University of Southern California Dissertations and Theses

An empirical evaluation of breast cancer treatment decision making

(USC Thesis Other)

An empirical evaluation of breast cancer treatment decision making

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

AN EMPIRICAL EVALUATION OF BREAST CANCER
TREATMENT DECISION MAKING
by
Carol E. Brown

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

August 2009

Copyright 2009 Carol E. Brown
ii
Acknowledgments
Many physicians provided me with assistance over the course of this research and
I am grateful to each of them for their contributions. In particular, Gail Lebovic, M.D.,
Christy Russell, M.D., and Kristin Skinner, M.D. assisted me with the development of the
breast cancer knowledge measure. Cathie Chung, M.D. and Dennis Holmes, M.D.
provided invaluable feedback that helped me turn the decision quality measure into a rich
and informative instrument. I would also like to express my sincerest appreciation to the
many participants who took the time to complete the instruments and provided supportive
feedback about their perceived importance of this research. It is heartwarming to know
that the potential beneficiaries of this line of research appreciate its value.
Gratefully, I acknowledge Brian Monroe who provided much needed technical
assistance with the development of the website that was used for the data collection. I
would like to thank the members of my committee, Richard John, Ph.D., Stephen Read,
Ph.D., and Jean Richardson, Ph.D., all of whom provided constructive feedback that led
to improvements in the final version of this dissertation. In particular, I am grateful for all
of the hard work that my advisor, David Walsh, Ph.D., contributed to this research. He
spent countless hours advising me and helping me develop not only this research, but also
the skills I need to succeed as a researcher, mostly the ability to think of the large,
important questions. Recognition is also due to John Horn, Ph.D., who provided feedback
and assistance in the early development of this research. Though he is no longer with us,
his enthusiasm for learning will always stay with me.
I am very grateful to my family for their love and support over the years. My
parents always supported my love of learning and made my education a top priority; I
iii
would not be where I am today without them. I only wish my granddad were here today
so I could share this accomplishment with him; I hope he would have been proud of this
work. Most importantly, I could not have completed this long process without Mike’s
love and encouragement. He kept me sane and motivated and would not let me give up
on this project or on myself. I would not be where I am today without him.
Finally, thank you, Bob Loblaw.
iv
Table of Contents
Acknowledgments............................................................................................................... ii
List of Tables ..................................................................................................................... vi
Abstract............................................................................................................................. vii
Chapter 1: Introduction....................................................................................................... 1
Background..................................................................................................................... 3
Desire for Participation ............................................................................................... 4
Desire for Information .............................................................................................. 11
Treatment Decisions ................................................................................................. 14
Information and Knowledge ..................................................................................... 17
Conclusion ................................................................................................................ 21
Statement of Research Purpose..................................................................................... 21
Chapter 2: Method ............................................................................................................ 23
Participants.................................................................................................................... 23
Materials ....................................................................................................................... 26
Demographic Measure .............................................................................................. 26
Breast Cancer Knowledge Measure.......................................................................... 28
Breast Cancer Decision Quality Measure ................................................................. 32
Treatment Selection .............................................................................................. 33
Reasoned Statements ............................................................................................ 35
Information Use .................................................................................................... 36
Participation Level................................................................................................ 37
Procedure ...................................................................................................................... 37
Data Analysis................................................................................................................ 39
Incomplete Data ........................................................................................................ 39
Chapter 3: Results............................................................................................................. 40
Mean Differences in Knowledge and Decision Quality ............................................... 40
Exploratory Factor Analysis of Information Use Scales .............................................. 42
Predicting Decision Quality.......................................................................................... 46
Surgical Treatment Deviation ................................................................................... 52
Surgical Rationale Deviation .................................................................................... 53
Adjuvant Treatment Deviation.................................................................................. 54
Adjuvant Rationale Deviation................................................................................... 55
Separate Regression Analyses for Patients and Non-Patients .................................. 56
Participation Level........................................................................................................ 57
Chapter 4: Discussion ....................................................................................................... 59
Mean Differences in Knowledge and Decision Quality ............................................... 59
Predicting Decision Quality.......................................................................................... 62
Participation Level........................................................................................................ 68
v
Importance of the Research .......................................................................................... 70
Limitations of the Research .......................................................................................... 72
Conclusions................................................................................................................... 74
References......................................................................................................................... 76
Appendix A....................................................................................................................... 82
Appendix B ....................................................................................................................... 84
Appendix C ....................................................................................................................... 89
vi
List of Tables

