HEALTH COMMUNICATION, 18(1), 23–39

Copyright © 2005, Lawrence Erlbaum Associates, Inc.

Mothers’ Involvement in Decision

Making During the Birthing Process:
A Quantitative Analysis of Women’s
Online Birth Stories
Carma L. Bylund
Department of Communication Studies
Department of Community and Behavioral Health
University of Iowa

This article reports on a study of 551 women’s birth stories that were posted on a birth
stories Web site. These online stories were analyzed for the communication and deci-
sion making that takes place between patients and clinicians during the birthing pro-
cess. In more than half of the stories, the women wrote about at least 1 decision that
was made. Further analyses were performed on the 285 stories in which decisions
were made. According to this analysis, overall, women were involved in decision
making about 57% of the time. The most frequent decision was about painkillers.
Making decisions about painkillers and having a midwife as a primary clinician pre-
dicted a woman’s increased involvement in decision making. Women’s involvement
in decision making correlated positively with the use of positive emotion words and
negatively with the use of negative emotion words in the online birth stories.

Birth stories are everywhere and nowhere. Seen in every movie theater but heard only
in brief gasps of attention in grocery store lines or parking lots, inculcated in prenatal
classrooms but shamed to the edges of conversation, birth stories permeate and haunt
our every day lives. (Pollock, 1999, p. 1)

These words hold some truth today. Birth stories are everywhere—everyone
has a birth story associated with him or her somewhere, whether that person knows

Requests for reprints should be sent to Carma Bylund, Department of Psychiatry & Behavioral Sci-
ences, Memorial Sloan Kettering Cancer Center, 1242 Second Ave., New York, NY 10021. E-mail:
carmabylund@yahoo.com
24 BYLUND

the story or not. Many parents, grandparents, siblings, and friends also hold dear
the narratives of a loved one’s birth. The proposition that birth stories are nowhere,
however, is no longer accurate. The Internet has provided a way for mothers and
fathers to offer their birth stories to anyone willing to read them. The first 100 re-
sults from a google.com search of the phrase “birth story,” provide links to at least
14 Web sites where a person can go to post a birth story for others to read. Add the
myriad other personal and business Web sites where birth stories have been posted,
and it is clear that there are currently many means, for Internet users at least, to read
and share birth stories.
Whether it is on the Internet or face to face, why do so many people, in most
cases women, willingly share birth stories? For some it may be for entertainment
value, to compare stories with other women or to connect with other women
(Soparkar, 1998). Others may see it as ritual—part of the “rite of passage of child-
birth” (Soparkar, 1998, p. viii). Still others may use the birth story as a means for
processing the experience (Simkin, 1992). For health communication scholars, the
birth story provides a means to better understand the woman’s experience with cli-
nician–patient communication during the birthing process. As such, this study ex-
amined clinician–patient communication and decision making through the quanti-
tative analysis of women’s online birth stories.
The use of unsolicited narratives from the Internet is a new, but growing,
method of inquiry, These narratives are a rich source of data for researchers (Rob-
inson, 2001). Although online health support groups and other Internet functions
such as e-mail groups can also serve as a means for sharing narratives about health,
unsolicited narratives are unique and have received little attention from health
communication scholars. These first-person accounts are posted for the evident
function of sharing a narrative with the readers. Such unsolicited narratives, al-
though often written informally with misspellings and grammatical errors, include
powerful expressions of experiences and emotions, as the authors of these narra-
tives may feel more comfortable sharing these than they would in a face-to-face in-
terview. In addition, unsolicited online narratives may allow for voices that have
previously been unheard (Robinson, 2001). The use of these narratives also allows
access to much larger sets of individual stories than can usually be attained through
conventional data-gathering techniques.

BIRTH STORIES

Fisher’s (1987) narrative theory argues that “humans are essentially storytellers”
(p. 64). Although Fisher’s definition of narrative includes more than just birth
stories, the characteristics of narrative he proposes are useful in thinking about
birth stories. Fisher argues that two criteria are important to apply in determining
a narrative’s rationality and, consequently, its ability to teach or persuade. The
 DECISION MAKING DURING THE BIRTHING PROCESS 25

narrative must be cohesive—free of inconsistencies, realistic, and meaningful.

