You are on page 1of 9


The Associations of Clinicians’ Implicit Attitudes About

Race With Medical Visit Communication and Patient
Ratings of Interpersonal Care
Lisa A. Cooper, MD, MPH, Debra L. Roter, DrPH, Kathryn A. Carson, ScM, Mary Catherine Beach, MD, MPH, Janice A. Sabin, PhD, MSW,
Anthony G. Greenwald, PhD, and Thomas S. Inui, MD

Racial/ethnic disparities in healthcare are docu-

Objectives. We examined the associations of clinicians’ implicit attitudes
mented across conditions, settings, diagnostic
about race with visit communication and patient ratings of care.
and treatment modalities, and dimensions of
Methods. In a cross-sectional study of 40 primary care clinicians and 269
technical quality,1 and ethnic minorities rate
patients in urban community-based practices, we measured clinicians’ implicit
interpersonal quality of care from physicians general race bias and race and compliance stereotyping with 2 implicit associ-
more negatively than do Whites.2---6 Minorities ation tests and related them to audiotape measures of visit communication and
experience poorer communication with physi- patient ratings.
cians,7,8 particularly in race-discordant patient--- Results. Among Black patients, general race bias was associated with more
clinician relationships.9,10 The Institute of Med- clinician verbal dominance, lower patient positive affect, and poorer ratings of
icine’s report “Unequal Treatment,” suggests interpersonal care; race and compliance stereotyping was associated with
disparities in healthcare emerge from longer visits, slower speech, less patient centeredness, and poorer ratings of
interpersonal care. Among White patients, bias was associated with more verbal
bias (or prejudice) against minorities; greater dominance and better ratings of interpersonal care; race and compliance
clinical uncertainty when interacting with mi-
stereotyping was associated with less verbal dominance, shorter visits, faster
nority patients; and beliefs (or stereotypes) held
by the provider about the behavior or health of speech, more patient centeredness, higher clinician positive affect, and lower
minorities.1(p9) ratings of some aspects of interpersonal care.
Conclusions. Clinician implicit race bias and race and compliance stereotyping
Biases may reflect explicit (conscious) biases are associated with markers of poor visit communication and poor ratings of
or implicit (unconscious) biases. There is evi- care, particularly among Black patients. (Am J Public Health. 2012;102:979–987.
dence that physicians have more negative doi:10.2105/AJPH.2011.300558)
explicit attitudes toward Blacks than towards
Whites, including stereotypes about nonad-
herence,11---13 and that negative explicit attitudes socioemotional and stylistic, but not necessarily METHODS
are associated with and mediate racial dispar- medically, focused). However, the pathway by
ities in physicians’ treatment decisions.13,14 which race-based compliance stereotyping af- Secondary data came from enrollment visits
Physicians’ implicit racial biases have been fects communication may be different from of clinicians and patients who participated in
linked to some treatment decisions in clinical that proposed for general racial bias because it 2 randomized clinical trials of interventions to
vignettes.15---17 Yet, little work has examined may relate to physicians’ professional obliga- enhance patient---provider communication and
how clinicians’ implicit racial attitudes affect tions to enhance patient adherence. Two im- outcomes for patients with hypertension (study
communication and patient experiences in plicit association studies have documented 1) and depression (study 2).21,22 In each study,
actual medical encounters.18 a physician pro-White bias regarding the con- clinicians randomized to a communication
We examined 2 implicit attitudes about race cept of the compliant patient.15,16 We reasoned skills training program (not focused on race)
among clinicians. The first relates to general that a race and compliance stereotype would received detailed individualized feedback on
racial bias; the second is specific to the medical have 2 communication-related consequences: their communication behaviors measured from
context, assessing racial bias regarding stereo- less positive emotional tone, reflecting frustra- a video-recorded encounter with a Black stan-
typing patient compliance. We explored these tion with a patient perceived as nonadherent; dardized patient. Information collected from
distinct measures of implicit bias because they and heightened attention to providing medical participants included a self-administered clini-
represent potentially different pathways to information. Given evidence that Black patients cian baseline survey (study 1: January 2002---
medical care (Figure A, available as a supple- are sensitive to pro-White bias,19,20 we hy- January 2003; study 2: June 2004---March
ment to the online version of this article at pothesized that both implicit measures are 2006) and patient interviews upon study en- We hypothesized that associated with more negative ratings of in- rollment (study 1: September 2003---August
clinicians’ implicit race bias is associated with terpersonal care in the visits of Black, but not 2005; study 2: October 2005---August 2006).
nonspecific aspects of communication (e.g., White, patients. We audiotaped patients in both studies with

May 2012, Vol 102, No. 5 | American Journal of Public Health Cooper et al. | Peer Reviewed | Research and Practice | 979

