where the parent does not respond to the physicians diagnosis of a cold, bronchitis, and no ear infection (lines 12).
(4.1) 1056 (Dr. 2) 1 2 3 4 5 6 7 DOC: -> Hes got uh good co:ld, little bronchitis_ -> <no ear infection. DOC: .h So I- Since youre traveling I will wanta put him on some medi- me- byu:h: medication but_ (0.2) DAD: I kinda [gured that. DOC: [Yeah. DOC: Cause she has uh little tendency to get into trouble.

Despite the lack of parent uptake to the diagnosis, the physician moves directly from the diagnosis into the treatment recommendation (line 3). This suggests that a lack of parent uptake following the diagnosis is not treated by the physician as problematic. If a response was due, in line with other interactional research, we would expect to see accounts for the diagnosis, a backing down from it, and other reactions (Heritage, 1984b; Pomerantz, 1984; see chapter 5). A very similar case is shown in Extract 4.2. In this case, there is even a small silence following the diagnosis in line 1, but the physician continues through a description of the infection and then progresses directly into treatment (line 6).
(4.2) 2002 (Dr. 6) 1 2 3 4 5 6 DOC: .hhh Uh:m his- #-# lef:t:=h ea:r=h, is infected, -> (0.2) DOC: .h is bulging, has uh little pus in thuh ba:ck,=h DOC: Uh:m, an its re:d, DOC: .hh So he needs some antibiotics to treat tha:t,

Parents do sometimes minimally acknowledge diagnoses. We can see this in Extract 4.3.
(4.3) 1183 (Dr. 1) 1 2 3 4 5 6 DOC: Well I think whats happened is is that she ha:s this: uh- (.) .h ear infection in her left ear?, MOM: -> [Mm:. DOC: [And well put her on some medicine and shell [be ne. MOM: [Okay.

But we can still observe that the physician does not indicate that acceptance or acknowledgment following the diagnosis delivery is required or even being sought.



In fact, although the mother does receipt the physicians diagnosis of an ear infection with a minimal acknowledgment Mm:. (line 3) (Gardner, 1997), the physician was already beginning his treatment recommendation, so the two occur in overlap. Thus, the acknowledgment was not being treated as a necessary prerequisite to his progressing to this next activity. Although sometimes, in response to diagnoses, parents offer more substantive acknowledgments such as Okay, as we have seen, it is more typical for parents to offer either no acknowledgment (e.g., 4.1, 4.2) or only minimal acknowledgment (e.g., Extract 4.3) in response. This is important because it suggests that physicians do not treat parent uptake of the diagnosis as mattering for whether and how they proceed. This, at least potentially, makes it difcult for parents to be truly involved in this phase of the visit in that, unlike the other two phases we examined, here there is no sequentially provided-for opportunity for them to respond. But parents do exploit the transition space (i.e., the space between the conclusion of one TCU and the beginning of a next, not necessarily a silence) in this location to initiate actions that delay the physicians progress to the treatment recommendation. The remainder of this chapter will focus on these actions as parents primary means of diagnosis resistance and thus of diagnosis-treatment outcome negotiation in this phase of the visit.

Diagnosis Resistance
As we will see in the next chapter, what constitutes resistance is very much dependent on the sequential context and what action is due next, if any. Resistance to the diagnosis takes the form of initiating a new sequence in a context where nothing more is relevant and where progress to the treatment recommendation activity appears preferred. Resistance generally involves calling into question or disafliating with the physicians diagnostic evaluation. This can be accomplished with three different sorts of sequence-initiating actions: newsmarks (Heritage, 1984a; Jefferson, 1981), questions about symptoms, and questions about the diagnosis. In all three cases, the action initiates a new sequence in an environment where movement to the next activity is preferred. As we will see, most forms of sequence expansionnewsmarks or full questionsare built to prefer reconrmation of the physicians previously stated position. However, despite this preference, these actions are clearly designed and treated as resistant because they obstruct the progress of the course of action that is under way. The turn design of the resistant actions suggests that parents generally work to mitigate the resistant action. This maintains respect for the physicians medical authority while nonetheless conveying a position of disafliation with him or her. Newsmarks According to Heritage and Se (1992: 390), newsmarks (e.g., Really? and It is?) treat the prior talk as news and, to varying degrees, promote further informings. In these data, I argue that their work in promoting further informing is disaligning because this defers a relevant next move to the treatment recommendation (in terms



of the overall structural organization of the visit). As an initial illustration of this, we can look at Extract 4.4. The rst point at which the diagnosis is possibly complete is not responded to at all. The physician then reviews a couple of the relevant symptoms and signs, and the mother receipts the second one (that her childs throat does look uh little bit re:d in lines 34) with what Heritage and Se (1992) term a marked acknowledgment. She responds with Right. (line 5). This portion of the diagnosis is conrmatory of the parents position that the childs condition is problematic, and this is particularly carried by the physicians stress on and use of the do-auxiliary (does). (See Stivers, 2005a, for a discussion of modied repeats in an immediately subsequent position.)
(4.4) 104 1 2 3 4 5 6 7 8 9 DOC: .hh Uh:m, I think probably hes- ya know has uh little viral col:d, His nose is uh little stuffy, DOC: .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesnt look like any[t h ing] much,=^Yeah, MOM: -> [Right. [Really,] DOC: An I think thuh redness is really -again like I say from dripping down thuh ba:ck, DOC: .hh Uh:m his chest is perfectly clear. Theres nothing in his lungs at all:.

However, when the physician then suggests that this redness is not actually signicant in line 4 (but it doesnt look like anything much,), which is a no-problem diagnosis, the mother receipts this with the newsmark Really,; this constitutes diagnosis resistance. It initiates a new sequence and further discussion of the news that the child does not have anything serious: In response, the physician conrms with ^Yeah, a prosodically strong turn design, and then an expansion of her theory of the etiology of the redness that further downplays the seriousness of the boys condition. The sequence initiated by the newsmark is brief. But the physician goes on to further justify the diagnosis with supporting physical examination ndings. This is built, with the and preface (Heritage & Sorjonen, 1994), as a continuation of the previous turn and thus not as a response to the newsmark. It is nonetheless notable that justications for diagnoses are common following resistance. Sometimes the response is even more minimal than that shown in Extract 4.4. In Extract 4.5, the physician answers the question posed by the newsmark and then immediately moves on to the next unrelated physical examination component.
(4.5) 1150 (Dr. 3) [the physician asked about the ears earlier, and the mother reported the child to be pulling at them indicating that they might be causing her pain] 1 2 3 DOC: Her ears are ne. MOM: -> Are they? A[h=hhh DOC: [Yeah. (ts) open up real big say ah:::,



The physician provides an online report (Heritage & Stivers, 1999) that the childs ears are ne. In response to this, the mother offers a partial questioning repeat Are they? (line 2), which initiates a new sequence. In response, the physician offers a minimal and, in fact, quiet Yeah. (line 3) before moving on to the next segment of the childs physical examination. Although this is an online diagnostic evaluation rather than a nal summative diagnosis, it indicates that there is no problem. Moreover, insofar as this is the very symptom that prompted the visit, it is unlikely that in other examinations the physician will nd additional problems. In this example, as in the previous one, we see that a newsmark does not necessarily engender a lengthy sequence, but it still initiates a sequence in an environment where the physician is simply progressing through the physical examination activity. In still other cases, physicians respond to newsmarks by revising their just-offered diagnosis. We can see this in Extract 4.6. Here, in response to the parents implied candidate diagnosis of yellow spots in the girls throat, the physician examines the throat. In his online diagnosis, he suggests that contrary to the parents implied theory, these are primarily blisters back there. (lines 12). Like the previous example, the spots are stated as the reason for the visit, so a no-problem evaluation of this symptom is tantamount to a no-problem summative diagnosis.
(4.6) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 DOC: Yeah:. You know actually what those a:re pr=h .hh are primarily blisters back there. MOM: -> Yea:h? DOC: Its almost like shes got cold sores in thuh back of er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: [And u:sually thatll go along with this just being viral. (.) MOM: -> [Really.= DOC: [#er-# DOC: =Y:eah.

In response, the mother offers a strongly question-intoned Yea:h? that calls the online diagnosis into question and is thus a rst instance of diagnosis resistance. This engenders further talk, though here it is not obviously responsive. Rather, in this position, the physician offers a slightly redone version of his prior diagnosis, leaving it equivocal as to whether it was designedly responsive to the Yea:h? or more of the diagnosis offered earlier. Either way, the physician, in lines 78, goes on to suggest that the diagnosis is viral. The move to labeling the diagnosis as viral may be part of the physicians effort to strengthen his diagnostic position by shifting from a report of his examination ndings to a diagnosis. In response, the mother offers a second newsmark, Really. which again initiates a new sequence, this time responded to with a reconrmation, Y:eah. (line 12).



I have asserted that newsmarks are the most minimal form of resistance. This concept is quite parallel to that laid out by Schegloff, Jefferson, and Sacks (1977): In cases of other initiated repair, there is a natural ordering based on their relative strength or power on such parameters as their capacity to locate a repairable (p. 369). In line with this, Drew (1997) has shown that open class repair initiators such as Huh? or What? communicate to the recipient that there is a repairable within the prior turn but leave it to the speaker to locate the precise trouble source. Similarly, a newsmark following a diagnosis does not specify which aspect of the diagnosis is being resisted but nonetheless communicates some problem with it. Physicians do not necessarily do much with newsmarks besides conrm the diagnosis. Often a Yeah or other single TCU response closes that sequence. But physicians may subsequently continue on their prior course of action with more informing, and this is arguably engendered by the newsmark-initiated sequence. However, even with nothing further, a reconrmation sequence still both delays the physicians progress to treatment recommendation and projects the possibility of more explicit disagreement. This is in contrast with uh huh or mm hm, which pass on the opportunity to do something more and treat a course of action as in progress (Schegloff, 1982). This is how we can analyze these newsmarks as accomplishing resistance and thus as a resource for negotiating the diagnostic and treatment outcome. At times, it is more explicit, as in the physicians response to the parents newsmark in Extract 4.7. During a physical examination of a girl who presented with complaints of ear pain and coughing, the physician offers an online no-problem evaluation of the girls cough (line 5). Because it is still during the physical examination, it is arguably not a nal diagnosis, but it nonetheless is diagnostic and projects no treatment. To this, the parent offers a newsmark No? that resists the physicians evaluation.
(4.7) 1075 (Dr. 1) [presented with complaints of ear pain and coughing] 1 2 3 4 5 6 7 8 9 DOC: Let me hear your cough. Can you cough for me?, (.) DOC: Go .h #huh# GIR: #huh huh huh#=.h DOC: Shes not too ba:d_ DAD: -> No?= DOC: =really:,=h right no:w, DOC: I dont:=think she needs an antibiotic_ [uh: .hh DAD: [Okay.

In response to this resistance, the physician suggests, I dont:=think she needs an antibiotic_. The physicians introduction of treatment discussion in an environment where he had yet to provide a nal diagnosis appears to be an early move to treatment. And the move to antibiotics in particular appears to treat the mothers No? as projecting disafliation with his projected no-treatment evaluation. The physicians strong assertion in line 8 (strong in part because of its placement) may display his lack of preparedness to overtly negotiate over antibiotics (though, of course, that is precisely what appears to be going on covertly).



Another piece of evidence that supports the claim that newsmarks project disagreement and are used to negotiate the visit outcome is that following a newsmark and reconrmation of the diagnosis, parents relatively commonly escalate to more explicit forms of diagnosis resistance. This can be seen in what follows in several of the cases we have looked at in this section. For example, if we return to the case shown in 4.4, we can see that following the sequence expansion engendered by the newsmark Really, the physician goes on to further account for the boys red throat and reconcile that with her no-problem diagnosis (lines 67), asserting that the redness is from dripping down the ba:ck,. This is further followed by the report of other null ndings: a clear chest and no fever. In response, the mother offers more full-scale resistance: a statement of concern about her sons moods that seems to be in line with her stance that her son has a treatable problem (lines 1314).
(4.8) SG104 [shown earlier in 4.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 DOC: DOC: MOM: DOC: DOC: MOM: DOC: MOM: -> DOC: MOM: -> DOC: -> MOM: -> DOC: .hh Uh:m, I think probably hes- ya know has uh little viral col:d, His nose is uh little stuffy, .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesnt look like any[thing ] much,= [Right. [Really,] =^Yeah,An I think thuh redness is really -again like I say from dripping down thuh ba:ck, .hh Uh:m his chest is perfectly clear. Theres nothing in his lungs at all[:. [Okay, .hh An I- ya know (you see) thuh fevers have gone dow:n, .hh uh:m_ I was just concerned cuz hes been so cranky an I thought well [there must be something= [^Well:=botherin im [that I cant: [Well (that) could be. I mean-= =see_ What will happen.<Let=me show you here.

((37 lines of ear anatomy explanation not shown)) 57 58 59 60 61 62 63 64 65 66 DOC: [.hh [An- An- An it can: #i:#=ya know an then it can open up again.=If he were older he might say to you my earsre popping. (.) MOM: Mm hm, DOC: -> Ya know an that just tells you its more of -> uh .h eustachian tube dysfunction. (.) MOM: [Oh:. DOC: -> [An that can bother them.

67 68 69 70


MOM: Yeah.= DOC: -> =That can bother them. DOC: -> .hh But he has no ear infections.=e- His ear is[( ) an (.) looks just (:ne.)

In contrast with newsmarks, which frequently receive minimal reconrmation, after fuller scale resistance, physicians more commonly and more explicitly either defend their diagnosis or back down. In Extract 4.8, in response to the mothers second resistant move (lines 1314/16/18), the physician concedes that there may be something bothering him but then returns to her diagnosis as she suggests that this is eustachian tube dysfunction (line 63) and reasserts that he has no ear infections and His ear is . . . looks just (:ne.) (lines 6970). Here, the physicians response to fuller scale resistance is to maintain her diagnosis, though it is not done immediately. The exact way that parents escalate varies. For instance, in Extract 4.9 the parent shifts from a newsmark to questioning the diagnosis, a stronger form of diagnosis resistance that we will discuss in more detail later.
(4.9) 1150 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 DOC: Her ears are ne. MOM: -> Are they? A[h=hhh DOC: [Yeah. (ts) open up real big say ah:::, (.) GIR: A[h::, DOC: [(What=youre tryin tuh say) ah::::. DOC: .hh Ah::::#:::.#=h DOC: Yeah? DOC: .h So unfortunately everything_ (.) .h I think shes probably jus got kind of the u thing (in dow).= MOM: =Okay:, (.) MOM: -> So you dont think theres:- its in thee ear.<I just didnt want thee [ear thing. DOC: [Yeah. (0.4) DOC: -> Her ea:rs look perfectly ne, so theres no ear thing starting there. and so .hh I think shes probly got thuh same viral thing that -everybody elses had in thuh house.

In response to the initial resistance Are they? the physician continues his physical examination and then offers an ofcial diagnosis (lines 910). Although initially the mother offers a provisional-sounding acceptance of this with Okay:, (line 11), she then goes on to offer fuller scale resistance of the diagnosis with a request for conrmation that the girl does not have thee ear thing. (lines 1314). By asking



about her daughters ears, the mother maintains a position that the girl may have an ear infection. Here, the resistance is formulated as an explicit question that makes relevant a response from the physician. The physician maintains and strengthens his position that the child has a nontreatable illness despite the mothers concern that her daughter has an ear infection, but this still reects that a negotiation of the diagnosis is underway. Yet a stronger form of resistance is to assert an alternative diagnosis. We can observe this pattern in Extract 4.10 (shown earlier as Extract 4.6). Following two newsmark-initiated sequence expansions and two rounds of physician response, the mother upgrades from this minimal form of sequence expansion to a stronger form of resistance: offering an alternative diagnosis in line 22.
(4.10) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 MOM: And then uh- I looked down her throat yesterdaylast ni:ght, an I could see thuh yellow:_ DOC: ^Okay. MOM: #spo:[t so:. ((trails off)) DOC: [.hh Well open up rea::l big. lets take uh look an (say-) say #Ah:::[:::.=hh GIR: [Ah::::=hh DOC: .hh (0.5) DOC: Yeah:. You know actually what those a:re pr=h .hh are primarily blisters back there. MOM: -> Yea:h? DOC: Its almost like shes got cold sores in thuh back of er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: [And u:sually thatll go along with this just being viral. (.) MOM: -> [Really.= DOC: [#er-# DOC: =Y:eah. DOC: .hh MOM: -> One v thuh teachers told me it might be stre:p so:[:_ DOC: [.mlk Yeah we are starting to see some strep so Im gonna culture just in case .hh shes got both going on at the same ti:me but- .hh when you see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it[s ya know like shes got blisters n MOM: [Oh yeah:_ MOM: Oh:::.



The mothers alternative diagnosis of strep throat both claims visit legitimacy and underscores her stance that her daughter needs treatment. More locally, this puts the physician in a position where he must deal directly with this diagnostic possibility. In this section, we have examined several instances of when diagnostic information is responded to with a newsmark. I have argued that these newsmarks do two primary things: First, they initiate a new sequence by making relevant at least conrmation or disconrmation of the diagnosis. Second, they disalign with the activity in progress and in doing so project disafliation with the diagnosis and/or the treatment implications of that diagnosis. The claim is that both through the disruption of the physicians progressivity to the next activity and through the projection of possible disagreement, the use of newsmarks following these diagnoses constitutes resistance. This was supported both in the way that physicians buttress their diagnoses and even their treatment recommendations in the face of newsmarks and also in the way that, commonly, full-scale resistance follows newsmark resistance sequences where no physician backing down is visible. Questioning an Examination Finding The second primary method of initiating a sequence and delaying a physicians progress toward the next activity following diagnosis is to question one of the examination ndings. Questioning the physician in this way is a stronger way of resisting than the use of a newsmark. Whereas the newsmark merely seeks reconrmation of what the physician has just stated, questioning an examination nding identies a problem area explicitly. Moveover, it is an area that is specically in the physicians domain of expertise and thus more strongly projects disagreement with the physician. An initial example of this type of resistance was shown in Extract 4.9. Another example is shown in Extract 4.11, where the physician makes a general statement that there has been a lot of cold and stuff going arou:nd, implying that this is what this boy has. She gives her diagnostic conclusion that the boy does not have a problem: aside from #thuh:# .hh thuh lotta mucus an stuff that he #ha:s,# he- n- he sounds ::ne, (lines 35). In response to this no-problem diagnosis, the mother inquires about the results of the just preceding lung examinationwhether he has chest congestion (line 6).
(4.11) 2081 (Dr. 8) [just following lung examination; parent presented BOY as having cold symptoms and a bad cough] 1 2 3 4 5 6 7 8 9 DOC: BOY: DOC: Theres been uh lotta cold and stuff [going arou:nd,=h [( ). but- n- .h ya know:, aside from #thuh:# .hh thuh lotta mucus an stuff that he #ha:s,# he- n- he sounds ::ne, <UhmMOM: -> His chest isnt congested. Is his [cheDOC: [Yeah::, h=its just all of his:: up here: actually not really lin:- down in his lu:ngs,



Questioning the examination nding is like responding with a newsmark in that both initiate new sequences and thus extend the diagnosis activity by making relevant a response. In this case, what we see is that in response to the parents question, the physician conrms that there is no chest congestion and then goes on to more fully defend her evaluation that there is no problem. We previously observed with newsmarks that they normally are designed to prefer conrmation that there is no problem. This is true here as well. In both cases, the physician is asked to reassert that there is no problem. This is a similar phenomenon to that discussed in the previous chapter: If there were a problem, the parent orients to it as something that would have been mentioned. In this case, the original question is designed to prefer a no-problem answer, but the mother immediately attempts to revise the preference design of the question (the shift from a negative declarative to an afrmative interrogative design). And this may reect her own difculty in requesting reconrmation that the symptom she thinks is problematic is not actually problematic. Arguably, then, the resistance in these questions is not so much the design of the questions as their position: that they are asked here at all. Because the preference within this activity is for parents not to intervene, the sheer presence of this kind of question, which seeks evidence and/or justication for the physicians diagnosis, is disaligning with the in-progress course of action. There is support for the idea that questions about examination ndings are treated as a stronger form of resistance than newsmarks. Following questions about symptoms more commonly than following newsmarks, physicians go on to provide additional justication for their diagnoses. In Extract 4.11, when the physician continues, she positively asserts that it is up here: actually, thereby rejecting the mothers idea of chest congestion with not really l- in:- down in his lu:ngs, (lines 89). Here, the physician responds to the parents resistance by ruling out an alternative problematic conditionlung congestionwhich would probably require medical treatment. When physicians offer accounts for their diagnoses, this reects their understanding of a parents inquiry as in some way doubting or problematizing the diagnosis: a domain of knowledge normally treated as resting solely with the physician. Thus, the claim that parents are treated by physicians as lobbying is not evidenced solely by cases where physicians back down from their diagnosis. Rather, cases where physicians defend their position are equally good evidence that they are treating the interaction as involving a negotiation. One way that progressivity to the next activity is visibly delayed is that physicians sometimes back up into another phase of the visit and redo their examination and/or diagnosis. In Extract 4.12, the physician can be observed to offer what is normally done as an online comment during the physical examination (Her ea:rs look perfectly ne, in line 17) and then moving to the diagnosis for the second time (notice that the rst time is in lines 910). And there is a built-in account for this diagnosis in that it is tied to everybody else in the house.
(4.12) 1150 (Dr. 3) [shown earlier in 4.9] 1 2 DOC: Her ears are ne. MOM: -> Are they? A[h=hhh

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20



[Yeah. (ts) open up real big say ah:::, (.) GIR: A[h::, DOC: [(What=youre tryin tuh say) ah::::. DOC: .hh Ah::::#:::.#=h DOC: Yeah? DOC: .h So unfortunately everything_ (.) .h I think shes probably jus got kind of the u thing (in dow).= MOM: =Okay:, (.) MOM: -> So you dont think theres:- its in thee ear.<I just didnt want thee [ear thing. DOC: [Yeah. (0.4) DOC: -> Her ea:rs look perfectly ne, so theres no ear thing starting there. and so .hh I think shes probly got thuh same viral thing that -everybody elses had in thuh house.

The mothers question you dont think theres:- its in thee ear. in line 13 is already, as discussed earlier, an upgrade on a prior resistant move in line 2. With the question, the mother directly queries the physicians examination of the girls ear. Although it is designed to request conrmation that there is no problem, by its sheer presence as an inquiry, it invites a revision of his prior statement. Following her resistant inquiry, the mother offers an account for that question with <I just didnt want thee ear thing. which may work to mitigate the challenge just below the surface of her prior question by suggesting that it is primarily motivated by her own concern, and not rooted in doubt about the physician or his competence. In response to the mothers resistance, the physician, similar to the previous example, buttresses his position in several ways. First, he upgrades his assessment of her ears from ne to perfectly ne, which treats the mothers resistance as specically calling the original evaluation into question. Second, by making the diagnostic implication of this evaluation explicit with his so-prefaced statement (line 18), he further strengthens his position that the girl does not have a treatable problem (Raymond, 2004). Third, by reusing the mothers phrase of ear thing to articulate the implication here, he more clearly rejects her implied diagnosis. Fourth, by restating his diagnostic evaluation, he moves out of a responsive, and in this case defensive, position where he is denying the existence of problems into a position where he can afrmatively state the girls problem and return the consultation to its activity trajectory of diagnosis delivery, followed by treatment recommendation (Byrne & Long, 1976; Robinson, 2003). Moreover, the shift from u to viral and the addition of same . . . thing that -everybody elses had in thuh house. do additional work to support his diagnosis. Viral is a more technical medical diagnosis that may be more difcult for the parent to challenge. Additionally, by invoking an earlier comment that others in the house-



hold have had viral infections, the physician also may inhibit further challenges by the mother. Next, by suggesting that the illness is thuh same viral thing that -everybody elses had (lines 1920), the physician invokes what is known in common with the motherthe epidemiology of the illness of other family membersand associates it with the condition of this child. Finally, whereas u projects no treatment by virtue of its being a viral illness, viral removes this intervening step, thus making a stronger claim that the illness is not treatable. In responding to this mothers resistant move, the physician thus deploys a range of resources to defend his own position that the child has a no-problem condition. But note that these are all defenses and treat the mothers inquiry as problematic, and thus as pressure. Like newsmarks, questions of physical examination ndings that are not met with a backing down by the physician also are at times pursued. We can see this in Extract 4.13. Here the physician offers an online comment that is both designedly responsive to the mothers concern of an ear infection and tantamount to offering a diagnostic evaluation. This diagnostic comment is shown in line 1. In line 2, the mother resists by calling the online nding into question through a repeat of it.
(4.13) 1083 (Dr. 1) [presented with a concern of ear infection] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 DOC: We^:ll, his ears are oka:y.= MOM: -> =His ears are o[kay. DOC: [Theyre not in[fected. MOM: [SHOW IM where you [think it hurts. DOC: [Wh- where does it hurt in your ear.=h (.) BOY: Ear. DOC: That one? (0.5) DOC: Okay. DOC: .h[h MOM: [He told em at school it was #hurting.# (0.4) DOC: .h #I mean its not even like=h [maybe uh= MOM: [Nothing. huh? DOC: =^little bit infected, DOC: .h Does it hurt when I go like tha:t?,=h BOY: Mm- yeah:, DOC: Oh does it, h DOC: Lemme look. (2.0) MOM: You cant see anything?, DOC: .hh hh .hh Mkay:, well look at the rest.

In response, the physician restates his diagnosis as an upgrade: He species his previous okay as an explicit evaluation that the ears are not infected (line 3), which



more explicitly rules out a treatable condition. Perkyl (1998) has shown that patients are less likely to resist diagnostic evidence to which they have no access. The mother in Extract 4.13 will have difculty resisting the physicians claim that the ears are not infected because it is a type of claim to which she has no epistemic access. In response, the mother asks her son to restate the location of his symptoms (lines 45). This further resists the diagnosis because it now solicits consideration, if not actual reinvestigation, of the childs symptoms. And it suggests that if there is no infection, the physician must reconcile the symptoms her son is experiencing with this lack of cause. Subsequently, the physician investigates the boys pain, rst by asking him about it (lines 6 and 9). When the mother resists still further with additional outside evidence that her son reported the symptom at school, the physician restates his diagnosis, this time more strongly. The strength here is done both with prosody and with the negation of any small possibility of infection: not even like=h maybe uh ^little bit infected, (lines 15/17). But before completion, the mother is resisting further with Nothing. huh? (line 16). It is notable that, similar to previous cases, while the form of this request for conrmation prefers a no, it is launched in the middle of an assessment that offers exactly this. Again, then, by the sheer presence of this inquiry, the mother pressures the physician to revise his assessment. And responsively, the physician investigates the childs symptoms: He both inquires about them (line 18) and reexamines the ear (line 2223). Finally, following still further inquiry from the parent (line 23), the physician does not specically respond to her but rather addresses the issue by announcing a move to continue a thorough physical examination (line 24). In this section, we have examined parent resistance that takes the form of a query of a childs symptom or a physicians examination nding. I have argued that this practice is stronger than a newsmark in terms of resisting the diagnosis. In each case, the physicians response to the question has not been a simple conrmation but has instead ruled out an alternative more problematic condition (4.11), restated the diagnosis (4.12), or further investigated the childs complaints (4.13). And these types of responses are very common, unlike the more minimal responses to newsmarks. In all cases, physicians either buttress and defend their position or open themselves up to revising their position (by moving toward a diagnostic concession or by restarting the examination). This suggests that both partiesphysicians as well as parentsare negotiating the diagnosis and treatment outcome of the childs visit. Calling into Question the Physicians Diagnostic Inference The nal type of diagnosis resistance to be discussed here involves the parent calling into question the physicians diagnostic evaluation. Of the three types of diagnosis resistance, this is the strongest because it questions a medical evaluation of the childs symptoms and signs: a domain over which the physician is normally treated as having sole responsibility and epistemic ownership. As an initial example of this practice, we can look at Extract 4.14. Here, the parents resistant action requests conrmation that the condition the physician characterized as involving a whee:ze is not bronchitis . . . (or anything,) (line 6). Although



designed to prefer disconrmation and offered cautiously (with the addition of or anything, at the end of the turn), the mother nonetheless questions the physicians diagnosis as involving a wheezing condition. The physicians announcement that shes- just got uh little bit of uh whee:ze_ may initially sound incomplete, but it is, from the parents perspective, enough to display a clear diagnostic alternative to her own concern of bronchitis.
(4.14) P202 (Dr. 8) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 DOC: I think youre all do:ne_ I can try giving you some medicine for that- (.) - uh:m=h tlk for thuh cou::gh, Is- shes- just got uh little bit of uh whee:ze_ an: sometimes- uhm .h[h MOM: -> [(Its) not: bronchitis though (or anything,) DOC: No::.=h Uhm_ .h Bu:t- after they have uh cold sometimes they can have this kind of uh wheezy cou:gh, DOC: -> Uhm, .hh (.) an:d I think probably:=your husband smoking inside the house is contributing to it so: he should really smoke outside, It makes:[their: uhm=h lungs more=h sensiti:ve=huh yeah. MOM: [I know:. ( big issue )

In her response to this resistant action, the physician rst conrms the negatively formulated question about the candidate diagnosis with No::. (line 7). Second, she reasserts her observation that the cough is wheezy, but this time it is embedded in a larger diagnosis of a cold (lines 79). By adding that smoking is a contributing factor (lines 1013), the physician may be understood to be not only defending her diagnosis but also addressing the mothers difculty in understanding a wheezing coughoften associated with allergies or irritants. Thus, the physician provides some insight into what might underlie the wheezing conditionthe fathers smoking. This analysis is in lines 1213, where the physician accounts for her explicit linking of smoking and wheezing. In this case, though, the basic practice is the same as we have seen: a sequence-initiating question is responded to rst by a disconrmation and then a justication of the diagnosis, which further displays the physicians analysis that her judgment had been called into question. Thus, as with the other forms of sequence expansion, we have evidence that questioning the diagnosis is heard as resistant and that it is working to negotiate the diagnostic outcome of the visit. The precise way that parents question the physicians diagnosis can vary. Whereas in Extract 4.14 the parent offered an alternative possible diagnosis as a question, in Extract 4.15, the parent offers an account for why she thought the illness was problematic, which adopts a stance that is at odds with the one the physician has adopted through his diagnosis that she has an acute gastroenteritis:, (line 1) that Doesnt look like its too signicant, (line 4). In terms of the mothers stance that



her childs condition is problematic, this is a blow to the legitimacy of her visit and certainly treats the illness as no problem. In response, the mother rst accepts the diagnosis with Okay: (line 6); however, this is prosodically built as preliminary to something more, and she goes on to state her concern with the amount of vomiting her daughter had the previous evening (lines 6/811). In the way it is framed as a concern, the mother is hearably defensive. Also, this symptom was earlier stated as the primary reason for the visit, so her raising it here is for the second time. Therefore, particularly with its placement just following the diagnosis, she hearably calls into question the physicians diagnosis of the condition. Additionally, this construction builds her daughters illness as a signicant one whether to justify her childs condition, ensure adequate medical investigation of her child, or present evidence for the treatability of that condition.
(4.15) 1059 (Dr. 2) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DOC: GIR: DOC: GIR: MOM: -> DOC: MOM: -> -> MOM: -> DOC: MOM: -> -> DOC: MOM: -> DOC: MOM: -> DOC: -> -> MOM: DOC: -> -> MOM: DOC: -> MOM: DOC: MOM: DOC: -> So it looks like she has an a[cute gastroenteritis:, [( ) ((about lollipop)) Doesnt look like its too sig[nicant, [( ) Okay: th- .hh th- I was con[cerned cause uh[(And expect er-). Usually: if- she:- when she=if she vo:mits: she(0.8) it- it doesnt last as long or as o:ften. (.) La[st night (was -) [(ts-) jus- jus uh few throw ups an thats thee end of it. #Yeah [thee-# [Yea:h_ Like I sa[y she was probably at least=h_ [(It) was uh little uIt was at least ten [ti:mes. [(s- s-) [more like: ten to twe:lve (yeah.) [Sh #uh:#=yeah. well she had very signicant=uh: (5.5) Signicant throw-up, but i:t=uh: [(Yeah [ .) [.hh [there wasnt uh whole lot of bile in it so shes not obstru- I dont (wanta th=[say) [No. .h[ likely be ob[structed #it_# [No. [No it was just once that [I saw [that. [.mlh [Or: uh. (2.8) But=uh:=h (0.5) Okay: every- An everything checks out :ne,



In response, the physician upgrades the mothers evaluation of the vomiting with very signicant (line 20). In this way, he works to reconcile his diagnosis of acute gastroenteritis:, his assessment of not too signicant, and the mothers narrative about her daughters illness experience. He then accounts for his evaluation by articulating part of the rationale that had previously been unavailable to the parent, though surely utilized by the physician: there wasnt uh whole lot of bile in it (line 23), thus that there is no bowel obstruction. Finally, following the mothers agreement that there was not in fact much bile, the physician reinvokes his earlier diagnosis by restating his examination ndings: An everything checks out :ne, (line 31). That the physician is reinvoking this here is also marked with the self-initiation of repair that appears to be primarily in order to insert the And preface that ties this back to the diagnosis he was providing prior to the mothers resistance. In contrast with the prior two examples, here the physician not only maintains his diagnosis but also responds to the parents resistance by ruling out an alternative, medically treatable, and much more problematic condition. Although the childs diagnosed condition might be treatable by, for example, intravenous uids for dehydration, here the physician orients to the girl as not in need of medical treatment. Another way that parents question the diagnosis is related to that shown in 4.15: Parents can offer up symptoms that are not easily reconciled with the diagnosis offered by the physician. This is shown in Extract 4.16. In this example, the girls visit is due to a rash. As the physician is examining her, she asks the question shown in line 1. In line 6, she offers an explanation for the conditionan informal evaluation but still potentially understandable as a diagnosis that is coordinated with a shift from examining the girl to writing information in the chart.
(4.16) 517 1 2 3 4 5 6 7 8 DOC: DAD: DOC: DOC: DAD: Any new soaps, new detergents, Uh::, (1.0) I don know. [Could be new soap. [((DOC moves from GIRL to chart)) Mm:. (.) That could be thuh rea[son. ((looks at DAD)) [That could be it?, (2.0) ((DOC nods))

((31 lines not shown talking about what soaps they have; physician writes in patients chart)) 40 41 42 43 44 45 46 47 48 DAD: GIR: DAD: GIR: DAD: DOC: DAD: DOC: Maybe no- no soap for uh week I guess?, Because= =hh= =[ma=[hh Use [Dove. [.hh hh .hh Use Dove?, Yeah. (.)

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63



[Thuh whi[te Dove. [(wah??) [O:kay. (2.0) DAD: -> But how come its not showing up on her: hand. DOC: -> Thats kinda wei:rd. ((moves from chart to GIRL)) DAD: Yeah_ (.) DOC: .hh= DAD: =I mean [cuz she (did) DOC: [I- Thats thuh rst thing I would do is [tuhDAD: [Okay. (0.2) DOC: (.hh) (Turn this way lemme) peek in thuh back.

Following an extensive detailing of the use of soaps in the family household, the father jokingly proposes a solution of no soap for uh week I guess?, (line 40). In response, the physician offers a serious recommendation Use Dove. (line 45). After acceptance of this recommendation (line 51), the father resists this diagnostic explanation for his daughters condition with a query about why, if it were a soap allergy, the rash would not be on his daughters hands as well, where she almost certainly was exposed to the soap. We have seen that in response to diagnostic resistance, physicians consistently respond by taking up the resistance and either defending or backing down. In this case, we see this again. In response to the fathers resistant query, the physician both acknowledges the apparent contradiction and moves back into a physical examination of the girl across line 54. After more examination, she recasts her earlier unmitigated recommendation (to replace the new soap with Dove soap) as thuh rst thing I would do. This downgrades her earlier recommendation from one that was nal to one that is a rst step. Thus, although she does not completely revise her diagnosis, she does still back down from it. In addition to negotiating the diagnosis, physicians also routinely treat diagnosis resistance as related to the treatment outcome. Although we will discuss the treatment implications in the next section, here I want to draw the connection between the two. We can see this in Extract 4.17, where, following no indication of a problem, the mother initiates a sequence with a request for reconrmation of the ndings (line 1). As is typical, this inquiry is designed to prefer a no, and this pattern is again present in her follow-up newsmark No? and then her question of the diagnosis itself: not even earache or. (line 7).
(4.17) 302509 1 2 3 MOM: MOM: He- he [doesnt have any[(thing)? [((head shake)) [((Doc washing hands))



4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19


[No. [((head shake)) No? Not- not even earache or. No. Ears are ne, the throats :ne? (.) Her che:st sounds okay, Okay, Its just maybe a viral (.) uhm_ Yeah cause hes been coughing a ^lot, a lo:t, and then=[I start(ed) giving him Robitussin,= [Mm hm, =but I want(ed) to make sure he doesnt have throa:t or ear infection,= =No. When did you start- giving the Robitussin.

In response to the inquiry about the diagnosis, the physician disconrms the alternative diagnosis and goes back over the boys physical examination ndings (lines 9 and 11). These ndings are accepted (line 12), but following a diagnosis (line 13) that suggests possible infection (though viral and thus untreatable), the diagnosis is resisted once again, rst by offering a problematic symptom (line 14) and then with an account for the mothers inquiries (lines 1718) and stating that she was concerned about a throat or ear infection. Like the other turns, this one is disconrmed (line 19), but here we see that the physician moves to what treatment was successful, thereby displaying her understanding that treatment may be one of the parents underlying issues and thus what the resistance is motivated by. This example also offers further evidence that parents treat types of resistance as ordered in strength, such that newsmarks often precede inquiries about physical examination ndings, which, more often than not, precede inquiries about diagnoses. In this section, we have examined a third type of sequence expansion for resisting the physicians diagnosis: calling the diagnosis into question. Like questioning examination ndings, this form of sequence expansion also typically engenders fullform responses to resistance, including a restatement of the diagnosis (4.14, 4.17), ruling out a more serious condition (4.15), accounting for the diagnosis (4.17), and further investigation (4.16). In all cases, we have observed that these practices are treated as resisting the diagnosis and as working to negotiate the diagnostic outcome of the visit.