The narrative also should have fidelity; in other words, it must seem reliable or
truthful. Fisher states that narration can involve either recounting, including his-
tory, biography, and autobiography, or accounting for, such as narratives that
give theoretical explanation and argument. Generally, most birth stories would
be of the recounting type, chronicling the events and feelings surrounding a
birth. Birth stories also have the potential for giving explanation and argument.
For instance, a child’s habitual lateness may lead to her mother telling the birth
story of that child—“You were 6 days late being born!”—offering such a story
as explanation for the child’s current behavior.
The birth story is a subset of this larger construct of narrative, which aside from
stories includes diaries, letters, memoirs, gossip, and more (Ochs & Capps, 1996).
According to Ochs and Capps, temporality and point of view are two important
facets of narrative. The temporality, or chronology, refers to the means by which
the teller can link certain events. This is not to say that narratives are told only in
chronological order; tellers often “shift back and forth in time as bits and pieces of
the tale and the concerns they manifest come to the fore” (Ochs & Capps, 1996, p.
24). For instance, a mother telling a child his or her birth story may begin by talk-
ing about the day the child was born, but then add in events that transpired before
that day as further explanation.
The point of view from which the narrative comes to the hearer is framed by the
perspective of the teller (Goffman, 1974). In using birth stories to study clini-
cian–patient communication, the importance of the point of view cannot be under-
stated. A study using narratives as a means for investigating reported communica-
tion phenomena is inherently from the teller’s perspective. However, using stories
as a means to study other communication phenomena is not without precedent
(e.g., VandeVusse, 1999; Zadoroznyj, 1999). VandeVusse used qualitative meth-
ods to analyze clinician–patient communication in a small sample of birth stories;
however, no study to date has used a quantitative method to analyze clinician–pa-
tient communication in a large sample.
Some researchers have examined the birth story in its own right. Birth stories
are naturally occurring stories in many families (Reese, 1996) and tend to be con-
sistent over the years (Simkin, 1992). A thematic analysis of the birth stories of 41
new mothers revealed five themes of such stories (DiMatteo, Kahn, & Berry,
1993): (a) the loss of autonomy and control; (b) unexpected physical pain; (c) un-
expected emotional reactions such as detachment, self-criticism, disappointment,
and sadness; (d) financial pressures; and (e) the effects of support during labor and
birth. In addition, Pollock (1999) organized birth stories by themes of “something
missing,” narrative rites, pain, and secrecy. The telling of birth stories may differ
depending on the audience or the teller. Mothers telling birth stories to younger
preschoolers (40 months) and to daughters tend to be more interpersonally fo-
cused, talking about the child in the context of others, than do mothers telling birth
26 BYLUND

stories to older preschoolers (58 months) and to sons. Similarities and differences
also exist in the ways that males and females tell birth stories (Reese, 1996). R. E.
Page (2002) argues that the context in which the story is told must also be recog-
nized when considering such sex differences.

CLINICIAN–PATIENT COMMUNICATION

The link between clinician–patient communication and important outcomes such

as the patient’s health status, adherence, and satisfaction has been well established
(e.g., Stewart et al., 1999). Although a similar relation between communication
and patient outcomes may be inferred to exist in clinician–patient relationships in
an obstetrics setting, there has been little focus on the actual communication in
such encounters (see Roter, Geller, Bernhardt, Larson, & Doksum, 1999, for the
only identified published study). However, birth stories provide insight into
women’s perceptions of these medical encounters and the communication that oc-
curs therein. In their stories, mothers frequently describe their level of involvement
in medical decision making during the birthing process. They also often describe
the sex and type of clinician(s) who participated in the birth. Following is a brief re-
view of findings in these areas as support for this study.

Shared Decision Making

Shared decision making between clinician and patient is receiving much attention
in scholarly research. Although some people do not believe that the benefits of pa-
tient participation in medical decision making have been sufficiently demonstrated
(Gattellari, Butow, & Tattersall, 2001; Guadagnoli & Ward, 1998), others have
found patients’ involvement in medical decision making to be associated with pos-
itive patient health outcomes (Greenfield, Kaplan, & Ware, 1985; Greenfield,
Kaplan, Ware, Yano, & Frank, 1988).
However, it is important to note that scholars measure shared decision making
in different ways (see Charles, Gafni, & Whelan, 1997). Most research measuring
decision making in the physician–patient interaction has focused on patient partic-
ipation (England & Evans, 1992; Street & Millay, 2001). This approach puts the
focus solely on the patient’s communicative actions during a medical encounter or
a medical decision. A considerably less frequent approach to studying shared deci-
sion making is to examine the degree to which a physician’s style of communica-
tion accommodates the involvement of the patient in the decision (Adams, Smith,
& Ruffin, 2001; Kaplan, Greenfield, Gandek, Rogers, & Ware, 1996). A third ap-
proach takes both the physician’s and patient’s communicative behaviors into ac-
count; that is, the analysis is focused on the interaction between the two rather than
the individual behaviors of the physician or the individual behaviors of the patient.
Because this approach uses the interaction as the analysis, it focuses on the actual
communication in the encounter rather than looking only at the physician’s behav-
 DECISION MAKING DURING THE BIRTHING PROCESS 27