their clinician at enrollment, and patients of this article at The IAT socioemotional, communication measures
completed a postvisit questionnaire. score is derived from the difference in average contained socioemotional and task-focused
We collected primary data after the clinician response time on the 2 sorting tasks.24 We medical elements (Figure A, available as a sup-
interventions via an Internet survey that in- scored the IATs according to published guide- plement to the online version of this article at
cluded 2 cognitive tests of implicit racial atti- lines26; possible scores ranged from –2 to +2,
tudes and stereotyping derived from the Im- and scores of zero indicated no greater pref- Patient perceptions of clinicians measured by
plicit Association Test (IAT). The median time erence for White relative to Black or no greater postvisit survey. We used measures of inter-
between the patient’s baseline interview and association of White than Black with the com- personal care related to continuity of care and
the clinician completing the online survey was pliant patient concept. Higher positive scores patient adherence in other studies33---40 and
9.4 months (range = –3.2---30.2 months). indicated greater preference for White or greater shown to differ by patient race3,5 and race
association of White with compliant patient. We concordance with clinicians.9 Patients rated
Study Variables gave participants the race attitude IAT and the their primary care clinician’s attitude toward
Independent variables were 2 measures of race and compliance IAT16 in randomly assigned them as (1) “My doctor likes me,”33 and (2) “My
implicit attitudes about race from clinicians’ order. The IAT is a stable measure of implicit doctor has a great deal of respect for me.”34
survey responses. Dependent variables, collected cognition when tested over time.27 Patients then rated how they felt about their
in the earlier 2 studies, included patient---clinician Medical visit audio recordings. We analyzed primary care clinician as (1) “I would recommend
communication measures from audiotapes of audiotapes of primary care visits with the widely this physician to a friend,” (2) “I like this doctor,”
medical visits and patient perceptions of the used Roter Interaction Analysis System, a coding (3) “I trust this doctor to look out for my best
clinician from postvisit surveys. system with demonstrated reliability and pre- interests,” and (4) “I have confidence in this
Measures of implicit attitudes about race. The dictive validity.28---30 This system assigns each doctor’s knowledge and skills.”9,33,35 Finally,
IAT is a computer-based, indirect measure of thought the patient and clinician express to we asked patients, if there were a choice be-
social cognition widely used in social psychol- mutually exclusive and exhaustive codes that can tween treatments, how often would this doctor
ogy to measure implicit attitudes and stereo- be combined to reflect categories of exchange, ask you to help make the decision?36 Five-point
types about race and other sociocultural phe- including functions of the medical interview.31 Likert scales were skewed toward positive re-
nomena.23 It measures the relative association We measured visit length (in minutes), speech sponses. We dichotomized responses as the top
strength between a pair of target concepts speed (the number of statements per minute), category (e.g., strongly agree) versus all others.
such as race (White vs Black) and a pair of clinician verbal dominance (ratio of clinician to Other variables. From clinicians we collected
attribute categories such as good versus patient statements), and patient centeredness age, gender, race, type of clinician (physician
bad.24---26 The IAT’s measure is predicated (ratio of the sum of psychosocial, rapport-build- vs nurse practitioner), board certification status,
on the assumption that concepts that the test ing, and facilitative behaviors by clinicians and location of training (United States or not), num-
taker readily associates will be sorted together patients representing the patient’s agenda to the ber of years in practice, and political identity
more quickly than are concepts that are weakly sum of biomedical questions, information giving, (self-report on a 6-point scale from strongly
associated. Whereas the race attitude IAT and closed-ended questions representing the conservative to strongly liberal). We collected
examines how much generic positive terms are clinician’s agenda).7,9 Verbal dominance is an self-reported measures from clinicians designed
associated with Black versus White faces, the indicator of the level of participation of the to parallel the targeted concepts in the IATs,
race and medical compliance IAT examines clinician relative to the patient in the dialogue, including preferences or feelings toward and
a more complex association of a race and with scores greater than 1 meaning the clinician perceived cooperativeness of Whites and
treatment adherence stereotype.15,16 verbally dominated the dialogue. Patient cen- Blacks.15---17 From patients, we collected age,
On the basis of previous IAT research, the teredness and verbal dominance have demon- gender, race, educational attainment, employ-
general race attitude IAT used words such as strated concurrent and predictive validity and ment status, insurance coverage, annual house-
joy, wonderful, and laughter to represent the have been linked to patient satisfaction and hold income, living arrangements (alone or not),
concept of good and agony, evil, and hurt to reported rapport with clinicians.28,30,32 and physical and mental health status (the
represent the concept of bad (Figure B, available Coders also rated the emotional tone of the Medical Outcomes Short Form 12).41
as a supplement to the online version of this dialogue on a 6-point scale (1 = low or none to
article at We constructed 6 = high) on global dimensions of patient pos- Statistical Analyses
the race and compliance patient IAT to measure itive affect (e.g., responsiveness) and clinician We used linear and logistic regression with
the association between race and the concept of positive affect (e.g., friendliness). We assessed generalized estimating equations to assess the
a “compliant patient.”16 The category of com- intercoder reliability (the Pearson r) on a 10% strength of associations between clinician pro-
pliant patient is represented by, for example, random sample of double coded tapes; re- White bias or stereotyping and our dependent
willing, reliable, and helpful. Words to repre- liability averaged 0.90 over clinician and 0.86 variables while accounting for the nesting of
sent the reluctant patient category are, for over patient verbal categories. Coder agree- patients within clinician.42,43 We calculated
example, reluctant, apathetic, and lax (Figure C, ment within 1 point on affect scales was 88% confidence intervals using robust empirical
available as a supplement to the online version to 100%. Except affective tone, which was only SE estimates. We identified covariates for

980 | Research and Practice | Peer Reviewed | Cooper et al. American Journal of Public Health | May 2012, Vol 102, No. 5