Implications of Diagnosis Resistance for Treatability

Parents resist physician no-problem diagnoses for a range of reasons. Generalizing from the cases we have, there appear to be three different but closely related motivations: First, parents may see a no-problem diagnostic evaluation as threatening the legitimacy of their visit. This invalidation is potentially embarrassing for a parent



because an unwarranted visit is possibly viewed as resulting from poor judgment as to the severity of the childs illness. Alternatively, parents may resist such an evaluation because they are concerned that the physician may have failed to notice a more serious diagnosis and believe that their child truly is sicker than the physicians diagnosis reects. A third motivation is that parents may feel that their child needs treatment and a no-problem diagnosis threatens the prospect of this outcome. A concern with treatability may in turn be motivated by a desire for a restful night, a need to return to work, a belief that the child is seriously ill, or a direct desire for antibiotics (which may itself be motivated by one of the previous issues). Sometimes a parents resistance may not clearly index just one of these three underlying motivations, and of course, these issues are often intertwined. I will argue that although each of the motivations outlined here is at times evident in a parents talk, physicians routinely understand parent resistance as indexing a desire for treatment for their child. In what follows, I show qualitative evidence for this, but quantitative evidence also exists. For instance, in the Seaside data, if there is no parent resistance, physicians report perceiving parents as expecting antibiotics only 7% of the time, whereas when parents have resisted a diagnosis, this jumps up to 20% of the time ( p<.05) (Stivers et al., 2003). And in a multivariate model, if parents resisted a viral diagnosis, the physician was 2.73 times more likely to report perceiving them as expecting antibiotics ( p<.001) (Stivers et al., 2003). But as discussed in chapter 1, it is possible for a physician to misinterpret a behavior as indexing a desire for treatment when it actually represents a concern about the legitimacy of the visit or a more serious condition. There is some support for this in the quantitative data as well, because there was no association between parents resisting a diagnosis and having reported an expectation for antibiotics. Although parents may have a variety of concerns and motivations that inform whether and how they resist, physicians typically treat diagnosis resistance as a mechanism for treatment negotiation. This is illustrated in the case shown previously in Extract 4.9. The way the physician formulates the diagnosis following parent resistance reects his orientation to the treatment implications of the diagnosis without saying anything about the treatment recommendation.
(4.18) 1150 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 DOC: MOM: DOC: Her ears are ne. Are they? A[h=hhh [Yeah. (ts) open up real big say ah:::, (.) GIR: A[h::, DOC: [(:What=youre tryin tuh say) ah::::. DOC: .hh Ah::::#:::.#=h DOC: Yeah? DOC: -> .h So unfortunately everything_ (.) .h I think shes -> probably jus got kind of the u thing (in dow).= MOM: =Okay:, (.)



13 14 15 16 17 18 19 20


So you dont think theres:- its in thee ear.<I just didnt want thee [ear thing. [Yeah. (0.4) Her ea:rs look perfectly ne, so theres no ear thing starting there. and so .hh I think shes probly got thuh same viral thing that -everybody elses had in thuh house.

Specically, in lines 910 the physician prefaces his diagnosis with unfortunately. This treats the girls illness of just . . . kind of the u thing as bad. But the only negative thing about this illness is the lack of treatment for it, so it is to this that the physicians turn design appears to be oriented. Not only do physicians treat diagnosis resistance as treatment implicative but also they treat parents as seeking a bacterial illness (and the corresponding antibiotics treatment). Evidence is shown in Extract 4.19. The physicians diagnosis was offered a bit earlier as I think probably hes- ya know has uh little viral col:d, (Extract 4.8; lines 12). The parent then resisted the diagnosis with a newsmark Really, and subsequently resisted the diagnosis again with I was just concerned cuz hes been so cranky an I thought well there must be something botherin im (Extract 4.8; lines 1314). At line 69, the physician is returning to her diagnosis. She has still not offered a treatment recommendation.
(4.19) 104 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 DOC: ???: MOM: .hh But he has no ear infections.=e- His ear is[( ) an (.) looks just (f[i:ne.??) [#hu:h .h huh# [Yeah I just thought- I thought- Cuz thuh fever (was gone) I said ^well .hh I dont need tuh take him (n) then he was so cranky this [morning (I thought we[ll: ya know)= [Yeah:. [^Well (see) =[ I better take him because_ ] =[(you listen)_Things can change.^] If all of uh sudden his fever(s) start going up agai:n? .hh He starts developing other symptoms, please let us know:. (.) [(Well yeah:.) [(But)/(What) I think what its doing now is resolving. .hh An- #e-# tincture of ti:me. (0.4) is basically usually what you need with uh co:ld. -> .h An viruses can start thuh same way that bacterial -> infections can.=They can be just as sick?



90 91 92 93



.hh They can have high fevers? (.) You know all that sort of thing. .hh What is important are lots of uids,

Just as the physician is returning to her no-problem diagnosis, the parent offers justication for bringing her son in for the visit (lines 7275/76). Then, at line 84, the physician suggests a prognosis for the illnessthat it is resolving and will continue to do so. With this, she offers a nonantibiotic treatment recommendation (tincture of ti:me.), which clearly suggests a self-limiting and self-healing process. But before offering a positive formulation of her treatment recommendation, she asserts that viruses can start thuh same way that bacterial infections can.=They can be just as sick? (lines 8889). Here, the physician does several things. First, with viruses she indexes the boys current illness, which has earlier been diagnosed as a viral col:d. Next, by comparing viruses with bacterial infections, she implies that this is the relevant contrast and treats the parent as having been concerned about a bacterial infection by suggesting that the way the boys illness started is the way a bacterial illness would; she states that they can start thuh same way. This formulation also supports the mothers position that her son is quite sick through her mention of They can be just as sick? (line 89). Importantly, the physician raises bacterial infections here in an environment where the parent had resisted a no-problem diagnosis. In this way, the physician conveys her understanding that the resistance was in search of a bacterial diagnosis. Additionally, the physician works to address the doctorability dimension of the resistance by suggesting that a viral illness may be equally problematic in experiential terms because patients can be equally sick. With the mothers reraising some justication of her reason for visiting, the two issues converge. Similarly, returning to the case shown in Extracts 4.6, we can see a second case where the viral-bacterial distinction is raised in the face of diagnosis resistance. Again, this offers evidence that physicians understand this to be being negotiated with parent resistance.
(4.20) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 MOM: And then uh- I looked down her throat yesterdaylast ni:ght, an I could see thuh yellow:_ DOC: ^Okay. MOM: #spo:[t so:. ((trails off)) DOC: [.hh Well open up rea::l big. lets take uh look an (say-) say #Ah:::[:::.=hh GIR: [Ah::::=hh DOC: .hh (0.5) DOC: Yeah:. You know actually what those a:re pr=h .hh are primarily blisters back there. MOM: -> Yea:h? DOC: Its almost like shes got cold sores in thuh



13 14 15 16

back of er throa:t. MOM: (Oh:[::.)/(Aw:::.) DOC: -> [And u:sually thatll go along with this just -> being viral.

In the face of diagnosis resistance to the physicians diagnosis (line 11), the physician suggests as part of his response that the symptoms are part of a viral condition. Specically, the physician orients to the parent as having implied that the condition was bacterial with the rejection component of the diagnosis in lines 1516. When the physician says thatll go along with this just being viral., he implicitly contrasts viral with another category that would be more signicant. This is done primarily through the use of just, which treats the given diagnosis as less signicant and possibly less treatable. Like other parent behaviors that are understandable as pressuring the physician for an antibiotic prescription, diagnosis resistance can affect treatment decisions and thus visit outcomes. Extract 4.21 shows evidence of a treatment recommendation being changed in the face of diagnosis resistance. In this case, the physician offers her diagnosis in lines 13. As part of this diagnosis, the physician offers a mitigated nonproblematic nding (uh little bit of uid on thuh right si:de, [line 2]). In response, the parent agrees with the physician about the uid (line 4). Before the completion of the physicians next TCU, the parent goes on to assert that the right ear is a recurrent problem (lines 45). Although the physician appeared to be offering her report of uh little bit of uid as part of an overall no-problem evaluation, the parents agreement with it treats it as a problematic nding with which to agree. In addition, by acknowledging the physicians turn unit by unit, she may be conveying incipient speakership (Jefferson, 1983) and, in this position, an indication of forthcoming resistance. (See Clayman & Heritage, 2002, for a related phenomenon in news interviewee responses to interviewer questions.)
(4.21) 2057 (Dr. 8) [Mom mentioned a concern of ear infection earlier] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DOC: .h Yeah. her ears arent infected. <She has uh little bit of uid on thuh right si:de, n I thi[nk thats probably whe[re_ MOM: -> [Yeah. [(its) -> (just) thuh one that always (dr ). DOC: Yea:h, DOC: .hh Uh:m,=h .mlkh So its been going really nonstop for two weeks, (.) DOC: You think,=[h MOM: [Uhm, well it- on and off: (as I say:.) (I mean she would like uh little bit.) DOC: Uh huh:, h DOC: -> .h But when you do:nt it- [seems to relapse? MOM: [( medicate her

16 17 18 19 20 21


an (.) Yeah. ) DOC: -> #An it comes back, okay.# DOC: .h Yeah:. I think she may have got uh #s:inus infection,=h DOC: Uh:m_ (2.0) When was the last time she took antibiotics:. ((11 lines of discussion of types of antibiotics not shown))

33 34 35


.hh Okay. Im gonna put er on Amoxicillin. which shes been on before probably::, Okay,

On the surface, lines 45 do not appear to be resistant. But they take up a portion of the physicians turn that, although problematic, was being designed to support a no-problem diagnosis. Moreover, it was a portion of the turn not designed for response. In this way, the mother conrms a problem that the physician had not designed her turn to identify as such. It is in this rather indirect way that the mothers turn accomplishes opposition to the physicians diagnostic evaluation. And the physician treats her as opposing a no-problem diagnosis insofar as she shifts from listing nonproblems and minor problems to querying the parent further. Following the mothers resistance, the physician agrees with Yeah. and then begins to summarize previous talk in the form of a request for conrmation with So its been going really nonstop for two weeks, (lines 78) and further pursuing response with the You think, (line 10). In this case, the sequence of questions and answers across lines 717 eventuates in a change in her diagnosis from what, across the physical examination, appeared to be headed for a no-problem evaluation to I think she may have got uh #s:inus infection,=h (lines 1819). This change in diagnosis implies a corresponding change in treatment because, although treatment had not yet been recommended, the treatment projected by the previous diagnostic trajectory was an over-the-counter remedy. By contrast, the treatment for sinusitis is usually antibiotics (line 33). In this case, the treatment is clearly negotiated, most obviously because the physician changes her diagnosis and corresponding treatment in the face of resistance. Her questions in lines 78 and 14 appear designed to foreshadow a treatment recommendation that would be in line with prescription treatment and a bacterial diagnosis rather than in line with the no-problem ndings she has reported throughout her examination. Finally, in support of the claim that diagnosis resistance is a type of pressure for antibiotic treatment in particular, we can look to Extract 4.22. In this case, the parents diagnosis resistance is an overt lobbying for antibiotics. The physician offers his diagnosis in lines 3/5. The mother receipts that with overt pressure for antibiotics (which will be the focus of chapter 6). She suggests that the diagnosis is in line with previous physicians who suggested the only way he got rid of his cold is tuh keep him on antibiotics for like twenty- uh: for: twenty one d^ays. (lines 1213).



(4.22) 1016 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 DOC: BOY: DOC: MOM: DOC: MOM: DOC: MOM: BOY: MOM: BOY: MOM: DOC: So:- let me hear your cough. Cough. #huh huh huh huh# So now- i- he has [this[He has uh cou:gh, co:ld, (Here-) He has uh cough, he has uh col:d, .h [Kay hold thi[s up. ((to BOY)) [.hh [An:d ya know i- i- thuh doct[ors in= [Ohhhha= =[in Sand City f- gured the only way] he got= =[ a a h h h = f h h h h ] =rid of his cold is tuh keep him on antibiotics for like twenty- uh: for: twenty one d^ays. Well ^sometimes thats an answer, only because_ thuh reason for that i:s is that- sometimes they have uh sinus infection.

Here, unlike previous examples, the mother takes an overt (rather than covert) position in favor of antibiotics, but it is still an instance of diagnosis resistancejust a very strong form of it. In response, the physician suggests that such a treatment would require a corresponding bacterial diagnosisa sinus infection. The physicians diagnosis had been a cold (line 5). But notice that the doctor is amenable in his response to the resistance: Sometimes thats an answer,. This example demonstrates that parents are oriented to diagnoses for their treatment implications. At least sometimes, parents exploit this opportunity to either indirectly pressure physicians for antibiotics by resisting a diagnosis that does not correspond to such treatment or, though less common, directly push for antibiotics in this location, where they initiate rather than respond to the treatment recommendation. And here it is clear that the parent is concerned with receiving antibiotics as a solution to her childs problem. This section has demonstrated that regardless of the variety of motivations that may underlie it, parent resistance to the diagnosis is consistently understood by physicians as a way of lobbying for antibiotic treatment. Primary evidence was that in response to diagnosis resistance, physicians typically either succumb to the pressure or ght it. Either way, they enter into a negotiation of it with parents.

This chapter has argued that once a physician offers a diagnosiswhether online or ofcialparents are no longer in a position of encouraging physicians to explore a particular diagnostic and/or treatment trajectory. Rather, if they are to effect change in the treatment trajectory and visit outcome, they must contend with the diagnosis presented by the physician. The primary mechanism through which parents negoti-



ate treatment outcomes in this sequential context is through querying the physician. I argued that questioning works because the action of diagnosis delivery does not make response conditionally relevant. Doctors and parents share an orientation to the diagnosis as within the sole knowledge domain of the physician. Both of these issues make parent uptake of the diagnosis dispreferred, and the structural organization of the visit and the subsequent activities also make actions that delay visit progress dispreferred. So sequence-initiating actions were shown to be a primary resource for parents to resist no-problem (and thus no-treatment) diagnoses. In particular, this chapter outlined three primary types of sequence initiations that parents use, which increase in strength: newsmarks, inquiries about physical examination ndings, and inquiries about diagnoses. Each of these makes relevant sequence closure and thus, at the very least, minimally delays the physicians progress in the visit. Evidence supported an analysis of these inquiries as resistant: Parents frequently escalate from quite minimal requests for conrmation to quite explicit challenges of the diagnosis if they encounter inadequate physician response to their initial resistance (e.g., failures to return to earlier examination ndings or to a reconsideration of the diagnosis). Moreover, physicians routinely respond defensively to inquiries by justifying their diagnoses, accounting for them, and/or backtracking to history taking or physical examination ndings. Diagnosis resistance was shown to be connected to the issues raised in chapter 1: legitimacy and treatability. At times, either or both of these issues may drive diagnosis resistance. On the one hand, as in the last case, parents can resist a diagnosis apparently in the service of obtaining antibiotic treatment. On the other hand, they may also resist for other reasons. Despite this, resistance appears to be treated by physicians as primarily about treatability and thus, even in cases where it was not intended to be pressure for antibiotic treatment, it may engender a shift in diagnosis or nal treatment outcome. This chapter documents a third phase in the visit, where negotiation of the diagnosis and treatment occur. Although some practices permeate different phases (e.g., the offering of particular diagnostic possibilities), other practices are related to the phase itself. In the reason for the visit phase, the practices had to do with how to answer a question. Question answering was also in play during the history-taking phase. By contrast, in the diagnosis phase, parents are not being offered any slot to speak. When they seize the opportunity by electing to speak after the diagnosis is offered (Sacks et al., 1974), they perform a rather assertive behavior, even though it is still quite indirect. Thus, although none of these behaviors is terribly frequent, and all are covert, we can nonetheless see a consistent pattern emerging: that parents have and make use of interactional resources at every step in the medical visit and thereby inuence the diagnostic and treatment outcome of the visit.

Treatment Resistance

n previous chapters, the argument has been that antibiotic treatment is being negotiated through behaviors that are not directly connected to treatment. Early in the visit, the parent is offered an opportunity to suggest an initial direction that the physician should follow in the investigation of the problem, as well as in the subsequent diagnosis and treatment. If parents offer a candidate diagnosis that their child has, for example, an ear infection, this suggests that the physician investigate a trajectory of conrming or disconrming this diagnosis and offering treatment in line with that diagnosis. In the history-taking phase, parents rely on the physicians questions to assess what diagnosis-treatment trajectory the physician is working toward and rely on the design of their responses to history-taking questions to push the physician away from that trajectory or toward an alternative trajectory. But there, too, we saw that this had implications for antibiotic treatment. In the last chapter, antibiotic treatment was observably being lobbied for through resistance to no-problem diagnoses. In most visits, parents are not offered an opportunity to simply state their treatment preference, and in the rare cases where physicians ask, parents seem taken aback and do not tend to respond, even if in other ways they appear to have preferences. Despite this, parents shape the treatment outcome through the behaviors we have seen thus far. But the phase where parents can most strongly affect their childs treatment is during the treatment recommendation phase because, in contrast to the other phases, here what is being negotiated becomes relatively more explicit than elsewhere in the visit. The treatment recommendation phase of the visit is also critical because it represents the second and nal structurally provided-for opportunity for parents to inuence the treatment decision. In this chapter, I will show that this is both because at this point the decision must be made and because, unlike diagnosis



deliveries, both physicians and parents treat treatment recommendations as a domain of shared, though not equal, epistemic rights and responsibilities. In this chapter, I rst examine the types of responses that parents give to treatment recommendations and build the case that parents display that they have rights to accept the proposed treatment recommendation. Second, I demonstrate that, in the absence of parent acceptance, both physicians and parents display that mutual agreement is required before the activity of treatment recommendation can progress to closure: Physicians actively pursue parent acceptance, and parents who do not receive a concession shift from passive to active displays of resistance to the treatment recommendation until consensus is reached. Third, insofar as parent acceptance is required as a condition of closing the treatment recommendation activity, parent resistance can lead almost directly to concessions and modications of the physicians treatment recommendation.

Treatment Recommendations: Proposals That Make Relevant Acceptance

Unlike diagnoses, treatment recommendations are oriented to by parents and physicians alike as proposals that normatively require parent acceptance for the physician to progress to the next phase of the visitclosing the encounter. (See Byrne & Long, 1976; Robinson, 2003; Waitzkin, 1991, for a discussion of acute medical encounter activity structure.) Evidence for this comes from multiple sources. We will discuss several types of evidence here. Parent Responses to Treatment Recommendations One type of evidence for the argument that treatment recommendations make conditionally relevant response is the way that parents respond to treatment recommendations. This is particularly striking when parent responses to treatment recommendations are compared with their responses to diagnosis deliveries. Broadly, both activitiesdiagnosis delivery and treatment recommendationinvolve the physician imparting medical knowledge to the parent. For this reason, we might expect that they would be responded to by parents in rather similar ways. This is not the case. Instead, as discussed in chapter 4, parents and physicians alike treat diagnoses as within the physicians domain of expertise, whereas they treat treatment recommendations as a domain of shared expertise: Parents typically respond to treatment recommendations, and acceptance is, as will be shown in later sections, oriented to as relevant. In this way, parents are treated as having an important role in the treatment decision. We can see the contrast between how parents respond to diagnoses and treatment recommendations in Extract 5.1.
(5.1) 2002 (Dr. 6) 1 2 DOC: .hhh Uh:m his- #-# lef:t:=h ea:r=h, is infected, -> (0.2)



3 4 5 6 7 8 9


.h is bulging, has uh little pus in thuh -> ba:ck,=h DOC: -> Uh:m, an its re:d, DOC: .hh So he needs some antibiotics to treat tha:t, DAD: => Alright. DOC: Mka:y, so well go ahead and treat- him: <he has no a- uh:m, allergies to any penicillin or anything.

Having just completed her examination of the child, the doctor here explains the childs diagnosis (lines 15). Although the doctor comes to possible turn completion most notably at the end of line 1 but also at the end of line 4 and at the end of line 5, the parent does not respond. By contrast, after the physician offers her treatment recommendation in line 6, the father accepts this with Alright. immediately upon possible completion of that TCU, and this is particularly notable because the intonation on the two TCUs is quite similar. The physician then announces, as an outcome of this, that well go ahead and treat- him: and then moving to what type of antibiotic to prescribe. A very similar acceptance token to that used in Extract 5.1 is shown in Extract 5.2. Although here the diagnosis is responded to, we can still see a big difference in the type of uptake the two announcements receive. The mother receipts the doctors diagnosis of an ear infection with Mm:. (line 3). This token offers only minimal acknowledgment of the diagnosis (Gardner, 1997).
(5.2) 1183 (Dr. 1) 1 2 3 4 5 6 DOC: Well I think whats happened is is that she ha:s this: uh- (.) .h ear infection in her left ear?, MOM: [Mm:. DOC: -> [And well put her on some medicine and shell [be ne. MOM: [Okay.

The parents response to the treatment recommendation is Okay. (line 6). This tokenparticularly with nal intonationaccepts the doctors recommendation. The consistency with which acceptance is offered to treatment recommendations and the consistency with which minimal or no response is offered to diagnoses suggest that parents hear treatment recommendations to make uptake relevant, whereas they do not apparently hear this for diagnoses. In this case, the parents two different receipt tokens offered in close proximity provide good evidence that parents orient to diagnoses and treatment recommendations as actions that make relevant different sorts of responses. Parent Rights to Respond Parents not only respond to treatment recommendations regularly but also act as though they have a right to accept these recommendations through the way they



design their responses. This provides a second type of evidence that treatment recommendations make acceptance relevant. We can see this exemplied in Extract 5.3. Here, the parents ultimate acceptance of the treatment recommendation is a full form agreement: Lets do that. (line 15). At line 1, the doctor offers his ndings during the chest examination of the child as uh little congested in his che:st, and appears to be moving into the nal diagnosis of the patient with Yeah I think- (line 2). At this point, the mother takes issue with the physicians mitigation of congested and asks a question about her sons more severe morning congestion (lines 5/7). When the physician moves to his treatment recommendation (line 11), the recommendation is offered in an unequivocal manner with we hafta; however, the mother nonetheless displays her orientation to it as a proposal to be accepted or rejected in line 15 with Lets do that.
(5.3) 1120 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 DOC: DOC: MOM: DOC: MOM: BOY: MOM: DOC: Well he sounds uh little congested in his che:st, Yeah I [think[Now its little:. [Yea:h. [.hh But why is it that- in thuh [morning= [#HUH huh# =its just [so:[Well thats because a:ll drips down the back of his throat, MOM: Uhhh ((sigh))/(0.2) DOC: I think we hafta put him on >an antibiotic<. #Eh: he just has- with his ea:r, BOY: Duh! DOC: an he has like a bronchitis in his chest, MOM: -> Lets do that. DOC: Bu:t=hell be :ne.=h

It is not just that she treats the proposal as something to be accepted. The mothers formulation Lets do that. is also stronger than Okay. or Alright.. Lets explicitly treats a decision about her childs treatment as shared. Although acknowledgment tokens such as Okay. and Alright. accept the treatment (Heritage & Se, 1992), the design of acceptance turns provides evidence that parents orient to the relevance of their stance toward the treatment recommendation. Notably, the parent does not acknowledge the diagnoses of an ear infection (noted earlier and indexed here with with his ea:r,) and bronchitis, despite their being just prior to the parents turn. In fact, her response Lets do that. addresses only the action of putting him on antibiotics.1 Although sometimes diagnostic evaluations are acknowledged with Okay, they are not routinely treated as proposals for acceptance or rejection. By contrast, treatment recommendations are routinely accepted with objects such as Okay. or Alright.; Lets do that.; Thats ne.; and assessments such as Good. 2 This



acceptance is treated by parents and physicians alike as relevant after the provision of the treatment recommendation. Physician Pursuits of Parent Acceptance That parents routinely accept physicians treatment recommendations but not diagnoses is one form of evidence that treatment is understood as a domain of joint responsibility and that parents participate in treatment decisions in a way that they do not participate in diagnosis deliveries. Further evidence lies in physicians pursuits of acceptance when none is forthcoming. If there is a normative constraint that makes parent acceptance of the treatment proposal relevant, then, in addition to active resistnace to the proposal, passive withholding of acceptance will also constitute resistance to the proposed treatment. Passive resistance was found by Heritage and Se (1992) in the context of community nurse visits to rst-time mothers. They showed that such resistance to health visitor advice involved unmarked acknowledgments such as mm hm, or yeah,. These objects, they argue, do not acknowledge or accept that talk as advice and thus do not constitute an undertaking to follow the advice offered (p. 395). And following these unmarked acknowledgments, health visitors commonly pursue fuller parent uptake. In the present data, we see an orientation to silence and unmarked acknowledgments as similarly withholding acceptance of the proposed treatment. First, when parent acceptance is not forthcoming following a physicians treatment recommendation (i.e., passive resistance), physicians typically pursue the parents acceptance, treating it as noticeably absent. Moreover, they do not move out of treatment recommendation (i.e., they do not initiate activity closure). These behaviors offer evidence that physicians orient to parent acceptance as normatively required. Pursuit of parent acceptance takes several formats, including offering a rationale for the treatment recommendation, offering evidence for the underlying diagnosis, returning to the examination ndings, and offering the parent a concessionary future action. An example of a physician pursuing parent acceptance when none is forthcoming is shown in Extract 5.4a.
(5.4a) 2043 (Dr. 8) ((BRO is older brother)) 1 2 3 4 5 6 7 8 9 10 11 12 DOC: MOM: DOC: DOC: DOC: (Is) its not infect:e:d, [Theres- uhm no uid or= [Mm. =anything, .hh An his lungs are completely clea:r_ Uhm_ (0.5) An hes not- breathing very fa:st, or har:d, Uhm_ .mlkh So I think hes just on his road to recovery:_ he just needs: another_ .h probly another week or so to get rid of thuh cou:gh completely, -> (Mm hm,) => Just lots of=ui:ds, Uhm he was probly uh little bit dizzy cuz he was:he had fever and he probly hadnt drank enough


13 14 15 16 17 18


=> at thuh ti:me, probly. DOC: => .h Uh:m_ .h So lots to dri::nk, DOC: and then uhm .mlk if he gets- - fever agai:n, (0.2) thou:gh uhm .h in thuh next two or three => day:s, .h uhm_ (0.2) we may need to see him ba:ck, DOC: in case he-n- does come down with something secondarily,

In line 1, the doctor offers a diagnostic evaluation that continues across lines 35. The parent minimally acknowledges this in line 2. At line 7, the doctor begins to detail her treatment recommendation: doing nothing for another week. The parent again offers only a minimal acknowledgment (line 9). In response, the physician expands her treatment recommendation by adding a second recommendation: uids. In contrast to her rst recommendation, this proposes something the parent can do. But here, too, the mother withholds acceptance. The physician next offers an account that supports her treatment recommendation to offer uids (line 10) by proposing that the symptom the parent reported as problematic was the result of dehydration, but the mother again passes on an opportunity to accept the proposal. The physician here pursues agreement by restating her treatment recommendation with So lots to dri::nk, (line 14). By redoing the treatment recommendation, she overtly renews the relevance of parent acceptance. When acceptance is, once again, not forthcoming, the physician slightly modies her proposal. Here, she suggests what the parent can do if the child fails to improvethe parent can bring the child back (line 17). Finally, in line 18 after the parent has, once again, passed on the opportunity to respond, the physician concedes that the boy may need different treatment in the future if he should come down with something secondarily,. Each of the physicians moves works to elicit parent acceptance of the existing treatment recommendation and thus displays the physicians orientation to the relevance of parent acceptance. Physicians pursue acceptance in a wide range of ways. In Extract 5.5a, the physician returns to earlier phases of the visit in order to pursue acceptance. The father withholds acceptance of the physicians recommendation to just watch i:t?.3
(5.5a) P201 (Dr. 7) 1 2 3 4 5 6 7 8 9 10 11 12 13 DOC: Unfortunately like most viruses we have to watch i:t? DOC: -> .hh becau:se- you know- she (can)/(could) have uh fever:: for another few days, and nothin el:se. -> (.) DOC: and jus- an be :ne, DOC: .hh Or else if she got uh fever an got wor:se, and: started limping actually at that time wed probably need er tuh come ba:ck, DOC: -> .hh But at this moment since theres no swelli:ng?, or theres no: .hh you know <nothing else, th-uh most important thing t do is tuh watch her. DOC: -> .hh So weve had a fe:w people right no:w that have had-



14 15 16


uh few of our kids are having tlk .h fever:s, for a few days, and not much other symptom:s. So can she go to preschool now?

In this interaction, the physician has diagnosed the girl with a viral infection. In line 1, she suggests that the best course of action is to watch i:t?. The father neither acknowledges nor accepts this recommendation, despite even the rising intonation. The physician expands her recommendation in lines 23 with an account for it: that the girl could easily have no other symptoms. The father does not accept this either (line 4). The physician here shifts to a discussion of a future plan. Similar to the doctor in Extract 5.4a, in lines 68, the physician suggests when the parent could reasonably return for another medical evaluation. Here, too, the father withholds acceptance. In lines 910, the physician returns to her previous physical examination ndings as further support for her diagnosis. Typically, when physicians retreat to previous activities, including restarting a verbal or physical examination or restating diagnostic ndings, physicians then proceed again through the remaining activity phases back to treatment recommendation. This is similar to what we observed in chapter 4, and this action occurs here, too. Having retreated to diagnostic ndings, the physician next restates her treatment recommendation to watch and see (lines 1112). By restating the treatment recommendation, the physiciansimilar to the physician in Extract 5.4arenews the relevance of the fathers acceptance. Yet here, too, none is forthcoming. In lines 1315, the physician offers a more generic rationale for her diagnosisthat several other children are having similar symptoms. Still, there is no acceptance from the father, though he initiates another sequence with a question about whether the girl can return to school. Yet more types of pursuits can be observed in Extract 5.6. Here, we can see the physician pursuing by repetition, the use of lists, and the use of overt requests for acceptance. At this point in the encounter, the physician has completed an in-ofce throat culture and is waiting for the results. She begins her treatment recommendation with suggestions that are irrespective of these culture results. Throughout this explanation, the parent says very little. At each single arrowed line, there is an opportunity for the parent to respond to the physicians recommendation: Acceptance is a relevant action. But in each case, the parent does not offer acknowledgment, let alone acceptance.
(5.6) 2020 (Dr. 6) 1 2 3 4 5 6 7 8 DOC: DOC: #Mkay:::.# so::,=h (0.5) Tlk=.h Lets see: what=thuh results of this i:s,=h while were waiting for tha:::t, DOC: .h So no matter what the result i:s, h she does ha:ve uh:m hh redness in er throa:t, an looks like she has pharyngitis, <whether its from bacterial -> or from virus, DOC: -> .hh So:: uhm I want her to do mouthwashes?,

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49


DOC: -> .h Gargling at ho:me?, DOC: -> Really deep gargling. (.) All the way back. => #Aghghghgh.# All thuh way back of thuh throat. okay:?, DOC: -> .hh Do it as many as- time as you can. (.) DOC: -> Three:_ four times uh day. Especially after eating. => Mkay, DOC: -> .h That clears it out an that makes it feel better. Mkay,=you can do it with salt water:, you can do it -> with Sco:pe, DOC: -> .hh whatever mouthwash: avor that she likes. DOC: -> .hh So lets do tha:t, DOC: => .hh Give er uh soft die:t?, Mkay:, Dont give her anything heavy, nothing oily:, -> French fries, (.) fried chicken_ hamburgers, DOC: => .hh Nothing spicy.=h for uh couple days. Okay:, DOC: .h Cuz its gonna hurt every time she swallows those -> kind uh stuff. DOC: -> .hh Lets give er lots of liquids at ho:me, (0.6) DOC: -> .hh Give er: water, jui:ce, whatever she wants to drink.=h DOC: -> Ice cream is okay:, That will make her feel better:, DOC: -> .h Popsicles, (.) DOC: -> That makes you feel better, DOC: => .h Mkay:?, DOC: -> .h Maybe some mashed potatoe::s, you know -> (so)/(its uh) soft diet. as uh general. (.) DOC: => Yogur:t, things like that. Nkay:, DOC: -> .hh Uh:m_ and youre just gonna have to rest. (.) DOC: You know?, (.) DOC: Shes gonna have to rest. MOM: Yeah.= DOC: =No more running arou:nd an- (.) ya know staying -> up la:te, an things like that. DOC: .h Youre just gonna have=t take lots of na:ps, -> an re:st, throughout thuh weekend. DOC: => .h Mkay:, ((Doc moves to look at rapid strep culture))

The physician pursues acceptance of her recommendations for mouthwashes (line 8), a soft diet (line 21), liquids (line 27), and rest (line 39). We can see this in several ways. First, similar to Extract 5.4a, she provides accounts for her recommendations (e.g., lines 16, 25, 30, and 33). She also restates her treatment recommenda-



tions (e.g., lines 1011, 3536, 43, and 4748). Third, she adds additional treatments (lines 21, 27, and 39). Fourth, she uses rising intonation at the end of TCUs, such as in lines 8, 9, and 21, which has been shown in other environments to be a resource for securing uptake (Sacks & Schegloff, 1979; Schegloff, 1996). That these locations were designedly in pursuit of acknowledgment can be seen, for example, in the doctors repeat of lines 8 and 9 in line 10 and the respeciciation with All the way back. also in line 10. There is still further pursuit in line 11, rst with the demonstration of gargling and second with the redoing, yet again, of All thuh way back of thuh throat. and then with a more direct request for acceptance with okay:?, Similarly, through the physicians use of three-part lists, she also hearably invites the parents uptake because these lists project completion and have been shown to be strongly designed for recipient uptake (Atkinson, 1984; Heritage & Greatbatch, 1986; Jefferson, 1990). For example, at the end of line 19, the doctor reaches the third item of her projected three-part list and thereby implicates conrmation. A similar list is in line 29, but as before, the parent does not offer any uptake. The physician actively pursues the parents acceptance through other means. For example, in the double arrowed lines, the doctor pursues acceptance with various forms of okay. The physician also switches from addressing the mother to addressing the child (see lines 33 and 39). This change appears to be designed to elicit acceptance, even if that is from the child.4 And in line 41, the physician pursues a response with You know? But it is not until line 44, after multiple pursuits and a change in addressee back to the mother, that the mother even minimally agrees with the doctors treatment recommendation of rest. In this section, I have shown that physicians work diligently to elicit parent acceptance, if it is not forthcoming, before closing the activity of recommending treatment. We saw that their pursuits of acceptance include extending the activity with accounts, returning to prior activities such as diagnostic ndings in support of the treatment recommendation, offering additional recommendations, pursuing acceptance with rising intonation, or, more explicitly, with variations on Okay?. A parents failure to accept constitutes withholding acceptance. Combined with the prior two sections, we now have substantial evidence that both parents and physicians treat the treatment recommendation as an activity that requires parent acceptance prior to moving forward with the visit. This is very different from the orientation shared by parents and physicians to the diagnosis delivery. This is important because it means that parents have additional resources for negotiating the treatment. Specically, whereas with the diagnosis, resistance could only take the form of actively impeding the physicians progress to treatment through the initiation of a new sequence, in the treatment recommendation, parents can impede the physicians progress to visit closure through inaction. So when parents fail to accept the treatment, this can be analyzed as treatment resistance as well. We might best understand this resistant behavior as passive treatment resistance following Heritage and Se (1992). As we have already observed, when parents passively resist a treatment recommendation, physicians pursue acceptance. What we can also observe is that in pursuing acceptance, physicians act as though they are in a negotiation. Thus, passive resistance is yet another resource parents have for pressuring physicians for treat-



ment. This argument relies on a normative structure of treatment recommendations to suggest that even doing nothing in a particular sequential environment can be a consequential form of participation and can affect treatment decisions. Further evidence of this is shown in the next section.

Active Resistance
Part of the evidence that withholding acceptance is resistant is from the parents side. If physicians do not alter their treatment recommendation in the face of passive resistance, parents routinely shift from passive to active resistance. Active treatment resistance closely parallels diagnosis resistance because it includes an action that questions or challenges the physicians treatment recommendation, including proposals of alternative treatments. These actions make relevant a response by physicians, and this feature differentiates active resistance from passive resistance. Heritage and Se (1992) found that this upgrading pattern was present in their advicegiving sequences as well. Sequences that included unmarked acknowledgments culminated in a more overt expression of resistance (p. 402).5 An example is shown in Extract 5.4b. We saw the rst component of the treatment recommendation activity in Extract 5.4a. Here, following the physicians indication of what sort of symptoms would cause her to review the childs case for treatment (lines 1518), the parent shifts from passive to active resistance of the physicians treatment recommendation.
(5.4b) 2043 14 15 16 17 18 19 20 21 23 23 24 25 26 27 28 29 30 31 32 DOC: .h Uh:m_ .h So lots to dri::nk, and then uhm .mlk if he gets- - fever agai:n, (0.2) thou:gh uhm .h in thuh next two or three day:s, .h uhm_ (0.2) we may need to see him ba:ck, in case he-n- does come down with something sec[ondarily, MOM: -> [(See c- cuz-) what I was worried about I [(wouldve)/(wouldnt)= DOC: [Mm hm, MOM: =normally_ (0.9) DOC: m- Bring [him in, MOM: [interpreted [this as a co- ya know= DOC: [Mm hm, MOM: =uh thing that [would run its course but- (.) this= DOC: [Mm hm?, MOM: =guy had thuh same thing and wound up on antibiotics cuz he got an infection. MOM: .hh[h DOC: [Whe:[re. MOM: [How can I prevent that. from happening.