ior or the patient’s behavior (Braddock, Fihn, Levinson, Jonsen, & Pearlman,
1997; Makoul, Arntson, & Schoefield, 1995). Accordingly, this study, although
from the woman’s perspective, uses an interaction approach by analyzing both cli-
nicians’ and patients’ communication behavior.
Other research has examined issues of involvement in decision making specifi-
cally in the birthing process. Control over decision making manifests itself in dif-
ferent ways during the birthing process. As examples, a woman may refuse the de-
cision that the clinician makes, she may initially oppose the decision but then go
along with it, or she may comply with the decision. There also might be shared
control through explanations or requests (VandeVusse, 1999).
Evidence of the effects on outcomes of a patient’s level of involvement in deci-
sion making during birthing is sparse. Although links have yet to be made between
a mother’s involvement in decision making during birth and health outcomes, stud-
ies have indicated a higher level of involvement to be related to patient satisfaction.
Israeli women who reported more involvement in decisions for obstetric interven-
tions self-reported more general satisfaction with their health care during labor and
delivery than those who reported less involvement (Berg et al., 2001). Patient satis-
faction during birthing is related to the sharing of information and feelings of con-
trol (Green, Coupland, & Kitzinger, 1990; Simkin, 1991). Mothers who report that
they were not given an active say in decision making during birth were less likely
to be satisfied with their care (Brown & Lumley, 1994). Furthermore, explanation
of procedures and mothers’ involvement in choosing among procedures were
ranked highly by women as being important to satisfaction with birth (Drew,
Salmon, & Webb, 1989). However, women do not expect to be involved in all deci-
sions. They understand that they will not be able to control all of the circumstances
surrounding the birth experience, yet they want quality medical care (Lazarus,
1997).
One may also argue that the quality of the overall experience may be measured
specifically through a woman’s emotional experience. Emotional well-being fol-
lowing the birth is related to a woman’s sense that she is being given information
and is in control (Green et al., 1990). VandeVusse (1999), using qualitative inquiry
with a sample of 33 birth stories, observed that mothers’ reported involvement in
decision making during birthing was related to more positive emotions expressed
in talking about the experience, whereas less involvement was related to the nega-
tive emotions expressed. This study builds on these findings, using quantitative
methods on a much larger set of data to test two relevant hypotheses:

H1: There will be a positive correlation between the amount of positive emo-
tions named in birth stories and the mother’s level of involvement in deci-
sions during the birthing process.
H2: There will be a negative correlation between the amount of negative emo-
tions named in birth stories and the mother’s level of involvement in deci-
sions during the birthing process.
28 BYLUND

Although most women report wanting to participate in decisions about their

care during labor (Waldenstrom, Borg, Olsson, Skold, & Wall, 1996), many factors
may affect the level of patient involvement in decision making. For instance, clini-
cal setting, age, and ethnicity all affected the level of perceived involvement in one
study (Berg et al., 2001). Furthermore, Zadoroznyj (1999) found that middle class
primiparous women (those giving birth for the first time) were more likely to take
an active approach toward the birthing process, having sought information and
made decisions on what they wanted to happen, whereas lower class women had a
“fatalistic orientation” (p. 276) toward their first births, evidenced as less involve-
ment in seeking information and participating in decisions. However, among
multiparous women (those giving birth for the second or subsequent time), this dif-
ference between social classes disappeared, with both lower and middle-class
women adopting active approaches toward childbirth. Zadoroznyj (1999) attrib-
uted this to the first birth acting as a socializing event for the lower class women
who then were better able to attempt to control facets of the birthing process along
with the better-resourced middle class women. Although this study does not mea-
sure social class, the overall effect of subsequent births should emerge. Thus, it is
hypothesized that

H3: Birth stories by primiparous women will show less mother involvement in
decision making than will multiparous women’s birth stories.

It is also important to remember that not all decisions are about the same topics;
the nature of the decision may affect the level of the woman’s involvement. Thus,
the following research question is posed:

RQ1: How does the type of decision affect a woman’s level of involvement in the
decision?

Clinician Sex and Type

A meta-analytic review by Roter, Hall, and Aoki (2002) reported some small but
significant differences in the communication of male and female clinicians. Spe-
cific to my study, female physicians were found to use more partnership-building
techniques with their patients than male physicians. However, in the only pub-
lished research to date systematically analyzing clinician–patient communication
in the obstetric setting, male obstetricians were shown to be more likely than fe-
male obstetricians to make partnership-building statements during clinic visits
(Roter et al., 1999). Thus, the following research question is posed:

RQ2: How does the sex of a woman’s primary clinician affect her involvement in
decision making?
 DECISION MAKING DURING THE BIRTHING PROCESS 29

Another point of interest in this study is how the type of clinician, either a physi-
cian or a midwife, is related to communication and decision making during the
birth process. The notions of a mother’s empowerment and control in the birth pro-
cess are associated with midwifery (L. Page, 1995). A statement on the American
College of Nurse-Midwives (n.d.)Web site reads, in part: “Making sound decisions
is key to good health care, and nurse-midwives want you to take an active role in
making the right decisions for you and your family.”
Empirical research supports this statement. In a study of Canadian midwives,
physicians, and nurses, midwives were more client centered in their attitudes than
were physicians and nurses (Blais et al., 1994). As part of being client centered,
midwives believed that women are not given the ability to participate fully in deci-
sions about labor and birth. Consequently, the final hypothesis is

H4: Women with midwives as their primary clinician will perceive their in-
volvement in decision making to be higher than will women with physi-
cians as their primary clinician.

METHODS

Sample
The original sample for this study consisted of 551 birth stories that were posted to
the Web site www.birthstories.com. This is a public Web site, requiring no regis-
tration for a person to post a birth story or to view the stories that have been
posted.1 The Web site is sponsored by iparentingmedia.com, a media group focus-
ing on parents that produces both on- and offline materials. The Web site gives
these instructions:

Thanks for visiting, and we’d love for you to share your story with us—and when
your story appears, share the link with your friends and family. Things to think about
when sharing your story: the duration of the various stages of your labor, how you felt
during the experience, how you reacted to and coped with labor and the birth, as well
as the baby’s gender, weight, and condition.