multivariate analysis from their known associ- attitude IAT scores, and White clinicians had a lower likelihood of perceiving the clinician as
ations with patient---physician communication higher implicit race and compliance IAT scores. participatory (17% lower; Table 4).
and included clinician gender; patient age, Clinician exposure to communication skills
gender, and education; and the mental com- training was not associated with lower IAT Race and Compliance Stereotyping
ponent of the Medical Outcomes Short Form scores. On explicit measures, clinicians per- and Interpersonal Care
12. We evaluated the effect of clinician race on ceived Whites as more cooperative patients Greater implicit stereotyping on the race and
outcomes with separate analyses stratified by than Blacks, but race attitudes (e.g., warmth, compliance IAT was linked to communication
clinician race. We analyzed the data with all preference) were neutral or favorable toward process: for Blacks, 20% longer visits and slower
patients in the models, including the IAT Blacks (Table A, available as a supplement to pace of the dialogue; for Whites, 21% shorter
measure, patient race, and the interaction of the the online version of this article at http://www. visits with a more rapid pace of the dialogue,
IAT measure and patient race as covariates. 25% less clinician verbal dominance, and higher
We stratified the results by patient race only. clinician positive affect. Additionally, the race
For communication measures, which are con- Patient Characteristics and compliance stereotype was associated with
tinuous and were analyzed with linear models Of 445 potentially eligible patients, 269 communication content. In Black patient visits, it
using generalized estimating equations, the (60%) had clinicians who completed the IAT was associated with less, and in White patient
estimate for “implicit bias” is the difference in survey. Greater proportions of patients in- visits more, patient-centered dialogue (Table 3).
the outcome associated with an increase in bias cluded in the analysis were high school grad- Higher levels of the race and compliance IAT
score of 0.5 (considered a moderate level of uates, were insured, and had an annual were linked to lower predicted probabilities of
bias). For patient perceptions of clinicians, household income greater than $35 000 than Blacks (12% lower) and Whites (27% lower)
which were dichotomous and analyzed with did those excluded. Smaller proportions were perceiving the clinician as involving them in
logistic models using generalized estimating assigned to or had clinicians who were assigned decisions, lower likelihoods of Blacks having trust
equations, we estimated the predicted proba- to patient-centered interventions. Similar pro- (10% lower) and confidence (8% lower) in the
bilities of having each outcome for a hypothet- portions of included and excluded patients clinician, and lower likelihood of Whites recom-
ical patient whose clinician exhibited no im- were audio taped at their enrollment visit. The mending the clinician (20% lower; Table 4).
plicit bias (a score of 0.0) and a similar sample was largely middle-aged women. More In analyses stratified by clinicians’ race, the
hypothetical patient whose clinician exhibited than two thirds were known to the clinician. associations of implicit bias or stereotyping with
a moderate level of implicit bias (a score of 0.5). Blacks were younger, were less likely to be most communication measures were similar re-
We performed analyses using SAS version 9.2 married, were more likely to have Medicaid gardless of the race of the clinician. However,
(SAS Institute, Cary, NC). insurance coverage, had poorer mental health the associations of implicit bias or stereotyping
status, and were less likely to be in race- with patient ratings were attenuated among
RESULTS concordant relationships with clinicians than Black patients seeing Black clinicians.
were Whites (Table 2). The interaction of clinician implicit bias with
Forty of the 63 clinicians (63%) with eligible patient race was not significant for any of the
patients (self-reported race Black or White) General Race Bias and Interpersonal Care communication behaviors (although there was
participated. There were no differences in de- More implicit bias on the race attitude IAT suggestive evidence for patient positive affect,
mographic and professional characteristics or was linked to 2 measures of communication P = .06). However, this interaction was statisti-
intervention assignments between clinician process: more clinician verbal dominance in cally significant for the following patient percep-
participants and nonparticipants. Table 1 the visits of Blacks (9% greater) and Whites tions: clinician respect, liking, and participatory
summarizes clinician characteristics and scores (11% greater) and a 0.10-point lower patient style, and patient liking and recommending the
on the 2 IAT measures. The mean scores positive affect score in the visits of Blacks clinician. The interaction of clinician implicit
(range) for participating clinicians were 0.26 (Table 3). For Black patients, higher levels of compliance stereotyping with patient race was
(–1.23 to 1.32) on the race attitude IAT and implicit bias on the race attitude IAT were also significant for visit length, speech speed, verbal
0.29 (–0.60 to 1.39) on the race and compliance linked to lower predicted probabilities of per- dominance, and clinician positive affect but not
IAT. The Cohen d, an effect size measure,44 ceiving respect from the clinician (15% lower), for any patient perceptions.
indicated a moderate implicit bias (0.54) for the liking the clinician themselves (14% lower),
race attitude IAT and a slightly stronger asso- having confidence in the clinician (9% lower), DISCUSSION
ciation of White race with compliance (0.70) and recommending the clinician to others
for the race and compliance IAT. The rela- (13% lower; Table 4). Clinicians, like everyone else, hold varying
tionship between the 2 IAT measures was For White patients, having a clinician with attitudes toward members of racial/ethnic mi-
weaker among non-Black than Black clinicians higher levels of general race bias was linked to norities. However, clinicians differ from others
(the Pearson q = 0.29 and 0.70, respectively). higher likelihoods of perceiving respect from regarding their professional role and code of
Clinicians who considered themselves more the clinician (12% higher) and believing they conduct. We hypothesized that general implicit
politically conservative had higher implicit race are liked by the clinician (6% higher) but racial bias would be related to socioemotional

May 2012, Vol 102, No. 5 | American Journal of Public Health Cooper et al. | Peer Reviewed | Research and Practice | 981

TABLE 1—Clinicians’ Characteristics and Implicit Association Test (IAT) Scores: The Race and
Relationship-Centered Care Study, Baltimore, MD, January 2002–August 2006

Mean 6SD, No. (%), Race Attitude IAT Race and Compliant Patient
Characteristic or Median (Range) Score,a Mean 6SD Pb IAT Score,a Mean (SD) Pb

All clinicians 40 (100) 0.26 60.49 .002 0.29 60.41 < .001
Age, y .82 .13
27–39 15 (38) 0.20 60.46 0.23 60.43
40–49 14 (36) 0.27 60.67 0.47 60.43
50–62 10 (26) 0.32 60.30 0.13 60.33
Gender .43 .12
Female 25 (62) 0.22 60.52 0.21 60.41
Male 15 (38) 0.35 60.45 0.42 60.40
Race .07 .01
White 19 (48) 0.32 60.49 0.47 60.33
Black 9 (22) –0.05 60.45 –0.01 60.39
Asian, including Indian subcontinent 12 (30) 0.41 60.45 0.20 60.43
Type of clinician .5 .93
Physician 36 (90) 0.28 60.51 0.29 60.42
Nurse practitioner 4 (10) 0.10 60.38 0.27 60.45
US medical graduate .06 .47
Yes 34 (85) 0.20 60.47 0.27 60.43
No 6 (15) 0.61 60.52 0.40 60.33
Board certified .53 .45
Yes 36 (90) 0.28 60.51 0.30 60.42
No 4 (10) 0.11 60.39 0.11 60.36
Specialty .34 .22
Internal medicine 31 (78) 0.31 60.50 0.33 60.41
Family medicine 9 (22) 0.13 60.46 0.13 60.39
Years since completing residency 13.4 67.3
Political identity .09 .37
Conservative 10 (29) 0.53 60.54 0.41 60.40
Liberal 25 (71) 0.26 60.35 0.27 60.41
Number of patients 6.5 (1–16)
Communication skills intervention assignment .19 .15
Yes 18 (45) 0.38 60.51 0.39 60.43
No 22 (55) 0.17 60.47 0.20 60.38

Note. Although 40 clinicians participated, 1 did not provide age data, 5 did not complete the political identity question, 1 did not complete the race IAT, and 1 did not complete the race and
compliant patient IAT.
The IAT scores do not have units. They are the standardized difference in response time to the 2 tasks (the difference in latency measures [measured in milliseconds] divided by the standard
The P values for the IAT scores for all clinicians compare the mean scores to a score of zero using a 2-sided 1 sample t-test. Other P values are from t-tests or analysis of variance (if more than 2
groups) to compare mean scores across physician characteristics.

and stylistic patterns of communication—those correct in hypothesizing distinct communica- As expected, the IAT measures were con-
common to social interactions regardless of tion pathways. As in previous work,15---17,45 the sistently associated with Blacks’ poor ratings of
professional setting—whereas implicit race and correlation between the 2 IAT measures was patient care. Unexpectedly, we observed both
compliance stereotyping would be related to relatively weak for non-Black clinicians, sug- positive and negative associations between
both medically focused communication and gesting that although related, the measures implicit racial attitudes and White patient
normative patterns of interaction. We expected may reflect different cognitive processes re- ratings of patient care. Although 1 study has
more implicit bias on both measures to have lated to race. Moreover, the measures were shown an association of implicit bias with
a negative influence on Black, but not White, associated with different indicators of visit Blacks’ negative ratings of care,19 this is the first
patients’ experiences. We were generally communication. study, to our knowledge, to demonstrate that