In this example, the mother is clearly concerned that her sons condition will become worse and that he will need further treatment if not given treatment now. This concern is not articulated until the doctor has provided more and more details about her recommendations for future actionunderstandable as pursuing uptake from the mother. It is only at this juncture that the mother explains what I was worried about (lines 1920). Framed in this way, her turn is formulated as an account: arguably an account for her prior passive resistance. The mother here actively resists the recommended treatment through the juxtaposition of her announcement that her other son is on antibiotics for an infection with her inquiry about how to prevent this child from having to deal with a similar infection. This announcement, and further the inquiry, challenges the physicians suggested treatment of watchful waiting and uids. That parents consistently upgrade from passive to active resistance in the face of a failure by physicians to modify their treatment recommendation is evidence that passive resistance is, from the parents perspective as well, a form of treatment negotiation. Active resistance, then, is a much stronger form of treatment negotiation. In the following example, the mother once again began her resistance passively and then here upgrades to active resistance (lines 2728). This case is particularly striking because the active resistance she moves to is an explicit inquiry about antibiotic treatment. Extract 5.5b follows Extract 5.5a shown earlier. The physician is, in line 25, returning to her previous ndings, having just responded to a parent question.
(5.5b) P201 ((8 lines omitted following 5.5a)) 25 26 27 28 29 30 31 DOC: She: doesnt have anything right no:w, any symptoms of mucus or vomiti[ng thats contagious. DAD: => [Are you gonna give her ana- antibiotics? DOC: Yeah- uh No: I dont have anything tuh treat right now for antibiotics. Her ears look really goo:d, .hh she has no sign of bacterial infection right no:w?,

The resistant inquiry: Are you gonna give her ana- antibiotics? makes relevant an answer, but physicians typically respond to such inquiries by not only responding to the question but also treating the actions as lobbying for antibiotics. In this way, the fathers question rather directly challenges the physicians own recommendation. (We will discuss this further in chapter 6.)

Antibiotic Negotiation
The social norm that treatment recommendations require parent acceptance has the consequence that active or passive resistance of a treatment recommendation puts the physician in a position of either working to convince a parent to accept the proposed treatment recommendation or offering the parent concessionseither pos-



sible or actual. In this way, treatment resistance can be seen as yet another resource for initiating a negotiation of the treatment decision. This is the most overt type of negotiating behavior we have seen, because through treatment resistance, parents take a position against the treatment they are being offered rather than merely urging physicians toward or away from a particular diagnostic or treatment trajectory. Parents usually resist over-the-counter, nonantibiotic treatment plans. Across the whole of these data, there were very few cases where parents could be understood to be resisting antibiotics. In most cases, the parents position is not fully on the surface of the interaction. However, it may be best exemplied by looking at an initial interaction where it is brought to the surface of the interaction. Here, after the physician offers his position against antibiotics in line 4, the father resists (lines 6/10/12/14/1718/20/23/25/27).
(5.7) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: -> I th:ink from what youve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I dont think antibiotics will ki:ll, (0.2) Well[Thee other[( ) >Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d Ive been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm, =sore [for (awhile- et- that) w:eek. [Uh huh, Right, an:d (.) I start taking thuh antibiotic (0.5) eh he ((cry)) Yesterday. Right, And it (.) seemed to take care of the problem. [(Well) thats why were doin a throat [culture. [( ) [Yeah. [is TUH SEE if they need antibiotics. [( ) Yeah yeah.

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: -> INF: DAD: -> DOC: DAD: -> DOC: DAD: DOC: DAD



32 33 34 35 36 37 38 39 40 41 42 43 44 45

DOC DAD DOC: => => INF: => DAD


(0.2) Cause <I dont th::ink they do. O[kay, [Now if you (.) absolutely insist_ I will give you antibiotics_ but [I dont think thats the right= [#eh::# medicine for em, No Im not saying- Im not saying it- (0.2) dont get me wrong but- Im sta- trying tuh tell you the [history of ( ) [I understand, I- I heard [you when you told me, [Yeah. I under[stand, [Uh huh,

In lines 23/25/27, the father states that antibiotics solved his own illness, thereby implying that they would be helpful for his two sons who are ill with the same thing (as he mentioned earlier in the encounter). In response, the doctor rst explains that this is a possibility, which is why he performed a throat culture to test for the strep bacteria. But in lines 3536 and 38, the doctor then goes on to offer to give the antibiotics against his medical judgment if the parent insists. I want to note three things about this case: First, the physician overtly treats the fathers participation, indeed his acceptance of the nonantibiotic recommendation, as both relevant and important. Second, he treats the fathers narrative about his own experience with antibiotics as applying pressure for antibiotics. In fact, he treats it as just one step short of insistence. We can see this when he says if you (.) absolutely insist (line 35), where he treats absolute insistence as having not yet occurred. However, as a condition that is being discussed, the physician further conveys his understanding that this is the behavioral trajectory the father has been on. Third, the physician here overtly acknowledges the impact of parent pressure: If the parent continues to lobby for antibiotics, he will provide them in spite of the fact that they would, in his opinion, be ineffective and thus inappropriate. More typically, the negotiation of treatment between parents and physicians is not brought to the interactional surface, but parent resistance can nonetheless be seen to initiate a negotiation, as evidenced by physician responses. For instance, we can return to a case shown earlier in Extract 5.5a. In the previous section, we observed that the father here shifts (line 2728) from passive to active resistance. Now I would like to focus on how the physician responds to this shift.
(5.5c) ((8 lines omitted following 5.5a)) 25 26 27 28 29 DOC: She: doesnt have anything right no:w, any symptoms of mucus or vomiti[ng thats contagious. DAD: => [Are you gonna give her ana- antibiotsics? DOC: Yeah- uh No: I dont have anything tuh treat right now

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58


-> for antibiotics. DOC: -> Her ears look really goo:d, DOC: -> .hh she has no sign of bacterial infection right no:w?, DOC: -> .tlkh and thats (what) shed get antibiotics fo:r. DOC: .hh So uh lotta times you can start out with uh virus like uh co:ld, (.) .h an:d if you- it goes on for uh while #uh:# bacteria (should) set in you can get uh secondary -> bacteria infection? and thats when you need antibiotics. DOC: .hh #But- y-# otherwi:se: since she doesnt have any source of an antibi- of uh bacterial infection?, -> that=uh=we just watch her. (0.3) DOC: right now. <Her ears look really good, <an her tubes are -> in too:. DOC: -> .h And theyre not draining or any[thing. DAD: => [It just means that=> ya know if she gets another fever we hafta bring her ba:ck, DOC: .hh Well what Ill do is she might still get uh fever: in thuh next couple uh #da:ys.# because: .h thats th way viruses wor:k?, you can have- you know (how have=you) if you have uh co:ld, you can get a fever for uh few da:ys? .hh And tha:t Since shes o:lder:, .h if somethings #uh# she (would) com[plain ( ) thuh symptom,= GIR: [( ). DOC: =then she would need tuh come back. DOC: .hh But what you ca:n do i:[sGIR: [Guess what.

At the beginning of this segment in the interaction, the father has not yet accepted the physicians proposal of no prescription treatment, and in overlap with the physicians reassertion that there is nothing really problematic wrong with his daughter, the father asks about antibioticsan alternative treatment proposal and a strong form of resistance (other examples of overt negotiation were shown in Extract 5.4b and 5.7). In response, the physician rst answers the question (line 29). She then goes on to account for her answer, stating that there is nothing to treat at this point (lines 2930). The father does not offer acceptance. Next, similar to how physicians respond to passive resistance, the physician restates a physical examination nding (line 31). Following no uptake again, the physician restates her diagnosis of no bacterial infection (line 32). The father still does not accept, and the physician expands her treatment recommendation, further ruling out the need for antibiotics (line 33). Again, there is no acceptance. At this point, the physician shifts to a scenario where treatment would be warranted. In this way, she intimates that she may make a concession in the future. However, there is still no parent acceptance, and in lines 3840, the physician restates her treatment recommendation.



Following 0.3 seconds of silence, the physician returns to restate additional physical examination ndings (lines 4243) in the face of no parent acceptance. This is further expanded in line 44. Here, the father again actively resists (lines 4547). Up to this point in the encounter, the physician has been working to secure parent acceptance of the current treatment recommendation to just watch the girl for a bit longer with the intimation that if things changed, she would be willing to treat her with antibiotics. However, she has not achieved parent acceptance, and in fact, her work has been met with increasingly stronger parent resistance. Here, the physician frames her response as a concession with Well what Ill do is. This does not reach completion before the girl initiates a sequence that the physician takes up. After the physician closes the sequence with the girl, she returns to offering a contingency plan (Mangione-Smith et al., 2001)a concession to the parentthat he could call rather than coming back in (lines 6570). The physician also intimates (in line 70 with talk to us and see:_) that the physician might be able or willing to take another course of action over the phone or as a result of the phone call, further suggesting concession to the fathers pressure. But she also maintains her stance in favor of the current treatment recommendation to watch the girl.
(5.5d) ((6 lines of conversation with GIRL not shown)) 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 DOC: -> DOC: -> DOC: DOC: -> DOC: -> DOC: .hh Uhm: mlk but usually what you can do is if over thuh next few days she might still get a fever of (a hundred an two) you can give us a call if youre concer:ned. .h And if it goes on more than tha:t, .hh she might need tuh come in #but:# .h most uh thuh time you can just uhm call us and talk to us and see:_= =If she has uh new symptom, breathing difculty:?, .hh if she had ear draina:ge, if she s:- did start tuh limp, then we would say she does need tuh come in. .hh But for uh child her a:ge, you cn get fever for uh few day:s, an: as long as she looks this goo:d, an no other symptoms, .hh ya know we just- well watch her. So like if she got uh fever this afternoo:n that doesnt mean she needs tuh come in right away:_ What I would do is like you did: this morning, .hh give er some Tyleno:l, If she .h looks great like thi:s then itsshes probably still just ghting off thuh virus. (0.3) Mka:y:? (0.8) Mkay.

-> DOC: -> -> DAD:

In lines 7173, the physician suggests that only certain circumstances would require the parent to return to the ofce. The parent still does not accept the treatment proposal. In lines 7476, the physician restates her treatment recommendation that we just- well watch her. However, this restatement still does not engender acceptance.



Following this, she also redoes her treatment proposal to include a recommendation of something the parent can do to be more proactive in lines 79806 and once again restates her diagnosis. This action is met with 0.3 seconds of silence. At this point, the physician overtly pursues acceptance with a heavily question-intoned Mka:y:? and, after a substantial delay, receives a quiet but minimal acceptance from the father (line 85). This case provides evidence that when physicians face treatment resistance from parents, they orient to this as initiating a negotiation of the treatment recommendation and work to secure parent acceptance. Although this physician did not ultimately modify her treatment recommendation from no antibiotics to antibiotics, she could nonetheless be seen to be making concessions to the parent, including suggesting that the parent call if the child got worse and suggesting that Tylenol would work to bring down the fever. Therefore, he would not need to return to the ofce. The negotiation of antibiotics can also be observed quite clearly in Extracts 5.8ac. Here, as part of her diagnosis, the doctor denies sinusitis7 (lines 12) and then moves into her treatment recommendation beginning in line 4. The mother endorses the doctors recommendations rst by inquiring about a decongestant that is in the general category of Sudafed, which was recommended by the doctor (line 8).
(5.8a) 2015 (Dr. 9) 1 2 3 4 5 6 7 8 9 10 DOC: ^Ya know, I probably_ (0.5) wouldnt call it sinusitis right now. (0.5) DOC: Uhm- h- What I would do: is keep up with thee uhm h over thuh counter- you know maybe like childrens Sudafe:d or something like that to help with thuh: thuh congestion in her nose. MOM: -> [Now shu- we should (be giving) her uh deconges[tant. DOC: [.hhh [hhh DOC: Yiea:h, I think that would probably help. ((35 lines of talk about different types of decongestant not shown)) 46 47 48 49 50 51 52 53 54 55 56 57 58 DOC: MOM: DOC: DOC: .hh Uh- but certainly if thuh fever goes up higher than just this low grade ninety nine or uh hundred. [Okay, [ . h h Uh:m tlk <Or if: uhm> (.) this doesnt seem to be going away.<but lets see=todays Thursday. .hh If she denitely has any more green stuff coming out. cuz I could just be hitting er at a good day where its drained down [thuh back [of her throat= [Right. [I was gonna say= =[an IM NOT SEEING ANYTHING. -> =[YOURE NOT SEEING it on her throat or anything. (.) ((Possible head shake by DOC))




59 60 61 62 63 64 65 66


Oka[y. [Uh uh. [Okay. [ .hh Uhm .h but: if ^its persisting: into next week. Ok[ay. [I mean_ Then I think we should see her ba:ck. [Okay. [I mean- and think about sinusitis.

Following the parents endorsement of the proposed treatment, the physician appeared to complete the treatment recommendation and to have moved into plans for future action: a move into closings.8 This can be seen here in line 4647/4953. However, in line 54, the mother offers a slightly premature Right. that may be working toward speaker transition (similar to Yeah as discussed by Jefferson, 1983), and then immediately upon possible grammatical completion, the mother initiates a turn of active resistance (line 54). Here, she requests conrmation of the physicians previous diagnostic ndings, thus asking the physician to retreat into an earlier activity (something we have seen in previous cases). In overlap, the physician shifts from her previous straightforward plan for future action to competitively addressing the mother as resistant. The competitiveness of the physicians talk is shown in that it is substantially louder than her other talk. Additionally, her second TCU an IM NOT SEEING ANYTHING. restates her examination ndings. Although we cannot conrm whether the parent responds visibly (e.g., with a nod or facial expression), after a micropause, the parent vocally accepts this with Okay. The physician returns to a future plan of when to see the girl back and further addresses the parents resistant action by stating that she might consider sinusitis at that time (line 66). Following this, the doctor offers another type of concession to the parentthat they could consider doing an x-ray of the sinuses. But rather than resulting in full acceptance, the mothers resistance to the current line of diagnosis and treatment escalates. Although she accepts the physicians position of not liking to x-ray children (line 71), she then goes on to offer a brief narrative about her older daughter, who was also without the classic sinusitis symptom of heavy nasal drainage but who apparently had a severe infection.
(5.8b) 67 68 69 70 71 72 73 74 75 DOC: .hh We can always get- just uh plain x ray: of thuh sinuses. An sometimes thats helpful whenon these equivocal things. <I dont like to x ray kids uh lo:t but- [.hh [Right. If were not su:re uhm if it doesnt sound really clear cu:t, .h we can- (.) >get an x ray.< Shes not really tender, .hh [over her sinuses either.


76 77 78 79 80 81 82 83 84 85


MOM: MOM: -> -> DOC: MOM: -> DOC: MOM: -> DOC: MOM: -> DOC:

[Yeah. My older daughter- I brought her in: This was two years ago an they s[aid well we dont think it- [Uh huh, We- she had no:ne uh thuh symptoms of sinusitis. [Uh huh:, [(ever.) An she had headaches for uh yea:r? Yeah. An when they nally x rayed her she was [totally blocked. [just socked in,

The mothers narrative is, like the previous examples of treatment resistance, positioned late in the counseling phase of the encounter and, furthermore, is positioned after several recommendations which the mother has endorsed. The mothers narrative conveys her position that she would like an x-ray of her daughters sinuses because in the past the x-ray revealed sinusitis with her older daughter. Like the situation the mother is in at this point in the encounter, in her narrative she relates doctors telling her we dont think it- (line 78), which appears on its way to denying sinusitis (precisely what this physician has done earlier in the encounter). Although this position is embedded in her narrative, it is nonetheless conveyed. This type of resistance is primarily focused on the diagnosis and ways to detect sinusitis. So unlike some of the other types of active treatment resistance we have seen, here a symptomatic treatment recommendation is being resisted through a challenge to the underlying diagnosis. In 5.8c, the doctor responds by dealing with both the mothers position in favor of an x-ray and her use of this as a vehicle for resisting a lack of prescription treatment. First, the doctor agrees with the mothers narrative as plausible (line 86). She then provides an account for a lack of drainage and having a sinus infection with its so blocked (line 90). However, she also asserts that the decongestants may allow drainage to begin. The doctors turn effectively disagrees with the mothers position in favor of an x-ray. But her opposition is embedded. She provides a condition under which she does recommend x-rays, and the contrast is built through the use of the additional modal will, along with its contrastive stress (line 88). Additionally, the doctor focuses her turn on how the treatment she has recommended may help: It may help allow the drainage to begin, if in fact the mother is right about her daughters condition. So this is another way that the physician displays her understanding that she and the parent are still in negotiation of the nal treatment decision.
(5.8c) 86 DOC: => Yeah:. 87 DOC: => .h Thats why especially- in kids (>whore<) complaining 88 => of headaches an things like that I will get an x ray. 89 MOM: Right.



90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110

DOC: => ^Because uh lotta times its so blocked at n(h)othings => dr(h)aining [ou(h)t. MOM: [Ri:ght. [Right. DOC: => [ . h h And what you may nd (.) => Her nose doesnt look particularly swollen er anything => inside but you may nd if you- consistently give er => thuh decongestants for uh couple days, .hh that it opens => up: thuh passages t thuh sinuses. DOC: => .hh An: [an youll start se[eing thuh stuff coming ou:t. MOM: [( ) [seeing it. MOM: [Okay. DOC: => [ . h h An then- I mean I would: at this point if she => comes back in on Monday h=er- or Tuesday and stuff=> I would have probly uh lower- since its been: (.) => continuing for all [this time_ .h]h to start her on= MOM: [Right. right.] DOC: => =some antibiot[ics. MOM: [Right. DOC: => <But I hate to- tuh put her on if she doesnt really => nee:d it. MOM: -> ^Okay. Thats ne.

The doctor also offers a possible concession in that she states that she would be willing to start the girl on antibiotics if the condition were to persist into the next week (lines 101104/106). The doctors response also addresses the mothers action as resisting her treatment recommendation. Specically, she again denies the need for antibiotics at this time (lines 108109). Having previously outlined her treatment recommendation and having moved from treatment recommendation into recommendations for other future action, such as when to bring the girl back to the ofce, this action is specically a return to her treatment recommendation. As such, it is hearably responsive to the parents resistance. The mother then accepts the doctors decision with ^Okay. Thats ne. (line 110). This acceptance is the most full acceptance provided throughout this phase of the encounter. Although ^Okay. alone might have been equivocal as a move to accept and close the sequence (especially given a context where okay has been used repeatedly at various junctures in the discussion), Thats ne. is much stronger as an acceptance of the doctors position, and further as taking a position of closing the sequence. This example again shows an elaborate negotiation of the treatment recommendation, including two concessions by the physician: (1) the offer to do an x-ray to conrm the mothers concern that it is sinusitis and (2) the physicians offer of antibiotics if the condition is not better by the following Monday or Tuesday. The latter concession especially appears to work, and the mother shifts from Okay and Rightacknowledgments that had been offered previously to a fuller form acceptance: ^Okay. Thats ne. (line 110).



Treatment Change as a Response to Resistance As with diagnosis resistance, the most extreme form of physician concession is to alter the treatment recommendation from no antibiotics to antibiotics. Although this is relatively rare, that it happens at all provides strong support for the power of treatment resistance as an interactional resource for negotiating treatment outcomes and, more generally, the orientation to parent acceptance of treatment proposals as required. An instance is shown in Extracts 5.9ad. In lines 12 of 5.9a, the physician recommends against antibiotics, but the parent does not accept. The physician expands her treatment recommendation against antibiotics in line 3 with an increment (Schegloff, 2001). The parent does not accept this either. The physician then afrmatively states that she would like to treat the girls eyes and give her some decongestant (lines 45). She provides a rationale for that recommendation in lines 67. The parent continues with passive resistance and then initiates active resistance during the 68 lines of talk that I have not shown here.
(5.9a) 2019 (Dr. 6) 1 2 3 4 5 6 7 DOC: -> DOC: -> DOC: -> -> DOC: .hh So: uh:m a- at this time I dont wanta commit er to: antibiotics. Like two weeks, or three weeks, or whatever:? .h I thi:nk Ill go ahead and treat her for the eye:s?, an I wanta give her some decongestant. So that would, suck out all that, um, secretions?=

((68 lines including passive and active resistance not shown)) 76 77 78 79 80 81 82 83 84 85 86 87 MOM: DOC: MOM: DOC: MOM: DOC: MOM: DOC: DOC: MOM: DOC: But anyway shes had low-grade temp [(an uhm), [Mm hm. (1.1) just really hasnt been hersel:f. Its- its- Its:= =M[m hm. [(ya know)/(even) more than: uhm (1.5) thee eye thi:ng. Uh huh:, <I mean I usually dont- I- I usually wait to bring her in at least until [( ). [You wait unti- Yeah:, .hhh Uh:[m[Cuz its such a big deal to come here [( ) [Yea:h,=h

In lines 76/78, she asserts that her child simply is not wellthat she is sick. The implicit claim appears to be that her daughter is sicker than the doctors treatment recommendation would suggest. This is pursued further and more explicitly in line 80, where she says its more than thee eye thi:ng.. This case further evidences that treatability can be separated from concerns of visit legitimacy: Here, the eye infection provides visit legitimacy but does not deal with the parents concern for a solu-



tion to the childs other symptoms and her desire for antibiotics. In lines 8283/86, the mother suggests that normally she is very troubles resistant (i.e., not a mother who rushes her child to the doctor). Although this is often associated with threats to visit legitimacy, here the implication is that her daughters illness is more serious than the doctors treatment recommendation would suggest and appears to be more of a means to address treatability. We can see the next component of the interaction in Extract 5.9b. Here, the physician begins a turn that is more concessionary in its design. She rst agrees with the parent with Yeah (line 87) and then with I mean: if you wa:nt ya know-, which begins to frame her forthcoming response as a concession to what the parent wants. Note that the parent has not yet stated anything that she wants or expects explicitly, but she has (1) passively resisted the physicians treatment recommendation by failing to accept it and (2) actively resisted the treatment recommendation by implying that her child is sicker than the doctor is prepared to recognize.
(5.9b) 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 MOM: DOC: MOM: DOC: DOC: MOM: DOC: [Cuz its such a big deal to come here [( ) [Yea:h,=h I mean: if you wa:nt ya know- I mean she looks.= =Can I at least have thuh prescription an Ill decide whether or not to ll it, i[n a couple day:s, [.tlk For the antibiotics[:? [Ye[ah. [Uh::m_ I really dont like to do tha:t, because: I mean .hh She doesnt look: like she has sinusitis:. Ya know?, (.) Uhm, if you really wanta be su:re we can go ahead and take: x rays to make su:re if its really opacify:, .hh cause unnecessary treatment for sinusitis: she can get resistant to uh lot of those antibiotics?, uh lot of those bugs. I mean. .hh An:d its- its not really good for her:. (1.0) So:: we try to minimi:ze ya know- treatment until its really necessary. (.)

DOC: -> DOC: -> -> DOC: -> -> DOC: -> ->

The concessionary frame is abandoned in favor of a less concessionary I mean she looks. that, given the no-problem physical examination that preceded this discussion, is likely to be heard as headed for another no-problem evaluation. This would be inconsistent with prescribing antibiotics. At this point, the mothers strongest form of treatment resistance comes: an overt request for antibiotics in lines 8990. The mothers request not only calls into question the treatment recommended so far but



also specically challenges the physicians assertion earlier in Extract 5.9a that she does not want to commit the girl to antibiotics at this point. The mothers request Can I at least have thuh prescription orients to the prescription as a minimal form of action and implies that it is signicantly less than actually treating the child. This is accomplished largely with at least. The second unit of her turn an Ill decide whether or not to ll it in a couple day:s, claims some measure of discretion (i.e., that she would not immediately ll the prescription and give her child antibiotics), as well as claiming that she would have the knowledge to determine whether and when to ll the prescription. The doctor denies her request in lines 9496 but offers the parent a concession: They could perform an x-ray that would potentially clarify whether the child should appropriately be treated for sinusitis. In addition, the physician cites the inappropriateness of treating this condition with antibiotics and the general need to avoid inappropriate prescribing as an account for her recommendation against antibiotics. The mother fails to accept either the physicians rejection of antibiotics or the concession. At each arrowed line, the mother passes on an opportunity to accept the physicians recommendation. The mother continues to actively resist across the next stretch of interaction. Here, after the doctor again returns to outline a situation in which she would concede and prescribe antibiotics (if the girl looks really -ba:d,), the mother asserts that her daughter never looks bad (lines 110/112). She then claims that her daughter is not herself, thus implying, again, that her daughter is sicker than the physician is recognizing and further justifying her seeking of antibiotics.
(5.9c) 105 106 107 108 109 110 111 112 113 114 DOC: So:: we try to minimi:ze ya know- treatment until its really necessary. (.) You know of course if shes s- you know looks really -ba:d, [then Ill go ahead. [(see she ne-) she never looks: ba:d. I mean [she can be really [Mm hm:, sick and she never looksMm hm[:, [You know: Ive taken her in here with:


((20 lines not shown: examples of girl not acting sick)) 135 136 137 138 139 140 141 142 143 MOM: MOM: DOC: MOM: DOC: MOM: DOC: MOM: [And plus its her (t=her:) uhm (0.6) tlk (0.4) Uh:hm_ (0.5) Whatm I tryin t say:_ Emotionally. (I [mean shes been) .hh (0.8) t- you know more n more= [Mm hm:, =tire:[d, [Mm [hm:, [And more n mo:re (.) upset easily_ [an stuff: [Mm hm, over thuh past couple weeks, [an its- its just been building=



144 DOC: 145 MOM: 146 DOC:

[Mm hm:, =an building an bui[lding. [Mm hm.

Throughout this chapter, I have argued that both parents and physicians are oriented to the treatment recommendation as an activity requiring agreement between the parent and the physician. We have seen a variety of evidence for this claim. In this case, we have seen the physician pursue agreement in several ways, including offering accounts for her treatment recommendations and offering an alternative course of actionthe x-ray. Throughout, the parent has been unyielding in her dissent, rst passively resisting and ultimately overtly requesting an alternative type of treatment. In the nal extract that I will show, the physician works to close the activity after what is now over 150 lines of discussion of the treatment recommendation. Note that if the mother had agreed readily to the treatment following the recommendation shown in Extract 5.9a, this activity might have closed within just a few lines. At this point, the physician offers yet another concessiona willingness to talk to the girls regular physician (lines 149; 152153; 160161).
(5.9d) 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 MOM: DOC: DOC: MOM: DOC: MOM: DOC: =an building an bui[lding. [Mm hm. .tlkhh Who: usually sees her. Doctor Hilton. .hh Uh:m lemme call him an see what he uhm says.= =Oh is h[e around (today?) [Okay? I dont know if hes arou:nd but Ill=lemmme try to call him. .hh because: uh:m_ Hes not [( ). [Tlk I really dont want to treat er. (0.5) Uhm but then Ive only seen her rst time. This is my rst time seeing her so I really dont know how she (.) you know i:s, .hh So let me call im an see: what he sugge:st, .h An the:n well go from there. (.) [Does that sound okay? [Okay. Sure, if you [can (reach) him it sounds great.




Even here, after proposing to call the childs regular doctor, the mother resists when the physician reraises her treatment recommendation in line 155. The mother still does not accept this (line 156). But when she proposes, as an alternative, that she



will see what he suggest, in line 159 and make a decision at that point (line 160), even though the mother does not accept this immediately, she does offer acceptance in line 163. The doctor pursues more explicit acceptance in line 163 with Does that sound okay? Then the mother more fully accepts (albeit provisionally) in line 165 with Sure, if you can (reach) him it sounds great.. Ultimately, the physician cannot reach the girls regular doctor, and she ends up prescribing for the girl, despite having diagnosed only conjunctivitis, having explicitly rejected a sinusitis diagnosis, and having repeatedly expressed a desire not to treat the girl with antibiotics (mentioned again in line 155 here). Similar to other concessions that physicians offer, this one is offered at a point when the parent has both passively and actively resisted the proposed treatment. In this case, the physician worked to convince the parent of a nonantibiotic treatment recommendation but was entirely unsuccessful. Despite the physicians strong position against prescribing, she is pressured through normative constraint that she must secure parent acceptance of the treatment recommendation in order to close this activity. When that is not forthcoming, the physician alters her recommendation to obtain the required acceptance. Although this case offers a rather extreme and overt example of the negotiation process and the possible outcome such a process can yield, that the outcome is negotiated is, as we have seen, not so unusual. In fact, there is no case in these data where a parent fails to accept the treatment and a physician nonetheless proceeds to activity and visit closure. Treatment resistance, like other behaviors we have looked at in this book, is not terribly common. Across analyses of both the Seaside and Metro data, parents resist 19% of nonantibiotic treatment recommendations (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; Stivers, Mangione-Smith, Elliott, McDonald, & Heritage, 2003). But as we have observed, qualitatively we can see that this behavior can have an important inuence on the diagnostic and treatment outcome of the visit. The reason appears to be that the behavior, like others we have looked at, places pressure on the physician for antibiotics and physicians respond to that pressure in various ways. At times, they defend themselves, but at other times they make concessions or even alter their treatment recommendation. Quantitative analyses also support this link. Parents are signicantly more likely to resist the diagnosis when the diagnosis is viral (21% vs. 6%; p <.001) (Stivers et al., 2003). And physicians were signicantly more likely to perceive parents as expecting antibiotics if they resisted their treatment recommendation (MangioneSmith et al., 2006; Stivers et al., 2003). In an analysis of the Metro data, physicians were 24% more likely to perceive parents as expecting antibiotics if they resisted than if they did not resist the treatment recommendation. Although any single behavior we have looked at may be present less than 10% of the time, what we have been able to see is that there are many opportunities for parents to inuence physicians diagnoses and treatment recommendations. In the Seaside data, there was some form of negotiation in a full 62% of total cases. In the Metro data, there was some form of negotiation in 42% of cases.9 A substantial number of cases in the data involve interactions where the illness will have been pronounced bacterial and treatable early on in the visit. Thus, a rate of 4060% is quite substantial. Additionally, most behaviors were not associated with each other. Across



both the Seaside and Metro data, only diagnosis resistance and treatment resistance were associated with each other (p <.01). Thus in general, these behaviors do not represent the same parents escalating across the visit. However, within a particular type of resistance, we can observe the same parents escalating (e.g., during diagnosis or treatment resistance as shown in chapters 4 and 5). At the same time, though, we see that parents are rather reticent in their use of such behaviors. We might not expect to see this, given that unlike the diagnosis or history-taking phase, parents are normatively required to be part of the treatment decision by accepting it. However, even in this context, parents remain largely covert in their resistance and very rarely make any overt statements about their treatment preferences. Their inuence is much more by pushing physicians away from a nonantibiotic form of treatment.

This chapter has shown that unlike diagnoses or any other activity in the medical visit, treatment recommendations are oriented to by doctors and parents alike as a domain of shared responsibility and epistemic rights. This was evidenced rst by the regularity with which parents accept treatment recommendations; second, through the observation that physicians pursue parent acceptance, if it is not forthcoming, prior to initiating closure of the medical visit; and third, through the observation that parents typically move from passive resistance to active resistance if physicians do not alter their treatment recommendation. Parents can exploit this normative structure to affect the treatment decision or outcome of the visit because by failing to accept the physicians recommendation, they effectively block the physician from being able to close the visit. When they actively resist the treatment recommendation, this blocks the physician from closing even more extremely because physicians must also deal with the initiation of new sequences, much like after diagnosis resistance, except that physicians must not only close these sequences but also secure parent acceptance: at minimum, a two-step process. When parents resist a nonantibiotic treatment, physicians systematically treat this as indexing a desire for antibiotics. They treat parents as having been in search of antibiotics, and thus as not getting what they want. Oftentimes, this appears to be what is at issue for the parents, as in Extract 5.9. But the more general issues of treatability and of legitimacy may also be involved. As was the case with most of the other behaviors, treatment resistance was not associated with parent reports of an expectation for antibiotics (Mangione-Smith et al., 2006; Stivers et al., 2003). Based on both this evidence and what we see interactionally, it appears that although parents who resist a nonantibiotic treatment recommendation are often resisting specically to lobby for antibiotics, they may also be resisting what represents a denial of a solution to their problem: a child who cannot sleep, a child who cannot go to school, a day care provider who will not accept the child until treated, or a mother who needs to return to work but is at home with the child until he or she is better. Thus, some parents may be resisting the lack of their childs treatability and not only the lack of antibiotics. Another possibility is that some parents who resist a nonantibiotic treatment



recommendation may be resisting the delegitimization of their visiteffectively a judgment against their decision to seek medical care. And as discussed in previous chapters, these issues can be conated: A doctors judgment that a child is not in need of treatment may imply that there was no need to waste time on a medical visit. It is often virtually impossible to tease apart these issues in real-time interaction. But physicians often fail to address either of the latter dimensions while focusing nearly exclusively on the former issue. This chapter showed several instances of more overt forms of negotiation. Although this behavior is relatively rare, it does sometimes occur. When it does, it is yet another way that parents inuence the visit. In general, such overt negotiation happens through one of the other practices and therefore represents only a very strong form of that behavior (e.g., a strong form of treatment resistance). In the next chapter, we will examine overt forms of negotiation and how physicians respond to them.

Overt Forms of Negotiation

his book has thus far been devoted to showing how parents play a role in the diagnostic and treatment outcomes of their childrens medical visits through covert communication practices. But as we have seen in the discussion of practices of mentioning additional symptoms, diagnosis resistance, and treatment resistance, parents do occasionally talk about antibiotics overtly. When they do this, they are quite explicitly lobbying physicians to prescribe antibiotics, whereas when they perform these behaviors without specically mentioning antibiotics, they lobby physicians more implicitly. In this chapter, I examine four main ways that parents overtly lobby for antibiotics and show evidence that parents and physicians alike treat antibiotics as under negotiation. Unlike the other practices that are largely conned to a single phase of the interaction, overt lobbying for antibiotics occurs in virtually all phases of the medical encounter.

Although overt parent lobbying for antibiotic treatment is unusual, it does occur, and by examining cases in which it is involved, we gain useful insights into the parentphysician negotiation process that hold across cases where the negotiation is entirely covert. In Extract 6.1, the parent and physician have already been in negotiation over antibiotics. The parent actually already has an antibiotic prescription from another physician for a prior illness. But that physician recommended against lling the prescription unless the child signicantly worsened. Here again, the physician has indicated that the child is basically okay, but the parent continues to bring up problems with the child. Apparently, she is looking not just for a prescription (which she has)



but for the physicians support of her using the medication. I show this case here because it offers evidence that physicians commonly nd themselves in opposition to parents on the topic of antibiotics, as conveyed by his statement that NOW youre on my si(h)de.
(6.1) 151205 1 2 3 4 5 6 7 8 9 10 11 12 13 DOC: DOC: MOM: DOC: MOM: MOM: DOC: MOM: DOC: DOC: Well she- she could have the same germ, (0.4) I mean_ Okay (0.2) But=uhm (.) shes ghtin it off pretty well_ [(Id say.) [Okay. Well I just dont like to give antibiotics if we dont have to.= =Okay well good. [just=uh:m Well I HAVE- I= [huh th=huh =NOW youre on my si(h)de.=huh

In this case, as noted earlier, the parent had just resisted the treatment recommendation by the physician and raised problematic symptoms. At line 1, the physician is attempting to address these issues and implying that no treatment is recommended with But=uhm (.) shes ghtin it off pretty well_ (line 6). In these data, it is rare that parents indicate that they dislike or do not want antibiotics (But compare Britten, Stevenson, Gafaranga, Barry, & Bradley, 2004). Usually when such a statement is made, it is actually still part of a negotiation, as in I hate to use antibiotics but. . . thereby underscoring the unique need for them in the present illness context (see Extract 6.8). The parents position with respect to antibiotics here appears to be very much this, because she has continued to press for endorsement of antibiotics. Here, the physician allows insight into the battle he is engaged in by noting that NOW youre on my side (in contrast with earlier). This statement reveals precisely what we see through a close analysis of covert cases: that physicians commonly feel that they are in battle with parents over issues of antibiotic prescribing. Thus, the argument throughout this book, that parents and physicians are frequently in tacit negotiation over antibiotics, is completely transparent in this case. Although parents do push physicians, it is not the case that they do so relentlessly. Rather, they show considerable self-restraint. Earlier, I argued that this restraint reects their own orientation to the domains of diagnosis and treatment outcomes as primarily the physicians domain of authority and that pressure intrudes into this domain. In fact, there is an even more fundamental way in which pressure is problematic because it represents a threat to the physicians negative face: his or her freedom of action (Brown & Levinson, 1987). In the world of acute medicine, the physician is typically allowed (and typically takes) the lead in what should be done for the patient, even in these more modern days of consumer medicine. When parents



bring a child to the physician, they come as the layperson to the expert and imply an acceptance of these roles (Freidson, 1970a, 1970b; Parsons, 1951; Starr, 1982). When they advocate for a particular outcome, they are vulnerable to being sanctioned for violating this implicit agreement about doctor-patient roles. Moreover, they may worry about the consequences of being understood by doctors to be intruding on their medical expertise and authority in relation to prescribing decisions. We can see these issues emerge in Extract 6.2. Here, the child collaboratively completes his fathers problem presentation by adding a candidate diagnosis (line 4). The father downgrades his epistemic authority to make such a diagnosis across line 6 with I believe that it: might be I don know. In spite of this, the physicians question Now where did you:- guys have that diagnosed at. treats the parent as out of line for having ventured a diagnosis on his own if it had not been veried by a health practitioner.
(6.2) 313 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 DOC: =Im sorry tuh interrupt. Now: we can go ahead(.)/(-) (1.0) DAD: Uhm, (.) On Friday (they)/(we)/(he) had uh ( -) uhm PT1: (ear infection.) ((running water as DOC washes his hands)) DAD: (ehb ah- I-) I believe that it: might be I don know. DOC: [An ear infection? DAD: [( ) DAD: Uh huh(.)/(,) (.) DAD: So: tht [thats what I want tuh make sure that his= DOC: [Kay:, DAD: =ear is okay. DOC: O[ka:y, DAD: [He was c- complaining of pain an:- .hh #uh:m# he (couldnt m- uh:=sleep,) #u# bcuz uh thuh pain. DOC: (That) was on Friday? DAD: That was (from)/(on) Friday. DOC: -> Now where did you:- guys have that diagnosed at. (1.0) DAD: Uh: hh We dont have (it)/(-) diagnosed that its infection [(we just-) he just was complaining (uhr-) DOC: [Oh:. Okay. DOC: =of pain.