Two coders selected every fifth story posted to two of the main sections of the Web
site—first births and second or more births. For consistency, stories not written by
the mother were skipped, and the coder selected the next fifth story.

1On advisement from the Human Subjects Office at my institution, informed consent was deemed

unnecessary because these were public data posted to a site that anyone could access without a pass-
word.
30 BYLUND

Coding Procedures
The author and three upper-level undergraduate communication studies students
worked as coders on this project.

Identification of decisions. Three coders worked together to develop rules

and a codebook for identifying decisions in birth stories. Coders considered only
decisions that were made between a health care provider or staff member and the
mother, father, or both and did not consider decisions that were made alone or just
with a spouse or partner or other family member or friend. Following coder train-
ing sessions, two coders established unitizing reliability on approximately 5% (n =
30) of the stories (Guetzkow’s U = .06 and 81% average agreement per encounter).
Each of the two coders subsequently identified decisions in about half of the birth
story sample, resulting in a total of 488 decisions over all the stories.

Involvement in decision making. A simple coding scheme was developed

to rate the involvement of the patient and the clinician in each identified decision.
This coding scheme had five levels, with the anchors defined as 1 (clinician’s own
decision) and 5 (patient’s own decision). Coding levels 2 and 4 were defined as
having either the patient or clinician initiate the decision, while the other partici-
pated in it, and coding category level 3 was defined to be a completely shared deci-
sion. Table 1 displays the categories of this coding scheme and examples of moth-

TABLE 1
Frequencies of Levels of Involvement in Decision Making

Level Frequency % Cumulative %

1—Clinician’s sole decision 192 39.3 39.3

“It was decided that I was going to have an emergency
c-section.”
2—Clinician initiates; patient participates 141 28.9 68.2
“My midwife told me I had two options. I chose the
Stadol.”
3—Shared decision 31 6.4 74.6
“My doctor and I had discussed pain relievers and we
had decided on a walking epidural.”
4—Patient initiates; clinician participates 106 21.7 96.3
“At the hospital, I asked the midwife to break my water.”
5—Patient’s sole decision 18 3.7 100
(This code was used when the mother seems to perceive
it as only her decision, e.g., she doesn’t say anything
about the clinician assenting.)
“I decided to take the whole hospital stay since I was
breastfeeding and wanted the nurses to help me.”
Total 488 100
 DECISION MAKING DURING THE BIRTHING PROCESS 31

ers’ statements (based on data) that fit into those categories. A codebook was
developed for this coding system, and coders attended training sessions. After
achieving reliability on approximately 20% (n = 100) of the decisions (kappa =
.80), each of the two coders rated the identified decisions using the shared deci-
sion-making coding scheme.

Sex and type. One coder read all the stories, noting whether the woman’s
primary clinician was a male doctor, a female doctor, or a midwife. To check accu-
racy, another coder did the same for about 9% (n = 50) of the stories (Cohen’s
kappa = .78).

Decision type. One coder evaluated each identified decision for its type us-
ing a simple categorical coding system. This coding scale had eight original cate-
gories. A decision categorized in a certain topic meant the decision could have any-
thing to do with that topic. For example, a decision about “pain medication”
included decisions about choosing among pain medications and choosing not to
have pain medication. To check accuracy, a second coder did the same for approxi-
mately 9% (n = 50) of the decisions (Cohen’s kappa = .93). After analysis, one cat-
egory was added into the miscellaneous category, leaving seven total categories.

Emotions. One coder read all stories, noting when 13 specific emotion words
were used. These emotion words were later subsumed into two categories: positive
emotions (happy, excited, proud, and glad) and negative emotions (anxious, sad,
disappointed, upset, frustrated, nervous, scared, terrified, and worried). Five emo-
tion word variables were computed for each birth story: number of positive emo-
tion words, number of negative emotion words, total of positive and negative emo-
tion words, ratio of positive emotion words to total emotion words, and ratio of
negative emotion words to total emotion words.

RESULTS

General Findings
Original sample. Of the 551 birth stories in the original sample, 52% con-
tained at least one decision. In this sample, there was enough information about
the clinician to code for type of clinician in 87% of the stories and to code for
clinician sex in 64% of the stories. For the study, stories with no decisions were
removed, leaving a study sample of 285 stories for analyses about involvement
in decision making. There were no significant differences between the stories
that were included in the study sample and stories that were not included based
on type of clinician, sex of clinician, or on primiparous versus multiparous birth.
32 BYLUND

The samples did differ significantly, however, on four of the five measures of
emotion words used, with the stories in the study sample having significantly
higher emotion word use (see Table 2). This is not unexpected, as stories that
would be longer and more detailed would be more likely to have decisions re-
ported as well as emotion words used.

Study sample. Of this study sample, the mean number of decisions per birth
story was 1.71 (SD = 1.08), and all but one story contained between one and five
decisions. Seventy-two percent of the stories were first birth stories. A higher per-
centage of stories in the study sample had enough information to be coded for cli-
nician type (93%) and sex (72%) than in the original sample. Of those that could be
coded, 88% of women used physicians and 12% used midwives. In addition, 56%
had male clinicians, and 44% had female clinicians. Of those who used a physician
rather than a midwife, twice as many had a male than had a female physician.
There was no significant difference between primiparous and multiparous
women’s use of a midwife, although there was a trend for multiparous women to
use midwives more often than primiparous women. Similarly, there was no signifi-
cant difference between the two types of women on sex of clinician. There was also
no difference in the number of decisions in primiparous and multiparous women’s
birth stories or in the numbers and types (positive or negative) of emotion words
that these two groups of women used.