982 | Research and Practice | Peer Reviewed | Cooper et al. American Journal of Public Health | May 2012, Vol 102, No. 5

TABLE 2—Patient Characteristics, Overall and by Race: The Race and Relationship-Centered Care Study,
Baltimore, MD, January 2002–August 2006

All Patients (n = 269), Black Patients (n = 213), White Patients (n = 56),

Characteristic, Frequency, % No. (%) or Mean 6SD No. (%) or Mean 6SD No. (%) or Mean 6SD Pa

Age, y 56.2 613.2 54.5 613.3 62.7 610.4 < .001

Female gender 192 (71) 156 (73) 36 (64) .19
High school graduate 210/260 (81) 165/204 (81) 45 (80) .99
Married 107/267 (40) 70/211 (33) 37 (66) < .001
Lives alone 58 (22) 52 (24) 6 (11) .03
Health care insurance
Has any health insurance 247 (92) 194 (91) 53 (95) .58
Has Medicaid 63/259 (24) 55/203 (27) 8 (14) .05
Has Medicare 87/256 (34) 67/200 (34) 20 (36) .75
Has private health insurance 163/257 (63) 125/201 (62) 38 (68) .53
Annual household income < $35 000 150/254 (59) 122/203 (60) 28/51 (55) .53
Employment status, working 111/267 (42) 95/211 (45) 16 (29) .03
Mental component score of the Medical Outcomes Short Form 12b 47.0 612.6 46.1 613.2 50.4 69.2 .006
Physical component score of the Medical Outcomes Short Form 12b 41.8 612.9 41.7 613.0 42.2 612.5 .83
Moderately well known by clinician 155/224 (69) 113/173 (65) 42/51 (82) .02
Seen by race-concordant clinician 66 (25) 42 (20) 24 (43) < .001

Note. If there are any missing data for the characteristic, the number of patients with data is specified.
P value from the Fisher exact test or 2-sample t-test.
There were 266 patients with data for the Medical Outcomes Short Form 12 (211 Blacks and 55 Whites).

both implicit bias and stereotyping are associ- of communication, including nonverbal be- handling of the concerns of Black patients.54---58
ated with directly observed medical visit com- haviors not detected by our coders, that Clinicians with more implicit compliance ster-
munication and patient perceptions of care. contributes to poorer patient perceptions. eotyping may be trying to compensate for what
The negative effect of implicit race bias for Social psychology studies show that implicit they perceive as greater mistrust from Black
Black patients is evident in communication attitudes “leak” during interactions through patients, or they may be making well-inten-
indicators (e.g., more clinician-dominated visit the inadvertent display of negative nonverbal tioned efforts to promote patient adherence
dialogue and lower coder ratings of patient behaviors48 and sentiments,49 even when in- or to not appear prejudiced. Conversely, the
positive affect during the visit) and a broad dividuals consciously endorse racial equality combination of slow speech and low patient
array of negative patient ratings. The effect of and are averse to any suggestion of racial centeredness may convey an authoritative tone
implicit stereotyping is also negative for Black bias.50 In previous studies, IAT scores reflecting that creates an overall negative impression on
patients. It is associated with lower levels of more racial bias predicted less speaking time, patients. Associations of slower speech with
patient-centered dialogue and lower patient less smiling, fewer social comments, less speech more biomedically focused and information-
ratings of trust and confidence in the clinician. fluency, and more speech errors among partic- dense visits, less patient centeredness, and
These findings are consistent with other studies ipants interacting with Black (than White) poorer interpersonal care ratings have been
demonstrating that Blacks are at greater risk experimenters.49,51 Blacks may be sensitive to reported in previous studies.59,60
than are Whites for narrowly biomedically these cues and use them in drawing conclu- In contrast to consistent negative findings for
focused visits with restricted patient input in sions about poor interpersonal treatment.20 Blacks, the effect of implicit race attitudes for
the psychosocial and lifestyle realm.7,28 Less positive affect among Black patients may White patients is largely, although not exclu-
Patient-centered communication is associ- also reflect “stereotype threat,” a phenomenon sively, positive. As implicit race bias increases,
ated with greater patient trust,10,46,47 which is whereby cues in one’s surroundings accentuate White patients report being more respected
in turn associated with adherence and conti- negative stereotypes associated with one’s and liked, and as implicit compliance stereo-
nuity of care.37,38,40 Thus, the differences in group and activate physiological and psycho- typing increases, coders rate the communica-
communication behaviors and patient ratings logical processes with negative influences on tion in their visits as more patient centered, less
of care in this study may have implications behavior.52,53 verbally dominant, and higher in clinician
for health outcomes. The communication The associations of implicit stereotyping positive affect. Some negative influences of
markers associated with implicit race attitudes with longer visits and slower speech might implicit bias are common to both Blacks and
may be a proxy for a more pervasive pattern suggest a more conscientious and thorough Whites. Clinicians with general race bias are

May 2012, Vol 102, No. 5 | American Journal of Public Health Cooper et al. | Peer Reviewed | Research and Practice | 983

TABLE 3—Implicit General Race Bias and Race and Medical Compliance Stereotyping as Predictors of Patient–Clinician
Communication by Patient Race: The Race and Relationship-Centered Care Study, Baltimore, MD, January 2002–August 2006

Implicit General Race Bias, Implicit General Race Bias, Implicit Race and Medical Implicit Race and Medical
Black Patients White Patients Compliance Stereotyping, Compliance Stereotyping,
(n = 131)a (n = 48) Black Patients (n = 135)a White Patients (n = 48)
Communication Behavior Mean Estimateb (95% CI) P Mean Estimateb (95% CI) P Mean Estimateb (95% CI) P Mean Estimateb (95% CI) P