Although physicians often let candidate diagnoses pass without explicit comment or sanction, parents nonetheless treat these actions as vulnerable to physicians sanctions for treading into their territory. In short, parents and physicians alike treat any type of movement into the domain of medicine (whether diagnostic or treatment related) to be somewhat of an intrusion into the doctors territory. It is likely for



this reason that parents only rarely and in special circumstances overtly advocate for antibiotic treatment. We might expect that although we see a general respect for physician expertise in the covert behaviors, once parents are prepared to lobby for antibiotics overtly, such deference would vanish. But this is not the case. In the next section, we examine evidence that, even in cases involving overt lobbying, parents remain reticent to enter into the physicians domain of authority.

Evidence for Parent Reticence

Perhaps the most obvious evidence for the claim that parents are reticent to overtly lobby for antibiotics is the infrequency of this type of behavior. Only 8% of the cases in the Metro and Seaside data involved any direct advocacy for antibiotics at all. And even when this occurs, it is not as direct as we might have envisioned. I have no cases of a parent saying, I came to get some antibiotics or I would like you to prescribe antibiotics. The design of turns that overtly lobby is quite different from this, as we will see. There is a second type of evidence that parents are reticent to overtly lobby for antibiotics. When they raise the question of antibiotics, parents often cite nonmedical circumstances as justication for competing with the physicians medical authority a trip out of town, a birthday party, the inconvenience of a return visit, or an upcoming holiday during which the childs condition could worsen. Parents also routinely appeal to other forms of life experienceother friends or family being treated in a similar way, past experience with a similar illness, and the like. Such appeals respect the physicians medical expertise through their restriction to nonmedical accounts for their behavior. As an example, we can look at Extract 6.3, in which a mother appeals to the fact that her son is having his fth birthday party shortly and she wants him to have antibiotics so that he will be well.
(6.3) 1035 (Dr. 2) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 MOM: -> -> DOC: MOM: -> -> [Im looking for: uh uh Im looking for a miracle from you. Okay::, heh heh Martin has: uh his very r:st major: (.) ve year old birthday party tomorrow, (1.0) MOM: -> And his temperatures been:=hh UH hundred an FI:VE, ((wail)) DOC: His birthday or is [(he) going tuh huh hah hah hah hah MOM: -> [His birthday. Tomorrow:. And hes so: -> ^sick and I NEED UH MIRACLE! [hhha ha huh huh: DOC: [ A h : : : [: . MOM: [I need a miracle.



15 16 17 18

MOM: -> -> -> ->

.hh Im h- .hh I know he probably just has thuh common cold but Im like praying he has a horrible bacterial infection in his ears and YOURE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC [ha ha ha ha ha ha

Later, I will look at this interaction more closely. Here, I want to point out only that the parent is treating her behavior as special, and she licenses it with her sons birthday. Although this case is quite extreme in the degree to which the mother treads into the physicians medical territory, even here we can see that the mother orients to her behavior as problematic. In this case, the mothers appeal is still indirect as evidenced by her use of need (lines 11 and 13) and praying (line 16) and her respect for the decision as still resting with the physician (Im looking for a miracle from you. in lines 12). Third, parents generally do not advocate for antibiotics directly. Rather, they treat the topic as delicate either by talking about it indirectly or by mitigating their lobbying actions. Just as we observed cautiousness in the way parents offer candidate diagnoses, so, too, can we see cautiousness when parents overtly lobby for treatment. We discussed this with respect to Extract 6.3, and we can see this again in Extract 6.4 in an even more overt case of lobbying.
(6.4) 161303 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 DOC: Uh: the virus irritates the big tubes and theres a lot of mucus. (0.5) [So: [Because in Oct^ober he ca:me and I believe it was also for= =Something similar li[ke this or:_ [Yeah: and he got- he was on antibiotics. [Yeah:. [He- He got on antibiotics an .hh you know uh:m I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time<BT- IVE= [No. (bt) SEE= =[NEVER HAD him o^ff.= =[it DEPENDS on the kind of infectio:n. Uh: the antibiotics work against what are called .hh bacterial infections. [So for example. . . [Uh huh,

DOC: MOM: -> DOC: MOM: -> -> DOC: MOM: -> -> -> DOC: MOM: DOC: DOC: MOM:

In this case, the mother topicalizes antibiotics in the context of a brief narrative about when her son was on antibiotics previously. She states rather strongly that she feels they make him feel better: - i- in a shorter time (lines 1112). Although it is strongly worded, the turn is hedged with I just feel and that antibiotics sort of



. . . make him. . . as well as with markers of hesitation (the inbreaths, cutoffs, and self-repairs). So, her turn design and thus her preference for antibiotics are mitigated because she does not baldly state what she wants. Across the covert practices we examined, we saw physicians treating antibiotics as desired by parents, but this is rarely brought to the surface of the interaction (compare Extract 5.7). A rare instance is shown in Extract 6.5. In this case, two children are being treated. The boy is receiving prescription treatment but not for antibiotics. The girl is receiving antibiotics. The physician had previously stated that the girl would get antibiotics and had told the father that the son needed clear uids but said nothing about antibiotics, nor did she say what kind of antibiotics the girl would receive. Parents sometimes refer to antibiotics by indirectly referring to the pink medicine (a reference to the bubble gumavored amoxicillin liquid commonly used as a rst-line antibiotic in the treatment of childhood illnesses). What is interesting is how the physician designs her treatment recommendation for the girl.
(6.5) 231910 1 2 3 4 5 6 7 8 DAD: (Whas) he getting any=uh=that pink medicine, that [antibiotic, DOC: [Im gonna give him thuh white one. huh? DAD: Thuh white one? DOC: Yeah. DAD: And for Leslie?, shoul sh- she have that pink one?, (0.5) DOC: Yeah Leslie can take thuh pink one.

In response to the fathers inquiry about antibiotic treatment for his son, the physician responds with a full form answer addressing only the color dimension but not the underlying question about antibiotics. The father advocates for the pink one again with his daughter, and this time the physician agrees. She says, Yeah Leslie can take thuh pink one. With the use of the modal can, the physician underscores her action as allowing the child to take the medication and thus to her role as a gatekeeper of a desirable commodity. Across the cases in this section, I have shown three things: (1) Antibiotics are desired by parents, (2) antibiotics are a negotiable commodity, but (3) parents treat lobbying for antibiotics (covertly or overtly) as treading into physician territory. The remainder of the chapter examines the specic practices through which overt negotiation is normally done.

Practices for Overtly Negotiating Antibiotics

Across the data, we observe four main communication practices to overtly lobby for antibiotics: (1) direct requests for antibiotic treatment, (2) statements of desire for antibiotic treatment, (3) inquiries about antibiotic treatment, and (4) mentions of past experience with antibiotic treatment. As was mentioned earlier, even when



table 6.1. Directness in Overt Lobbying for Antibiotics

Requires Physician Response + Overtly Favors Antibiotics Does Not Overtly Favor Antibiotics Direct Requests Inquiries about Antibiotics Requires No Physician Response Statements of Desire Mentions of Past Experience with Antibiotics

parents overtly lobby for antibiotics, there are varying degrees of directness. This is exemplied in table 6.1, which shows the two main indices of directness evident in these four practices. Actions that both overtly adopt a stance in favor of antibiotics and require a physicians response are the most direct form of overt lobbyingdirect requests. Actions that neither overtly favor antibiotics nor require a physicians response are least direct though nonetheless explicitmentions of past experience with antibiotics. Statements of desire and inquiries about antibiotics fall closer to a midpoint in their directness because statements of desire take a position in favor of antibiotics but do not require a physicians response; conversely, inquiries about antibiotics do not overtly take a position in favor of antibiotics but do require a physicians response. In general, the more direct forms of communication are the least frequent. In what follows, we will examine each of these types of communication behaviors in some detail. Requests for Antibiotics One way that parents initiate a negotiation of antibiotic treatment is to request it. As noted earlier, this is the most direct form of overt pressure observed in these data. It both straightforwardly indicates a parents preference for antibiotics and obligates the physician to respond to the request. However, it is also the least frequent practice, occurring in only two cases in the entire corpus of 882 visits. The example shown in Extract 6.6 was previously shown as an instance of treatment resistance in chapter 5 because it is positioned following a treatment recommendation. I will not go through a full analysis of this case because it was analyzed earlier, but what is most striking about this case for the present purpose is (1) that a clear request is made at all given the scarcity of such requests in the data and (2) that although the mother does make such a request, she still orients to this action as an intrusion into the physicians domain of expertise. At this point in the visit, the doctor had already diagnosed the girl with conjunctivitis. The mother has resisted this diagnosis by inquiring about whether the treatment would be similar to the way one would treat sinuses (apparently indexing sinusitisa bacterial condition typically treated with antibiotics). At the end of the physicians response to the mothers question, the physician states that she does not want to give the girl antibiotics (lines 13) but wants to treat her eyes and give her a decongestant (lines 45). In this environment (i.e., shortly following the physicians



recommendation for an alternative type of treatment) the mother explicitly raises antibiotic treatment as a possibility.
(6.6) 2019 (Dr. 6) 1 2 3 4 5 6 7 DOC: DOC: DOC: DOC: .hh So: uh:m a- at this time I dont wanta commit er to: antibiotics. Like two weeks, or three weeks, or whatever:? .h I thi:nk Ill go ahead and treat her for the eye:s?, an I wanta give her some decongestant. Uh:m, .hh- Ya know i- She doesnt look like uh:m (.) Why dont we go ahead and try the decongestant rst.

((26 lines of discussion of treatment not shown)) 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 MOM: <I mean I usually dont- I- I usually wait to bring her in at least until [( ). DOC: [You wait unti- Yeah:, DOC: .hhh Uh:[mMOM: [Cuz its such a big deal to come here [( DOC: [Yea:h,=h I mean: if you wa:nt ya know- I mean she looks.= MOM: -> =Can I at least have thuh prescription an Ill decide -> whether or not to ll it in a couple day:s, DOC: For the antibiotics? MOM: Ye[ah. DOC: [Uh::m_ I really dont like to do tha:t, because: I mean .hh She doesnt look like she has sinusitis:. Ya know?, Uhm, if you really wanta be su:re we can go ahead and take: x rays to make su:re if its really opacify:, cause unnecessary treatment for sinusitis: she can get resistant to uh lot of those antibiotics? a lot of those bugs. I mean. DOC: .hh An:d its- its not really good for her:. (1.0) DOC: So:: we try to minimi:ze ya know- treatment until its really necessary.

The mothers request is clear. The format Can I . . . have . . . makes conditionally relevant a granting or denial. Although the format is direct on one level, the request is for thuh prescription and thus neither names the drug nor the class of treatment (i.e., antibiotics) directly. By discriminating between a prescription and the drug itself, the request formulation treats the prescription as a more minimal type of treatment outcome than she would ideally have. With at least, she treats this as less than a request for antibiotics and goes on to indicate that she is not actually requesting antibiotics at this moment but would like a prescription that she could ll when necessary. Thus, she maintains some degree of circumspection, even here.



More transparently than in cases of covert communication behaviors, in this extract there is evidence that both the parent and the physician are oriented to the prescription of antibiotics as negotiable. First, the mother requests a prescription after the physician has recommended against antibiotics (lines 12). By doing this at all, she treats the prescribing decision as something that she can inuence, rather than something that is decided on by the physician and that she must accept without discussion. Although the physician denies the mothers request, there is still evidence that she, too, orients to the prescription as negotiable: The physician does not deny the request outright but states a preference against the requested course of action (I really dont like to do tha:t,). This does not yet state that she will not but states her preference, thereby treating this as a discussion and a negotiation of a nal decision about the girls treatment. Furthermore, she provides an account for this preference that appears designed to enlist the agreement and support of the parent because: I mean .hh She doesnt look like she has sinusitis:. Ya know?, (lines 4647). Finally, she offers the parent an alternative course of actionan x-ray (line 4749)to determine if the girl in fact has a sinus blockage. This course of action was not previously recommended but is offered here as a compromisea way to investigate the condition that would mandate the desired antibiotic treatment. By offering this procedure as an alternative (explicitly marked as such with if you really wanta be su:re), the physician treats the decision on a course of action as one that is being made jointly by the parent and physician rather than one that is handed down from the physician to the parent. By adding if you really wanta be su:re, the physician overtly indexes her response as acquiescing to the parents pressure. Although requests are rare, that they happen at all is another excellent piece of evidence for the orientation by parents and physicians alike to the negotiability of treatment and of antibiotics in particular. This case also demonstrates that even in very extreme cases like this, parents still orient to norms of behavior that give physicians primary responsibility for and authority over treatment decisions. Stating a Desire for Antibiotics Parents more frequently state a desire for antibiotics than ask for them, but statements of desire are also quite rare. They are relatively direct because they convey pressure by overtly indicating a treatment preference. Unlike requests, though, they do not make a granting or denial by the physician conditionally relevant. For this reason, statements of desire for antibiotics are somewhat less direct than requests. Despite this, physicians treat them as applying signicant pressure for antibiotics (much like requests do). We can see this illustrated in Extract 6.7, part of which was shown in Extract 6.2. Here, the mother asserts that she is hoping for antibiotics very early in the encounter.
(6.7) 1035 (Dr. 2) 1 2 DOC: MOM: Are we ready::. Hi: Doctor Sa:[nders,

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40


DOC: DOC: MOM: DOC: MOM: DOC: MOM: DOC: MOM: -> -> DOC: MOM: -> -> MOM: -> DOC: MOM: -> -> DOC: MOM: MOM: -> -> -> -> DOC: MOM: MOM: DOC: MOM: DOC: MOM: DOC: DOC:

[W- well hi:. How are you guy::s.=h .h[h [Well:We have two victims he:[:re. [Two victims. We added one when I picked [Sara up from schoo:l. [Thats oka:y?, Actually: we wouldnt be here Doctor Sanders except=h Youre ta[king the trip. huh huh huh [Im looking for: uh uh Im looking for a miracle from you. Okay::, heh heh Martin has: uh his very r:st major: (.) ve year old birthday party tomorrow, (1.0) And his temperatures been:=hh UH hundred an FI:VE, ((wail)) His birthday or is [(he) going tuh huh hah hah hah hah [His birthday. Tomorrow:. And hes so: ^sick and I NEED UH MIRACLE! [hhha ha huh huh: [ A h : : : [: . [I need a miracle. .hh Im h- .hh I know he probably just has thuh common cold but Im like praying he has a horrible bacterial infection in his ears and YOURE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC [ha ha ha ha ha ha ha ha. [Ha ha ha ha ha ha ha ha .hh But I know its not gonna h(h)appe(h)n. .hh #But I just said (.) you know what, .hh (.) Before I cancel this party, (1.0) .hh Tlk=[hahhh. [#Yeah:,# I- I- i- - hh. I- Im afraid I cantAhh[h h a h a ha h a ] hih hih hih(s). [d(h)o th(h)uh mir(h)acle.] #I mean Id ah# cause he- I think hes gonna probably come down with the u:, ya know?,

The mother states that she is praying he has a horrible bacterial infection in his ears and YOURE GONNA C:URE IT WITH TWO DOSES OF ANTIBIOTIC (lines 2729). That the mother is praying for this outcome strongly communicates her desire for antibiotics. But she differentiates between what she wants and the facts of her sons illness. This is one way in which she maintains respect for the physicians authority in the matter. Her desire for a bacterial diagnosis and antibiotics is couched as something that has yet to be decided (with YOURE GONNA). Additionally, by



prefacing her preferred diagnosis (a horrible bacterial infection) with a more likely alternative diagnosis, he probably just has thuh common cold (lines 2627), she treats her desired outcome as unlikely. This recognizes both that her son may not be able to be effectively treated with antibiotics and that she understands this. With her subsequent admission But I know its not gonna h(h)appe(h)n. (line 31), she treats her desire for a bacterial diagnosis and antibiotics as being in the doctors hands, which is, in turn, based in facts. But she nonetheless communicates her position as one strongly in favor of antibiotics. The parents animated talk also conveys her orientation to the delicacy of her action. Haakana (2001) shows that laughter is commonly used when patients report misdeeds or otherwise engage in delicate actions. Here, the misdeed appears to be the very strong pressure she is applying for antibiotic treatment, rather than allowing the physician the space to arrive at his own recommendation. Besides laughter, which inltrates much of this interaction, there are also large shifts in amplitude (e.g., 29 31) and large pitch uctuations (e.g., lines 2122). These may, like laughter, convey the parents orientation to her actions as problematic. The parent also displays an understanding of the differences between viral and bacterial diagnosis and the fact that bacterial diagnoses are treatable with antibiotics. This understanding is not uncommon among parents, especially in the Seaside data from which this extract comes. In response, the doctor both denies the treatment and rejects the diagnosis being suggested. First, he says, Im afraid I cant- d(h)o th(h)uh mir(h)acle. This matches the affective tone of the parents talk and is tted to her earlier I NEED UH MIRACLE! (line 22). The declination of ability to do the miracle conveys the doctors position as unable to cure the illness. But we can also observe that the physician treats miracle here not so much as indexing antibiotics as a miracle drug but as a biblical miracle. By shifting his orientation to miracle, the physician avoids a head-on conict with the parent while simultaneously conveying that the boy does not have an ear infection and that he therefore will not prescribe the desired two doses of antibiotics. That this is implicit in this turn is supported by the next turn, where the doctor suggests, I think hes gonna probably come down with the u:, ya know?, (lines 3940). Here the doctor implies that the mothers prefatory diagnosis is correct and suggests that antibiotics would be inappropriate. This implicitly rejects the mothers preferred diagnosis. The doctor also works to communicate this diagnosis as, so far as possible, afliated with the mother. This is done both affectively and by offering this position in close proximity to the mothers own downplaying of the likelihood of the miracle to occur. Although there is an effort to acknowledge the nal decision as the doctors, the mother nonetheless clearly states her preference, and the physician addresses that. That this appears to be working to pressure the physician is most apparent across lines 1516/18/2122, where the mother offers her account for wanting her son to have an ear infection and wanting antibiotic treatmentthat he is having a birthday party. This case also illustrates that in the very opening of the visit, even prior to a move by the physician to establish the reason for the visit, parents can initiate a negotiation of antibiotics. As was mentioned at the outset of this chapter, overt negotiation behaviors are unlike the covert behaviors we examined in chapters 25 because



they are not restricted to particular phases within the visit. Statements of a desire for antibiotics can occur at virtually any point in the visit, though they are more likely to occur, as most overt forms of negotiation do, later in the visit. We can see an example of this in Extract 6.8. The mother has already resisted the physicians treatment recommendation and resists it again in lines 4/6. With the and preface, she builds continued resistance in lines 68 with a statement that supercially opposes antibiotics but in this case appears to be building a contrast between not liking antibiotics generally but X. The physician works to afliate with the dimension of the parents resistance that is opposed to antibiotics in building his defense, stating again that antibiotics are unnecessary and then, as we have seen is common in responding to treatment resistance, expanding on the diagnosis. This time he details the prognosis of the illness as he has seen it with other children.
(6.8) 161303 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 DOC: Uh: and ^usually they do just ne with: just a good (.) decongestant cough medicine. (0.2) Well Ive been [giving him a (.) decongestant= [(and uh:) =[cough medicine<an I.hh I really dont like =[Yeah. antibiotics on him? [but [No that- thats the point is: if: ya know he had like uh .hh uh: some kids come in with uh strep throa:t and then- (0.4) they need the antibiotic_<but- this thi:ng uh: (0.8) its it just doesnt_ they dont need it.=hh Okay:, .hh Its a SElf-ROOted (0.4) illne:ss, (0.4) uh: the virus: uh: starts_ (0.5) uh:- more or less up here and then it works dow:n, Okay, So: they ^tend to uh (0.8) start with a sore throat, and then they work into hoarseness and coughing. Okay, Uh: the virus irritates the big tubes and theres a lot of mucus. (0.5) [So: [Because in Oct^ober he ca:me and I believe it was also for= =Something similar li[ke this or:_ [Yeah: and he got- he was on antibiotics. [Yeah:. [He- He got on antibiotics an .hh you know uh:m







32 33 34 35 36 37 38 39


I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time<BT- IVE= [No. (bt) SEE= =[NEVER HAD him o^ff.= =[it DEPENDS on the kind of infectio:n. Uh: the antibiotics work against what are called .hh bacterial infections. [So for example. . . [Uh huh,

Following this, the mother again resists, using a form we will see discussed later on: mentioning past experience with antibiotics. Here, I want to focus on the mothers comment that I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter time (lines 3233). This is a clear statement in favor of antibiotics when the parent and physician are negotiating what kind of treatment is to be given. On the surface of this interaction is also an account for why the mother wants antibiotics: because she thinks they will help her son feel better sooner. So we gain insight into why parents may pressure physicians for antibiotics in situations that are clinically inappropriate. In addition, we have another instance of a parent who crosses the implicit agreement between physicians and parents/patients to respect the physicians cultural authority and role as medical expert. This section discussed a second practice through which parents overtly lobby physicians for antibiotics. Statements of desire for antibiotics were shown to be, on the one hand, less direct than requests for antibiotics, but on the other hand, they are nonetheless quite a strong form of pressure because they make the parents preference so explicit. We also observed that, once again, even in these fairly high-pressure environments, parents are oriented to their actions as treading into medical territory and show this through downgrading and mitigating their statements. Inquiring about Antibiotics One of the more common ways that parents lobby for antibiotics involves inquiring about them. Parents who ask about antibioticsexplicitly introducing them into the interactioninvite discussion of them and require a physician to respond to this treatment possibility. In all of these ways, such inquiries come to exert pressure on the physician. Inquiries occur in all phases of the visit. As an illustration of this, consider one rather unlikely place: during the history-taking phase. It is an unlikely place because, as we saw in chapter 3, this activity is generally comprised of physician questions and parent responses. But even during history taking, such inquiries occur.
(6.9) 221814 1 2 3 4 DOC: MOM: Any vomiting or diarrhea?, (0.2) Uhm no. (0.8)

5 6 7 8 9 10 11 12 13 14 15



She has been [like picking at her ea:r though. [(Yeah okay_) Okay. (2.0) Do you think she needs antibiotics?= =Let me see. (0.5) #Okay.# Yeah:, (0.5) Has she have any serious medical problems since birth. O^h: n^o:. nothing like that- Nothing at all.

The physician is in the middle of taking the girls history at line 1. The mother answers the question and adds an additional problematic symptom about the childs ear. Following the physicians acknowledgment of this, the mother inquires about antibiotics (line 9). The physician defers answering with Let me see (line 10), and then following acceptance of that (line 12), he proceeds with history taking (line 13). Inquiries such as this one are hearable as lobbying for antibiotics, even though they do not overtly take a position in favor of antibiotics, because they convey what the parent is thinking about. They are typically treated as revealing a parents expectation, hope, or desire, even though this is not stated. This becomes more transparent when the inquiries are positioned, as they most commonly are, following a treatment recommendation or at a point in the encounter where a nonantibiotic treatment recommendation has been implied (e.g., if the child had ear pain and the physician indicates throughout the examination that there is no evidence of any infection) or stated. Because of this positioning, these inquiries can be understood as resistant to the actual or implied treatment proposal. An example was shown in Extract 5.5, part of which is reproduced here in Extract 6.10. The physician has explained to the father that his daughter has a viral infection and has recommended watching it. This recommendation is essentially a no-treatment recommendation. After answering several questions related to when the child can return to school and other treatment, the physician returns to the childs symptoms and treatment. In line 4, the father inquires as to whether the doctor is going to prescribe antibiotics. The question is designed in a way that shows that the father is oriented to this as the doctors decision. But following the doctors explanation, the father, at least retroactively, shows his question to have also been an indication of his own preference, with the account being that they would hafta bring her ba:ck, (line 2122).
(6.10) P201 (Dr. 7) 1 2 3 4 5 6 DOC: She: doesnt have anything right no:w, any symptoms of mucus or vomiti[ng thats contagious. DAD: => [Are you gonna give her ana- antibiotics? DOC: Yeah- uh No: I dont have anything tuh treat right now for antibiotics. Her ears look really goo:d, .hh she has



7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

no sign of bacterial infection right no:w?, .tlkh and thats (what) shed get antibiotics fo:r. .hh So uh lotta times you can start out with uh virus like uh co:ld, (.) .h an:d if you- it goes on for uh while #uh:# bacteria (should) set in you can get uh secondary bacteria infection? and thats when you need antibiotics. DOC: .hh #But- y-# otherwi:se: since she doesnt have any source of an antibi- of uh bacterial infection?, that=uh=we just watch her. (0.3) DOC: right now. <Her ears look really good, <an her tubes are in too:. DOC: .h And theyre not draining or any[thing. DAD: => [It just means that=> ya know if she gets another fever we hafta bring her => ba:ck, DOC: .hh Well what Ill do is she might still get uh fever: in thuh next couple uh #da:ys.# because: .h thats th way viruses wor:k?, you can have- you know (how have=you) if you have uh co:ld, you can get a fever for uh few da:ys? DOC: .hh And tha:t Since shes o:lder:, .h if somethings #uh# she (would) com[plain ( ) thuh symptom,= GIR: [( ). DOC: =then she would need tuh come back. DOC: .hh But what you ca:n do i:[sGIR: [Guess what. DOC: DOC:

At the beginning of this extract, the father has not yet accepted the physicians proposal of over-the-counter (i.e., nonprescription) treatment with a token such as Okay., Alright., or Thats ne. as is typically the case when parents agree to physician proposals of treatment. This may already be passively resisting the recommended treatment (Heritage & Se, 1992). However, in asking about antibiotics, the parent shifts to actively displaying his position in favor of antibiotic treatment. As mentioned earlier, raising antibiotics at this point in the interaction suggests resistance to the alternative nonantibiotic treatment plan that was suggested. Second, this inquiry is positioned in what appears to be midturn during the physicians explanation of the childs symptoms (line 4). In this sense, the father treats a no-problem explanation (lines 12) as problematic in itself. Finally, by inquiring about an alternative treatment, the father explicitly displays himself to be disinclined toward the previous no-treatment recommendation. After answering the parents question and providing an account for the negative answer, the physician reasserts that since she doesnt have any source of an antibi- of uh bacterial infection?, that=uh=we just watch her.. At line 15, the father, having passed on several prior opportunities to accept the physicians recommendation, again does not respond, and in the face of 0.3 seconds of silence, the physician



adds an increment to her turn and again offers evidence in support of her position against treating the girl: Her ears look really good. In partial overlap with a similar statement in line 20, the father offers what is effectively an account for his withholding of acceptance: It just means that- ya know if she gets another fever we hafta bring her ba:ck, a statement that reects an underlying assumption that another fever would be bacterial and in need of treatment. Moreover, similar to Extract 6.3, the fathers account for why this is problematic is based in nonmedical contingencies (i.e., inconvenience) rather than in biomedical issues. Returning to the doctor is an inconvenience, and this inconvenience is offered as an account for his implied preference for receiving antibiotic treatment during this encounter. Once again, this also displays the fathers orientation to antibiotic treatment as negotiable because his raising inconvenience as an account shows him to be oriented to his preferences as potentially affecting the nal prescribing outcome and thus to the decision as not wholly based on medical evidence. Following this sequence, the physician launches what appears to be a solution to this problem with What Ill do is. . . and in line 32 What you ca:n do i:s-. But after an interruption in which the physician interacts with the child, she returns to suggest that the father could in fact call and would probably not have to bring his daughter back, concluding with: as long as she looks this goo:d, an no other symptoms, .hh ya know we just- well watch her. (data not shown). The question in extract 6.10, like Extract 6.9, is designed to prefer a Yes. (For relevant discussions of question design, see Boyd & Heritage, 2006; Pomerantz, 1984, 1988.) If a treatment recommendation has been made or implied, as is the case in Extract 6.10, it is much more common for these inquiries to prefer a No. But both designs accomplish the same action in terms of raising antibiotics for discussion, requiring a physicians response, and conveying a preference for antibiotics. The No preference structure may do more to orient to the physicians medical authority and to his or her conveyed trajectory toward a no-antibiotic treatment plan. We can see an example of this turn design in Extract 6.11. Here, after the doctor has diagnosed a cold and suggested Triaminic or Robitussin (both over-the-counter cold remedies), and furthermore after the doctor has stated that the illness should just go away by itself (lines 13), the parent questions the need for antibiotics (line 42).
(6.11) 2081 (Dr. 8) 1 2 3 4 5 6 7 DOC: #An thuh- (0.2) an thuh cough and that kind of stuff- it should hopefully just go away [by itself in the next #wee:k,# [( ) .hh And: you should- uhm_ (0.2) if youre feeling really ba:d, (kh ta-), (.) call your (o ) cuz. . .


((30 lines discussing mothers illness not shown)) 38 39 DOC: Tell her: (.) tuh pinch it. ((re: bloody noses)) (0.2)



40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63


Like h pinch it as much as she ca:n for ve minutes an then= MOM: -> =So [he dont even need any antibiotics then. DOC: [itll go awa:y, DOC: #Hm mm[:.# MOM: [Okay so Im may need that in writing. (.) MOM: <[I mean- [I believe you and youre thuh doctor= BOY: [( ) DOC: [-Hh .hh MOM: =[(but she-). DOC: =[.HH for thuh:: for thuh [what. Mmhmph. MOM: [(she -) MOM: I dont know [but she just- thats what she said. DOC: [hm hm MOM: He needs antibiotics. [and (Im like okay.) DOC: [Ha ha ha ha ha ha ha [ha ha ha ha ha ha ha ha ha ha ha ha ha MOM: [( ) She wont listen to me so- and [Im leaving m with er] for a week= DOC: [ha ha ha ha ha ha ha ha] MOM: [so shell be paranoid. DOC: [h hu .hh DOC: AH! ha ha huh. .h

The mother rst seeks conrmation that her son does not need antibiotics (line 42). Despite the built-in structural preference for agreement, this actionlike those in Extracts 6.9 and 6.10reopens a discussion of treatment and turns the talk to antibiotics specically. Additionally, the turn initial So treats this as the upshot of the doctors earlier explanation, which did not mention antibiotics, and thus displays the mothers orientation to the doctor as having mentioned all of the medication likely to be suggested. After the doctor answers the mothers question in the negative with a simple #Hm mm:.# (line 44), the mother indicates a desire for antibiotics even more strongly by requesting the physicians recommendation in writing (line 45). This request is done on the behalf of a third party (the grandmother), which works to distance the mother from the challenging nature of the action. However, it nonetheless advocates for antibiotic treatment. Subsequently, although the physician treats this challenge in a joking way, it is notable that the mother does not afliate with this stance by laughing with the physician (Jefferson, 1979, 2004). In this case, we again see the mothers orientation to her preferences as having legitimate impact on the physicians prescribing decision because she orients to her position as potentially altering the treatment outcome. We observed this pattern across the covert practices shown in previous chapters. In these more overt practices, it is all the more transparent: Parents view themselves as both being able and having a right to participate in their childs treatment, even if they also display an orientation



to this as a delicate action, but this is counterbalanced with concerns of treading into physician territory. Mentions of Past Experience with Antibiotic Treatment A nal relatively common practice parents use to explicitly communicate pressure for antibiotics is to mention a past experience with antibiotic treatment. This behavior, like inquiring about antibiotic treatment, is less direct than requests or statements of desire. In fact, mentioning a past experience with antibiotics is arguably the least direct of the methods discussed because, although it conveys a position in favor of antibiotics and overtly raises antibiotics for discussion, it neither explicitly states that preference (and therefore can be treated as simply providing information) nor makes a physicians response conditionally relevant. Nonetheless, physicians routinely treat parents who mention past experience with antibiotics as lobbying for antibiotic treatment. For example, in Extract 6.12, the mother has already presented the childs problem (data not shown), and in line 1 the physician is initiating the physical examination. It is during the physical examination that the mother mentions her prior experience with antibiotics.
(6.12) 1113 (Dr. 1) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 DOC: MOM: So:- Lets take uh listen to er che:st, (Alright), (.) MOM: -> Remember she- she:=uhm_ had something like this: in -> December? DOC: Uh huh, (0.5) GIR: Hhh.=.h MOM: -> (n) She was on an antibiotic. (1.0) DOC: ^Doo doo. ((to girl)) GIR: Ksh:::::, uh. DOC: #Yeah:# Well I think that she probably got:=similar type of thi:ng, ya know [some sort of a secondary-= MOM: [Mm hm:, DOC: =.hh uh: respiratory infection in her ches:t, like #uh bronchitis an_#

The mothers assertion in lines 45/9 that her daughter had a similar problem before and was treated with an antibiotic conveys a position that the prior treatment was successful and that her daughter is in need of the same treatment again this time. Initially, such behavior might not appear to be lobbying for antibiotics. But no statement from the mother was due here. It is a self-initiated report that poses the puzzle of what the mother is doing with this. Thus, although the report might not appear to be lobbying for antibiotics, it is doing something special by virtue of its position in this



sequential context. Also, mentions of precedents have previously been found to be a resource for pressuring interlocutors. For example, Kleinman, Boyd, and Heritage (1997) and Heritage, Boyd, and Kleinman (2001) show that in interactions between physicians and physician-insurance reviewers, tympanostomy tube surgery was more likely to be approved if the physician cited a prior use of tubes. In the present case, by reporting this previous instance of antibiotic use, the parent implies that this illness is similar to that one and thus that this treatment should also mirror that. With this action, the parent also treats this information as potentially inuencing the doctors treatment recommendation with respect to antibiotics. The physician initially does not respond but examines the girl (lines 1012). Following that examination, the physician agrees with the parent and suggests that this illness is in fact a similar type of thi:ng, making antibiotics an appropriate treatment (line 13). Subsequently, the physician goes on to prescribe antibiotics. Parent mentions of past experience also sometimes invoke their own illnesses. We saw an illustration of this in Extract 5.7 (reproduced here in 6.13). Having already fully analyzed this case in chapter 5, I will not redo that here. What I want to focus on here is that part of the parents resistance is to state how antibiotics were helpful in his own past experience, and because he believes the illnesses of his sons are the same as his own illness, the father conveys that antibiotics are his preferred treatment and thereby pressures the physician to prescribe antibiotics.
(6.13) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: DOC: DAD: -> I th:ink from what youve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I dont think antibiotics will ki:ll, (0.2) Well[Thee other[( ) >Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d Ive been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm =sore [for (awhile- et- that) w:eek. [Uh huh Right, an:d (.) I start taking thuh antibiotic (0.5)

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: ->

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38


INF: eh he ((cry)) DAD: -> Yesterday. DOC: Right, DAD: -> And it (.) seemed to take care of the problem. DOC: [(Well) thats why were doin a throat [culture. DAD: [( ) [Yeah. DOC: [is TUH SEE if they need antibiotics. DAD [( ) Yeah yeah. (0.2) DOC Cause <I dont th::ink they do. DAD O[kay, DOC: [Now if you (.) absolutely insist_ I will give you antibiotics_ but [I dont think thats the right= INF: [#eh::# medicine for em,

This practice may further emphasize the parents role in his sons health care because it treats the childrens personal history as unique and potentially distinct from the more generic illness histories that physicians may use to gauge whether a child requires antibiotic treatment. For example, a parent whose child routinely moves from viral colds to ear infections will cite this as sufcient reason for the physician to prescribe the medication when the child has only a viral cold. Although both the physician and parent may say that antibiotics are generally not needed for such viruses, when the parent introduces this aspect of the childs history, they claim that this makes this childs needs different from those of the typical child with a cold. Another example was shown in Extract 6.8. There the mention of past experience with antibiotics was positioned after a nonantibiotic treatment recommendation and was part of resistance to this recommendation. Earlier, we focused on the parents statement of desire for antibiotics, which came just following her mention of prior experience with antibiotics some months previously. But, we can also observe that with this mention she suggests that antibiotics are appropriate and desired this time.
(6.14) 161303 1 2 3 4 5 6 7 8 9 10 11 MOM: DOC: Okay, Uh: the virus irritates the big tubes and theres a lot of mucus. (0.5) [So: [Because in Oct^ober he ca:me and I believe it was also for= =Something similar li[ke this or:_ [Yeah: and he got- he was on antibiotics. [Yeah:. [He- He got on antibiotics an .hh you know uh:m




12 13 14 15 16 17 18 19


I just feel that antibiotics sort of .hh -:make him feel better: - i- in a shorter [time<BT- IVE= [No. (bt) SEE= =[NEVER HAD him o^ff.= =[it DEPENDS on the kind of infectio:n. Uh: the antibiotics work against what are called .hh bacterial infections. [So for example. . . [Uh huh,

Here, we see evidence that the mothers mention of past experience with antibiotics is lobbying for antibiotics through what she does immediately subsequently: state a desire for antibiotics by saying how she feels that antibiotics make him feel better sooner. We have now examined four ways that parents work to lobby physicians for antibiotic prescriptions. These formulations vary in their directness, but all explicitly raise antibiotics for negotiation, and all convey the parents position as in search of antibiotics. Through the action of lobbying for antibiotics, parents adopt a stance toward the prescribing decision as one that resides not solely with the physician but rather one that is properly shared by the parent and physician and can thus be negotiated in and through the interaction.