Involvement in Decision Making

The average decision making score was 2.22 (SD = 1.27). As seen in Table 1, the
majority of decisions were coded as the clinician’s full decision, followed by clini-
cian initiates, patient participates. About 22% of the decisions were coded as pa-
tient initiates, clinician participates. Very few decisions were coded as either hav-

TABLE 2
Differences in Emotion Word Usage

Mean

Emotion Variable Nonstudy Samplea Study Sampleb t Valuec

POS Positive emotion words .38 .64 –3.39**

NEG Negative emotion words .48 .76 –3.34**
TOT Sum of POS and NEG .86 1.40 –4.16**
POS divided by TOT .37 .37 n.s.
NEG divided by TOT .41 .42 –2.43*

Note. n.s. = not significant.

an = 266. bn = 284; missing data account for difference from n = 285 study sample. cdf = 548.

*p < .05. **p < .01.

 DECISION MAKING DURING THE BIRTHING PROCESS 33

ing equal participation or as being the patient’s full decision. Table 3 shows the
types of decisions that were made. Most decisions were about pain medication
(31.6%) and induction (28.7%).

Emotions. Women used positive and negative emotion words in their birth
stories. Some examples of the women’s own words may help to illustrate how
emotion words are used in these stories. One mother expressed her joy during the
birthing process when she wrote, “I got to the hospital at 5 p.m. and after getting
into a room and checked in I was dilated to 8 centimeters! WOW! I was very ex-
cited!!” (No. 216). Another mother spoke of her fears during the birthing process,
when she felt the baby coming and told her husband to get the nurse. She wrote:
“He got the nurse and she wasn’t coming fast enough. I wanted to cry. I was scared
the baby was going to come out of me with no one there to help it” (No. 197). Posi-
tive and negative emotion words can also be located very closely in a story. One
mother wrote of her joy and fear during birthing. The nurse had told her she could
start pushing the baby out and she wrote: “I was so happy. I was about to see my
baby … The doctor pulled and pulled and he finally came out, his cord was
wrapped around his neck and he was not breathing. I was terrified. I finally heard
his cry and began crying myself. I was so happy!” (No. 205).
Concerning the mothers’ emotions, the first two hypotheses predicted a relation
between the type of affect expressed in the birth stories and the woman’s level of
involvement in decision making. This was first measured by analyzing the number
of positive emotion words and the number of negative emotion words in women’s
stories for a correlation with the woman’s involvement with decisions. Neither cor-
relation was significant. However, when the stories with no emotion words were
removed from the sample (n = 110, 38.6%), a small but significant negative corre-
lation between the number of negative emotion words and the women’s involve-
ment emerged (r = –.17; p < .05), providing support for H2 that, although signifi-
cant, has a small effect size (3%).

TABLE 3
Decision-Making Types and Scores

Type of Decision Frequency % M SD

Pain medication 154 31.6 3.26 1.11

All other decision types 334 68.4 1.73 1.03
Going to and from or staying at hospital 29 5.9 2.00 1.07
Induction 140 28.7 1.55 0.89
C-section 61 12.5 1.87 1.09
Aid in getting baby out 19 3.9 1.53 0.84
Pushing 13 2.7 1.77 0.60
Miscellaneous 72 14.8 1.92 1.25
Total 488 100 2.22 1.27
34 BYLUND

The second method of testing H1 and H2 was to examine both positive and neg-
ative emotion words as a ratio of total emotion words. For this analysis, stories
with no emotion words were removed. Pearson correlations indicated a positive
correlation between the positive emotion word ratio and involvement in decision
making (r = .127; p < .05) and between the negative emotion word ratio and in-
volvement in decision making (r = –.127; p < .05). H1 and H2 appear to have some
support, again with a small effect size (about 2%).

Birth number. A difference in the amount of involvement a mother had in de-

cision making was expected to be greater for the multiparous than the primiparous
women. The results showed the general direction expected (average scores of 2.43
and 2.24, respectively, on the involvement in decision making scale) but a t test was
not significant. Thus, H3 was not supported.

Decision type. RQ1 asked how the type of decision would influence the
women’s involvement in the decision. Descriptive statistics (Table 3) showed that
women’s involvement levels in decisions about pain medication were much higher
(M = 3.26; SD = 1.11) than in any other type of decision, grouped together (M =
1.73; SD = 1.03) or separately. A t test comparing women’s involvement levels in
decisions about pain medication and their involvement levels in other decisions re-
vealed a highly significant difference (t = 14.85, p < .0001; df = 486).