Verbal dominance ratio .05 .01 .14 .02

No implicit bias 1.47 (1.29, 1.66) 1.27 (1.09, 1.49) 1.48 (1.28, 1.71) 1.56 (1.26, 1.94)
Implicit bias 9% (0, 19) 11% (2, 21) 10% (–3, 24) –25% (–42, –4)
Visit length, minutes .36 .14 .02 .001
No implicit bias 14.0 (11.7, 16.7) 17.1 (14.6, 19.9) 13.3 (11.2, 15.7) 19.5 (16.0, 23.8)
Implicit bias 7% (–7, 22) 8% (–17, 3) 20% (3, 40) –21% (–31, –9)
Speech speed, statements per minute .23 .83 .02 .001
No implicit bias 25.5 (24.1, 26.9) 23.7 (21.6, 25.8) 25.8 (24.6, 27.0) 21.6 (19.4, 23.8)
Implicit bias –0.76 (–2.02, 0.50) 0.25 (–2.05, 2.55) –1.75 (–3.25, –0.25) 3.9 (1.6, 6.3)
Patient centeredness ratio .63 .37 .06 .02
No implicit bias 1.66 (0.95, 2.37) 0.70 (0.58, 0.82) 1.97 (1.00, 2.94) 0.60 (0.45, 0.74)
Implicit bias –0.10 (–0.51, 0.31) –0.05 (–0.17, 0.07) –0.93 (–1.91, 0.04) 0.15 (0.02, 0.28)
Clinician positive affect .14 .78 .35 .02
No implicit bias 3.63 (3.51, 3.75) 3.38 (3.31, 3.45) 3.60 (3.49, 3.72) 3.30 (3.20, 3.40)
Implicit bias –0.10 (–0.23, 0.03) –0.01 (–0.07, 0.05) –0.06 (–0.19, 0.07) 0.12 (0.02, 0.21)
Patient positive affect .04 .87 .53 .09
No implicit bias 3.39 (3.30, 3.49) 3.31 (3.19, 3.43) 3.36 (3.26, 3.45) 3.24 (3.10, 3.38)
Implicit bias –0.10 (–0.19, –0.00) 0.01 (–0.09, 0.10) –0.04 (–0.16, 0.08) 0.11 (–0.02, 0.24)

Note. CI = confidence interval. Adjusted for clinician gender and patient age, gender, education, and the mental component of the Medical Outcomes Short Form 12.
We excluded 1 observation from speech speed and 1 from patient centeredness because they were extreme outliers.
We estimated the means while holding all other covariates at their means. The estimate for “no pro-White bias” for verbal dominance and visit length is the geometric mean from the generalized
estimating equations (GEE) model for a bias score of zero; the estimate for “implicit bias” is the percentage change in verbal dominance and visit length associated with a 0.5-point increase in the
bias score. For all other variables, the estimate for “no implicit bias” is the mean outcome score from the GEE model for a bias score of zero; the estimate for “implicit bias” is the change in the
outcome associated with a change in bias score of 0.5 (considered a moderate level of bias).

more verbally dominant in the visits of all Limitations performance. However, studies of perfor-
patients, and both patient groups perceive The study has some limitations. The sam- mance bias conclude that recordings have
clinicians with race and compliance stereotyp- pling of clinicians and patients was not ran- little systematic effect on performance.64---67
ing as less likely to involve patients in treatment dom but contingent on participation in one of Patients were well known by their clinicians
decisions. A more conservative political orien- the earlier studies. Participating clinicians may and may have given socially desirable re-
tation is a correlate of implicit bias in our sample, be more motivated about communication sponses to survey items; however, it is un-
and previous work suggests that politically skills and caring for minority patients than are likely this would have differed according to
conservative doctors work in more traditional other physicians; they are also from an urban the implicit attitudes of clinicians.
area with a high representation of Blacks, Although the coders achieved high reli-
settings and may be less open to partnership
where interracial relationships may differ ability and identified some racial differences
relationships with their patients.61---63 To explore
from other areas. The study sample included in patient and clinician affect linked to implicit
whether the association of implicit bias and
mostly Black patients seeing White clinicians; attitudes, both are White women and may be
stereotyping with clinician verbal dominance
we might have seen different results with less sensitive to biased clinicians’ subtle neg-
might be a consequence of a more traditional,
a larger sample of White patients. However, ative nonverbal vocal cues that might be
authoritarian approach to the doctor---patient because most ethnic minorities in the United evident to a Black coder.20 Measuring the IAT
relationship, we adjusted for clinicians’ political States receive care from White physicians, some months after the patient interview
ideology and found it partially explained this these findings may be relevant to a large makes it conceivable that the patient interview
finding, but only among Black patients, perhaps proportion of ethnic minority primary care influenced IAT measures, rather than the
because we had limited statistical power in our patients. Knowledge that the visit was being reverse. However, this alternative inter-
smaller sample of White patients. recorded may have biased clinician pretation is rendered implausible by, in

984 | Research and Practice | Peer Reviewed | Cooper et al. American Journal of Public Health | May 2012, Vol 102, No. 5

TABLE 4—Implicit General Race Bias and Race and Medical Compliance Stereotyping as Predictors of Patient Perceptions of
Clinician, by Patient Race: The Race and Relationship-Centered Care Study, Baltimore, MD, January 2002–August 2006

Implicit General Race Bias, Implicit General Race Bias, Implicit Race and Medical Implicit Race and Medical
Black Patients White Patients Compliance Stereotyping, Compliance Stereotyping,
(n = 191)a (n = 55) Black Patients (n = 197)a White Patients (n = 55)
Predicted Probabilitiesb Predicted Probabilitiesb Predicted Probabilitiesb Predicted Probabilitiesb
Patient Perceptions (95% CI) P (95% CI) P (95% CI) P (95% CI) P

Clinician would ask patient to help decide treatment .48 .002 .02 .04
No implicit bias 31.3 (22.8, 41.2) 33.2 (17.9, 53.3) 35.2 (27.8, 43.4) 46.3 (26.5, 67.3)
Implicit bias 28.3 (22.3, 35.1) 16.6 (7.3, 33.5) 23.0 (16.8, 30.7) 19.4 (6.8, 44.1)
Clinician respects him or her .001 < .001 .15 .47
No implicit bias 50.2 (38.6, 61.7) 14.2 (7.0, 26.5) 44.1 (33.7, 55.1) 27.9 (15.5, 44.9)
Implicit bias 34.9 (27.2, 43.5) 26.5 (18.7, 36.0) 37.0 (28.7, 46.2) 21.6 (8.3, 45.5)
Clinician likes him or her .25 .003 .41 .23
No implicit bias 28.5 (20.4, 38.2) 2.4 (0.5, 9.9) 26.5 (18.8, 35.9) 15.1 (5.0, 37.8)
Implicit bias 23.8 (17.9, 31.0) 8.0 (3.0, 19.5) 23.2 (16.7, 31.3) 6.6 (1.8, 21.4)
Likes clinician < .001 .12 .75 .12
No implicit bias 46.6 (37.5, 55.9) 22.8 (11.9, 39.1) 39.8 (31.0, 49.3) 42.1 (27.1, 58.7)
Implicit bias 32.7 (26.2, 39.9) 31.4 (20.8, 44.5) 38.4 (31.7, 45.5) 31.6 (22.5, 42.5)
Trusts clinician .29 .49 .02 .17
No implicit bias 71.2 (64.3, 77.3) 81.3 (61.7, 92.2) 74.4 (67.3, 80.5) 83.1 (64.2, 93.1)
Implicit bias 67.6 (60.6, 73.8) 76.9 (68.9, 83.4) 64.0 (56.5, 70.9) 73.9 (64.2, 81.7)
Has confidence in clinician .007 .25 .05 .15
No implicit bias 80.1 (72.9, 85.8) 83.6 (69.0, 92.1) 79.3 (72.2, 85.0) 83.9 (66.4, 93.2)
Implicit bias 71.4 (64.1, 77.7) 79.0 (70.7, 85.4) 71.2 (63.8, 77.7) 74.9 (66.0, 82.0)
Would recommend clinician .001 .51 .18 .03
No implicit bias 47.3 (38.6, 56.1) 32.3 (17.4, 51.8) 42.4 (32.8, 52.7) 43.8 (24.0, 65.8)
Implicit bias 34.4 (27.1, 42.5) 29.1 (19.4, 41.2) 36.3 (29.4, 43.8) 23.7 (15.5, 34.5)