Physician Response to Parent Pressure

Across the interactions we have examined in this chapter, we have seen through the ways that the physicians respond to parents overt lobbying for antibiotics that they hear parents as pressuring for, and initiating (or even escalating) a negotiation for, antibiotics, although the exact response is, of course, related to the parents form of overt lobbying. For instance, in Extract 6.5, after the parent advocated for pink medicine, the physicians recommendation for antibiotics was designed as accommodating the parent with her use of can have. In Extracts 6.6 and 6.10, the physicians displayed their orientations to being in negotiation with parents through the way their denials were formulated, as well as through their defenses. In Extract 6.7, the physician formulates himself as regretful that he may not be able to perform the miracle of curing the child. Finally, in Extract 6.8 (shown again in 6.14), the physician rejects the parents account for wanting antibiotics (that they would help him feel better faster). Of course, as with other practices, perhaps the best evidence that overt negotiation inuences diagnostic and treatment outcomes is cases where this is transparent. One case was analyzed in detail in Extract 5.9ad. There, after multiple types of resistance, including very strong overt lobbying for antibiotics, the physician agrees to prescribe antibiotics, going against her earlier diagnosis of a viral infection and going against her earlier recommendation against antibiotics. In chapter 5, we were focused on treatment resistance. Here, we can observe that the change to offering antibiotics occurred after the parent shifted to overt lobbying for antibiotics.



We can observe a similar shift in Extract 6.15. Here, the physician has done a throat culture to test for strep throat. If the culture shows bacteria, then antibiotics would be appropriate. When physicians perform a strep test, they sometimes prescribe antibiotics with the idea of stopping them if the results are negative. Alternatively, they provide a prescription only if the results are positive. So although the results are not back, there is nonetheless a prescribing decision to be made. In lines 12, the physician tells the mother to call tomorrow about eleven:, when she can get the results and the prescription if one is necessary.
(6.15) 1061 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 DOC: DOC: MOM: DOC: MOM: DOC: DOC: Mlk=Okay, so well culture her, n=you call tomorrow about eleven:, .hh= =Alright [so were not starting her= [If her culture is=[ o n a n y t h i n g, [today?, =[were not gonna start i- [.hh Well I tell you what I cn do is let me give you uh sample of something just so we dont (miss) uh day. .hh In case: ya know it is positive then we= =ya know well have ( ).

-> -> => => => =>


In response to the physicians directive (lines 12), the parent requests conrmation that were not starting her on anything, today?, (lines 4/6). This works to renew a discussion of medication. In response, the physician offers what he frames as a concession with Well I tell you what I cn do and then gives the mother samples of an antibiotic (line 9). As was the case in Extract 6.14, a physicians prescribing decision is changed from no antibiotics to antibiotics in and through the interaction with the parent. In a nal example, the mother inquires about antibiotics at a point where the physician has indicated that the only treatment needed is for the childs wheezing.
(6.16) 181412 1 2 3 4 5 6 7 8 9 10 DOC: Cause hes wheezing. So go every four hours, and then (d) you need cough medicine or you have some at home. Uhm (0.5) I dont know exactly. No. I havent had any for a while. Okay so I get you some anyway. Okay, (1.0) Would antibiotics be:_ Uh::: he=u=Does he have any temperature,=Yeah




11 12 13 14 15 16 17



he got a little bit tiny_ Yea:[h, [Little tiny bit. So I go ahead put him on. I really want to make sure we get this out of his system before he goes back to school Monday, (0.2) Yeah_

The physician is detailing the treatment for the childs wheezing: so go every four hours (with a breathing machine that the family has at home) and then recommends cough medicine. She adds, or you have some at home. indicating either an over-thecounter cough medicine or a prescription medication that she has prescribed for this child previously. From the physicians subsequent offer to get you some anyway. it appears that it is a prescription medication. In any case, the parent answers but then asks about antibiotics as an additional type of treatment not previously offered or implied: Would antibiotics be:_. The physician initially indicates predisagreement with Uh::. The delay of an answer is hearable as predisagreement because it defers agreement and implies that agreement is contingent on that sequence. (See Schegloff, 2006, for a discussion of pre-second insert sequences.) However, after determining that a slight fever has been present, the physician does give antibiotics (line 13). Although it is not framed as agreeing to do this, the physicians recommendation is directly responsive to the parents inquiry, and all evidence suggests that an antibiotic was not otherwise going to be recommended. Thus, as in other cases we have seen, overt lobbying can successfully be used to negotiate a change in prescription from no-antibiotics to antibiotics.

This chapter has focused on the least common form of parent pressure for antibiotics: overt pressure. As with all other types of negotiating practices, parents do not overtly pressure physicians very often. But when they do, this inuences physicians and is therefore an important type of parent behavior. This chapter has demonstrated that overt pressure for antibiotics is typically done through one of four interactional practices and that all are treated by physicians as lobbying for antibiotics and all initiate (or escalate) a negotiation of antibiotics. In addition, though, this chapter showed that parents do not treat negotiation, particularly in this overt form, lightly. Rather, they treat pressuring physicians as delicate business and do so in ways that specically maintain an orientation to the physicians authority over the medical (and particularly the diagnostic and treatment) domain. This is an important dimension of the interaction because it helps to account for why overt pressure is relatively uncommon, whereas covert pressure is relatively more common and is therefore the primary vehicle through which parents convey their treatment preferences, as well as other concerns that affect diagnostic and treatment outcomes. This chapter also demonstrates that parents have interactional resources to pressure physicians for antibiotics during any activity within the medical visit. In this



chapter, we saw overt negotiation used during the opening of the visit prior to the reason for the childs visit, during the middle of history taking, during the physical examination, and, most commonly, following the treatment recommendation. Taken together, overt negotiation practices and covert negotiation practices offer a wide array of weapons with which to battle physicians who do not believe a child has a bacterial diagnosis or needs antibiotic treatment. And physicians do, as we saw quite explicitly in Extracts 6.1 and 6.13, perceive themselves to be in battle with parents at times over this issue. We might begin to wonder what physicians can do about this problem and whether our only hope is parent education. The next chapter turns to the issue of physician behavior.

Physician Behavior That Inuences Parent Negotiation Practices

hysicians seeing a child for whom antibiotics are not an appropriate treatment frequently face a difcult task: to secure parent acceptance of a no-problem diagnosis and a nonantibiotic treatment recommendation. The task is difcult because it denies parents a treatment that they may have wanted. To make matters worse, this outcome is vulnerable to being understood as indicating that parents judgment was awed when they decided to visit the physician. Thus physicians also need to balance maintaining visit legitimacy with appropriate medical decision making. Because a lack of prescription treatment is, in itself, a threat to the legitimacy of the visit, each physician behavior that indicates the physicians trajectory toward a no-antibiotics diagnosis and treatment is potentially threatening to visit legitimacy as well. But as Maynard (2003) has documented with respect to bad news, one very good resource for securing parent acceptance of bad news is to forecast the news in order to assist in the recipients realization of the news. It is ironic that parents treat a viral diagnosis as bad news because in many ways, physicians are delivering good news: Children who are not in need of antibiotics generally have self-limiting viral infections that will resolve in 7 to 10 days. But as we have seen in earlier chapters, telling parents that their child must simply be sick and that there is no cure is generally viewed as bad news. For these reasons, interactional resources that forecast that a child does not need antibiotics and help physicians maintain the legitimacy of the childs visit are important. This chapter will discuss several types of physician resources and focus on three: online commentary, presentation of the diagnosis, and presentation of the treatment recommendation.



Resources That Forecast the Diagnosis

Physicians rely on a variety of resources to forecast that the child has a viral or noproblem diagnosis. Perhaps one of the earliest resources for this is to reject a parents bacterial candidate diagnosis. Examples of this behavior were shown in chapter 2 in the discussion of immediate responses to candidate diagnosis problem presentations. For instance, in Extract 7.1 the mothers candidate diagnosis of sinus infection (line 10) is rejected with Not necessarily:, (line 13), despite the fact that the physician has not yet begun any verbal or physical examination of the child.
(7.1) SG615 [shown earlier in Extract 2.5; 2.13] 10 11 12 13 14 15 16 MOM: DOC: MOM: DOC: So I was thinking she had like uh sinus in[fection= [.hhh =er something.= =Not necessarily:, Thuh basic uh: this is uh virus basically:, an=uh: .hh (.) thuh headache seems tuh be:=uh (0.5) pretty prominent: part of it at r:st uh: (0.2) .hh

The countering behavior works to shape the parents expectations that the childs illness will not be diagnosed as bacterial. But it does little to legitimate the childs visit (and thus the parents decision to schedule the visit). Even when physicians suggest an alternative diagnosis, this may still do little to validate the legitimacy of the visit if the proposed diagnosis is a no problem one because this implies nontreatability. When appropriate, mentions of diagnostic tests that can be done may be at least partially validating. An example of this is shown in Extract 7.2. In this case, following a bacterial candidate diagnosis but before any verbal or physical examination, the physician rejects the diagnosis of strep and counterproposes uh viral process (line 57).
(7.2) SG316 [full presentation shown in 2.6; also shown in 2.14] 31 32 33 34 35 36 MOM: -> [(I-) I thought (0.5) maybe I better just- <I dont -> know if ya know strep has secondary er anything like -> that I wasnt sure.But he hasnt had thuh fever er thuh nausea er anything that he[s had before. DOC: [O:kay:, DOC: .hh (Goo:d?,)

((16 lines not shown including joking about BOY having day off but not MOM)) 53 54 55 56 DOC: DOC: => => => O:kay:, .hh Well:, (.) o:ne good thing is: that- uhm (0.5) strep infections:- respond really well tuh amoxicillin. .hh so wh:ile he may not have strep any more (.) he could still



57 58 59 60 61 62 63 64 65 66 67

DOC: => have- uh viral process going on, he could still have just => residual sore throa:t, .h[h dry weather kind of things, .hhh MOM: [(Okay.) DOC: Uhm: besides having an actual infection so we can always look at those issues, .hh an then if you want we can also just retest his throat. (.) DOC: An make sure theres no more strep there too. (.) MOM: Well (you it) kinda depends on what you- what you [think. DOC: [Mkay,

After countering the parents concern of strep (lines 5456), including accounting for it (5455), as well as counterproposing an alternative (lines 5758), the physician offers to retest the boys throat (lines 6164). Testing the throat may help validate the legitimacy of the visit. However, the presentation is critical. Here, it is framed as a concession to the parent with if you want, which may reduce its success in terms of validation. We can see that the parent, with this formulation, treats it as a sanction of her intrusion into the physicians domain of authority with it kinda depends on what you- what you think. (line 66). Outright rejections of diagnostic proposals during history taking also forecast that a bacterial diagnosis is unlikely. We can see an example of this in Extract 7.3.
(7.3) SG506 [shown as 3.21] 8 9 10 11 12 13 14 15 16 17 18 GMA: -> -> -> GIR: DAD: GIR: DOC: => => So thats why I=uhm (1.5) we decided to bring im in because (0.4) with thuh temperature. (.) (I know ) somet[hings gonna be wrong. [Daddy. [Stop that. [( ) Well actually uh regular cold can give you uh high fever for: three day:s. (0.4) DOC: => As long as they act okay then actually its- it: => (0.8) may go away by itself, (its) just uh virus.

Just prior to this, the grandmother stated that the boy commonly gets ear and throat infections. She provides support for her possible diagnosis of ear or throat infection by citing his temperature. With these behaviors, the grandmother adopts a stance toward the illness as treatable and thus, at least for the physician, as bacterial. It is this connection that the physician quarrels with in her response, which contends that uh regular cold (line 14) can produce the same symptoms and (its) just uh virus. (line 18). Here again, the physician is denitively forecasting a nontreatable diagnosis. On the one hand, this is useful because it conveys to the parent (or in the last case, the



grandparent) early on that the child may have a no-problem, no-treatment outcome on the way. However, what remains as an interactional problem is that in forecasting this the physician simultaneously threatens the legitimacy of the childs visit and the childs treatability. Both of these issues can motivate parents to pressure physicians for antibiotics. The remainder of this chapter will look at three behaviors that deal with these issues, though the balance of forecasting, legitimating, and treatability is slightly different with each. First, we will examine physicians use of no-problem online comments1 (Heritage & Stivers, 1999). Second, we will examine alternative practices for delivering the viral diagnosis. And third, we will examine two practices for delivering nonantibiotic treatment recommendations. Online Commentary In earlier work, online commentary was dened as an action that describes or evaluates the signs physicians are encountering during the physical examination (Heritage & Stivers, 1999). This behavior affords the parent some access to the physicians diagnostic reasoning. However, although they are relevant to, and may foreshadow, the physicians nal diagnostic evaluation, online comments are quite distinct from the nal or ofcial diagnosis in two dimensions: content and positioning. In terms of content, online commentary differs from an ofcial diagnosis in that it does not contain inferential reasoning in the form of conclusions about the patients medical condition. Rather, online commentary simply formulates the sensory evidence that is available to the physician in the course of the medical examination. In terms of positioning, online commentary occurs during the physical examination of the patient. In contrast, diagnostic evaluation is typically produced as a distinct action constituting a discrete activity within the consultation (Byrne & Long, 1976; Heath, 1992; Perkyl, 1998). Thus, diagnostic evaluation is almost invariably spatially and temporally separated from the examination activity, and it is offered at its termination. Online commentary can be characterized not only in terms of its content and position relative to other forms of diagnostic talk but also in terms of its design. There are several key features. First, in contrast to nal diagnoses, which are, as discussed in chapter 4, treated as a central activity in the consultation, physicians and patients and/or parents treat online commentary as, at best, an intermission in and subordinate to the activity of physical examination that is under way. Second, online comments can be divided into two broad types: 1. Online commentary that describes signs that are present but mild. These comments are normally mild, downgraded, or qualied (e.g, Thats a little bit red back there, or there may be a little bit of lymph node swelling on this side compared to the other side). 2. Online commentary that describes the absence of signs. This commentary is often mitigated by the use of evidential formulations, such as I dont see any uid, which make reference to the sensory evidence from which observations come (Chafe & Nichols, 1986).2



Third, online comments addressing both present and absent signs take two primary formats: (1) reports of observations, such as I dont see any uid, Little bit re:d, and I dont see any drainage, or (2) assessments of what is observed, such as Your ears look goo:d or This one looks perfect. In the report format, the physician does not overtly evaluate the signicance of the observation for the patients health status but leaves it to parents to draw their own conclusions about it. In the assessment format, the physician provides less insight into the examination but overtly draws evaluative conclusions. As an example, we can look at Extract 7.4a, which shows a fairly complex management of a no-problem evaluation across an entire examination. In it, we see the pediatrician carefully balancing maintaining the legitimacy of the visit and rmly resisting any implied expectation for antibiotic medication. This consultation with an 11-year-old girl and her mother took place on a Monday afternoon, and the child had already missed most of her school day. The problem presentation involves a candidate diagnosis shown in chapter 2 as Extract 2.4. The pediatricians examination of the girl commences after a brief history taking that reveals no surprises with respect to the condition but where the mother did conrm that the illness had been passed from one family member to the next, which seems to underscore her claim that this child is ill. The physical examination begins in Extract 7.4a with the primary complaint: the girls ears.
(7.4a) 305 [problem presentation shown in Extract 2.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 DOC: PAT: DOC: -> MO?: DOC: -> -> PA?: DOC: PAT: DOC: DOC: PAT: DOC: DOC: MOM: MOM: DOC: -> -> MOM: Which ears hurting or are both of them hurting. (0.2) Thuh left one, Okay. This one looks perfect, .hh (U[h:.???) [An:d thuh right one, also loo:ks, (0.2) even more perfect. ( ) Does it hurt when I move your ears like that? (0.5) No:. No?, .hh Do they hurt right now? (2.0) Not right now but they were hurting this morning. They were hurting this morning? (0.2) M[ka:y, [(Youve had uh-) sore throat pain? (Yes) Lets check your throat. <.hh=I=uhm: theres- I dont really see uh lotta uid build up. Some[times youll= [(Yeah.

24 25 26


DOC: -> =see tha:t ya know, MOM: Uh huh, DOC: -> (But) I dont see it right no:w,

The left ear is the rst to be examined. The physicians online commentary at line 4, uttered during the otoscopic examination of the ear, embodies the evidential formulation that is common when online commentary is deployed to counteract patient claims. A second online comment, uttered during the examination of the right ear, is built as a cumulative addition to his rst and is deployed to similar effect. During further examination of the ears, the physician asks about current pain symptoms (lines 9, 12, and 13), nally eliciting a response (line 15) in which the girl defends herself (and her mothers prior claim, the candidate diagnosis shown in Extract 2.4) against the possible inference that her symptoms are mild or nonexistent. At this point, the mother intervenes with a question, directed to her daughter, about sore throat pain (line 19), which the physician treats as raising another problem for evaluation (line 21). As he prepares for examination, he returns with further online commentary about the state of the girls ears (lines 21, 22, 24, 26), which embodies an evidentially formulated denial of the kind of uid buildup characteristically associated with middle ear infections. This denial includes two other mitigating elements. In characterizing uid buildup as something you sometimes see, the pediatrician implies that this is not a criterial feature of an ear infection or ear pain. Further, in adding that he does not see it right now, he leaves open the possibility that uid buildup may have been present in the past or may appear in the future. At this stage in the consultation, the effect of the physicians online commentary is to deny the existence of the main signs associated with the girls chief complaint and thus what would have been support for the mothers candidate diagnosis. Similar to denying a candidate diagnosis outright, the physician forecasts that a no-problem, nontreatable diagnosis is likely. But online commentary differs in two key ways: First, it proceeds incrementally, so the parent is gradually (rather than suddenly) brought to the realization that the diagnosis is likely to be nonproblematic and nontreatable. Second, online commentary at least claims to be supported by signs that are observable (or observably missing). Whereas an outright rejection simply asserts the physicians authority, online commentary claims only the authority to observe what otherwise independently exists, and thus the authority is supported by facts. If we return to this case, we can see how this practice works to support, rather than undermine, the legitimacy of the visit. The examination now proceeds to the girls throat.
(7.4b) 305 21 22 23 24 25 26 DOC: MOM: DOC: MOM: DOC: Lets check your throat. <.hh=I=uhm: theres- I dont really see uh lotta uid build up. Some[times youll= [(Yeah. =see tha:t ya know, Uh huh, (But) I dont see it right no:w,



27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

DOC: PAT: DOC: -> DOC: -> MOM: DOC: -> MOM: DOC: -> -> DOC: MOM: DOC: MOM: DOC:

(0.4) Uh: lets see. Say ah:?, (0.5) (uh_??) (1.0) Thats uh little bit red back there, (0.2) I dont see anything: (0.4) Yeah. Very good. Thank you. Huh h[uh huh (.hh) [I dont see anything (.) that looks infected. Reall[y, Okay. [Uh: in thuh sense that were: looking at bacterial, strep throat kinda thing(s). .h[h [O[kay. [Lemme listen to ya. (1.5) Could it be [from allergie:s, [Take uh deep breath, Sit up straight?,

This examination, initiated at line 21 (and 28), eventuates in a slight change of procedure in the physicians use of online commentary. At line 32, he observes a positive (but mild) symptom that could validate the childs complaint of sore throat pain and, with it, the decision to make the medical visit. Notably, this observation, which is supportive of the mothers decision to seek medical care for her child, is not downgraded via evidential formulation. Subsequently, having completed the examination and while preparing to listen to the girls lungs, he produces a more comprehensive online assessment that is also evidentially formulated: I dont see anything (.) that looks infected. (line 36). Subsequent to the mothers resistant Really (line 37), he qualies his previous assessment with a grammatically tted increment (lines 3839) (Schegloff, 2001), in a way that allows that the child may still have some kind of infection while still eliminating the prospect of a bacterial infection and, by implication, the prospect of antibiotic treatment. In this exchange, then, we see both that the physician is working to forecast a no-problem diagnosis and that he is working to validate the mothers decision to seek medical help. In this case, the physician is still facing resistance through the Really,, but note that the mother also moves toward acceptance with Okay., particularly in line 41, where this comes following his explicit invocation of bacterial illness. The mothers response to this outcome at line 44 is to maintain her position that her daughter has a medically treatable problem by raising the prospect of a further conditionallergies. Insofar as this inquires into a different diagnosis of the problem, it displays her acquiescence to the physicians rejection of strep throat as a diagnosis. After an uneventful lung examination, the pediatrician moves to examine the girls lymph nodes. This examination is shown in Extract 7.4c.

(7.4c) 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77



Does it hurt when you breathe in deep like that? (1.4) ((Patient shakes head)) DOC: No:? (0.2) DOC: How bout- under your chinny chin chin.<Lets see. (1.5) DOC: -> No(w) there may be uh little bit of lymph no:de swelling -> on this side com[pared to the other side, MO?: [(Yeah.) (.) DOC: -> On thuh [left side, MOM: [Oh:: okay. DOC: .hh So: it would loo:k hh like she is:=uhm (.) probly ghting some (.) viral: upper respiratory kinda stuff, DOC: .hh More on thuh left than on thuh right, which can account for some pain maybe,

The physicians identication of lymph no:de swelling (line 68) gives implicit support to the patients claim that she has experienced pain primarily on the left side. Subsequently, the pediatrician makes this explicit in lines 7677 as part of the diagnosis, which he begins with the upshot formulating so at line 74. The nal diagnosis is offered in two parts separated by an extended compliment about the cleanliness of the girls ears:
(7.4d) [shown in Extract 2.21] 74 DOC: 75 76 DOC: 77 78 MOM: .hh So: it would loo:k hh like she is:=uhm (.) probly ghting some (.) viral: upper respiratory kinda stuff, .hh More on thuh left than on thuh right, which c[an account for some pain maybe, [Okay. ((13 lines of ear compliment sequence removed)) 92 93 94 95 96 97 98 99 100 DOC: Uh:- I would tell you though I dont hhh (.) I dont see anything that requires like antibio:tics er anythi:ng, but certainly sympto[matic treatment might be in order, [Mm. .hh O[kay. [Uhm: anything from vaporizers tuh maybe some chloraseptic kinda stuff for thuh [throat, lozenges might be better, [Oh:. Okay.


The rst part of the problem discussion (the diagnostic evaluation in lines 7477) is supportive of the mother and daughters decision to seek medical care and draws on



earlier online comments (e.g., Extract 7.4c, lines 6869) that were also supportive of that decision. The second half (lines 9293) builds from the adverse online commentaryespecially that in (7.4b)and then clearly rejects antibiotic treatment (line 93) in favor of symptomatic over-the-counter remedies (line 94/9899). And as the physician moves from diagnosis to treatment, he begins with the phrase I would tell you though (line 92). In this way, he builds the recommendation as contrastive with the notion that viral conditions require antibiotic treatment, and hence in contrast to any position the mother might hold in favor of antibiotic treatment, without contradicting her earlier claim made both by her presence at the clinic and the candidate diagnosis (Extract 2.4) that the child is sick. His use of the evidential formulation (I dont see anything. . .) revives the relevance of the observations reported in his earlier online comments and reinvokes their signicance as evidence for the position he is currently taking. Across this sequence, the mother responds to both the supportive and adverse aspects of the diagnostic evaluation with an acknowledgment token okay, which accepts the physicians evaluation. At line 100, this acceptance becomes more marked with the addition of Oh (Heritage, 1984a; Heritage & Se, 1992). Subsequent to this, the mother discusses the merits of several commercial remedies in a cordial way and without contesting any aspect of the physicians conclusions. This interaction proceeds from a situation in which both the patient and her mother initially viewed the patients complaint as signicant and were defensive (extract 7.4a, lines 15, 19; extract 7.4b, line 44) when that stance seemed to be threatened, to a situation in which they acquiesced to the physicians nonantibiotic treatment without active resistance or signs of disappointment. I would argue that the primary reason for this is that the legitimacy of the visit has been maintained, at least partially. Online commentarys role in the progressive construction of powerful support for the physicians nal diagnostic evaluation in this case seems clear. Parents are in a position where contesting the online comments is very difcult because they represent a physicians authoritative way of seeing, hearing, feeling, or assessing. And when online commentary is used, inferential resources for the diagnostic evaluation and the lack of antibiotic treatment are built up incrementally. The result is a more persuasively formulated case for the nal evaluation and recommendation than would be obtained without the use of online commentary. Moreover, when online comments have been used, it is more difcult for parents to contest nal diagnoses without exhuming the online observations that they have already let pass. But as Heritage (2006) discusses, this use of online commentary not only ramps up the physicians claim to authority by offering unchallengeable observations but also is evidence formulating (Perkyl, 1998), and in this way it makes the physicians diagnosis a bit more accountable. In sum, then, the use of online commentary is an interactional resource with which physicians can build a case for a no-problem diagnosis and thus a case for no treatment, while still reassuring patients of the rightness of their decision to seek medical assistance. Because parents are more likely to be satised when physicians make use of this resource, it appears to be an excellent means for diminishing resistance to no-problem, no-antibiotic treatment outcomes.



Presentation of No-Problem, No-Treatment Conditions A second interactional domain I will focus on is how physicians present their diagnoses and treatment recommendations. Physicians most commonly formulate their noproblem diagnoses and no-antibiotic treatment recommendations as announcements, but there are still quite different ways that they can be, and are, formulated. When we examine cases where physicians make use of these alternative ways of formulating diagnoses and treatment recommendations, we can observe that parents respond differently depending on the format and that this is consequential for whether and how they resist no-problem diagnoses and nonantibiotic treatment recommendations. The remainder of this chapter will focus on a parallel discussion of one dimension of diagnosis delivery formats and treatment recommendation formats: whether they frame the announcement afrmatively or negatively. Background The diagnosis delivery and treatment recommendation activities are, as discussed earlier, rather different. In chapters 4 and 5, we observed that the normative structure of the treatment recommendation activity was such that parent acceptance was due upon completion of the physicians initial treatment recommendation, whereas the diagnosis delivery made no such response conditionally relevant. And although physicians did not pursue parent acceptance of a diagnosis, if acceptance was not forthcoming following a treatment recommendation, physicians routinely pursued, or did interactional work to secure, this acceptance. Given the problematic nature of parent resistance (especially to the treatment but also to the diagnosis), the question to be addressed here is whether the way that physicians deliver their diagnoses and treatment recommendations promotes or inhibits parent resistance. This section will rst describe the main formats that are observed in each activity. From this point, we will see that the implications and explanations of these formats vary and then discuss each in turn. Afrmative Announcements When the physician delivers a diagnosis or a treatment recommendation, the most common turn design involves an afrmative announcement of (in the case of the diagnosis) what the condition is or (in the case of treatment) what should be done for the patients problem. As we will discuss more later, physicians sometimes offer multiple diagnostic statements, but in the Metro data, 55% of rst diagnoses and 54% of all diagnoses (whether rst or subsequent) are afrmative announcements. So across the board, this is the dominant type of diagnosis. An illustration of an afrmative diagnosis is shown in Extract 7.5. As the physician completes his physical examination (lines 15), he moves into a summary of his examination ndings (56) and then the diagnosis that: kind of viral stuff. (lines 1011). The physician does not specify a condition name, but as noted briey earlier, such diagnoses are very rare in the context of nonbacterial URI diagnoses, so this is not unusual. What I want to focus on is that the condition is an announcement, and it is delivered afrmatively.



The stress on is appears designed to deliver the diagnosis as conrmatory of something the mother had said (Stivers, 2005a), though the mother had not previously asserted a claim that this was a viral infection. This delivery may do additional work to secure parent acceptance because if it is built as conrmatory, however contrived, it may be more likely to be accepted.
(7.5) 323 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 DOC: I see that (tooth coming in but not) (0.2) DOC: any other ones, (0.4) DOC: ( ) (.) An she looks uh little congested_ But I dont see anything wrong with her ears or thro^a:t. MOM: (Oh good.) (0.7) MOM: Oh: my g[oodn(h)e:ss. DOC: -> [S:o I would say: that this is that: kind -> of viral stuff. MOM: Okay.= DOC: =An I would keep doing thuh clear uids: (.) for twelve hours,= MOM: =(Mm [hm,) DOC: [An the:n (.) start thuh banana:s, ri:ce, applesau:[ce, toast, (kind of diet_) MOM: [Okay.

Positive announcements are not necessarily delivered denitively. For instance, in Extract 7.6 the physician downgrades the epistemic certainty of the diagnosis with probly (a reduced version of probably). As the extract begins, the physician is reaching the conclusion of his physical examination as he transitions into a diagnosis of the primary reason for the visit: the cold symptoms (see lines 56). The diagnosis is offered in line 7: That stuff is probly caused by uh virus. and restates this on its way to treatment in lines 89.
(7.6) 324 1 2 3 4 5 6 7 8 9 10 DOC: .h But=(uhm) uh lotta kids do:_ when they= especially when they cut their molars they- pull at their ear, MOM: Yeah:. [(No,) DOC: [Uhm, (0.2) Now as far as=uh (0.5) fe:ver, hh (0.4) co:ld, (0.4) congestio:n, runny nose, -> That stuff is probly caused by uh virus. DOC: .hh So=uhm (0.2) .tlkh he- hes- ya know hes probly got uh virus that hes gonna get over:=an hes probly (getting it) over already.

11 12 13 14 15



#Mm hm,# Ya know getting over it already_ On his ow:n. (0.2) .h An you guys (could) of course give him something for congestio:n like uh little- .h [( -)

The nal example in Extract 7.7 shows a diagnosis that is offered without any epistemic downgrading or mitigation but nonetheless gives insight into the physicians stance toward the diagnosis as not what she assumes the parent wants to hear with her use of the turn initial Unfortunately.
(7.7) 215 1 2 3 4 5 6 7 DOC: -> -> MOM: MOM: DOC: MOM: Unfortunately he caught uh bad co:ld. He caught thuh [one I think that [Da:d had. [(ah:-) [(Yes.) And he gave it to me too: [an- (thuh whole family.) [<Oh n Thank you very ^mu(h)ch= =huh huh huh .huh [(.huh huh .huh .huh) .hhh [Thuh whole family.

The cases shown in Extract 7.57.7 illustrate the most common diagnosis delivery format. This format is even more commonly used for treatment recommendations. In the Metro data, when treatment recommendations are given, 94% of all rst recommendations are done with afrmative announcements, and 93% of all treatment recommendations (no matter how many) are afrmative announcements. We can see examples of announcements of treatment recommendations if we return to extracts 7.5 and 7.6. In both cases, following the afrmative diagnosis, the physician moves to afrmatively recommend treatment as well. In 7.5, this is in lines 1314/1516, where the physician suggests keeping the child on clear uids and then moving to mild solid foods. In 7.6, after suggesting that the child will get better on his own (and thus indicating that no treatment is necessary), he goes on to offer an afrmative treatment recommendation of something for the congestion (lines 1415). Although the last word of this recommendation is not clear, and the physician goes on to later further specify a recommendation, here I want to note only that in both of these cases a treatment recommendation for a particular course of action is offered. Afrmative announcements of treatment do not necessarily involve medication. This is illustrated in Extract 7.8.
(7.8) 100101 1 2 3 DOC: MOM: So I think hes (just has) a co:ld_ an its just making him feel kinda of crummy. (Yeah:.)

((10 lines not shown - expansion of diagnosis))



14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

DOC: MOM: DOC: -> -> -> DOC: -> -> -> -> DOC: -> -> DOC: DOC: -> MOM:

A:lright. and Ill stand away from you so you dont (ha[fta me anymore.) [(Hah hah hah .hh heh heh) Uhm_ (0.2) but:=uh: (.) I think thuh best thing tuh do at this point (would)/(will) jus be try elevating his head at night. give him plenty of liquids to drink. .hh you cn- .h if you cn put something other- under one end of the ma:ttress, so thuh whole ma:ttress is on uh sla:nt or under one one end of thuh bed, so thuh whole bed is tilted, .hh that way his >heads up a little bit<. That helps thuh mucus tuh drain better. (.) for him. Right.

Here, the physician offers her diagnosis in lines 12. After providing some evidence for this in terms of the fever (data not shown), she outlines the treatment she recommends for the boy in lines 1725. The physicians recommendation is framed as a recommendation of what should be done: elevating the boys head at night and giving him plenty of liquids. Although there is no medication being recommended, the treatment recommendation is nonetheless formatted afrmatively. This section has shown that one way that both announcements of nal diagnoses and announcements of treatment recommendations are delivered is through an afrmative turn design. This turn format is not the only turn format used by physicians. It is certainly the most common format when there is a bacterial diagnosis and treatment to be prescribed. In no-problem and no-treatment scenarios, it is not uncommon for the physician to offer no comment on the diagnosis at all, particularly if online comments were delivered during the examination. However, many physicians do go on to formulate nal diagnoses. Because, as we saw in chapter 4, this is an area where parent resistance can be consequential, when physicians deliver diagnoses of no-problem conditions, it is important to understand whether and how turn design affects parent behavior. We now turn to a second primary turn format for diagnosis deliveries and treatment recommendations. Negative Announcements Both diagnoses and treatment recommendations can be formulated to announce the action negatively: what the condition is not or what is not recommended. With respect to the diagnosis, this may take the form of ruling out either a single diagnosis or a class of diagnoses. In Extract 7.9, a bacterial infection, as a class of infections, is ruled out (lines 1415). Although the ruling out is part of a response to the naming of a problem following the physical examination, it nonetheless offers a ruling out diagnosis that in turn the parent can deal with as such.



(7.9) 383507 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Josephs so wa^:rm. I mean: (.) you know thats not usually how he i^s. MOM: .hh Ay^! DOC: Uh:m (.) MOM: Eh! (.) DOC: But you havent measured uh f- youve checked him:, and he hasnt had a fever, MOM: No:, n[o (fever but) I gave him= INF: [((screaming throughout)) MOM: =Pedi[acare today:_ DOC: [Why dont- He may be- His temperature may be a little bit more than usual?, (0.5) but -> (0.4) uhm (0.5) >yknow< I dont see any -> bacterial infection. MOM: Oh my:_ MOM:

Most commonly, single diagnoses are ruled out, as we can see in Extract 7.10 in lines 12. Here, the physician rules out any infection and then goes on to specically rule out ear and throat infections by stating that they are ne. There is no afrmative statement here about what the child does have, although she does go on to account for the sore throat with the cough. That the physician is done (at least for the moment) is evidenced by her just subsequent move into treatment recommendation at line 8.
(7.10) 302501 1 2 3 4 5 6 7 8 DOC: MOM: DOC: MOM: DOC: DOC: So with Clarissa right now she doesnt have any infection. Ears and throats: ne. Okay. So=uhm:: she just has- you know this- the throat hur:ts because of the coughing. [Uh huh [(an)/(it) usually irritates the throat. Ill give her another coughing formula the same- an:=

As we saw before, the announcements can vary in terms of epistemic certainty and other dimensions of stance. For instance, whereas the diagnosis in Extract 7.10 is delivered without mitigation or downgrading as simple fact, the diagnosis in Extract 7.11 includes Im not really worried about as opposed to the simple factual statement She doesnt have. Here, the physician reports on her physical examination ndings regarding the lung examination and then rules out a diagnosis of bronchitis before moving on to a treatment recommendation.
(7.11) 160706 1 DOC: I hear- I dont hear the mucus sitting in there uh lot,



2 3 4 5 6 7 8 9 10 11



I mean I get a sense (th) its rattling a little bit but Not a whole bunch, Mm hm, So Im not really worried about a bronchitis; Mkay_ .hh Uhm (1.0) But she is uh little- taking a little to what we say exhale, to get the air ou:t, ((Nods))/(.) So Im gonna help her out with a different cough syrup. Mkay.

In the Metro data, ruled out diagnoses comprised 18% of rst diagnoses and 21% of all diagnoses. Like negative diagnoses, treatment recommendations are sometimes delivered against a class of treatment, though most commonly they rule out a particular treatment. As an example of the former, we can look at Extract 7.12.
(7.12) 100114 1 2 3 4 5 6 7 8 9 10 DOC: Yeah her lungs sound great. (0.5) DOC: .h So- Im pretty sure this is all just viral. (.) DOC: What I will d[o is uh culture of her throat just to be= MOM: [Okay. DOC: -> =uh hundrd percent su:re_ [but I dont think we need to= MOM: [(Okay_) DOC: -> =put her on any medica[tion_ MOM: [Okay.

Here the physician moves from an afrmative diagnosis in line 3 to a test recommendation (lines 5/7) and to her treatment recommendation, I dont think we need to put her on any medication_ (lines 7/9). In contrast to the afrmatively designed recommendation, in this example the physician announces what she recommends against (any medication) rather than asserting what should be done. As mentioned in previous chapters, when physicians use terms like medicine, medication, or treatment, parents routinely understand them to be indicating antibiotics, and physicians generally appear to be intending that reference. Conversely, when physicians talk about symptomatic treatment, they are indicating that they will not be prescribing. It does though here suggest not just Not X but no treatment at all. It is much more common, though, that physicians recommend against particular medications, with antibiotics being the most common treatment that is recommended against.3 Of course, this is most common when the diagnosis is not bacterial. In the Metro data, among cases with an initial nonbacterial diagnosis, this type of treatment



recommendation format occurred 16% of the time (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). We can see an example of this in Extract 7.13.
(7.13) 150607 1 2 3 DOC: -> But in the meanti::me no:: antibiotics or anything yet. DOC: Okay?, MOM: Yeah.

In line 1, the physician names antibiotics as an instantiation of medication that is unnecessary at this time (with or anything). Also note that when acceptance of the treatment recommendation is not immediately forthcoming after line 1, it is pursued by the physician in line 2. This displays, at minimum, the physicians orientation to the treatment having been completed at line 1. This section has shown that both diagnostic announcements and treatment recommendations can be designed negatively by ruling out a diagnosis or treatment possibility. If one goal of pediatrician-parent interactions in these contexts is to minimize parent resistance to no-problem diagnoses and nonantibiotic treatment recommendations, then the next question becomes, Do these different turn designs make a difference for how parents receive diagnoses and treatment recommendations and whether they resist? The next sections will examine these issues. Parent Responses to Diagnoses In chapter 4, we observed that oftentimes diagnoses are not responded to. So unlike following treatment recommendations, following diagnoses, only active resistance is notable. Silence following a diagnosis does not have the same import as it does following a treatment recommendation.4 These data show no statistical association between a negative diagnosis and parent diagnosis resistance. One account for this is that negative diagnoses in isolation are relatively rare. Thus, parents may view them as generally on their way to a positive diagnosis: a feature that is different from treatment recommendations in no small part because there is no normative orientation to response as relevant following an initial diagnosis. Despite this lack of distributional evidence, diagnosis resistance is nonetheless arguably engendered by a ruled out diagnosis. We can see an example in Extract 7.14. Following an initial negative diagnosis, the parent is resistant. She initiates a sequence that delays the physicians progress to the treatment recommendation. Initially, this appears to resist the implied nontreatment, but as the mother elaborates, she appears to be resisting the diagnosis in that she goes on to say that her son is not himself. The physician treats the mother as quarreling with the diagnosis and, on the one hand, works to legitimize the mothers claim that the child is ill (line 10) while, on the other hand, also defending the diagnosis with more negative diagnoses.
(7.14) 383507 1 2 MOM: Josephs so wa^:rm. I mean: (.) you know thats not usually how he i^s.