Clinician sex and type. The next two areas of interest addressed the effect
of clinician sex and type on women’s involvement in decision making. RQ2 asked
how the sex of the women’s primary clinician would affect her level of involve-
ment in decision making. A first comparison of female clinicians (physicians and
midwives) and male clinicians (physicians) resulted in a general direction of more
involvement with female clinicians (average scores of 2.44 and 2.23, respectively,
on the involvement in decision making scale), but a t test was not significant. When
examining only birth stories in which the primary clinician was a physician, there
was virtually no difference in the amount of involvement in decision making
whether the physician was a man or a woman (average score of 2.23 for both men
and women).
H4 predicted a greater amount of involvement in decisions for women with
midwives as their clinicians than for women with physicians. A t test showed
strong support for this hypothesis (t = –2.8, p < .01, df = 39.3). Mothers with mid-
wives as their clinicians had an average involvement score of 2.81 (SD = 1.09),
whereas mothers with physicians as their clinicians had an average involvement
score of 2.23 (SD = 1.14).
It was important to ensure that the findings of RQ1 did not confound the support
for H4. That is, if midwives give their patients more opportunity to make decisions
about pain medication, and patients were more involved with pain medication de-
 DECISION MAKING DURING THE BIRTHING PROCESS 35

cisions anyway, then the finding that patients of midwives were more involved in
decision making could be misleading. Further analysis of the type of decisions in
stories revealed that there were fewer decisions about pain medication reported in
stories with a midwife clinician (25.4%) than in stories with a physician (32.3%),
although this was not a significant difference. Therefore, the finding that women
with midwives are more involved in decisions appears to be independent of the
type of decisions that are made.

DISCUSSION

Women’s birth stories are rife with examples of communication and medical deci-
sion making. According to the analyses of these birth stories, experiences with de-
cision making during birthing were not remarkably different for primiparous and
multiparous women. For both groups, more than two thirds of the decisions were
either the clinician’s sole decision or were initiated by the clinician. The fact that
less than 4% of decisions were made only by the patient is not surprising. In fact,
making the coding system a continuum anchored at one end by the patient as sole
decision maker may be an unusual way to measure involvement in decision mak-
ing or shared decision making. Others have looked at the joint or shared decision as
the anchor (e.g., VandeVusse, 1999). What is surprising in this study is that only
6% of decisions were seen as having equal participation between clinician and pa-
tient. This may be due to the fact that most decisions, although some may see them
as equal, actually are initiated by the clinician or the patient (as in coding catego-
ries 2 and 4). Perhaps a better measure of actual shared decision making for this
study, then, is the 57% of decisions that fit in coding categories 2 to 4, as these deci-
sions all included some sort of patient involvement. Using such a definition, it is
fair to say that, overall, women were involved in a majority of the decisions re-
ported here.
What predicts a woman’s involvement in decision making? Although previous
research has found multiparous women to be more involved in decision making
than primiparous women, that difference was not supported in this study. In addi-
tion, the sex of the clinician was shown to have no significant effect on the
mother’s level of involvement. In fact, when looking only at physicians, stories
with male and female physicians had almost identical levels of mother involve-
ment in decision making. On the one hand, this finding contradicts previous re-
search that has shown female clinicians generally to be more partnership oriented
in their communication with their patients (Roter et al., 2002). Perhaps, as Roter et
al. (1999) noted, the pressure male obstetricians feel to compete with a growing
popularity for female obstetric and gynecological physicians has required them to
improve their interpersonal skills. On the other hand, this study’s finding on clini-
cian sex also contradicts communication research specific to the obstetrician–
36 BYLUND

patient interaction that showed male physicians to use more partnership-building

behaviors (Roter et al., 1999). Perhaps the context of that study, measuring com-
munication in an office setting, was different enough from this study, which exam-
ined communication during the entire birthing process, that the pendulum did not
swing the other way.
One problem with coding a primary clinician in the story was that the coding
did not always capture which clinician was working with the woman to make the
decision. (e.g., a mother’s primary clinician could have been a midwife, but a deci-
sion may be made between the mother and the midwife’s backup obstetrician).
This ambiguity, and the difficulty of ascertaining whether physicians were men or
women in 28% of the stories, may or may not have contributed to the lack of sup-
port for this hypothesis. The presence of a midwife as a primary clinician, how-
ever, was a significant predictor of a woman’s involvement in decision making.
Causality may be difficult to parcel out in this relationship—that is, women who
want to be involved in decision making may be more likely to choose midwives,
who in turn, are more likely to offer more involvement in decision making. Not
coding which clinician was making the decision might not matter as much with
women who use midwives, as they may be more likely to be involved in decision
making regardless.
The biggest predictor of a woman’s involvement in decision making, however,
was the type of decision. The mother’s involvement was more likely when the de-
cision was about pain medication than about anything else in the birthing process.
Decisions about pain medication usually do not affect the birth of the baby,
whereas the other types of decisions do. Perhaps mothers are so involved in the de-
cisions about pain medication decisions because those are about the mother’s com-
fort, rather than the mother’s or baby’s health. In other decisions, the clinician’s
medical knowledge is presumably greater than the mother’s, which may affect the
involvement in the decision. This finding highlights the need for research on
shared decision making in the medical encounter to examine the effects of context
on these decisions (Bylund & Imes, in press).
Adding to other findings on the greater satisfaction and more positive emotions
related to patient involvement in decision making during birth (Berg et al., 2001;
Brown & Lumley, 1994; Green et al., 1990; Simkin, 1992), this study found that a
women’s involvement in decision making was related positively to the proportion
of positive emotions a woman named, as well as negatively related to both the
number and proportion of negative emotions a woman named, in stories that had
emotions named. Although small in magnitude, these findings may reinforce the
importance of patient involvement in medical decision making. However, these re-
lations are correlational in nature and it could be argued that a woman’s positive or
negative outlook on the birthing experience may influence how involved she is in
the process, especially because emotion words were used continually throughout
the woman’s constructed birth narrative and cannot be measured simply as an out-
 DECISION MAKING DURING THE BIRTHING PROCESS 37