Note. CI = confidence interval. Adjusted for clinician gender and patient age, gender, education, and the mental component of the Medical Outcomes Short Form 12.
A few patients did not respond to some of the outcomes (minimum n = 186 for Black and n = 54 for White patients).
We estimated the predicted probabilities while holding all other covariates in the model constant at their mean. These are the probabilities of having the outcome for a hypothetical patient whose
clinician exhibited no implicit bias (a score of 0.0) and for a similar hypothetical patient whose clinician exhibited a moderate level of implicit bias (a score of 0.5).

combination, the known stability of IAT Conclusions upholding the elimination of healthcare dispar-
measures,27 their limited malleability,45,68 and Notwithstanding these limitations, this study ities as a local, national, and global priority. j
the lack of evidence for effects of order of informs future research and interventions tar-
administration of IAT measures on relevant geting health professionals to reduce health- About the Authors
behaviors.18 We have not included an analysis care disparities. Theoretically based interven- Lisa A. Cooper and Mary Catherine Beach are with the
of the association of clinicians’ self-reported tion strategies—which increase clinicians’ Department of Medicine, Johns Hopkins University School
of Medicine, Baltimore, MD. Debra L. Roter is with the
race attitudes with communication behaviors awareness and understanding of the basis of Department of Health Behavior and Society at the Johns
or ratings of care. However, IAT measures bias and help them develop cultural sensitivity, Hopkins Bloomberg School of Public Health, Baltimore,
have greater validity than do self-report mea- patient-centered communication, and partner- MD. Kathryn A. Carson is with the Department of
Epidemiology, Johns Hopkins Bloomberg School of Public
sures in predicting stereotyping behavior.18,24 ship-building in the patient---clinician relation- Health. Janice A. Sabin is with the Department of Medical
Because the study included multiple compari- ship—will enable clinicians to reduce their Education and Biomedical Informatics, Center for Clinical
sons, the possibility of statistical type I error reliance on social categories when clinically and Epidemiological Research, University of Washington,
Seattle. Anthony G. Greenwald is with the Department of
exists; however, this is unlikely because analy- irrelevant.69 Interpersonal bias in healthcare is
Psychology, University of Washington. Thomas S. Inui is with
ses were conceptually driven and grounded only 1 of the manifestations of racial discrim- the Regenstrief Institute at Indiana University School of
in previous literature, most of the observed ination in our society; however, health pro- Medicine, Indianapolis.
Correspondence should be sent to Lisa A. Cooper, MD,
associations are in the expected directions, fessionals can serve as influential advocates for
MPH, Johns Hopkins University, Welch Center for Pre-
and findings across related measures are social justice70 by encouraging open discourse vention, Epidemiology, and Clinical Research, 2024 East
consistent. about the existence of bias in healthcare and Monument Street, Suite 2-500, Baltimore, MD 21287

May 2012, Vol 102, No. 5 | American Journal of Public Health Cooper et al. | Peer Reviewed | Research and Practice | 985