3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28



.hh Ay^! Uh:m (.) Eh! (.) But you havent measured uh f- youve checked him:, and he hasnt had a fever, No:, n[o (fever but) I gave him= [((screaming throughout)) =Pedi[acare today:_ [Why dont- He may be- His temperature may be a little bit more than usual?, (0.5) but (0.4) uhm (0.5) >yknow< I dont see any bacterial infection. Oh my:_ [so what do you think I should do to uhm_ [So_ (Dont kick tha::t?,)((to older child)) keep him_ hh (cuz) I was so worried about him cause hes just not him. hes kinda .h Hes sick. He- he is si:ck. I- do not dispute that. [(but that-) [hah=hah=hah=hah=hah_ Is it an ear infection or a pneumonia or [(something sinus,) I dont see that. [(lookit) theres no ear infection. I dont see any of that. ((Knock on door; DOC answers it))

Earlier, it was claimed that although negative diagnoses are not necessarily clearly resisted more frequently than afrmative diagnoses, they do, by design, make resistance easier. This is probably because, in stating what the child does not have, the physician implies that the child is not ill and thus risks delegitimizing the visit. The same problem occurs in the treatment recommendation. However, there, because parent acceptance is due immediately upon completion of the recommendation, we can observe that negative treatment recommendations are more likely to be resisted. Parent Resistance to Recommendations against Particular Treatment Although diagnosis resistance is nearly as common as active treatment resistance, treatment resistance is much more common when passive treatment resistance is taken into account. In addition, it arguably represents a bigger problem in that parents are more likely to persist with resistance and escalate in the context of treatment discussions because physicians will not, as previously shown, move to the next activity without parent acceptance.



Parents are more likely to actively resist a treatment recommendation if its rst component involves a ruling out rather than an afrmative recommendation. This is supported both quantitatively and qualitatively. Quantitatively, in the Metro data, parents were 24% more likely (39% vs. 15%) to resist a physicians treatment recommendation when physicians ruled out the need for antibiotics than when they formulated the nonantibiotic treatment recommendation afrmatively (p <.001) (Mangione-Smith et al., 2006). Qualitatively, this pattern is very visible, as is the same pattern with passive resistance. As a rst example, we can look at Extract 7.15, which we previously analyzed as an example of treatment resistance in chapter 5.
(7.15) 322803 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 DOC: DAD: DOC: DAD: DOC: DOC: DAD: DOC: DAD: DOC: DAD: -> I th:ink from what youve told me (0.2) that this is pro:bably .h uh kind of (0.2) virus infec[tion, [Uh huh, (0.4) th:at I dont think antibiotics will ki:ll, (0.2) Well- [( ) [Thee other>Go=ahead_< Yeah. .hh ( ) I had it- I had thuh symp[toms [I understand. Three weeks ago. [Right. [.hh An:d Ive been taking thuh over the counter cough [( ) [(Good_) Uh s- ( ) coughing syrup, Nothing take away .hh Especially my sor- my [th- my throat was real= [Mm hm =sore [for (awhile- et- that) w:eek. [Uh huh Right, an:d (.) I start taking thuh antibiotic (0.5) eh he ((cry)) Yesterday. Right, And it (.) seemed to take care of the problem. [(Well) thats why were doin a throat [culture. [( ) [Yeah. [is TUH SEE if they need antibiotics.

-> -> -> ->

DOC: DAD: -> -> DOC: DAD: -> DOC: DOC: DAD: -> INF: DAD: -> DOC: DAD: -> DOC: DAD: DOC:

Following the diagnosis (lines 12), the physician offers a recommendation against antibiotics (line 4). Passive treatment resistance begins immediately subsequently (line 5), and then the father escalates to active resistance. Well projects disagreement (Pomerantz, 1984). Following the physicians >Go=ahead_< (line 8), the



father continues with active resistance in lines 9/11/1314/1617/19/22/24/26. This constitutes resistance because across this stretch of talk, the father builds a case for why antibioticsthe treatment the physician recommended againstmay be effective: They were effective in treating a similar illness that the father recently had. A second case is shown in Extract 7.16. Here, the physician offers a recommendation against antibiotics in line 1. She pursues agreement to this proposal in line 2 and receives this in line 3. But immediately following the agreement, the mother inquires about further testing of the child, which the physician agrees with. Although the sequence is possibly closed at this point, and the physician initiates a move to closing in line 8 with an inquiry to the child about the visit not being so scary,, in overlap, the mother initiates active resistance (line 9).
(7.16) 150607 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DOC: DOC: MOM: MOM: DOC: MOM: MOM: DOC: MOM: -> -> -> -> DOC: But in the meanti::me no:: antibiotics or anything yet. Okay?, Yeah. Did you want her tuh get that ultra sound? Yes I want her [to get thuh ultra-sound too. [Okay. Okay. ( ) ((to girl)) [Alright see: [( ) not so scary, [So [Should we- bring her? Eh- Should we bring her i:n? See e- my husband gets just rea:l insistent that(.) theres somein wrong with her because she keeps getting sick. So: Shes been sick for (how is-) well last time was thuh urwas thuh kidney. (Or well-) .hh both time was urin- urinary problem ri:ght?

In line 9, the mother begins her turn with another inquiry about future action: Should we bring her i:n?. This is then immediately accounted for with an upgrade to resistance by invoking her husband (lines 1012). As resistance, the parent claims that what the physician has offered them up until this point cannot be defended to a third party. The physician addresses the mothers action as resistance by restarting an investigation of the childs illness. As mentioned earlier, it is not only how but also whether a parent responds that shapes whether and how physicians continue with treatment recommendations. We can see this exemplied in Extract 7.17.
(7.17) 170812 1 2 3 4 5 DOC: 1-> As you know theyre viral infections, so theres 1-> no point in any a- any ant- antibiotics. (0.5) DOC: 2-> Simply control thuh cou:gh with .hh whatever 2-> your favorite cough medicine is,



The physician rules out antibiotics in lines 12. It is only following a lack of uptake of this recommendation that a recommendation for other treatment is offered (lines 45). This pattern comprised the majority of the cases where a recommendation against treatment was followed by a recommendation for particular treatment. We have seen evidence that ruling out antibiotics is more likely to engender parent resistance than afrmatively recommending nonantibiotic treatment. But the question remains as to why that would be the case and what this reveals about parents concerns in the medical visit. One analysis is that, like negative diagnoses, ruling out antibiotic treatment undermines the legitimacy of the visit. Additionally, by suggesting no treatment, the physician implies that the condition is untreatable. After all, these children do have illness symptoms for which the parents are looking for a solution. I will argue that a sufcient treatment recommendation necessarily involves advice about a solution for these symptoms. By denition, treatment recommendations that rule out treatment do not involve this advice. In the next section, I discuss what a sufcient treatment recommendation looks like and provide evidence that this is what underlies the pattern discussed earlier that shows more parent resistance to ruling out treatment recommendations. Sufcient Treatment Recommendations As we have discussed earlier, when parents arrive at a decision to bring their child to the pediatrician, this generally means that they no longer feel able to self-manage the childs problem. They turn to pediatricians for expertise and a solution to the problem. The treatment recommendation represents the latter. Here, we consider what constitutes a solution or a sufcient treatment recommendation. We will see that parents generally treat a treatment recommendation as insufcient (in the sense of not providing a solution) if it (1) fails to provide an afrmative action step, (2) is nonspecic, or (3) minimizes the signicance of the problem. By contrast, a sufcient treatment recommendation asserts a specic next action step afrmatively and treats it as a wholly legitimate recommendation, thereby treating the patients problem as legitimate. Adult patients in the acute care context display an orientation to receiving a treatment recommendation as a relevant next activity following a diagnosis delivery (Robinson, 2003). One type of evidence for this is that when treatment is not immediately forthcoming, patients pursue a treatment recommendation. In these data, this pattern is also present: Parents ask about and thus pursue a treatment recommendation if no treatment recommendation is offered. But what is more prominent is that some treatment recommendations proposed by physicians are responded to by parents as insufcient. So the absence of a treatment recommendation may be part of a more signicant normative orientation by parents to the relevance of having a sufcient treatment recommendation offered by physicians. This section will be concerned with providing evidence in support of this claim. By denition, all cases of recommendations against antibiotics fail to provide an afrmative next action step, and I argue that it is for this reason that they are more likely to be resisted. Evidence that it is the lack of an afrmative action step that is at issue comes rst from the observation that parents pursue this dimension of treatment following recommendations against particular treatment. We can see this in Extract



7.18. Following the physicians recommendation against an antibiotic (line 5), the mother inquires about a medication that she can provide (line 9).
(7.18) 322708 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DOC: .hh So: I think its just (.) one uh thuh (.) thi:ngs: kids get one thing after another sometimes, MOM: M[kay. DOC: [Nothing serious here, DOC: .mh Nothing that I can see that an antibiotic would help, MOM: Okay; (.) DOC: [Uh:m MOM: -> [So uh:m (.) should I continue with thuh Tyleno:l? er_ DOC: Tylenol if hes uncomfortable. (.) DOC: [With fever n (0.2) headache. MOM: [(kay) DOC: or anything [like that. MOM: [(Okay.)

Although in some cases parents actively advocate for antibiotics (as shown in chapter 6), in other cases, resistance and other such behaviors may be rooted in a concern for receiving specic treatment. In this case, we can observe that the parent does not resist the lack of an antibiotic per se but displays her concern, and thus the root of this resistance, to be the lack of an afrmative next action step. We can see that the mother does not resist the lack of a prescription because Tylenol is an over-the-counter recommendation. Rather, she pursues a concrete afrmative recommendation. That it is not always a prescription that parents are looking for is further supported by Extract 7.19. After the physician recommends against one over-the-counter cold medication (lines 12), the mother inquires about another form of nonprescription treatment that she could offer her daughter (line 8).
(7.19) 170802 1 2 3 4 5 6 7 8 9 10 11 DOC: -> Uh::m o- nl- unfortunately we probably cant give her -> stuff .hh like Sudafed. (.) DOC: Because thatd crank her blood pressure up_ an we dont need tha:t. MOM: Right. (1.0) MOM: -> Okay: so give her Tylenol?,= DOC: =Yeah. (0.2) DOC: for discomfort.



The parent displays her concern to be not with the denial of Sudafed or with the failure of the physician to recommend antibiotics but rather with what the mother can do for her child. Note, too, that in all cases, the parents questions are designed to prefer a yes answer and so to prefer the proposal of an afrmative course of action. Parents appear to be oriented not primarily to the relevance of receiving a treatment recommendation but rather to receiving an afrmative next action step. These cases show that parents are oriented to a minimally sufcient treatment recommendation as necessarily including such an action step, and without an afrmative next action step, parents treat the recommendation as insufcient. Parent Responses to Recommendations for Particular Treatments Generally, when physicians recommend nonantibiotic treatment, if they do it afrmatively, it is not resisted. This pattern is illustrated in Extract 7.20.
(7.20) 150614 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 DOC: DOC: Looks like he has a co:ld,=h Its just uh virus, not uh bacteria;=his lungs sound really good,=its just .h all irritation up here;= =(and)/(that) hes coughing thuh- .h throat looks uh little red_ but theres no puss or anything; .hh ear is just uh little (.) slightly pi:nk and .h its uh combination for with thuh stuffy no:se_ .hh so=w:e have=to .h clear thuh nose. Ya know like ((exhaling noise))/(0.2) reduce thuh congestions that will help him uh lot. [.hh [>Okay.<= =An Im gonna give you some cough medicine that has some decongestant in it. ((whispering))/ ((DAD nods)) M[kay. [That will help him out.

DOC: 1-> DOC: 1-> DOC: DAD: => DOC: 2-> 2-> BOY: DAD: => DOC:

Following a diagnosis of a cold (line 1) and the explication of the evidence for that diagnosis (lines 27), the physician afrmatively recommends nonantibiotic treatment: cough medicine. Besides being formatted afrmatively, the physician also formats her recommendation specicallyshe recommends a type of cough medicine (lines 1314). Earlier, I mentioned that specicity was the second criteria of a sufcient treatment recommendation. Although the cough medicine may or may not turn out to be prescription, what appears to be important for whether resistance is likely to be engendered is that the recommendation is both specic and for a concrete next action step. Here, the cough medicine is not named, but the physician states that she is gonna give you some and species that it has some decongestant in it.. Both of these aspects of the turn indicate that she has in mind a particular medication, and in



this way she is being specic in her recommendation. This is subsequently accepted both visibly (line 15) and verbally (line 16). A similar example was shown in Extract 7.8. Earlier, we saw that the physician diagnosed a cold and recommended elevation and uids in an afrmative formatted treatment recommendation. The physicians treatment recommendation offers the parent concrete afrmative action steps. Although she slightly minimizes her recommendation with just in line 18, in general, her recommendation is formulated as an unmitigated positive announcement of what needs to be done, in spite of the fact that the recommendation does not involve antibiotics. There is a moment of delay in acceptance (line 26) that leads to the physicians use of an increment to recomplete the recommendation (line 27) (Schegloff, 2001), but there is both acknowledgment at line 28 and acceptance (Oh okay) following the (possibly) responsive explanation of the illness and justication for the treatment (data not shown). A similar case is shown here in Extract 7.21.
(7.21) 100115 1 2 3 4 5 6 7 8 DOC: If her uhm if her cough: does really become croupy tonight, (0.4) just be sure you use the coo- in fact you can use cool mist hu[midier anyway in the room. [Okay./((nodding)) That will help her feel more comfortab[le. [((nods)) Keep thuh nose open. Okay.


Here again, the physician delivers her treatment recommendation beginning in line 1. It is initially projected as a conditional with if, but this is revised in line 2 with in fact and then in line 3 with anyway. The recommendation is afrmative and specica cool mist humidier. These cases are not only afrmative. In them, physicians satisfy all of the conditions outlined earlier for sufcient treatment recommendations: They are afrmative, specic, and nonminimized treatment recommendations. Because recommendations for treatment by denition satisfy the criteria of being afrmative, this may explain why they are less likely to be resisted. But the other requirements for sufciency are also important. Evidence for this is that when recommendations for treatment are resisted, they typically fail on one of the latter two dimensions: Typically, either they involve a vague or nonspecic treatment recommendation, or the physician minimizes the treatment recommendation. We can see an example of this in Extract 7.22. The physician recommends against antibiotics and for medicine for the girls stuffy nose, cough, and fever, but he fails to offer any specic treatment recommendation. At line 19, the mother resists the recommendation with just over the counter?,; in response, the physician offers to check off my favorites for you, and goes on to provide a specic treatment plan. This revision is accepted by the parent (line 25), providing evidence that a physicians specicity is important to parents in their orientation to what constitutes a sufcient treatment recommendation.



(7.22) 383412 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 DOC: .hh Which is goo:d that means that she doesnt need any antibiotics. (.) DOC: because this is probably, (.) caused by uh virus, DOC: .hh [an:d=eh as you may kno:w antibiotics dont kill= MOM: [Mm hm, DOC: =viruses. [soMOM: [Mm hm, DOC: .hh uh: and this is- (p)/(.) uh lotta kids this, -> [(is) pretty common;=so .hh treatment will be: MOM: [Mm hm, DOC: =you know medicine- thatre gonna make her comfortable and treat her symptoms. so .hh you cd get her medicine thats gonna make her nose less stuffy an make it less runny, an uh medicine for thuh cou:gh?, DOC: .hh An:d=uh you know something for thuh fever like (youve)/(weve) been doing, DOC: .hh [uhMOM: [Just over thuh counter?, (.) MOM: jisDOC: Over thuh counter_ yeah, DOC: I have some here_ Ill- Ill check off my favorites for you, MOM: Okay,

Similarly, recommendations for treatment are also more likely to be resisted if they fail on the third dimension of recommendation sufciency: minimization. We can see an example of this in Extract 7.23. The physician does not initially receive acceptance when he recommends against antibiotics. He proceeds to recommend Robitussin afrmatively, but he presents his overall recommendation in a downgraded manner with just. In contrast with Extract 7.8 where one component of the treatment was downgraded, here the just downgrades and minimizes the entire recommendation. The recommendation here implies a contrast between stronger treatment and Robitussin, thereby treating his recommendation of Robitussin as a minimal sort of treatment. The passive resistance in line 4 is addressed by the physician with one account for the lack of uptakethat the parent has tried this medicine already and found it to be inadequate.
(7.23) 151213 1 2 3 4 DOC: DOC: DOC: -> => She probably doesnt need antibiotics. I mean uh: most of these are viral. .h uhm hh .h Id- just give her Robitussin, (0.7)



5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Have you- (.) given her any of tha[t, [Uh:m Ive given her the Tylenol: an stuff but .h she just seems so uncomfortable when she cou:ghs; that [ya know_ [Uh huh_ I feel like I need to give her something uh little bit Yeah. .h Right. well I- (s-) she denitely needs cough medicine tuh thin out thuh mucus. DOC: so that when she does cough (that) itll come up; (0.5) DOC: .h [an thenMOM: [And then her congestion? DOC: Pardn me, MOM: => Also her congestion. (.) DOC: Yeah. (.) DOC: -> Well, .hh they have uh Robitussin PE: that (could-) MOM: Sh sh sh sh sh [sh sh sh ((to child)) DOC: [take care of tha:t. MOM: => Oh okay.

DOC: -> MOM: => => DOC: MOM: => DOC:

In response, the mother resists the treatment further by asserting that her daughter is uncomfortable and needs something a little bit, which idiomatically suggests that the likely next term was stronger. The physician responds to the parents resistance by agreeing that cough medicine is necessary, but he does not provide a specic responsive recommendation that indicates an upgrade on his prior recommendation. The mother then again resists the treatment by inquiring about treatment for her daughters other symptom: congestion. With this inquiry, similar to other examples shown earlier, the mother conveys her perception that the physicians prior suggestion was insufcient. In response, the physician offers a specic recommendation Robitussin PE (line 22), and this is accepted with Oh okay. in line 25. In the treatment recommendation phase of the visit, offering prescription treatment such as antibiotics is one way that physicians show parents that they were correct in having sought medical attention. Physicians legitimize the childs visit through the prescription of medication. As mentioned earlier, offering no treatment is, conversely, understandable as delegitimizing the visit. When a treatment is afrmatively and specically recommended, this may do work to counter this problem. But when physicians generalize to whatever your favorite cough medicine is or orient to the treatment as minimal or arbitrary, this creates or perhaps intensies the problem parents face when the physician offers no treatment recommendation or recommends against a particular treatment because with a minimized or vague treatment plan (at least in the URI context), physicians both delegitimize the patients visit and fail to provide a solution, thereby underscoring that the condition is not treatable. This section has argued, then, that treatment resistance is best avoided through the use of sufcient treatment recommendations: recommendations that are afrma-



tive in design, specic, and not minimizing. Thus, this provides a second resource physicians can make use of in securing parent acceptance of a nonantibiotic treatment recommendation. Diagnosis Delivery Although diagnosis resistance was not clearly associated with negative diagnoses, what can be observed is that afrmative diagnoses, like afrmative treatments, can nonetheless fail to be optimally designed. Like treatments, parents are more likely to resist diagnoses that are minimized. This suggests that whereas treatability is primarily at issue with respect to treatment recommendations, the maintaining of legitimacy may be more of an issue with respect to diagnoses, though both are clearly issues throughout the visit. This section examines cases involving diagnoses that are minimized to provide evidence for the claim that minimized diagnoses are more likely to engender resistance than non-minimized but still afrmative diagnoses. As an initial example, we can look at Extract 7.24. The physician afrmatively identies blisters and cold sores at the end of a physical examination and then moves into an afrmative but minimized diagnosis in lines 78.
(7.24) 1126 (Dr. 3) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 DOC: MOM: DOC: MOM: DOC: -> -> MOM: => DOC: DOC: DOC: MOM: => => DOC: Yeah:. You know actually what those a:re pr=h .hh are primarily blisters back there. Yea:h? Its almost like shes got cold sores in thuh back of er throa:t. (Oh:[::.)/(Aw:::.) [And u:sually thatll go along with this just being viral. (.) [Really.= [#er-# =Y:eah. .hh One v thuh teachers told me it might be stre:p so:[:_ [.mlk Yeah we are starting to see some strep so Im gonna culture just in case .hh shes got both going on at the same ti:me but- .hh when you see: (you know)/(any uh) those #uh:# (thuh)/(that) white stuff you see back there is- is really not: like pus pus but it[s ya know like shes got blisters n [Oh yeah:_ Oh:::.


The minimizer just treats the illness as less signicant than it might be if it were not viral. And in response, the parent rst resists with a newsmark Really. Following



reconrmation, the parent goes on to challenge the physicians diagnosis by counterproposing a bacterial diagnosis: strep throat. Another way in which this diagnosis is presented as a minimal or less signicant one is that whereas bacterial illnesses are typically named strep throat or sinusitis or are labeled as infections as in sinus infection or ear infection, viral illnesses, are often not called infections. This further delegitimizes the visit because there is greater risk that the parent will hear the physician to be implying that the child is not actually ill. Another example of a minimized diagnosis is shown in Extract 7.25. Here, the physician announces the diagnosis afrmatively, but similar to 7.24, it is minimized in uh little viral col:d,; this minimization is accentuated with the contrastively prefaced but it doesnt look like anything much, which is resisted with the newsmark Really, in partial overlap.
(7.25) 104 [shown earlier in 4.4] 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 DOC: -> -> DOC: -> MOM: => DOC: DOC: MOM: DOC: MOM: => => DOC: MOM: => DOC: MOM: => DOC: .hh Uh:m, I think probably hes- ya know has uh little viral col:d, His nose is uh little stuffy, .hh Uh:m (.) tl=His throat does look uh little bit re:d [but it doesnt look like any[thing ] much,=^Yeah, [Right. [Really,] An I think thuh redness is really -again like I say from dripping down thuh ba:ck, .hh Uh:m his chest is perfectly clear. Theres nothing in his lungs at all[:. [Okay, .hh An I- ya know (you see) thuh fevers have gone dow:n, .hh uh:m_ I was just concerned cuz hes been so cranky an I thought well [there must be something botherin im= [^Well:=[that I cant: =[Well (that) could be. I mean-= =see_ What will happen.<Let=me show you here.

After the physician details justication for her claim that there is no problem, the mother offers more substantive resistance with I was just concerned cuz hes been so cranky. . .. Even when physicians do not actively minimize the presentation of a viral illness, it is vulnerable to being heard as minimized because of the absence of a true diagnostic label. This is shown in Extract 7.26.
(7.26) 161303 1 2 DOC: Its- Its=uh: .h Its something thats - b- been going through the schoo:ls the last two tuh three wee:ks. like

3 4 5 6 7 8 9 10 11 12 13



crazy:, .h ^I dont know if she had the same thing or not. .hh Its possible.=Some kids its very mi:ld an then others get it an uh ( ) theyre really hard like this [but- .h Its uh c- its uh virus, Its uh kind of [Mm: hm, u:, It- h- he doesnt seem to have an infection? (.) .hh Uh yi=y: its uh viral infection. [<See (what) the problem is .hh you know you think . . . [A viral infection.

The critical point here is that the diagnosis is presented as something thats - b- been going through the schoo:ls and then its uh virus and its uh kind of u:, which is as close to a diagnostic label as the physician gets here. And the physician treats this as indicating that the physician found no infection (line 9): something that she treats as problematic. It is interesting that the cutoff of c- in line 6 may represent the initiation of repair on what would have been uh cold, in which case uh virus upgrades the diagnosis to a medical category. But even so, it is problematic. These cases offer evidence that diagnoses, like treatment recommendations, are less likely to be resisted if they are delivered in such a way that they do not make light of the childs illness or the treatment recommendation but rather underscore the legitimacy of the parents decision to bring the child in for medical evaluation. Thus, just as the last section asserted the importance of formulating the treatment recommendation afrmatively, specically, and without minimization, this section argues that diagnoses will be less likely to be resisted if they are delivered without minimization. The principles of supporting visit legitimacy, addressing treatability, and forecasting a nonantibiotic outcome can be handled successfully in ways that are quite different than have been outlined here. Extract 7.27 provides an illustration of this. What appears to be more important than the exact format of the counseling is that the basic principles are oriented to by the physician. In this case, the physician packages his diagnosis and treatment recommendation together, projected as a multiunit explanation from the outset with Theres some good news, and theres some bad news.
(7.27) 211701 1 2 3 4 5 6 7 8 9 DOC: Okay. .h Theres some good news, and theres some bad news. Okay, uhm the good news is she doesnt look like she doesnt look like she has any kind of a bacteria i[nfection.=okay,=So she doesnt need any antibiotics. [Okay Okay, Her ears look <fanta:stic,> M[m hm, [Okay so no ear infection, .h doesnt look like uh strep




10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 42






throat at a:ll, Okay. <Okay_< and it also- she doesnt have any kind of a pneumonia. (.) Okay. No asthma, or no wheezing either [i- her lungs sound= [(mm hm,) =absolutely fantastic. Thats thuh good news. (Mk[ay:,) [The bad news is she does have an infection though. [Okay; [Okay, and the infection is mainly in her no:se and thats why she (i)s picking it(h) <Do we have kleenex [in here?, [He:re_ I ha:ve. Oh okay. [((getting kleenex out and handing it to girl)) [Uhm (1.2) But the- (uh) she does have an infection and its the virus in her nose. .hh The way yu=we get rid of viruses is not by antibiotics; Okay, .h but uhm actually what we do is we give (a) lot of ui:ds, Mm [hm:, [Okay, very very important, .h a lot of rest,=h very very important, to keep her body strong so she can ght that in[fection, [Uh huh .hh An then also (.) [we try to get <r:id of this:> (.)] [(( gesturing like nose picking ))] [as much as possible. Mkay, cuz only time will= [Mm hm, =actually=uh get rid of the viru[s also. [Okay.

Relative to the concern of minimization, what can be seen here is that the no-problem diagnosis and nonantibiotic treatment recommendation are explicitly identied as good news. The illness the child has is labeled bad news, and this validates the visit and the parents decision to come. The illness not only is not minimized but also has its signicance heightened by calling it an infection. Moreover, no antibiotics is also explicitly identied as good news, whereas the offering of uids is not identied as bad news but provided as an afrmative, specic, and nonminimized solution to the infection. This case suggests that a range of formal variations can be used by physicians in successfully dealing with parents in the delivery of no-problem diagnoses and nonantibiotic treatment. What this section has asserted is that in the counseling



phase of the visit, diagnoses and treatment recommendations will be less likely to engender resistance if they validate the childs illness and provide a solution to the childs problem.

In the context of routine viral upper respiratory infections, if physicians are to work to avoid inappropriate prescribing, they must attempt to also avoid parent interactional behaviors that pressure them for antibiotics. This chapter has argued that physicians who forecast a no-problem diagnosis and a nonantibiotic treatment to parents may help to get parents to accept this outcome, especially if this outcome is managed in such a way that the legitimacy of their medical visit is maintained and the treatability of the child is acknowledged. This chapter discussed several resources for forecasting a no-problem diagnosis and focused on three resources that both forecast this news and work to maintain the legitimacy of the visit and/or address the treatability of the child. First, online commentary delivered during the physical examination forecasts a no-problem diagnosis and a no-treatment recommendation. This practice not only forecasts the news but also, by providing parents with insight into how a noproblem diagnosis was reached, assists parents in accepting this outcome. Second, when physicians deliver a viral diagnosis with mitigation, parents were shown to be more likely to resist the diagnosis. Diagnoses that are delivered afrmatively and do not minimize the childs illness were argued to be optimal in securing parent acceptance of the nonantibiotic treatment outcome. Finally, treatment recommendations, like diagnoses, can be formulated positively or negatively. This chapter demonstrated that parents are less likely to resist a treatment recommendation that is formulated afrmatively, provides a specic next action step or solution to the childs problem, and does not minimize this solution. In this way, this resource both addresses the legitimacy of the visit and provides a solution to the problem the parent has sought help fortreatability. The prior chapters in this book demonstrated the important role that parents play in these pediatric encounters in terms of shaping the visit outcome. Although physicians hold the pen and thus, in the end, are the party truly accountable for inappropriate prescribing, this outcome has been shown to be very much an interactional product. Chapters 26 showed a number of ways in which parents mobilize interactional resources that are likelyfor a variety of different reasonsto have an impact on the prescribing decision. This chapter showed several interactional resources that some physicians make use of to successfully retard parent resistance and thus steer clear of an interactional environment that is more likely to lead to inappropriate antibiotic prescribing.


hen physicians in developed countries like the United States prescribe antibiotics to children who they believe have a viral upper respiratory tract infection, they generally know that they are committing a medical error. But what we have seen throughout this book is that the diagnosis and treatment of upper respiratory tract infections is not simply the result of applying a clinical algorithm. Rather, the diagnosis and treatment are arrived at in and through a moment-by-moment interaction with the parents and children. Recall Mangione-Smith and colleagues nding that where physicians perceive that parents expect an antibiotic, they are signicantly more likely to prescribe it for a viral condition (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006; Mangione-Smith, McGlynn, Elliott, Krogstad, & Brook, 1999). Thus, although it may sound rather straightforward to say no antibiotics for a viral upper respiratory tract infection, a close look at pediatric interactions like those shown throughout this book suggests that it is actually much more difcult to deny a sick child and the parents who simply want their child to feel better. So, at the root of a large-scale global health problem, as well as a classic social dilemma, is a micro-level problem in social interaction.

Summary of Findings
In chapters 25, we examined four phases of the acute pediatric visit and observed that in each both parents and physicians have interactional resources for negotiating the diagnostic and treatment outcome of the visit. These are summarized in table 8.1.




table 8.1. Parent Interactional Practices That Pressure Physicians for Antibiotics
Interactional Practice Candidate diagnosis presentation Primary Phase of Use Reason for the visit Frequency of Use 26% 9% 12% 17% 19% 8%

Mention of additional symptoms History taking Mention of alternative diagnoses History taking Diagnosis resistance Active treatment resistance Overt pressure Diagnosis delivery Treatment recommendation Treatment recommendation

We rst observed that during the problem presentation, parents can take a stance toward the childs illness as bacterial through their use of a candidate diagnosis. We saw that when parents introduce their childrens problems with a candidate diagnosis, as opposed to simply presenting their childrens symptoms, physicians treat them as seeking conrmation of their view of the childs illness but, more important, as looking for antibiotic treatment for it. Parents may use a candidate diagnosis for reasons other than communicating a desire for antibiotics: They may wish to underscore why they thought the problem required a medical visit or be genuinely concerned that the illness is serious. But physicians tend to focus on the desirability of antibiotics, as evidenced by their responses to candidate diagnoses either immediately or later in the visit. This shows that candidate diagnoses are one interactional resource through which parents can participate in and shape the diagnosis and treatment outcome. During the history-taking phase of the visit, we observed that with each question physicians ask, they display whether they are taking a stance toward that symptom as problematic or not. Thus, with each question, parents gain insight into what sort of trajectory the physician is on: toward a problem diagnosis and treatment or toward a no-problem diagnosis and nonantibiotic treatment. In this phase, we further observed that parents, particularly on hearing a question or series of questions that convey the physician to be on a no-problem trajectory, mention additional symptoms to push physicians away from their current no-problem trajectory, or they mention alternative possible diagnoses to encourage physicians toward an alternative trajectory. Again, these interactional practices may be motivated by a desire for antibiotics, by a concern that the childs illness is more serious than the physician appears to be treating it, or by a concern that their child needs some form of help to get well and they do not know what to do. Physicians typically treat these behaviors, like candidate diagnoses, as primarily indexing a desire for antibiotics, so by mentioning additional problematic symptoms or alternative possible diagnoses during history taking, parents can also inuence the visit outcome.



Next, when physicians deliver their diagnoses, we observed that this afrmatively takes a position toward the illness as problematic or not. If the physician takes a position that the illness is nonproblematic, this action potentially threatens both the legitimacy of the childs medical visit and the treatability of the child. Thus, the parent may, for either or both reasons, be concerned that they are not going to receive any solution to their childs problem, though this concern may be motivated by various issues: fears of having wasted the physicians time, because they desire antibiotics, or because they feel that the physician has not understood the childs illness in that the child is clearly sick and in need of treatment. Their primary resource for negotiating the outcome of the visit in this phase is to resist the diagnosis. We saw that diagnosis resistance could be accomplished quite minimally because virtually any sequence-initiating action delays progress of the ongoing physical examination (in the case of online diagnosis) or to treatment (as in the case of ofcial diagnoses). But parents must actively resist if they are to take issue with this phase of the visit because response is not normatively required. This phase is, in some ways, particularly interesting; it is where the physicians expertise and authority are, at least in theory, at their highest. Empirically, the lack of a normative response to a diagnosis delivery is in line with this theoretical position. However, parents do nonetheless intrude into this domain, though they do so most commonly in ways that downplay this intrusion and rarely directly question the diagnosis. Regardless of the directness, diagnosis resistance is yet another interactional resource for shaping the diagnosis and treatment decision. Resistance during the diagnosis phase is rather different from what we observed in the treatment phase, where physicians and parents are far more oriented to a shared responsibility and domain of authority. This was evidenced by the normative organization that makes parent acceptance of a suggested treatment conditionally relevant before a physician will initiate a move to the next activity or to closing the encounter. In line with this, but in contrast with the diagnosis context, here parents who fail to respond at all are treated as resisting. Therefore, a broader range of resistance to treatment recommendations, including both passive and active types of resistance, is available to parents. This phase is also the nal phase where there is a structurally provided opportunity for visit outcome negotiation, and thus it is perhaps not surprising that the negotiation in this phase is typically the most overt, including questions about whether antibiotics are necessary. But even here, parents behaviors may be driven by different factors. Although a desire for antibiotics is at times plain, at other times, the same collocation of issues that can motivate behaviors at other points in the visit are analyzable here: Parents may be pushing for treatment because they feel that the legitimacy of their visit has been threatened; they may feel that their child is seriously ill and will get still sicker if not properly treated; they may feel that without medicine their child will be sick longer, that they will be out of work longer, that the child will be out of day care longer, and so on. But again, physicians tend to primarily address treatment resistance as a behavior driven by a desire for antibiotics. In chapter 6, we examined situations when negotiation is most transparent: when parents overtly lobby for antibiotics. We saw that overt lobbying took the form of



one of four main practices that varied in their directness but all pushed overtly for antibiotics and could occur at any point in the visit. We saw examples occurring at the very opening of the medical encounter, during history taking, amid the physical examination, and through to the treatment recommendation. We also saw that diagnosis resistance or treatment resistance could at times involve overt lobbying for antibiotics, though most of the time it remained covert. Examining the more overt cases of physician-parent negotiation over antibiotics provided yet further support for the claims that true negotiation is going on, despite the fact that a vast majority of the negotiation goes on covertly. Observing that negotiation is occurring and how it is being done is only part of the story. If we want to attempt to reduce inappropriate prescribing, we must also understand what is driving these behaviors so we understand how best to reduce them. We dealt with the complexity behind parent behaviors most in chapter 7, which showed that physicians could minimize parent resistance to no-problem visit outcomes with behaviors that worked to maintain the legitimacy of the visit, forecast to the parent early in the visit that their child might not have a condition that is treatable with antibiotics, and addressed the general treatability of the childs illness. Forecasting the bad news scenario that a condition is not problematic and that there will be no antibiotics helps ease parents into this outcome and may therefore help them accept it (Maynard, 2003). This may then be why behaviors such as rejecting a candidate diagnosis or addressing additional symptoms or diagnostic possibilities are useful. However, these behaviors do nothing to maintain the legitimacy of the visit. By contrast, online commentary, when used as discussed here, can work to forecast a no-problem diagnosis and a no-antibiotics treatment early in the visit before an ofcial diagnosis is reached. It also provides parents with insight into why the diagnosis is what it is and reassures parents that each possibly concerning symptom was investigated. Moreover, online commentary also give parents insight into what minor problems the child has that may have caused the child discomfort or the symptoms that led the parent to schedule the visit. Thus, online commentary can also work to maintain the legitimacy of the visit at exactly the point where there was a possible threat to its legitimacy. Similarly, when physicians formulate diagnosis announcements straightforwardly and afrmatively, they avoid threatening the legitimacy of the visit. By contrast, diagnoses that minimize the illness, no matter how well intentioned, do threaten visit legitimacy. Chapter 7 took the position that physicians delivering no-problem diagnoses would do best to present the diagnosis as legitimate without minimization because this, at the very least, does not actively threaten the legitimacy of the visit any more than null ndings do by denition. Finally, we observed that, when physicians deliver treatment recommendations, both negative announcements and positive announcements that are vague or nonspecic or that minimize the treatment are problematic and more likely to meet with parent resistance. This is perhaps the strongest evidence that parents are not necessarily primarily driven by a desire for antibiotics but that antibiotics embody a cure for their childs illness, and a cure is what they are in search of. If a physician can offer them an alternative solution, they may be more willing to try it and not resist than if the physician denies them this.