come. Undoubtedly, the missing link in this area of research is the connection be-
tween the patient’s involvement in decision making and positive health outcomes
for the baby or mother. Although one can draw on evidence of good clinician–pa-
tient communication as being related to positive patient outcomes in general to
support such a claim, published research has yet to demonstrate such a relation
between birthing and decision making.
Although the use of publicly posted birth stories provided a large sample for
this study, the choice of sample also limits the findings of the study. First, there was
no demographic information available about these women. In addition, using the
Web site’s designation of “first” and “second or more” birth stories may also be
problematic, as stories seemed to be occasionally placed in the wrong category by
the Webmaster, although the coders corrected for some instances of that error.
Third, aside from counting the emotions named in the women’s birth stories, there
were no outcome data about the births, the baby’s or the mother’s health, or the
mother’s satisfaction with the birth (Bramadat & Driedger, 1993). Finally, general-
izing this sample to birthing women in general is risky. Women who post their birth
stories to the Internet are likely to be of a higher level of education and income than
those who do not use the Internet (Robinson, 2001). In addition, they may be prone
to be more active in their health care, which may affect their involvement level in
medical decision making.
Future research might reap the benefits of the large amounts of birth story data
on the Internet by contacting the women who post their e-mail addresses and ask-
ing them to complete demographic, outcome, and other self-report information, af-
ter gaining the approval of the human participants board and, in some cases, the
Web site owner or discussion board moderator. This may facilitate investigation of
the link between involvement in the decision-making process in birthing and im-
portant outcomes of the birthing process. Examining differences between unsolic-
ited Internet narratives and other methods of narrative gathering would be fruitful
in understanding more about this data source. Future studies on birthing and com-
munication, in general, might also focus specifically on one type of decision made
in the birthing process, such as cesarean section, recognizing the importance of
context on medical decision making.
This study is an important step in understanding more about the understudied
clinician–patient communication in the context of birthing. The relationship be-
tween a woman and her clinician(s) during pregnancy and the birthing process is
distinctive from other clinician–patient relationships for at least two important rea-
sons. First, the relationship between a pregnant or birthing woman and her clini-
cian(s) does not exist because of a patient’s ill health. Granted, patients do visit
physicians for routine and preventive care when they are healthy, but primarily, pa-
tients see physicians when they have problems, whereas that is not the primary rea-
son for the pregnant or birthing woman’s relationship with the clinician. Second,
the relationship involves a third individual (the baby) in a way that other clini-
38 BYLUND

cian–patient relationships do not. These factors make this context distinctive, and
health communication scholars have much to add to the understanding of this
unique medical relationship.

ACKNOWLEDGMENTS

Carma Bylund is now at the Department of Psychiatry & Behavioral Sciences, Me-
morial Sloan Kettering Cancer Center, New York.
The author thanks Anna Artrip, Joe Kiehn, and Lesley Christensen for their
work as coders on this project, and Greg Lincoln for his help with data analysis. An
earlier version of this article was presented at the 2003 Western States Communi-
cation Association Convention.

REFERENCES

Adams, R., Smith, B. J., & Ruffin, R. E. (2001). Impact of the physician’s participatory style in asthma
outcomes and patient satisfaction. Annals of Allergy, Asthma & Immunology, 86, 263–271.
American College of Nurse-Midwives. (n.d.). Thinking of using a nurse-midwife? Retrieved May 8,
2005, from http://www.midwife.org/focus
Berg, A., Yuval, D., Ivancovsky, M., Zalcberg, S., Dubani, A., & Benbassat, J. (2001). Patient percep-
tion of involvement in medical care during labor and delivery. Israel Medical Association Journal, 3,
352–356.
Blais, R., Lambert, J., Maheux, B., Loiselle, J., Gauthier, N., & Famarin, A. (1994). Controversies in
maternity care: Where do physicians, nurses, and midwives stand? Birth, 21, 63–70.
Braddock, C. H., Fihn, S. D., Levinson, W., Jonsen, A. R., & Pearlman, R. A. (1997). How doctors and
patients discuss routine clinical decisions: Informed decision making in the outpatient setting. Jour-
nal of General Internal Medicine, 12, 339–345.
Bramadat, I. J., & Driedger, M. (1993). Satisfaction with childbirth: Theories and methods of measure-
ment. Birth, 20, 22–29.
Brown, S., & Lumley, J. (1994). Satisfaction with care in labor and birth: A survey of 790 Australian
women. Birth, 21, 4–13.
Bylund, C. L., & Imes, R. S. (2005). Communication and medical decision making in context:
Choosing between reasonable options. In E. B. Ray (Ed.), Case studies in health communication
(2nd ed., 69–80). Mahwah, NJ: Lawrence Erlbaum Associates, Inc.
Charles, C., Gafni, A., & Whelan, T. (1997). Shared decision-making in the medical encounter: What
does it mean? (Or it takes at least two to tango). Social Science and Medicine, 44, 681–692.
DiMatteo, M. R., Kahn, K. L., & Berry, S. H. (1993). Narratives of birth and the postpartum: Analysis
of the focus group responses of new mothers. Birth, 20, 204–211.
Drew, N. C., Salmon, P., & Webb, L. (1989). Mothers’, midwives’ and obstetricians’ views on the fea-
tures of obstetric care which influence satisfaction with childbirth. British Journal of Obstetrics and
Gynaecology, 96, 1084–1088.
England, S. L, & Evans, J. (1992). Patients’ choices and perceptions after an invitation to participate in
treatment decisions. Social Science & Medicine, 34, 1217–1225.
Fisher, W. R. (1987). Human communication as narration: Toward a philosophy of reason, value and
action. Columbia: University of South Carolina Press.
 DECISION MAKING DURING THE BIRTHING PROCESS 39