(e-mail: Reprints can be ordered at of care, and concordance of patient and physician race. explicit attitudes are related but distinct constructs. Exp by clicking the “Reprints” link. Ann Intern Med. 2003;139(11):907---915. Psychol. 2007;54(1):14---29.
This article was accepted October 31, 2011. 10. Street RL Jr, O’Malley KJ, Cooper LA, Haidet P. 26. Greenwald AG, Nosek BA, Banaji MR. Under-
Understanding concordance in patient---physician rela- standing and using the Implicit Association Test: I. An
Contributors tionships: personal and ethnic dimensions of shared improved scoring algorithm. J Pers Soc Psychol. 2003;85
L. A. Cooper acquired the data; provided administrative, identity. Ann Fam Med. 2008;6(3):198---205. (2):197---216.
technical, and material support; and supervised partici-
11. van Ryn M, Burke J. The effect of patient race and 27. Egloff B, Schwerdtfeger A, Schmukle SC. Temporal
pant recruitment, data collection, and data analysis. L. A.
socio-economic status on physicians’ perceptions of pa- stability of the Implicit Association Test—anxiety. J Pers
Cooper, D. L. Roter, M. C. Beach, J. A. Sabin, A. G.
tients. Soc Sci Med. 2000;50(6):813---828. Assess. 2005;84(1):82---88.
Greenwald, and T. S. Inui contributed to the conceptu-
alization and design of the study. L. A. Cooper, D. L. 12. Lutfey KE, Ketcham JD. Patient and provider 28. Roter DL, Stewart M, Putnam SM, Lipkin M Jr, Stiles
Roter, M. C. Beach, J. A. Sabin, A. G. Greenwald, and assessments of adherence and the sources of disparities: W, Inui TS. Communication patterns of primary care
K. A. Carson participated in analysis and interpretation of evidence from diabetes care. Health Serv Res. 2005;40(6 physicians. JAMA. 1997;277(4):350---356.
data. L. A. Cooper, D. L. Roter, M. C. Beach, J. A. Sabin, pt 1):1803---1817. 29. Roter DL, Larson S. The relationship between
and K. A. Carson drafted the article. L. A. Cooper and 13. Bogart LM, Catz SL, Kelley JA, et al. Factors residents’ and attending physicians’ communication dur-
T. S. Inui obtained funding. K. A. Carson performed the influencing physicians’ judgments of adherence and ing primary care visits: an illustrative use of the Roter
statistical analysis. All authors participated in critical treatment decisions for patients with HIV disease. Med Interaction Analysis System. Health Commun. 2001;13
revision of the article for important intellectual content. Decis Making. 2001;21(1):28---36. (1):33---48.
14. van Ryn M, Burgess D, Malat J, Griffin J. Physicians’ 30. Mead N, Bower P. Measuring patient-centredness:
Acknowledgments perceptions of patients’ social and behavioral character- a comparison of three observation-based instruments.
This work was supported in part by the Fetzer Founda- istics and race disparities in treatment recommendations Patient Educ Couns. 2000;39(1):71---80.
tion Relationship-Centered Care Research Network for men with coronary artery disease. Am J Public Health. 31. Lazare A, Putnam SM, Lipkin MJ. Three functions
(grant 2005), the National Heart, Lung, and Blood 2006;96:351---357. of the medical interview. In: Lipkin M, Putnam S,
Institute (grants R01HL69403 and K24HL083113), the 15. Green AR, Carney DR, Pallin DJ, et al. Implicit bias Lazare A, eds. The Medical Interview: Clinical Care,
Agency for Healthcare Research and Quality (grant among physicians and its prediction of thrombolysis Education, and Research. New York: Springer-Verlag;
R01HS013645), and the Implicit and Unconscious Cogni- decisions for Black and White patients. J Gen Intern Med. 1995:3---19.
tion Research Fund at University of Washington. L. A. 2007;22(9):1231---1238. 32. Paasche-Orlow M, Roter D. The communication
Cooper also gratefully acknowledges the support of the
16. Sabin JA, Rivara FP, Greenwald AG. Physician patterns of internal medicine and family practice physi-
MacArthur Foundation Fellows Program.
implicit attitudes and stereotypes about race and quality cians. J Am Board Fam Pract. 2003;16(6):485---493.
of medical care. Med Care. 2008;46(7):678---685. 33. Hall JA, Horgan TG, Stein TS, Roter DL. Liking in
Human Participant Protection 17. Sabin J, Nosek BA, Greenwald A, Rivara FP. the physician---patient relationship. Patient Educ Couns.
The Johns Hopkins institutional review board approved Physicians’ implicit and explicit attitudes about race by 2002;48(1):69---77.
all study protocols. All patients provided written in- MD race, ethnicity, and gender. J Health Care Poor 34. Beach MC, Roter DL, Wang NY, Duggan PS, Cooper
formed consent. All clinicians provided either oral or Underserved. 2009;20(3):896---913. LA. Are physicians’ attitudes of respect accurately per-
written informed consent.
18. Greenwald AG, Poehlman TA, Uhlmann E, Banaji ceived by patients and associated with more positive
MR. Understanding and using the Implicit Association communication behaviors? Patient Educ Couns. 2006;62
References Test: III. Meta-analysis of predictive validity. J Pers Soc (3):347---354.
1. Smedley BD, Stith AY, Nelson AR, eds. Unequal Psychol. 2009;97(1):17---41. 35. Kao AC, Green DC, Davis NA, Koplan JP, Cleary PD.
Treatment: Confronting Racial and Ethnic Disparities in Health 19. Penner LA, Dovidio JF, West TV, et al. Aversive Patients’ trust in their physicians: effects of choice,
Care. Washington, DC: National Academies Press; 2003. racism and medical interactions with Black patients: continuity, and payment method. J Gen Intern Med.
2. Saha S, Arbelaez JJ, Cooper LA. Patient---physician a field study. J Exp Soc Psychol. 2010;46(2):436---440. 1998;13(10):681---686.
relationships and racial disparities in the quality of health 20. Livingston RW, Drwecki BB. Why are some in- 36. Kaplan SH, Gandek B, Greenfield S, Rogers W,
care. Am J Public Health. 2003;93(10):1713---1719. dividuals not racially biased? Susceptibility to affective Ware JE. Patient and visit characteristics related to
3. Doescher MP, Saver BG, Franks P, Fiscella K. Racial conditioning predicts nonprejudice toward Blacks. Psy- physicians’ participatory decision-making style. Results
and ethnic disparities in perceptions of physician style chol Sci. 2007;18(9):816---823. from the Medical Outcomes Study. Med Care. 1995;33
and trust. Arch Fam Med. 2000;9(10):1156---1163. 21. Cooper LA, Roter DL, Bone LR, et al. A randomized (12):1176---1187.

4. Malat J. Social distance and patients’ rating of health- controlled trial of interventions to enhance patient-- physician 37. Schneider J, Kaplan SH, Greenfield S, Li W, Wilson
care providers. J Health Soc Behav. 2001;42(4):360---372. partnership, patient adherence and high blood pressure IB. Better physician---patient relationships are associated
control among ethnic minorities and poor persons: study with higher reported adherence to antiretroviral therapy
5. Boulware LE, Cooper LA, Ratner LE, LaVeist TA, protocol NCT00123045. Implement Sci. 2009;4:7. in patients with HIV infection. J Gen Intern Med. 2004;
Powe NR. Race and trust in the health care system. Public 19(11):1096---1103.
22. Cooper LA, Ford DE, Ghods BK, et al. A cluster
Health Rep. 2003;118(4):358---365.
randomized trial of standard quality improvement versus 38. McGinnis B, Olson KL, Magid D, et al. Factors
6. Ngo-Metzger Q, Legedza AT, Phillips RS. Asian patient-centered interventions to enhance depression related to adherence to statin therapy. Ann Pharmacother.
Americans’ reports of their health care experiences. care for African Americans in the primary care setting: 2007;41(11):1805---1811.
Results of a national survey. J Gen Intern Med. 2004;19 study protocol NCT00243425. Implement Sci. 2010; 39. O’Malley AS, Sheppard VB, Schwarts M,
(2):111---119. 5:18. Mandelblatt J. The role of trust in use of preventive
7. Johnson RL, Roter D, Powe NR, Cooper LA. Patient 23. Greenwald AG, McGhee DE, Schwartz JL. Measur- services among low-income African-American women.
race/ethnicity and quality of patient---physician commu- ing individual differences in implicit cognition: the Im- Prev Med. 2004;38(6):777---785.
nication during medical visits. Am J Public Health. plicit Association Test. J Pers Soc Psychol. 1998;74 40. Safran DG, Montgomery JE, Chang H, Murphy J,
2004;94(12):2084---2090. (6):1464---1480. Rogers WH. Switching doctors: predictors of voluntary
8. Street RL Jr, Gordon H, Haidet P. Physicians’ 24. Nosek BA, Greenwald AG, Banaji MR. Under- disenrollment from a primary physician’s practice. J Fam
communication and perceptions of patients: is it how they standing and using the Implicit Association Test: II. Pract. 2001;50(2):130---136.
look, how they talk, or is it just the doctor? Soc Sci Med. Method variables and construct validity. Pers Soc Psychol 41. Ware J Jr, Kosinski M, Keller SD. A 12-item short-
2007;65(3):586---598. Bull. 2005;31(2):166---180. form health survey. Construction of scales and prelimi-
9. Cooper LA, Roter D, Johnson R, Ford D, Steinwach 25. Nosek BA, Smyth FL. A multitrait-multimethod nary tests of reliability and validity. Med Care. 1996;34
DM, Powe NR. Patient-centered communication, ratings validation of the Implicit Association Test: implicit and (3):220---233.