Parent pressure and physician responses to it are reections of a societal-level problem, and it is partially the society as a whole that is responsible for the current state of affairs. In chapter 1, we analyzed the use of antibiotics in the face of the bacterial resistance problem as a commons dilemma. Jared Diamonds Collapse: How Societies Choose to Fail or Succeed (2005) offers numerous examples of how some societies destroy themselves through poor decision making, often with respect to commons dilemmas. One of the latter chapters asks the question Why do some societies make disastrous decisions? The problem we are discussing here is a disaster in the making: We are quickly reaching a time when a great many illnesses will no longer be curable with antibiotics, and we will probably once again risk losing the lives of especially vulnerable members of our societychildren, the elderly, and the immunocompromisedto such illnesses. According to Diamond, four levels of decit cause societies to fail in solving their problems: failure to anticipate a problem; failure to perceive a present problem; failure to try to solve a perceived problem; failure to solve a problem despite attempts. In our present case, many (but by no means all) parents do not perceive the problem at any level and thus are in no position to contribute to a solution. This, then, is one level of our problem. As long as parent pressure remains a part of the equation, a problem remains at both the individual and societal levels. Thus, parents must become aware of the problem so that they can then become part of the solution. The primary means thus far for heightening parent awareness of the bacterial resistance and inappropriate antibiotic-prescribing problem has been public health campaigns (e.g., Finkelstein et al., 2001; Madle, Kostkova, Mani-Saada, Weinberg, & Williams, 2004; Perz et al., 2002; Wheeler et al., 2001). Various international and national associations have web resources for educating consumers, and research about the impact of this is going on (Madle et al., 2004). But most efforts focus on the difference between bacterial and viral conditions and what antibiotics can and cannot treat. Although this is clearly relevant, it does not entirely handle the problem for multiple reasons. One reason that was observable in these data, as well as in my own conversations with parents, is that even parents who understand that antibiotics do not cure viral infections sometimes pressure physicians for antibiotics. This, then, reects a critical disconnect between medical thinking and parent thinking. The disconnect appears to be rooted in an understanding that viral infections are less serious than bacterial infections. As I mentioned before, physicians have, to a big extent, exacerbated this problem in the way they talk about viral infections. What parents have learned is that if their children have minor symptoms, then they have viral infections, whereas if their children are very sick, then they probably have bacterial infections and need antibiotics. So parents who believe their child is very sick will pressure for antibiotics and will be unlikely to accept that their child has a viral illness because the child is sicker than they have been taught is associated with such illnesses. Even if public health campaigns address this issue, as long as physicians reinforce this model through their own way of talking about viral infections, the problem will persist.



Related to this issue is a disconnect between illness models that parents and the medical community may have. Parents across these data view fevers and pain as symptoms of infection that require treatment, most often antibiotic treatment. These symptoms are, for them, not associated with minor colds but with major bacterial infections or at least infections (by which they mean illnesses treatable with antibiotics). Once again, many physicians and public health campaigns miss opportunities to teach parents that these symptoms exist in both types of infections. Thus, both physicians and public health campaigns must work to address parent models of illness. Both need to make the connection clearer between prescribing for a cold and the risks to other community members. And both need to incorporate a way of discussing viral infections that no longer minimizes either the illness itself or the ability we have to help children who suffer from these infections. Returning to Diamonds four levels of decit, physicians are, at least intellectually, aware of the problem. They anticipated it as early as 1945, and the literature reects a broad awareness of the scale of the problem of bacterial resistance, as well as of the link with inappropriate antibiotic prescribing. Still, a range of issues here contribute to a continuing problem: First, although physicians know about the problem, they cannot actually see it. The slow trend toward less effective antibiotics conceals the problem in the same way that Diamond argues global climate change was concealed. This makes it easier to do the wrong thing under pressure. Probably more important, though, is the failure of physicians to attempt to solve the problem. Much research attempts to modify physician behavior in prescribing antibiotics (Avorn & Soumerai, 1983; Bauchner & Philipp, 1998; Belongia et al., 2001; Coley et al., 2000; Davis, Thomson, Oxman, & Haynes, 1995; Doyne et al., 2004; Finkelstein et al., 2001; Perz et al., 2002). And the education on both the parent and physician sides appears to be contributing to a reduction in antibiotic use (Finkelstein et al., 2003). But it is still clear that physicians continue to prescribe inappropriately. Our discussion of why physicians commit this medical error has shown that they are responding to perceived pressure by parents in a very understandable prosocial way: At least sometimes, they succumb to the perceived pressure. But this book also points to a range of interactional lessons. First, although parent behaviors such as using a candidate diagnosis or mentioning additional problematic symptoms may feel like pressure for antibiotics, they are not always motivated by this goal. Physicians may need to explicitly register when parents use these behaviors but then attempt to sort out whether the behavior is driven more by parent anxiety about the illness, a desire for a solution or something to do for their child, or a desire for antibiotics. In all cases, the use of online commentary has been shown to help prepare parents for a no-problem, no-antibiotics treatment outcome while maintaining the legitimacy of the visit. Moreover, nonminimized, afrmative, and specic diagnoses and treatment recommendations may help physicians manage parents who use these behaviors. Second, across-the-board use of afrmative, nonminimized, and specic diagnoses and treatment recommendations reduces parent resistance yet also educates parents to the idea that viral illnesses are real illnesses with real treatments. Based



on how physicians talked about viral illnesses in these data, parents were receiving consistent reinforcement of the idea that viral illnesses were minor relative to bacterial illnesses and were not treatable. Physicians can implicitly educate parents that viruses are every bit as serious as bacterial illnesses and are treatable (i.e., something can be done even if they cannot be cured), and this may, in time, reduce parent pressure for antibiotics. Third, although physicians do need to educate parents, the place to educate them about the role of antibiotics in upper respiratory tract infections is after parents have received an afrmative, specic, and nonminimized diagnosis and treatment recommendation. Only then are parents in a position to learn. Prior to receiving this, parents are far more likely to hear it as an insufcient treatment recommendation. Thus, ruling out the need for antibiotics and explicating the connection between viral infections, bacterial infections, and antibiotics should be done following the initial treatment recommendation. Returning once again to Diamonds four levels, we can recognize that for the reasons described previously, parents and physicians have not been trying to solve the problemwhether due to lack of perception of a problem or because of pressure. But as has been mentioned, at the societal level, there are efforts being made to curb antibiotic prescribing for viral infections. We can ask the question, Why have these efforts not yet worked? This returns us to our commons dilemma. For societal-level efforts to work, both physicians and parents must not only perceive the problem but also perceive themselves to be part of the solution. Physicians may suffer from perceiving the prescribing of antibiotics in individual cases to be either (1) worth the damage done to society or (2) not truly contributing sufciently to the problem to actually constitute wrongdoing. With respect to the former, physicians who perceive themselves to be at risk of alienating a parent by not prescribing, or who cannot get the parent to leave without prescribing, may well feel inclined to prescribe, even though they know that they should not. As one physician put it to me, If I dont give them the antibiotic they want, theyll just go next door. With respect to the latter, the perceived problem is spread out so diffusely that individuals may not sufciently perceive their actions to matter. These dimensions are critical to solutions to commons dilemmas. If people feel that their actions will not have any discernible effect on the situation, then they will not cooperate with what is better for the group (Kollock, 1998). We can also speculate on another dimension of this problem that is commonly at issue with commons dilemmas: group identity. Physicians and parents alike, even if they perceive and understand the problem of inappropriate antibiotics prescribing, may nonetheless prioritize their own needs over societys if they fail to perceive themselves as intimately connected with the society. Kramer and Brewer (1986a, 1986b) show that in simulated commons dilemmas, subjects are more willing to exhibit personal restraint if they identify as members of a group. Diamond highlights this as well (2005). Kollock notes that one reason individuals may be more willing to cooperate if they feel they are part of a group is that a collective social identity may increase member altruism (Kollock, 1998). Alternatively, it may be more strategic (e.g., see Karp, Jin, Yamagishi, & Shinotsuka, 1993). As Kollock summarizes, the



effects of group identity may stem from a belief in the interdependencies of group members and expectations of reciprocity among the members (1998: 198). Even when reciprocity is not logically possible, the expectation is sometimes manifest (e.g., Karp et al., 1993). We can only speculate that whether parents and physicians feel themselves to have a stronger collective identity with their community is an important factor here. Parents who recognize the impact of their actions on society and who feel a strong sense of collective identity may be less likely to push for antibiotics. Similarly, physicians who believe that a child is suffering from a viral illness and therefore should not be prescribed antibiotics may be less likely to do so if they feel their actions will adversely affect the society. If physicians do not believe that their actions matter, then, as with similar types of social dilemmas, they may be more likely to prioritize their individual patient. If they are receiving pressure, then succumbing to that pressure, for interactional reasons, may be more attractive. It is certainly possible that a sense of collective identity is playing a role in this dilemma. With all of these factors, and especially this last factor, we would expect to see quite different rates of prescribing internationally, nationally, and regionally. And this is certainly the case, as we saw in chapter 1. To date, there has been no study that attempts to account for why some countries have lower rates of inappropriate antibiotic prescribing, though these rates have, as discussed earlier, been viewed as a primary contributor to the rate of bacterial resistance in the country. Among the many possible contributing factors outside what we discussed in the last section are health policy guidelines, economic factors such as drug costs, cultural attitudes toward illnesses, the role of physicians, and medication, as well as the role that various health care providers such as pharmacists play in the health care system. But in addition to these might be, as discussed in the last section, the degree to which individuals feel that they are part of a local community. Whether or not this is present in a given culture, this attitude should be present in medical schools, and physicians who begin to bring this way of thinking into their practices may make it possible for parents to see social connections that they might not otherwise see. And these issues do not take away from the legitimacy of their visit at all, a further benet to their inclusion in any discussion of antibiotic prescribing. Just as the problem of antimicrobial resistance is multifaceted, any solution to the problem will be as well. One alteration that might make a difference in countries such as the United States and Australia would be a push at the level of the medical school to encourage physicians in visits such as those examined in this book to maintain a focus on what their actions do to the larger society and to the vulnerable members of the population in particular. Instead of telling parents about the differences between viral infections and bacterial infections, perhaps parents need to hear stories like the one told by Alexander Fleming in his Nobel Prize acceptance speech, quoted in chapter 1. Finally, the problems facing developing countries with respect to bacterial resistance are as much the problems of the Americans as our problems are to developing countries like Bangladesh and Vietnam because, unlike humans, bacteria do not know nor do they respect national or regional boundaries. International travel makes



it possible for a country like The Netherlands to be quite effective in changing antibiotic prescribing health policies within their borders and yet still face problems with bacterial resistance. Currently, we are on track for a disastrous outcome: a return to the era of no cure for many bacterial infections. Our challenge is to change both our policies and our way of thinking about antibiotic prescribing.

This page intentionally left blank


1. Temporal and Sequential Relationships

A. Overlapping or simultaneous talk is indicated in a variety of ways. Separate left square brackets, one above the other on two successive lines with utterances by different speakers, indicates a point of overlap onset, whether at the start of an utterance or later. Separate right square brackets, one above the other on two successive lines with utterances by different speakers, indicates a point at which two overlapping utterances both end, where one ends while the other continues, or simultaneous moments in overlaps that continue. B. Equal signs ordinarily come in pairs: one at the end of a line and another at the start of the next line or one shortly thereafter. They are used to indicate two things: 1. If the two lines connected by the equal signs are by the same speaker, then there was a single, continuous utterance with no break or pause, which was broken up to accommodate the placement of overlapping talk. 2. If the lines connected by two equal signs are by different speakers, then the second followed the rst with no discernible silence between them or was latched to it. C. Numbers in parentheses indicate silence, represented in tenths of a second; what is given here in the left margin indicates 0.5 seconds of silence. Silences may be marked either within an utterance or between utterances.

[ [ ] ]





D. A dot in parentheses indicates a micropause, hearable but not measured; ordinarily less than 0.2 seconds.

2. Aspects of Speech Delivery (Including Intonation)

A. The punctuation marks are not used grammatically, but to indicate intonation. The period indicates a falling, or nal, intonation contour, not necessarily the end of a sentence. Similarly, a question mark indicates rising intonation, not necessarily a question, and a comma indicates continuing intonation, not necessarily a clause boundary. A combined question mark and comma indicates a rise stronger than a comma but weaker than a question mark. An underscore following a unit of talk indicates level intonation. The semicolon indicates that the intonation is equivocal between nal and continuing. B. Colons are used to indicate the prolongation or stretching of the sound just preceding them. The more colons, the longer the stretching. C. A hyphen after a word or part of a word indicates a cutoff or self-interruption. D. Underlining is used to indicate some form of stress or emphasis, either by increased loudness or higher pitch. The more underlining, the greater the emphasis. E. The degree sign indicates that the talk following it was markedly quiet or soft. When there are two degree signs, the talk between them is markedly softer than the talk around it. G. The caret indicates high pitch. Within a word, it indicates a pitch rise and fall. H. The combination of more than and less than symbols indicates that the talk between them is compressed or rushed. Used in the reverse order, they indicate that a stretch of talk is markedly slowed or drawn out. The less than symbol by itself indicates that the immediately following talk is jump-started, that is, sounds like it starts with a rush. I. Hearable aspiration is shown where it occurs in the talk by the letter h the more hs, the more aspiration with each h representing approximately 0.1 seconds. The aspiration may represent breathing, laughter, or the like. If

. ? , ?, _ ; :: word

two ^ >< <> < hhh



it occurs inside the boundaries of a word, in may be inclosed in parentheses to set it apart from the sounds of the word. If the aspiration is an inhalation, it is shown with a dot before it.

3. Other Markings
A. Double parentheses are used to mark transcribers descriptions of events, rather than representations of them. Thus ((cough)), ((sniff)), ((telephone rings)), and the like. (word) B. When all or part of an utterance is in parentheses, or the speaker identication is, this indicates uncertainty on the transcribers part but represents a likely possibility. ( ) Empty parentheses indicate that something that could not be understood is (or is possibly) being said. (( ))

This page intentionally left blank


Chapter 1 1. Although many countries have public health campaigns to educate patients about antibiotics, those who do not understand that there are social consequences to their use certainly will not recognize a decision about antibiotic usage as a social dilemma at all. For those patients who do, and certainly for most physicians, antibiotic usage does represent such a dilemma. 2. Although patients and parents are acting on a sort of commons dilemma, in actuality antibiotics are not rational at the individual level if they do not help. But the point is that if they are perceived as helpful at the individual level but costly at the collective level, then we do, in fact, have a commons dilemma. 3. See Foster (1974) for an interesting discussion of key differences between a sociology of medicine and an anthropology of medicine perspective. Chapter 2 1. Parents can, even in the very opening of the visit, overtly pressure physicians for antibiotics (see chapter 6, Extract 6.7), but to do so they must buck the normative phase structure of the visit and initiate a course of action that counters the one normally followed in acute care visits. This is vanishingly rare in these data. 2. The percentage actually rises to 26% when implied candidate diagnoses are taken into account. This will be discussed later in the chapter. Here, I begin with a discussion of the explicit candidate diagnoses. 3. In linguistics, the term marked is used in many different ways. Here, when I refer to unmarked, I mean the normal, usual, or default way of doing some action. When I use the term marked, I mean that the form departs from the usual. This use is discussed by Levinson (1983, 1987, 2000).




4. For a discussion of rst versus second position and epistemic rights, see Heritage (1998), Heritage and Raymond (2005), and Stivers (2005a). Chapter 3 1. Also see Heinemanns discussion of preference and polarity in Danish (2005) and Koshiks discussion in another institutional context (2002). 2. Also closely related is Sackss original idea of preferencethat events can be characterized in such a way as to rank an activity such as dinner as including talk (a rst preference invitation), as opposed to a lower activity such as talk (which does not include dinner) (Sacks, 1992, pp. 367369). And Drews maximal properties of description (1992) partially overlaps with a concept called minimization (Levinson, 1987). All of these ideas circle around the fundamental observation that speakers and hearers do not state everything possibleall events are partially describedbut which absences are understood in what ways is contingent on a number of factors, including presumptions interactants make about the relevant circumstances surrounding the activity. Chapter 5 1. The placement of the treatment recommendation prior to the diagnosis may help to avoid further sequences of parent resistance. 2. Although this claim requires more investigation, it appears that the period-intoned Okay and Alright are treated as doing acceptance, whereas these objects said with comma or level intonation may be offering only acknowledgment. Beach (2001) has explored prosodic variation in the token okay. The period-intoned Okay. or Alright. appears to be a more minimal form of acceptance than the Thats ne type of assessment. It may be that fuller forms of acceptance are treated as optimal in this environment and therefore lead most directly to activity closure. 3. Here and in some of the later examples, I do not have access to video to determine if there is visible behavior during these silences. However, as will be shown particularly in the next section, if there were any visible behavior such as nodding during these spaces, the physicians do not treat this as sufcient to accept the recommendation but rather treat it as indicating that the activity is (or should be) continuing (Schegloff, 1982). 4. If the physician had elicited agreement from the childwhich is explicitly sought this might have helped in gaining a somewhat coerced acceptance by the mother. It is in this sense that I see this as a practice for pursuing parent acceptance. 5. This ordering parallels the larger principle of interaction that conict-producing turns are typically delayed. Preference organization in second pair part deliveries displays this principle. There, delays of various sorts precede dispreferred responses, thus allowing the just prior speaker an opportunity to reformulate the rst pair part and thereby remove the relevance of the dispreferred second pair part (Heritage, 1984b; Pomerantz, 1984). A similar pattern is observable with respect to shifts from passive to active treatment resistance. 6. A similar instance of this shift from a recommendation to do nothing to a recommendation to do something was seen in Extract 5.4a, where the physician rst suggested waiting and then suggested lots of uids. Although only an incremental shift in position, this change may nonetheless be responsive to parent resistance of a do nothing approach to managing the childs illness. 7. This was a concern implied early in the visit. The mother offered only the symptom of a runny nose as her problem presentation. However, apparently drawing on the patients chart, the doctor offered for conrmation a description of the color of the discharge. Built into this



turn is the doctors understanding that the discharge is currently greenish, thus conveying the understanding that the parent is worried about sinusitis. 1 2 3 4 5 6 7 MOM: DOC: DOC: Shes [had uh r- runny nose off an on for about two weeks. [Uh huh ( ) Okay:_ (0.8) ((DOC hearably erasing something)) DOC: -> An:d initially: was it kinda clea:r? an then it -> started [g e t t i n g t h i s: greenish co[lor, MOM: -> [Initially clear but it got (.) green. [Right.

8. Here, the physician offers a straightforward plan for what should be done in the future. In previous cases where doctors introduced plans similar to this, it was offered by way of suggesting when they would reconsider treatment or by way of addressing when the parent would need to return to the ofce. In Extract 5.8a, by contrast, the plan is a move toward closing the encounter by providing a straightforward future action plan. 9. This difference is probably informed by two factors: Coding was slightly more conservative in the Metro data set, so some behaviors that were counted as initiating a negotiation in the Seaside data set were not counted in the Metro data. In addition, though, the latter data were much more heterogeneous in terms of ethnicity and socioeconomic factors. This is important because, for instance, African Americans did not resist the treatment in any case in the Metro data, whereas other ethnic groups did (Mangione-Smith, Elliott, Stivers, McDonald, & Heritage, 2006). Chapter 7 1. Problem online commentary is also used (Mangione-Smith, Stivers, Elliott, McDonald, & Heritage, 2003). This chapter will not address this practice because although it forecasts a diagnosis, it generally forecasts a bacterial diagnosis and thus is outside the realm of this chapter. All references to online commentary here will refer to no-problem online commentary. 2. Note that this is a way of downgrading the claim as well (Chafe & Nichols, 1986). The claim I dont see any uid is not as strong as There isnt any uid because it leaves open the possibility of there being uid that is unseen. 3. Sudafed, an over-the-counter medication, and prescription allergy medications were other examples of medication that were ruled out. This was not at all frequent. 4. This is the basic argument of noticeable absences observed by Schegloff (1968). At any given point, any number of actions may be absent, but most of them were not relevant for the sequence in progress, so they are not noticeably absent. Here, the point is that because nothing is relevant following a diagnosis, silence does not mark a noticeable absence in this context.

This page intentionally left blank


Adam, D. (2002). Global antibiotic resistance in Streptococcus pneumoniae. Journal of Antimicrobial Chemotherapy, 50, 15. Albrich, W. C., Monnet, D. L., & Harbarth, S. (2004). Antibiotic selection pressure and resistance in Streptococcus pneumoniae and Streptococcus pyogenes. Emerging Infectious Diseases, 10, 514517. Aronoff, S. C. (1996). Antimicrobials in children and the problem of drug resistance. American Family Physician, 54, 44. Athreya, B. H., & Silverman, B. K. (1985). Pediatric Physical Diagnosis. Norwalk, CT: Appleton-Century-Crofts. Atkinson, J. M. (1984). Public speaking and audience responses: Some techniques for inviting applause. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 370409). Cambridge: Cambridge University Press. Avorn, J., & Solomon, D. H. (2000). Cultural and economic factors that (mis)shape antibiotic use: The nonpharmacologic basis of therapeutics. Annals of Internal Medicine, 133, 128135. Avorn, J., & Soumerai, S. B. (1983). Improving drug-therapy decisions through educational outreach. A randomized controlled trial of academically based detailing. New England Journal of Medicine, 308, 14571463. Baquero, F., Baquero-Artigao, G., Canton, R., & Garcia-Rey, C. (2002). Antibiotic consumption and resistance selection in Streptococcus pneumoniae. Journal of Antimicrobial Chemotherapy, 50, 2738. Barden, L. S., Dowell, S. F., Schwartz, B., & Lackey, C. (1998). Current attitudes regarding use of antimicrobial agents: Results from physicians and parents focus group discussions. Clinical Pediatrics, 37, 665672. Bartoloni, A., Cutts, F., Leoni, S., Austin, C. C., Mantella, A., Guglielmetti, P., et al. (1998). Patterns of antimicrobial use and antimicrobial resistance among healthy children in Bolivia. Tropical Medicine and International Health, 3, 116123.




Bates, B., Bickley, L. S., & Hoekelman, R. A. (1995). Physical Examination and History Taking (6th ed.). Philadelphia: Lippincott. Bauchner, H., & Philipp, B. (1998). Reducing inappropriate oral antibiotic use: A prescription for change. Pediatrics, 102, 142145. Beach, W. (2001). Some prosodic displays of incongruity in interaction. Paper presented at the National Communication Association, Atlanta, GA. Beckman, H. ,& Frankel, R. (1984). The effect of physician behavior on the collection of data. Annals of Internal Medicine, 101, 692696. Beckman, H. B., Markakis, K. M., Suchman, A. L., & Frankel, R. M. (1994). The doctorpatient relationship and malpractice: Lessons from plaintiff depositions. Archives of Internal Medicine, 154, 13651370. Bell, N. (2002). Antibiotic resistance: The Iowa experience. American Journal of Managed Care, 8, 988994. Bell, R. A., Wilkes, M. S., & Kravitz, R. L. (1999). Advertisement-induced prescription drug requests: Patients anticipated reactions to a physician who refuses. Journal of Family Practice, 48, 446452. Belongia, E. A., Knobloch, M. J., & Kieke Jr., B. A. (2005). Impact of statewide program to promote appropriate antimicrobial drug use. Emerging Infectious Diseases, 11, 912920. Belongia, E. A., Sullivan, B. J., Chyou, P. H., Madagame, E., Reed, K. D., & Schwartz, B. (2001). A community intervention trial to promote judicious antibiotic use and reduce penicillin-resistant Streptococcus pneumoniae carriage in children. Pediatrics, 108, 575583. Blanchard, C. G., Labrecque, M. S., Ruckdeschel, J. C., & Blanchard, E. B. (1988). Information and decision-making preferences of hospitalized adult cancer patients. Social Science and Medicine, 27, 11391145. Borzo, G. (1997). Consumer drug ads booming: FDA reviews restrictions. American Medical News, 40, 1, 37. Bosu, W. K., & Ofori-Adjei, D. (1997). Survey of antibiotic prescribing pattern in government health facilities of the Wassa west district of Ghana. East African Medical Journal, 74, 138142. Boyd, E. A. (1997). Constructing histories and negotiating care: Professional discourse during medical peer review. Unpublished dissertation, University of California, Los Angeles. Boyd, E. A. (1998). Bureaucratic authority in the company of equals: The interactional management of medical peer review. American Sociological Review, 63, 200224. Boyd, E. A., & Heritage, J. (2006). Taking the patients medical history: Questioning during comprehensive history-taking. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 151 184). Cambridge: Cambridge University Press. Braddock, C. H., Edwards, K. A., Hasenberg, N. M., Laidley, T. L., & Levinson, W. (1999). Informed decision making in outpatient practice: Time to get back to basics. Journal of the American Medical Association, 282, 23132320. Braithwaite, A., & Pechere, J. C. (1996). Pan-European survey of patients attitudes to antibiotics and antibiotic use. Journal of International Medical Research, 24, 229238. Britten, N. (1994). Patient demand for prescriptions: A view from the other side. Family Practice, 11, 6266. Britten, N., Jenkins, L., Barber, N., Bradley, C. P., & Stevenson, F. A. (2003). Developing a measure for the appropriateness of prescribing in general practice. Quality and Safety in Health Care, 12, 246250. Britten, N., Stevenson, F., Gafaranga, J., Barry, C., & Bradley, C. (2004). The expression of aversion to medicines in general practice consultations. Social Science and Medicine, 59, 14951503.



Britten, N., Stevenson, F. A., Barry, C. A., Barber, N., & Bradley, C. P. (2000). Misunderstandings in prescribing decisions in general practice: Qualitative study. British Medical Journal, 320, 484488. Britten, N., & Ukoumunne, O. (1997). The inuence of patients hopes of receiving a prescription on doctors perceptions and the decision to prescribe: A questionnaire study. British Medical Journal, 315, 15061510. Brody, D. S. (1980). The patients role in clinical decision making. Annals of Internal Medicine, 93, 718722. Brody, D. S., Miller, S. M., Lerman, C., Smith, M. D., & Caputo, C. (1989). Patient perception of involvement in medical care: Relationship to illness attitudes and outcomes. Journal of General Internal Medicine, 4, 506511. Brody, D. S., Miller, S. M., Lerman, C. E., Smith, D. G., Lazaro, C. G., & Blum, M. J. (1989). The relationship between patients satisfaction with their physicians and perceptions about interventions they desired and received. Medical Care, 27, 10271035. Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge: Cambridge University Press. Butler, C. C., Kinnersley, P., Prout, H., Rollnick, S., Edwards, A., & Elwyn, G. (2001). Antibiotics and shared decision making in primary care. Journal of Antimicrobial Chemotherapy, 48, 435440. Bybee, J., Perkins, R., & Pagliuca, W. (1994). The evolution of grammar. Chicago: University of Chicago Press. Byrne, P. S., & Long, B. E. L. (1976). Doctors talking to patients: A study of the verbal behaviours of doctors in the consultation. Exeter: Royal College of General Practitioners. Cars, O., Mlstad, S., & Melander, A. (2001). Variation in antibiotic use in the European Union. Lancet, 357, 18511853. Cassileth, B. R., Zupkis, R. V., Sutton-Smith, K., & March, V. (1980). Information and participation preferences among cancer patients. Annals of Internal Medicine, 92, 832836. Centers for Disease Control, National Center for Health Statistics. (2004). National Ambulatory Medical Care Survey, 2004 [Data le]. Available from CDC Web site, http://www. Chafe, W., & Nichols, J. (1986). Evidentiality: The linguistic coding of epistemology. Norwood, NJ: Ablex. Clayman, S., & Heritage, J. (2002). The news interview: Journalists and public gures on the air. Cambridge: Cambridge University Press. Clayman, S. E., Elliott, M., Heritage, J., & McDonald, L. (2006). Historical trends in questioning presidents 19532000. Presidential Studies Quarterly, 36, 561583. Clayman, S. E., Heritage, J., Elliott, M., & McDonald, L. (in press). When does the watchdog bark? Conditions of aggressive questioning in presidential news conferences. American Sociological Review. Clift, R. (2001). Meaning in interaction: The case of actually. Language, 77, 245291. Cockburn, J., & Pit, S. (1997). Prescribing behavior in clinical practice: Patients expectations and doctors perceptions of patients expectationsA questionnaire study. British Medical Journal, 315, 520523. Cohen-Cole, S. A. (1991). The medical interview: The three-function approach. St. Louis: Mosby. Coley, K. C., Skledar, S. J., Fine, M. J., Yealy, D. M., Gleason, P. P., Ryan, M. L., et al. (2000). Changing physician prescribing behavior: The community-acquired pneumonia intervention trial. American Journal of Health System Pharmacy, 57, 15061510. Coulter, A. (1997). Partnerships with patients: The pros and cons of shared clinical decisionmaking. Journal of Health Services Research Policy, 2, 112121.



Cristino, J. M. (1999). Correlation between consumption of antimicrobials in humans and development of resistance in bacteria. International Journal of Antimicrobial Agents, 12, 199202. Dagan, R. (2000). Clinical signicance of resistant organisms in otitis media. Pediatric Infectious Disease Journal, 19, 378382. Davis, D. A., Thomson, M. A., Oxman, A. D., & Haynes, R. B. (1995). Changing physician performance: A systematic review of the effect of continuing medical education strategies. Journal of the American Medical Association, 274, 700705. Deber, R. B. (1994). Physicians in health care management. The patient-physician partnership: Decision making, problem solving and the desire to participate. Canadian Medical Association Journal, 151, 423427. Deeks, S. L., Palacio, R., Ruvinsky, R., Kertesz, D. A., Hortal, M., Rossi, A., et al. (1999). Risk factors and course of illness among children with invasive penicillin-resistant Streptococcus pneumoniae. The Streptococcus pneumoniae Working Group. Pediatrics, 103, 409413. Diamond, J. (2005). Collapse: How Societies Choose to Fail or Succeed. New York: Viking Penguin. Doern, G. V., & Brown, S. D. (2004). Antimicrobial susceptibility among community-acquired respiratory tract pathogens in the USA: Data from PROTEKT US 200001. Journal of Infection, 48, 5665. Doyne, E. O., Alfaro, M. P., Siegel, R. M., Atherton, H. D., Schoettker, P. J., Bernier, J., et al. (2004). A randomized controlled trial to change antibiotic prescribing patterns in a community. Archives of Pediatric and Adolescent Medicine, 158, 577583. Drew, P. (1992). Contested evidence in a courtroom cross-examination: The case of a trial for rape. In P. Drew & J. Heritage (Eds.), Talk at Work: Social Interaction in Institutional Settings (pp. 470520). Cambridge: Cambridge University Press. Drew, P. (1997). Open class repair initiators in response to sequential sources of trouble in conversation. Journal of Pragmatics, 28, 69101. Drew, P. (2006). Misalignments in after-hours calls to a British GPs practice: A study in telephone medicine. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 416444). Cambridge, England: Cambridge University Press. Drew, P., & Heritage, J. (1992a). Analyzing talk at work: An introduction. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 365). Cambridge: Cambridge University Press. Drew, P., & Heritage, J. (Eds.). (1992b). Talk at Work. Cambridge: Cambridge University Press. du Pr, A. (2000). Communicating about Health: Current Issues and Perspectives. Mountain View, CA: Mayeld. Dunnell, K., & Cartwright, A. (1972). Medicine Takers, Prescribers, and Hoarders. London: Routledge and Kegan Paul. Elwyn, G., Edwards, A., & Kinnersley, P. (1999). Shared decision-making in primary care: The neglected second half of the consultation. British Journal of General Practice, 49, 477482. Emanuel, E. J., & Emanuel, L. L. (1992). Four models of the physician-patient relationship. Journal of the American Medical Association, 267, 2221. Emerson, C. (1983). Bakhtin and Vygotsky on internalization in language. Quarterly Newsletter of the Laboratory of Comparative Human Cognition, 5(1), 913. Ende, J., Kazis, L., Ash, A., & Moskowitz, M. A. (1989). Measuring patients desire for autonomy: Decision making and information-seeking preferences among medical patients. Journal of General Internal Medicine, 4, 2330.



Evans, B. J., Kiellerup, F. D., Stanley, R. O., Burrows, G. D., & Sweet, B. (1987). A communications skills programme for increasing patients satisfaction with general practice consultations. British Journal of Medical Psychology, 60, 373378. Faden, R. R., Becker, C., Lewis, C., Freeman, J., & Faden, A. I. (1981). Disclosure of information to patients in medical care. Medical Care, 19, 718733. Falloweld, L., Hall, A., Maguire, G. P., & Baum, M. (1990). Psychological outcomes of different treatment policies in women with early breast cancer outside a clinical trial. British Medical Journal, 301, 575580. Feikin, D. R., Schuchat, A., Kolczak, M., Barrett, N. L., Harrison, L. H., Lefkowitz, L., et al. (2000). Mortality from invasive pneumococcal pneumonia in the era of antibiotic resistance, 19951997. American Journal of Public Health, 90, 223229. Fidler, D. P. (1998). Legal issues associated with antimicrobial drug resistance. Emerging Infectious Diseases, 4, 169177. Finkelstein, J. A., Davis, D. A., Dowell, S. F., Metlay, J. P., Soumerai, S. B., Rifas-Shiman, S. L., et al. (2001). Reducing antibiotic use in children: A randomized trial in 12 practices. Pediatrics, 108, 17. Finkelstein, J. A., Metlay, J., Davis, R. I., Rifas, S., Dowell, S. F., & Platt, R. (2000). Antimicrobial use in dened populations of infants and young children. Archives of Pediatrics and Adolescent Medicine, 154, 395400. Finkelstein, J. A., Stille, C., Nordin, J., Davis, R. I., Raebel, M. A., Roblin, D., et al. (2003). Reduction in antibiotic use among US children, 19962000. Pediatrics, 112, 620 627. Fischer, T., Fischer, S., Kochen, M. M., & Hummers-Pradier, E. (2005). Inuence of patient symptoms and physical ndings on general practitioners treatment of respiratory tract infections: A direct observation study. Retrieved from http://www.biomedcentral. com/1471-2296/6/6. Fisher, S. (1991). A discourse of the social: Medical talk/power talk/oppositional talk? Discourse and Society, 2, 157182. Fisher, S., & Todd, A. D. (Eds.). (1983). The Social Organization of Doctor-Patient Communication. Washington, DC: Center for Applied Linguistics. Fleming, A. (1929). On the antibacterial action of cultures of a penicillium, with special reference to their use in the isolation of B. inuenzae. British Journal of Experimental Pathology, 10, 226236. Fleming, A. (1945). Penicillin: Nobel Lecture. Retrieved from laureates/1945/eming-lecture.pdf. Foster, G. M. (1974). Medical anthropology: Some contrasts with medical sociology. Medical Anthropology Newsletter, 6, 16. Frankel, R. (1984). From sentence to sequence: Understanding the medical encounter through microinteractional analysis. Discourse Processes, 7, 135170. Freidland, I. R. (1995). Comparison of the response to antimicrobial therapy of penicillinresistant and penicillin susceptible pneumococcal disease. Pediatric Infectious Diseases Journal, 14, 885890. Freidson, E. (1970a). Profession of Medicine: A Study of the Sociology of Applied Knowledge. Chicago: University of Chicago Press. Friedson, E. (1970b). Professional Dominance. Chicago: Aldine. Froom, J., Culpepper, L. J. M., Green, L. A., de Melker, R. A., Grob, P., Heeren, T., et al. (2001). A cross-national study of acute otitis media: Risk factors, severity, and treatment at initial visit. Report from the International Primary Care Network (IPCN) and the Ambulatory Sentinel Practice Network (ASPN). Journal of the American Board of Family Parctice, 14, 406417.



Frosch, D. L., & Kaplan, R. M. (1999). Shared decision making in clinical medicine: Past research and future directions. American Journal of Preventive Medicine, 27, 1139 1145. Gani, L., Arif, H., Widjaja, S. K., Adi, R., Prasadja, H., Tampubolon, L. H., et al. (1991). Physicians prescribing practice for treatment of acute diarrhoea in young children in Jakarta. Journal of Diarrhoeal Disease Research, 9, 194199. Gardner, R. (1997). The conversation object Mm: A weak and variable acknowledging token. Research on Language and Social Interaction, 30, 131156. Gill, V. T. (1998). Doing attributions in medical interaction: Patients explanations for illness and doctors responses. Social Psychology Quarterly, 61, 342360. Gomez, J., Banos, V., Ruiz Gomez, J., Herrero, F., Nunez, M. L., Canteras, M., et al. (1995). Clinical signicance of pneumococcal bacteraemias in a general hospital: A prospective study 19891993. Journal of Antimicrobial Chemotherapy, 36, 10211030. Gonzales, R., Malone, D. C., Maselli, J. H., & Sande, M. A. (2001). Excessive antibiotic use for acute respiratory infections in the United States. Clinical Infectious Diseases, 33, 757762. Gonzalez, R., Steiner, J. F., & Sande, M. A. (1997). Antibiotic prescribing for adults with colds, upper respiratory tract infections and bronchitis by ambulatory care physicians. Journal of the American Medical Association, 278, 901904. Greenberger, N. J., & Hinthorn, D. R. (1993). History Taking and Physical Examination: Essentials and Clinical Correlates. St. Louis: Mosby. Greeneld, S., Kaplan, S. H., Ware, J. E., Yano, E. M., & Frank, J. L. H. (1988). Patients participation in medical care: Effects on blood sugar control and quality of life in diabetes. Journal of General Internal Medicine, 3, 448457. Gums, J. G. (2002). Assessing the impact of antimicrobial resistance. American Journal of Health System Pharmacy, 59, S4S6. Guyon, A. B., Barman, A., Ahmen, J. U., Ahmen, A. U., & Alam, M. S. (1994). A baseline survey on use of drugs at the primary health care level in Bangladesh. Bulletin of the World Health Organization, 72, 265271. Haak, H. (1988). Pharmaceuticals in two Brazilian villages: Lay practices and perceptions. Social Science and Medicine, 27, 14151427. Haakana, M. (2001). Laughter as a patients resource: Dealing with delicate aspects of medical interaction. Text, 21, 187219. Halkowski, T. (2006). Realizing the illness: Patients narratives of symptom discovery. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 86114). Cambridge: Cambridge University Press. Hamm, R., Hicks, R., & Bemben, D. (1996). Antibiotics and respiratory infections: Are patients more satised when expectations are met? Journal of Family Practice, 43, 5662. Harbarth, S., Albrich, W., & Brun-Buisson, C. (2002). Outpatient antibiotic use and prevalence of antibiotic resistant pneumoncocci in France and Germany: A sociocultural perspective, CDC 8(12), Retrieved from: Harbarth, S., & Samore, M. H. (2005). Antimicrobial resistance determinants and future control. Emerging Infectious Diseases, 11, 794801. Hardin, G. (1968). The tragedy of the commons. Science, 162, 12431248. Haug, M., & Lavin, B. (1983). Consumerism in Medicine: Challenging Physician Authority. Beverly Hills, CA: Sage. Heath, C. (1981). The opening sequence in doctor-patient interaction. In P. Atkinson & C. Heath (Eds.), Medical Work: Realities and Routines (pp. 7190). Aldershot, England: Gower.