Gattellari, M., Butow, P. N., & Tattersall, M. H. (2001). Sharing decisions in cancer care. Social Science
and Medicine, 52, 1865–1878.
Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA:
Harvard University Press.
Green, J. M., Coupland, V. A., & Kitzinger, J. V. (1990). Expectations, experiences, and psychological
outcomes of childbirth: A prospective study of 825 women. Birth, 17, 15–24.
Greenfield, S., Kaplan, S., & Ware, J. E. (1985). Expanding patient involvement in care. Annals of In-
ternal Medicine, 102, 520–528.
Greenfield, S., Kaplan, S., Ware, J. E., Yano, E. M., & Frank, H. J. L. (1988). Patients’ participation in
medical care: Effects on blood sugar control and quality of life in diabetes. Journal of General Inter-
nal Medicine, 3, 448–457.
Guadagnoli, E., & Ward, P. (1998). Patient participation in decision-making. Social Science and Medi-
cine, 47, 329–339.
Kaplan, S. H., Greenfield, S., Gandek, B., Rogers, W. H., & Ware, J. E. (1996). Characteristics of physi-
cians with participatory decision-making styles. Annals of Internal Medicine, 124, 497–504.
Lazarus, E. (1997). What do women want? Issues of choice, control, and class in American pregnancy
and childbirth. In R. E. Davis-Floyd & C. F. Sargent (Eds.), Childbirth and authoritative knowledge:
Cross cultural perspectives (pp. 132–158). Berkeley: University of California Press.
Makoul, G., Arntson, P., & Schoefield, T. (1995). Health promotion in primary care: Physician-patient
communication and decision making about prescription medicines. Social Science & Medicine, 41,
1241–1254.
Ochs, E., & Capps, L. (1996). Narrating the self. Annual Review of Anthropology, 25, 19–43.
Page, L. (1995). Change and power in midwifery. Birth, 22, 227–231.
Page, R. E. (2002). Evaluating in childbirth narratives told by women and men. Discourse Studies, 4,
99–116.
Pollock, D. (1999). Telling bodies: Performing birth. New York: Columbia University Press.
Reese, E. (1996). Conceptions of self in mother-child birth stories. Journal of Narrative and Life His-
tory, 6, 23–38.
Robinson, K. M. (2001). Unsolicited narratives from the Internet: A rich source of qualitative data.
Qualitative Health Research, 11, 706–714.
Roter, D. L., Geller, G., Bernhardt, B. A., Larson, S. M., & Doksum, T. (1999). Effects of obstetrician
gender on communication and patient satisfaction. Obstetrics and Gynecology, 93, 635–641.
Roter, D. L., Hall, J. A., & Aoki, Y. (2002). Physician gender effects in medical communication: A
meta-analytic review. Journal of the American Medical Association, 288, 756–764.
Simkin, P. (1991). Just another day in a woman’s life? Part 1: Women’s long-term perceptions of their
first birth experience. Birth, 18, 203–210.
Simkin, P. (1992). Just another day in a woman’s life? Part II: Nature and consistency of women’s
long-term memories of their first birth experience. Birth, 19, 64–81.
Soparkar, A. A. (1998). The telling of childbirth stories. Unpublished doctoral dissertation, University
of Massachusetts.
Stewart, M. A., Brown, J. B., Boon, H., Galajda, B. A., Meredith, L., & Sangster, M. (1999). Evidence
on doctor-patient communication. Cancer Prevention & Control, 3, 25–30.
Street, R. L., & Millay, B. (2001). Analyzing patient participation in medical encounters. Health Com-
munication, 13, 61–73.
VandeVusse, L. (1999). Decision-making in analyses of women’s birth stories. Birth, 26, 43–50.
Waldenstrom, U., Borg, I., Olsson, B., Skold, M., & Wall, S. (1996). The childbirth experience: A study
of 295 new mothers. Birth, 23, 144–153.
Zadoroznyj, M. (1999). Social class, social selves and social control in childbirth. Sociology of Health
and Illness, 21, 267–289.