986 | Research and Practice | Peer Reviewed | Cooper et al. American Journal of Public Health | May 2012, Vol 102, No. 5

42. Zeger SL, Liang K. Longitudinal data analysis for 61. Mechanic D. Politics, Medicine, and Social Science.
discrete and continuous outcomes. Biometrics. 1986;42 New York: John Wiley & Sons; 1974.
(1):121---130. 62. Whitney SN, Brown BW Jr, Brody H, Alcser KH,
43. Liang K-Y, Zeger SL. Longitudinal data analysis using Bachman JG, Greely HT. Views of United States physi-
generalized linear models. Biometrika. 1986;73(1):13---22. cians and members of the American Medical Association
44. Cohen J. Statistical Power Analysis for the Behavioral House of Delegates on physician-assisted suicide. J Gen
Sciences. 2nd ed. Hillsdale, NJ: Erlbaum; 1988. Intern Med. 2001;16(5):290---296.

45. Nosek BA, Smyth FL, Jansen JJ, et al. Pervasiveness 63. Waitzkin H. Information-giving in medical care. J
and correlates of implicit attitudes and stereotypes. Eur Health Soc Behav. 1985;26:81---101.
Rev Soc Psychol. 2007;18(1):36---88. 64. Inui TS, Carter WB, Kukull WA, Haigh VH. Out-
come based doctor---patient interaction analysis: I. Com-
46. Zolnierek KB, Dimatteo MR. Physician communi-
parison of techniques. Med Care. 1982;20(6):535---549.
cation and patient adherence to treatment: a meta-anal-
ysis. Med Care. 2009;47(8):826---834. 65. Redman S, Dickinson JA, Cockburn J, Hennrikus D,
Sanson-Fisher RW. The assessment of reactivity in direct
47. Fiscella K, Meldrum S, Franks P, et al. Patient trust: is
observation studies of doctor---patient interactions. Psy-
it related to patient-centered behavior of primary care
chol Health. 1989;3(1):17---28.
physicians? Med Care. 2004;42(11):1049---1055.
66. Pringle M, Stewart-Evans C. Does awareness of
48. Dovidio JF, Kawakami K, Gaertner SL. Implicit and
being video recorded affect doctors’ consultation behav-
explicit prejudice and interracial interaction. J Pers Soc
ior? Br J Gen Pract. 1990;40(340):455---458.
Psychol. 2002;82(1):62---68.
67. Coleman T, Manku-Scott T. Comparison of video-
49. Richeson JA, Shelton JN. Brief report: thin slices of
recorded consultations with those in which patients’
racial bias. J Nonverbal Behav. 2005;29(1):75---86.
consent is withheld. Br J Gen Pract. 1998;48(427):971---
50. Dovidio JF, Penner LA, Albrecht TL, Norton WE, 974.
Gaertner SL, Shelton JN. Disparities and distrust: the
68. Joy-Gaba JA, Nosek BA. The surprisingly limited
implications of psychological processes for understanding
malleability of implicit racial evaluations. Soc Psychol.
racial disparities in health and health care. Soc Sci Med.
69. Burgess D, van Ryn M, Dovidio J, Saha S. Reducing
51. McConnell AR, Leibold JM. Relations among the
racial bias among health professionals: lessons from
Implicit Association Test, discriminatory behavior, and
social-cognitive psychology. J Gen Intern Med. 2007;22
explicit measures of racial attitudes. J Exp Soc Psychol.
70. Medical professionalism in the new millennium: a
52. Steele CM, Aronson J. Stereotype threat and the
physician charter. Ann Intern Med. 2002;136(3):243---
intellectual test performance of African Americans. J Pers
Soc Psychol. 1995;69(5):797---811.
53. Burgess DJ, Warren J, Phelan S, Dovidio J, van Ryn
M. Stereotype threat and health disparities: what medical
educators and future physicians need to know. J Gen
Intern Med. 2010;25(suppl 2):S169---S177.
54. Wiggers JH, Sanson-Fisher R. Duration of general
practice consultations: association with patient occupa-
tional and educational status. Soc Sci Med. 1997;44
55. Gross DA, Zyzanski SJ, Borawski EA, Cebul RD,
Stange KC. Patient satisfaction with time spent with their
physician. J Fam Pract. 1998;47(2):133---137.
56. Wilson IB, Kaplan S. Physician---patient communi-
cation in HIV disease: the importance of patient, physi-
cian, and visit characteristics. J Acquir Immune Defic
Syndr. 2000;25(5):417---425.
57. Howie JG, Porter AM, Heaney DJ, Hopton JL. Long
to short consultation ratio: a proxy measure of quality
of care for general practice. Br J Gen Pract. 1991;41
58. Morrell DC, Evans ME, Morris RW, Roland MO.
The “five minute” consultation: effect of time constraint
on clinical content and patient satisfaction. Br Med J (Clin
Res Ed). 1986;292(6524):870---873.
59. Erby LH, Roter D, Larson S, Cho J. The rapid
estimate of adult literacy in genetics (REAL-G): a means
to assess literacy deficits in the context of genetics. Am
J Med Genet A. 2008;146A(2):174---181.
60. Roter DL, Larson SM, Beach MC, Cooper LA.
Interactive and evaluative correlates of dialogue sequence:
a simulation study applying the RIAS to turn taking
structures. Patient Educ Couns. 2008;71(1):26---33.

May 2012, Vol 102, No. 5 | American Journal of Public Health Cooper et al. | Peer Reviewed | Research and Practice | 987