Heath, C. (1992). The delivery and reception of diagnosis and assessment in the general practice consultation. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 235267). Cambridge: Cambridge University Press. Heinemann, T. (2005). Where grammar and interaction meet: The preference for matched polarity in responsive turns in Danish. In A. Hakulinen & M. Selting (Eds.), Syntax and Lexis in Conversation: Studies on the Use of Linguistic Resources in Talk-in-Interaction (pp. 375402). Amsterdam: John Benjamins. Helman, C. G. (1978). Feed a cold, starve a fever: Folk models of infection in an English suburban community, and their relation to medical treatment. Culture, Medicine and Psychiatry, 2, 107137. Heritage, J. (1984a). A change-of-state token and aspects of its sequential placement. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action (pp. 299345). Cambridge: Cambridge University Press. Heritage, J. (1984b). Garnkel and Ethnomethodology. Cambridge: Polity Press. Heritage, J. (1998). Oh-prefaced responses to inquiry. Language in Society, 27, 291334. Heritage, J. (2002). The limits of questioning: Negative interrogatives and hostile question content. Journal of Pragmatics, 34, 14271446. Heritage, J. (2006). Revisiting authority in physician-patient interaction. In M. Maxwell, D. Kovarsky, & J. Duchan (Eds.), Diagnosis as Cultural Practice (pp. 83102). New York: Mouton de Gruyter. Heritage, J., Boyd, E., & Kleinman, L. (2001). Subverting criteria: The role of precedent in decisions to nance surgery. Sociology of Health and Illness, 23, 701728. Heritage, J., & Greatbatch, D. (1986). Generating applause: A Study of rhetoric and response at party political conferences. American Journal of Sociology, 92(1), 110157. Heritage, J., & Lindstrm, A. (1998). Motherhood, medicine and morality: Scenes from a medical encounter. Research on Language and Social Interaction, 31, 397-438. Heritage, J., & Maynard, D. (2006). Communication in Medical Care: Interactions between Primary Care Physicians and Patients. Cambridge: Cambridge University Press. Heritage, J., & Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in assessment sequences. Social Psychology Quarterly, 68, 1538. Heritage, J. & Robinson, J. (2006a). Accounting for the visit: Giving reasons for seeking medical care. In J. Heritage & D. Maynard (Eds.), Communication in Medical Care: Interactions between Primary Care Physicians and Patients (pp. 4885). Cambridge: Cambridge University Press. Heritage, J. & Robinson, J. D. (2006b). The structure of patients presenting concerns: Physicians opening questions. Health Communication 19(2), 89102. Heritage, J., & Se, S. (1992). Dilemmas of advice: Aspects of the delivery and reception of advice in interactions between health visitors and rst time mothers. In P. Drew & J. Heritage (Eds.), Talk at Work (pp. 359417). Cambridge: Cambridge University Press. Heritage, J., & Sorjonen, M.-L. (1994). Constituting and maintaining activities across sequences: And-prefacing as a feature of question design. Language in Society, 23, 129. Heritage, J., & Stivers, T. (1999). Online commentary in acute medical visits: A method of shaping patient expectations. Social Science and Medicine, 49, 15011517. Hibbard, J. H., & Weeks, E. C. (1985). Consumer use of physician fee information. Journal of Health and Human Resources Administration, 7, 321335. Himmel, W., Lippert-Urbanke, E., & Kochen, M. M. (1997). Are patients more satised when they receive a prescription? The effect of patient expectations in general practice. Scandinavian Journal of Primary Health Care, 15, 118122. Holmberg, S. D., Solomon, S. L., & Blake, P. A. (1987). Health and economic impacts of antimicrobial resistance. Reviews of Infectious Diseases, 9, 10651078.



Hogerzeil, H. V., Bimo, Ross-Degnan, D., Laning, R. O., Ofori-Adjei, D., Santoso, B., et al. (1993). Field tests for rational drug use in twelve developing countries. Lancet, 342, 14081410. Horn, L. (1989). A Natural History of Negation. Chicago: University of Chicago Press. Hossain, M. M., Glass, R. I., & Khan, M. R. (1982). Antibiotic use in a rural community in Bangladesh. International Journal of Epidemiology, 11, 402405. Hui, L., Li, X. S., Zeng, X. J., Dai, Y. H., & Foy, H. M. (1997). Patterns and determinants of use of antibiotics for acute respiratory tract infections in children in China. Pediatric Infectious Disease Journal, 16, 560564. Humphrey, N. K. (1988). The social function of intellect. In R. Byrne & A. Whiten (Eds.), Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes, and Humans (pp. 1326). Oxford: Clarendon Press. Igun, U. A. (1994). Reported and actual prescription of oral rehydration therapy for childhood diarrhoeas by retail pharmacists in Nigeria. Social Science and Medicine, 39, 797806. Indalo, A. A. (1997). Antibiotic sale behaviour in Nairobi: A contributing factor to antimicrobial drug resistance. East African Medical Journal, 74, 171173. Jacobs, M. R., Felmingham, D., Appelbaum, P. C., Grneberg, R. N., & Group, T. A. P. (2003). The Alexander Project 19982000: Susceptibility of pathogens isolated from community-acquired respiratory tract infection to commonly used antimicrobial agents. Journal of Antimicrobial Chemotherapy, 52, 229246. Jefferson, G. (1979). A technique for inviting laughter and its subsequent acceptance/declination. In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology (pp. 7996). New York: Irvington Publishers. Jefferson, G. (1981). The Abominable Ne?: A Working Paper Exploring the Phenomenon of Post-Response Pursuit of Response. Occasional Paper No. 6, Department of Sociology, University of Manchester, Manchester, England. Jefferson, G. (1983). Notes on a systematic deployment of the acknowledgement tokens yeah and mm hm. In G. Jefferson (Ed.), Two Papers on Transitory Recipientship (pp. 118). Tilburg University: Tilburg Papers in Language and Literature 30. Jefferson, G. (1988). On the sequential organization of troubles-talk in ordinary conversation. Social Problems, 35(4), 418441. Jefferson, G. (1990). List construction as a task and interactional resource. In G. Psathas (Ed.), Interaction Competence (pp. 6392). Washington, DC: International Institute for Ethnomethodology and Conversation Analysis; University Press of America. Jefferson, G. (2004). A note on laughter in male-female interaction. Discourse Studies, 6, 117133. Kaiser, L., Lew, D., Hirschel, B., Auckenthaler, R., Morabia, A., Heald, A., et al. (1996). Effects of antibiotic treatment in the subset of common-cold patients who have bacteria in nasopharyngeal secretions. Lancet, 347, 15071510. Kaplan, S. H., Greeneld, S., & Ware, J. (1989). Assessing the effects of physician-patient interactions on the outcomes of chronic disease. Medical Care, 27, S110126. Karp, D., Jin, N., Yamagishi, T., & Shinotsuka, H. (1993). Raising the minimum in the minimal group paradigm. Japanese Journal of Experimental Social Psychology, 32, 231240. Kassirer, J. P. (1994). Incorporating patients preferences into medical decisions. New England Journal of Medicine, 330, 18951896. Kleinman, L., Boyd, E., & Heritage, J. (1997). Adherence to prescribed explicit criteria during utilization review: An analysis of communications between attending and reviewing physicians. Journal of the American Medical Association, 278, 497501. Kollock, P. (1998). Social dilemmas: The anatomy of cooperation. Annual Review of Sociology, 24, 183214.



Korsch, B., Gozzi, E. K., & Francis, V. (1968). Gaps in doctor-patient communication. Pediatrics, 42, 855871. Koshik, I. (2002). A conversation analytic study of yes/no questions which convey reversed polarity assertions. Journal of Pragmatics, 34, 18511877. Kozyrskyj, A. L., Dahl, M. E., Chateau, D. G., Mazowita, G. B., Klassen, T. P., & Law, B. J. (2004). Evidence-based prescribing of antibiotics for children: Role of socioeconomic status and physician characteristics. Canadian Medical Association Journal, 171, 139 145. Kramer, R. M., & Brewer, M. B. (1986a). Effects of group identity on resource use in a simulated commons dilemma. Journal of Personality and Social Psychology, 38, 504516. Kramer, R. M., & Brewer, M. B. (1986b). Social group identity and the emergence of cooperation in resource conservation dilemmas. In H. A. M. Wilke, D. M. Messick, & C. Rutte (Eds.), Experimental Social Dilemmas (pp. 205234). Frankfurt: Verlag Peter Lang. Kravitz, R. L., Bell, R. A., & Franz, C. E. (1999). A taxonomy of requests by patients (TORP): A new system for understanding clinical negotiation in ofce practice. Journal of Family Practice, 48, 872878. Kravitz, R. L., Epstein, R. M., Feldman, M. D., Franz, C. E., Azari, R., Wilkes, M. S., et al. (2005). Inuence of patients requests for direct-to-consumer advertised antidepressants: A randomized controlled trial. Journal of the American Medical Association, 293, 19952002. Kunin, C. M., Lipton, H. L., Tupasi, T., Sacks, T., Scheckler, W. E., Jivani, A., et al. (1987). Social, behavioral, and practical factors affecting antibiotic use worldwide: Report of task force 4. Review of Infectious Diseases, 9, S270S284. Lerner, G. (1996). On the semi-permeable character of grammatical units in conversation: Conditional entry into the turn-space of another speaker. In E. Ochs, E. A. Schegloff, & S. Thompson (Eds.), Interaction and Grammar (pp. 238276). Cambridge: Cambridge University Press. Levine, M. N., Gafni, A., Markham, B., & MacFarlane, D. (1992). A bedside decision instrument to elicit a patients preference concerning adjuvant chemotherapy for breast cancer. Annals of Internal Medicine, 117, 5358. Levinson, S. C. (1983). Pragmatics. Cambridge: Cambridge University Press. Levinson, S. C. (1987). Minimization and conversational inference. In M. Papi & J. Verschuren (Eds.), The Pragmatics Perspective (pp. 61129). Amsterdam: J. Benjamins. Levinson, S. C. (2000). Presumptive Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press. Levy, S. B. (1992). The Antibiotics Paradox: How Miracle Drugs Are Destroying the Miracle. New York: Plenum. Lindstrm, A. (1995). Or -constructed inquiries as a resource for probing the relevance of prior talk in Swedish conversation. Paper presented at The Georgetown Linguistics Society meeting, Washington, DC. Little, P., Dorward, M., Warner, G., Stephens, K., Senior, J., & Moore, M. (2004). Importance of patient pressure and perceived pressure and perceived medical need for investigations, referral, and prescribing in primary care: Nested observational study. Retrieved from http:// 10&RESULTFORMAT=&author1=Little&andorexactfulltext=and&searchid=1& FIRSTINDEX=0&sortspec=relevance&resourcetype=HWCIT Madle, G., Kostkova, P., Mani-Saada, J., Weinberg, J., & Williams, P. (2004). Changing public attitudes to antibiotic prescribing: Can the Internet help? Informatics in Primary Care, 12, 1926.



Mainous, A. G., Cheng, A. Y., Garr, R. C., Tilley, B. C., Everett, C. J., & McKee, M. D. (2005). Nonprescribed antimicrobial drugs in Latino community, South Carolina. Emerging Infectious Diseases, 11, 883888. Mainous, A. G., Hueston, W. J., & Clark, J. R. (1996). Antibiotics and upper respiratory infection: Do some folks think there is a cure for the common cold? Journal of Family Practice, 42, 357361. Mamun, K. Z. (1991). Prevalence and genetics of resistance to commonly used antimicrobial agents in faecal enterbacteriaceae from children in Bangladesh. Unpublished Ph.D. Thesis, University of Liverpool, Liverpool, England. Mangione-Smith, R., Elliott, M. N., McDonald, L., & McGlynn, E. A. (2002). An observational study of antibiotic prescribing behavior and the Hawthorne effect. Health Services Research, 37, 16031623. Mangione-Smith, R., Elliott, M. N., Stivers, T., McDonald, L., Heritage, J., & McGlynn, E. A. (2004). Racial/ethnic variation in parent expectations for antibiotics: Implications for public health campaigns. Pediatrics, 113, e385e394. Mangione-Smith, R., Elliott, M. N., Stivers, T., McDonald, L. L., & Heritage, J. (2006). Ruling out the need for antibiotics: Are we sending the right message? Archives of Pediatric and Adolescent Medicine 160, 956952. Mangione-Smith, R., McGlynn, E. A., Elliott, M. N., Krogstad, P., & Brook, R. H. (1999). The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics, 103, 711718. Mangione-Smith, R., McGlynn, E. A., Elliott, M. N., McDonald, L., Franz, C. E., & Kravitz, R. L. (2001). Parent expectations for antibiotics, physician-parent communication, and satisfaction. Archives of Pediatric and Adolescent Medicine, 155, 800806. Mangione-Smith, R., Stivers, T., Elliott, M. N., McDonald, L., & Heritage, J. (2003). The relationship between online commentary use and prevention of inappropriate antibiotic prescribing by pediatricians. Social Science and Medicine, 56, 313320. Maynard, D. W. (2003). Bad News, Good News: Conversation Order in Everyday Talk and Clinical Settings. Chicago: University of Chicago. McCaig, L., Besser, R. E., & Hughes, J. M. (2002). Trends in antimicrobial prescribing rates for children and adolescents. Journal of the American Medical Association, 287, 3096 3102. McCaig, L. F., Besser, R. E., & Hughes, J. M. (2003). Antimicrobial drug prescriptions in ambulatory care settings, United States, 19922000. Emerging Infectious Diseases, 9, 432437. McCaig, L. F., & Hughes, J. M. (1995). Trends in antimicrobial drug prescribing among ofce-based physicians in the United States. Journal of the American Medical Association, 273, 214219. McCormick, A. W., Whitney, C. G., Farley, M. M., Lyneld, R., Harrison, L. H., Bennett, N. M., et al. (2003). Geographic diversity and temporal trends of antimicrobial resistance in Streptococcus pneumoniae in the United States. Nature Medicine, 9, 424430. Melker, R. A., & Kuyvenhoven, M. M. (1994). Management of upper respiratory tract infections in Dutch family practice. Journal of Family Practice, 38, 353357. Memish, Z. A., Venkatesh, S., & Shibl, A. M. (2003). Impact of travel on international spread of antimicrobial resistance. International Journal of Antimicrobial Agents, 21, 135142. Mendonca, P. J., & Brehm, S. S. (1983). Effects of choice on behavioral treatment of overweight children. Journal of Social Clinical Psychology, 1, 343358. Metlay, J. P., Shea, J. A., Crossette, L. B., & Asch, D. A. (2002). Tensions in antibiotic prescribing. Journal of General Internal Medicine, 17, 8794.



Mishler, E. (1984). The Discourse of Medicine: Dialectics of Medical Interviews. Norwood, NJ: Ablex. Nava, J. M., Bella, F., Garau, J., Lite, J., Morera, M. A., Marti, C., et al. (1994). Predictive factors for invasive disease due to penicillin-resistant Streptococcus pneumoniae: A population-based study. Clinical Infectious Diseases, 19, 884890. Neu, H. C. (1992). The crisis in antibiotic resistance. Science, 257, 10641073. Okeke, I. N., Laxminarayan, R., Bhutta, Z. A., Duse, A. G., Jenkins, P., OBrien, T. F., et al. (2005). Antimicrobial resistance in developing countries: Part I: Recent trends and current status. Lancet, 5, 481493. Ong, L., de-Haes, J. C., Hoos, A., & Lammes, F. (1995). Doctor-patient communication: A review of the literature. Social Science and Medicine, 40, 903918. World Health Organization (1999). Containing Antimicrobial Resistance: Review of the Literature and Report of a WHO Workshop on the Development of a Global Strategy for the Containment of Antimicrobial Resistance. Geneva: World Health Organization. Orr, P. H., Scherer, K., MacDonald, A., & Moffatt, M. (1993). Randomized placebo controlled trials of antibiotics for acute bronchitis: A critical review of the literature. Journal of Family Practice, 36, 507512. Otters, H. B. M., van der Wouden, J. C., Schellevis, F. G., van Suijlekom-Smit, L. W. A., & Koes, B. W. (2004). Trends in prescribing antibiotics for children in Dutch general practice. Journal of Antimicrobial Chemotherapy, 53, 361366. Pachter, L. M. (1994). Culture and clinical care: Folk illness beliefs and behaviors and their implications for health care delivery. Journal of the American Medical Association, 271, 690694. Palmer, D. A., & Bauchner, H. (1997). Parents and physicians views on antibiotics. Pediatrics, 99, 862863. Pantell, R. H., Steward, T. J., Dias, J. K., Wells, P., & Ross, A. W. (1982). Physician communication with children and parents. Pediatrics, 70, 396402. Papper, S. (1970). The undesirable patient. Journal of Chronic Diseases, 22, 777779. Parsons, T. (1951). The Social System. New York: Free Press. Pendleton, D. (1983). Doctor-patient communication: A review. In D. Pendleton & J. Hasler (Eds.), Doctor-Patient Communication (pp. 553). New York: Academic. Pennie, R. A. (1998). Prospective study of antibiotic prescribing for children. Canadian Family Physician, 44, 18501856. Perkyl, A. (1998). Authority and accountability: The delivery of diagnosis in primary health care. Social Psychology Quarterly, 61, 301320. Perz, J. F., Craig, A. S., Coffey, C. S., Jorgensen, D. M., Mitchel, E., Hall, S., et al. (2002). Changes in antibiotic prescribing for children after a community-wide campaign. Journal of the American Medical Association, 287, 31033109. Peyrot, M., Alperstein, N. M., van Doren, D., & Poli, L. G. (1998). Direct-to-consumer ads can inuence behavior: Advertising increases consumer knowledge and prescription drug requests. Marketing Health Services, Summer, 2732. Pinto, M. B., Pinto, J. K., & Barber, J. C. (1998). The impact of pharmaceutical direct advertising: Opportunities and obstructions. Health Marketing Quarterly, 15, 89101. Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. M. Atkinson & J. Heritage (Eds.), Structures of Social Action: Studies in Conversation Analysis (pp. 57101). Cambridge: Cambridge University Press. Pomerantz, A. (1988). Offering a candidate answer: An information seeking strategy. Communication Monographs, 55, 360373.



Quagliarello, A. B., Parry, C. M., Hien, T. T., & Farrar, J. J. (2003). Factors associated with carriage of penicillin-resistant Streptococcus pneumoniae among Vietnamese children: A rural-urban divide. Journal of Health Population Nutrition, 21, 316324. Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1972). A Grammar of Contemporary English. London: Longman. Radyowijati, A., & Haak, H. (2003). Improving antibiotic use in low-income countries: An overview of evidence on determinants. Social Science and Medicine, 57, 733744. Raymond, G. (2003). Grammar and social organization: Yes/no interrogatives and the structure of responding. American Sociological Review, 68, 939967. Raymond, G. (2004). Prompting action: The stand-alone so in sequences of talk-in-interaction. Research on Language and Social Interaction 37, 185218. Reeder, L. G. (1972). The patient-client as a consumer: Some observations on the changing professional-client relationship. Journal of Health and Social Behavior, 13, 406412. Reichler, M. R., Allphin, A. A., Breiman, R. F., Schreiber, J. R., Arnold, J. E., McDougal, L. K., et al. (1992). The spread of multiply resistant Streptococcus pneumoniae at a day care center in Ohio. Journal of Infectious Diseases, 166, 13461353. Robinson, J. D. (1998). Getting down to business: Talk, gaze, and body orientation during openings of doctor-patient consultations. Human Communication Research, 25, 97123. Robinson, J. D. (2003). An interactional structure of medical activities during acute visits and its implications for patients participation. Health Communication, 15, 2757. Roter, D., & Hall, J. (1992). Doctors Talking with Patients/Patients Talking with Doctors: Improving Communication in Medical Visits. Westport CT: Auburn House. Roter, D. L. (2002). The Roter Method of Interaction Process Analysis. Retrieved from www. Ruusuvuori, J. (2000). Control in the medical consultation: Practices of giving and receiving the reason for the visit in primary health care. Unpublished doctoral dissertation: University of Tampere. Sacks, H. (1992). Lectures on Conversation, Vol. 2 (Fall 1968Spring 1972). Oxford: Blackwell. Sacks, H., & Schegloff, E. A. (1979). Two preferences in the organization of reference to persons and their interaction. In G. Psathas (Ed.), Everyday Language: Studies in Ethnomethodology (pp. 1521). New York: Irvington. Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50, 696735. Sanchez-Menegay, C., & Stalder, H. (1994). Do physicians take into account patients expectations? Journal of General Internal Medicine, 9, 404406. Sargent, J., & Welch, H. G. (2001). Reducing unnecessary antibiotic use in primary care: Hard rules, soft calls. Effective Clinical Practice, 4, 136138. Schegloff, E. A. (1968). Sequencing in conversational openings. American Anthropologist, 70, 10751095. Schegloff, E. A. (1972a). Notes on a conversational practice: Formulating place. In D. Sudnow (Ed.), Studies in Social Interaction (pp. 75119). New York: Free Press. Schegloff, E. A. (1972b). Sequencing in conversational openings. In J. J. Gumperz & D. Hymes (Eds.), Directions in Sociolinguistics: The Ethnography of Communication (pp. 346380). New York: Holt, Rinehart and Winston. Schegloff, E. A. (1977). Identication and recognition in interactional openings. In I. de Sola Pool (Ed.), The Social Impact of the Telephone (pp. 415450). Cambridge, MA: MIT Press.



Schegloff, E. A. (1982). Discourse as an interactional achievement: Some uses of uh huh and other things that come between sentences. In D. Tannen (Ed.), Analyzing Discourse (Georgetown University Roundtable on Languages and Linguistics 1981) (pp. 7193). Washington, DC: Georgetown University Press. Schegloff, E. A. (1996). Turn organization: One intersection of grammar and interaction. In E. Ochs, S. Thompson, & E. Schegloff (Eds.), Interaction and Grammar (pp. 52133). Cambridge: Cambridge University Press. Schegloff, E. A. (2000). On granularity. Annual Review of Sociology, 26, 715720. Schegloff, E. A. (2001). Increments: Where they are and what they do. Paper presented at the Linguistic Institute, Santa Barbara, CA. Schegloff, E. A. (2006). A primer for Conversation Analysis: Sequence Organization. Cambridge: Cambridge University Press. Schegloff, E. A., Jefferson, G., & Sacks, H. (1977). The preference for self-correction in the organization of repair in conversation. Language, 53, 361382. Schulman, B. A. (1979). Active patient orientation and outcomes in hypertensive treatment. Medical Care, 17, 267280. Schwartz, B. (1999). Preventing the spread of antimicrobial resistance among bacterial respiratory pathogens in industrialized countries: The case for judicious antimicrobial use. Clinical Infectious Diseases, 28, 211213. Schwartz, R. H., Freij, B. J., Ziai, M., & Sheridan, M. J. (1997). Antimicrobial prescribing for acute purulent rhinitis in children: A survey of pediatricians and family practitioners. Pediatric Infectious Disease Journal, 16, 185190. Sheridan, M. (1994). Parents reporting of symptoms in their children: Physicians perceptions. Hawaii Medical Journal, 53, 216217. Sleath, B., Svarstad, B., & Roter, D. (1997). Physician vs patient initiation of psychotropic prescribing in primary care settings: A content analysis of audiotapes. Social Science and Medicine, 44, 541548. Smolinski, M. S., Hamburg, M. A., & Lederberg, J. (Eds.). (2003). Microbial Threats to Health: Emergence, Detection, and Response. Washington, DC: Institute of Medicine. Starr, P. (1982). The Social Transformation of American Medicine. New York: Basic Books. Steffen, R., & Lobel, H. O. (1996). Travel medicine. In G. C. Cook (Ed.), Mansons Tropical Diseases (Vol. 20, pp. 407420). London: W. B. Saunders. Stevenson, F. A., Greeneld, S. M., Jones, M., Nayak, A., & Bradley, C. P. (1999). GPs perceptions of patient inuence on prescribing. Family Practice, 16, 255261. Stivers, T. (1998). Pre-diagnostic commentary in veterinarian-client interaction. Research on Language and Social Interaction, 31, 241277. Stivers, T. (2000). Negotiating antibiotic treatment in pediatric care: The communication of preferences in physician-parent interaction. Doctoral Dissertation. University of California, Los Angeles. Stivers, T. (2001). Negotiating who presents the problem: Next speaker selection in pediatric encounters. Journal of Communication, 51, 131. Stivers, T. (2002a). Participating in decisions about treatment: Overt parent pressure for antibiotic medication in pediatric encounters. Social Science and Medicine, 54, 11111130. Stivers, T. (2002b). Symptoms only versus candidate diagnoses: Presenting the problem in pediatric encounters. Health Communication, 14, 299338. Stivers, T. (2005a). Modied repeats: One method for asserting primary rights from second position. Research on Language and Social Interaction, 38, 131158. Stivers, T. (2005b). Non-antibiotic treatment recommendations: Delivery formats and implications for parent resistance. Social Science and Medicine, 60, 949964.



Stivers, T. (2005c). Parent resistance to physicians treatment recommendations: One resource for initiating a negotiation of the treatment decision. Health Communication, 18, 4174. Stivers, T. (Submitted). Stance, alignment and afliation during story telling: Nodding as a token of preliminary afliation. Stivers, T., & Heritage, J. (2001). Breaking the sequential mould: Answering more than the question during comprehensive history taking. Text, 21, 151185. Stivers, T., Mangione-Smith, R., Elliott, M. N., McDonald, L., & Heritage, J. (2003). Why do physicians think parents expect antibiotics? What parents report vs what physicians perceive. Journal of Family Practice, 52, 140148. Strong, P. (1979). The Ceremonial Order of the Clinic. London: Routledge. Swenson, S. L., Buell, S., Zettler, P., White, M., Ruston, D. C., & Lo, B. (2004). Patient centered communication. Journal of General Internal Medicine, 19, 10691079. Szasz, P. S., & Hollender, M. H. (1956). A contribution to the philosophy of medicine: The basic model of the doctor-patient relationship. Archives of Internal Medicine, 97, 585 592. Terasaki, A. (2004). Pre-announcement sequences in conversation. In G. Lerner (Ed.), Conversation Analysis: Studies from the First Generation (pp. 171223). Washington, DC: University Press of America. Thomas, K. (1999). Prospective multicentre hospital surveillance of Streptococcus pneumoniae disease in India. Lancet, 353, 12161221. Thompson, S. C., Pitts, J. S., & Schwankovsky, L. (1993). Preference for involvement in medical decision-making situational and demographic inuences. Patient Education and Counseling, 22, 133140. Todd, A. D., & Fisher, S. (Eds.). (1993). The social organization of doctor-patient communication. Norwood, NJ: Ablex. Todd, J. K., Todd, N., Damato, J., & Todd, W. A. (1984). Bacteriology and treatment of purulent nasopharyngitis: A double blind, placebo-controlled evaluation. Pediatric Infectious Disease, 3, 226232. Tuckett, D., Boulton, M., Olson, C., & Williams, A. (1985). Meetings between Experts: An Approach to Sharing Ideas in Medical Consultations. London: Tavistock. U.S. Department of Health and Human Services, Public Health Service (2000). Healthy People 2010: Understanding and Improving Health (2nd ed.). Washington, DC: U.S. Government Printing Ofce. Uppal, R., Sarkar, U., Giriyappanavar, C. S., & Kacker, V. (1993). Antimicrobial drug use in primary health care. Journal of Clinical Epidemiology, 46, 671673. Vinson, D. C., & Lutz, L. J. (1993). The effect of parental expectations on treatment of children with a cough: A report from the ASPN. Journal of Family Practice, 37, 2327. Vuckovic, N., & Nichter, M. (1997). Changing patterns of pharmaceutical practice in the United States. Social Science and Medicine, 44, 12851302. Waitzkin, H. (1991). The Politics of Medical Encounters. New Haven, CT: Yale University Press. Wald, E. R., Guerra, N., & Byers, C. (1991). Upper respiratory tract infections in young children: Duration of and frequency of complications. Pediatrics, 87, 129133. Warren, J. (1998, December 28). Too much of a good thing. San Diego Union Tribune, pp. E1, E3. Watanabe, H., Sato, S., Kawakami, K., Watanabe, K., Oishi, K., Rikitomi, N., et al. (2000). A comparative clinical study of pneumonia by penicillin-resistant and -sensitive Streptococcus pneumoniae in a community hospital. Respirology, 5, 5964.



Watson, R. L., Dowell, S. F., Jayaraman, M., Keyserling, H., Koczak, M., & Schwartz, B. (1999). Antimicrobial use for pediatric upper respiratory infections: Reported practice, actual practice and parent beliefs. Pediatrics, 104, 12511157. Weber, J. T., & Courvalin, P. (2005). An emptying quiver: Antimicrobial drugs and resistance. Emerging Infectious Diseases, 11, 791793. West, C. (1984). Routine Complications: Troubles with Talk between Doctors and Patients. Bloomington: Indiana University Press. Whalen, J., Zimmerman, D. H., & Whalen, M. R. (1988). When words fail: A single case analysis. Social Problems, 35(4), 335362. Whalen, M., & Zimmerman, D. H. (1987). Sequential and institutional contexts in calls for help. Social Psychology Quarterly, 50, 172185. Wheeler, J. G., Fair, M., Simpson, P. M., Rowlands, L. A., Aitken, M. E., & Jacobs, R. F. (2001). Impact of a waiting room videotape message on parent attitudes toward pediatric antibiotic use. Pediatrics, 108, 591596. Whitney, C. G., Farley, M. M., Hadler, J., Harrison, L. H., Lexau, C., Reingold, A., et al. (2000). Increasing prevalence of multidrug-resistant Streptococcus pneumoniae in the United States. New England Journal of Medicine, 343, 19171924. World Health Organization. (2000). Overcoming antimicrobial resistance. World Health Organization Report on Infectious Diseases 2000. Retrieved August 18, 2006, from http:// Wise, R., Hart, T., Cars, O., Streulens, Helmuth, R., Huovinen, P., et al. (1998). Antimicrobial resistance is a major threat to public health. British Medical Journal, 317, 609610.

This page intentionally left blank


Additional symptoms, mention of, 6367, 7176, 186 Alternative diagnosis, mention of, 6770, 186 Afliation, 41, 53, 81, 141142, 147 Afrmative announcements, 164167, 169, 188. See also Diagnosis delivery; Treatment recommendation Antibiotics history of, 45 inquires about, 143148 mentions of past experience with, 148151 preference for, 139, 144 rate of use, 3, 7, 9, 12 regulation of, 6 relevance of, 39 requests for, 10, 126, 137139 statement of desire for, 139143, 150 use in livestock, 5 Antimicrobial resistance, 4, 5 determinants of, 58 rates of, 6, 8, 9 Australia, 192 Avorn, J., 7 Bacterial resistance. See Antimicrobial resistance

Bangladesh, 192 Barber, J. C., 11 Beach, W., 200n Belgium, 8 Bell, R. A., 11, 12 Boyd, E. A., 25, 53, 55, 149 Brewer, M. B., 191 Britten, N., 21 Byrne, P. S., 23, 70 Candidate Diagnosis. See Problem presentation, candidate diagnosis Centers for Disease Control (CDC), 6, 13 Children in the pediatric visit, role of, 16 Clayman, S. E., 52 Commons dilemma, 78, 191, 199n. See also Social dilemma Concession, 75, 125, 146 Conditional relevance, 79, 87, 106114, 148, 164, 187 Conrmation, 46, 47, 48, 49, 82, 165 Consumerist movement, 11 Contingency plan, 75 Conversation analysis, 1315 Counseling phase, 4349 Course of action, 52 Cultural authority, 78, 79 Culture, 7




Danish, 200n Developing countries, 56 Diagnosis delivery, 78, 164171, 180184, 187 online, 77, 83 Diagnosis resistance, 77104, 180184, 187 Diamond, J., 189 Differential diagnosis, 52, 58 Direct to consumer advertising, 4, 1112 Discourse analysis, 1314 Doctorability, 17, 18, 53 Domain of expertise, 10, 28, 88, 106, 129, 132, 137, 153, 187188 Drew, P., 52, 84, 200n Epistemic territory, 14, 29, 48, 78, 89, 92, 106, 129, 133136. See also Domain of expertise Escalation, 153, 171, 172 Evidential formulation, 160, 163, 201n Fisher, S., 14 Fleming, A., 3, 5, 7, 192 Foster, G. M., 199n France, 8, 9 Franz, C. E., 11 Germany, 8 Gill, V., 28 Guidance model, 13 Haakana, M., 74, 141 Halkowski, T., 18 Hall, J., 11, 19 Hardin, G., 78 Haug, M., 11 Heath, C., 78 Heinemann, T., 200n Helman, C. G., 78 Heritage, J., 14, 28, 52, 53, 55, 70, 78, 79, 8182, 113, 114, 149, 163, 200n History taking, 5176, 186 Hong Kong, 9 Humphrey, N. K., 8 Illness model, 50, 190 India, 6 Israel, 9 Japan, 9

Jefferson, G., 16, 84 Kleinman, L., 149 Kollock, P., 191192 Korsch, B., 13, 16 Koshik, I., 200n Kramer, R. M., 191 Kravitz, R. L., 11, 12 Lavin, B., 11 Legitimacy, 1721, 39, 50, 54, 62, 69, 97, 124125, 155158, 160, 163, 179184, 187188, 192 Levinson, S. C., 5556, 199n Lifeworld agenda, 14, 53. See also Mishler, E. Long, B. E. L., 23, 70 Malpractice, 7 Mangione-Smith, R., 185 Markedness marked, 53, 78, 199n unmarked, 27, 38, 49, 164167 Maynard, D., 14, 155 Medical agenda, 14 Medical authority, 133, 153. See also Cultural authority Medical error, 7, 185 Mexico, 9 Mishler, E., 14 Misuse of antibiotics, 5 in developed countries, 68, 9 in developing countries, 56 Negative announcements, 167170. See also Diagnosis delivery; Treatment recommendation Negotiation, 23, 187 of diagnosis, 4050, 7276, 96 of treatment, 7276, 84, 102, 120, 124 129, 131154 Netherlands, The, 8, 193 Newsmarks, 8188, 92, 99, 180181 Oh-preface, 53, 163 Online commentary, 91, 158163, 188, 190, 201n Overt negotiation, 103, 131154, 187 Parent expectations, 9



Parent pressure motivations for, 98 physician citation of, 9, 10 physician response to, 117, 151154 Parsons, T., 11, 18 Patient satisfaction, 11 Patient/parent participation, 1013, 50, 53, 78 Perkyl, A., 78, 92 Physician perceptions disconnect with parent expectations, 10 of parents, 10, 190 See also Parent expectations Pinto, J. K., 11 Pinto, M. B., 11 Portugal, 8 Power, 14 Prescribing rates. See Antibiotics, rate of use Presuppositions, 52, 53, 70 Problem presentation, 2350 Problem presentation candidate diagnosis, 24, 2836, 3950, 67, 133, 156, 159, 186, 199n symptoms-only, 24, 2527, 3639 Progressivity, 71, 79 Prosody, 57, 92, 94, 113, 122, 165 Public health campaigns, 189190, 199n Question design negative polarity item, 54, 55 optimization, 53, 54, 55, 56 preference, 53, 54, 55, 63, 79, 89, 200n problem attentiveness, 55, 56 recipient design, 53 Questioning the diagnosis, 9297 an examination nding, 8892 Raymond, G., 53, 200n Roter, D., 11, 12, 13, 19 Roter Interaction Analysis System (RIAS), 13 Ruling out treatment. See Negative announcements Ruusuvuori, J., 24, 28 Sacks, H., 84, 200n

Schegloff, E. A., 84, 201n Se, S., 8182, 113, 114 Sequence expansion, 63, 71, 75, 8197 Shared decision making, 11 Sheridan, M., 19 Sleath, B., 12 Social action, 14 Social dilemma, 7, 185, 191, 199n Solomon, D. H., 7 Sorjonen, M., 70 Spain, 8, 9 Stance, 53, 76 Stance adjustment of, 75 towards diagnosis, 55, 60, 85, 93, 166 towards problem, 39, 48, 52, 61, 68, 78, 163, 186 towards treatment, 57, 59, 119, 147, 151 Streptococcus pneumoniae, 3, 6 Strong, P., 18 Structural organization of the visit, 23, 79, 82 Svarstad, B., 12 Switzerland, 8 Todd, A. D., 14 Treatability, 17, 2021, 2527, 29, 39, 50, 53, 54, 97104, 124125, 179, 187 Treatment change, 124129, 152153 recommendation, 106114, 164184, 174176 Treatment resistance, 105130, 114115, 163, 171180, 187 active, 114130 passive, 109, 113, 124, 145 Troubles resistance, 17, 18, 19, 59, 61, 125, 149 United Kingdom, 8 Vietnam, 192 West, C., 14 Wilkes, M. S., 12 World Health Organization (WHO), 5, 6