Table 1: Participant Characteristics .............................................................................. 24
Table 2: Patient Characteristics..................................................................................... 25
Table 3: Other Experiences with Breast Cancer ........................................................... 27
Table 4: Item Analysis of Knowledge Measure............................................................ 30
Table 5: Mean Differences between Patients and Non-Patients on Breast Cancer
Knowledge and Decision Quality Scores ....................................................... 41
Table 6: Factor Structure of Surgical Information Use Items....................................... 43
Table 7: Mean Differences between Patients and Non-Patients on Surgical
Information Factor Scores............................................................................... 45
Table 8: Factor Structure of Adjuvant Information Use Items ..................................... 47
Table 9: Mean Differences between Patients and Non-Patients on Adjuvant
Information Factor Scores............................................................................... 49
Table 10: Correlations between Surgical Outcome Variables and Predictor
Variables ......................................................................................................... 50
Table 11: Correlations between Adjuvant Outcome Variables and Predictor
Variables ......................................................................................................... 51
Table 12: Summary of Simultaneous Regression Analyses for Variables Predicting
Surgical Treatment Decision Quality (N = 142)............................................. 53
Table 13: Summary of Simultaneous Regression Analyses for Variables Predicting
Surgical Assessments of Reasoning (N = 142)............................................... 54
Table 14: Summary of Simultaneous Regression Analyses for Variables Predicting
Adjuvant Treatment Decision Quality (N = 142) ........................................... 55
Table 15: Summary of Simultaneous Regression Analyses for Variables Predicting
Adjuvant Assessments of Reasoning (N = 142) ............................................. 56

vii
Abstract
Research in the area of breast cancer treatment decision making has typically focused
separately on two issues: 1) what patients know about breast cancer and 2) what surgical
treatments they undergo. The current research seeks to examine the link between
knowledge and treatment decision making. To this end, two instruments were developed.
The first instrument was a measure of breast cancer knowledge and the second was a
measure of treatment decision making ability that assessed decision quality and causal
reasoning skills. Participants were 142 women (53 breast cancer patients, 89 non-
patients). Patients performed significantly better than non-patients on the knowledge
instrument and each of 4 decision quality outcomes. Results from regression analyses
showed multiple predictors of decision quality, including breast cancer knowledge,
patient status, age, and various information use factor scores. Results from this research
provide a clearer picture of what participants know about breast cancer, the quality of
their decisions, the importance of information and how information use is related to the
decisions that participants make. In the future, these instruments can be used to determine
patients’ readiness to make their own breast cancer treatment decisions. Additionally, the
knowledge gained from this research can be used to design educational interventions that
teach patients what they need to know to make informed treatment decisions.
1
Chapter 1: Introduction
Breast cancer patients and advocates have fought for many years to assure that
patients are informed about, and involved in, their treatment decisions. One reason for
this effort is that breast cancer often has multiple treatment options that result in similar
survival outcomes. For example, a mastectomy has nearly identical survival rates as a
lumpectomy with radiation (Fisher et al., 2002).
1
However, these treatments can vary
substantially in the invasiveness of the procedure, the side effects, and the time involved
in seeking treatment. When there are multiple treatments that lead to similar survival
outcomes, it is often the goals and values of the informed patient that determine which
treatments are most appropriate. In the current medical system, breast cancer patients are
typically expected to take an active involvement in their treatment decisions. The
reasoning is that patients who are actively involved will make treatment choices in line
with their goals and values, thus leading to greater satisfaction with the treatment process.
However, it must first be determined if patients are willing and able to take on this active
involvement in such a complex decision making process.
Over the past two decades, considerable research has been conducted to determine
patients’ desire for involvement in various aspects of treatment, such as decisional
control and desire for information. Research in the area of decisional control has mixed
results, with patients’ desires ranging from active roles in which the patient retains
complete control and final selection of treatment to passive roles where the physician is

1
A modified radical mastectomy is a surgical procedure where the affected breast, some lymph nodes from
under the arm, and the lining over the chest muscles are removed. A simple mastectomy (also called a total
mastectomy) is a surgical procedure to remove the affected breast. In contrast, breast conserving treatments
remove the cancer, but not the whole breast. A lumpectomy is the most common type of breast conserving
treatment. It consists of removing the tumor and a small amount of normal tissue around it (National
Institutes of Health, 2000).
2
completely in charge. Regardless of patients’ decision making preferences, researchers
generally concluded that most breast cancer patients wanted large amounts of information
about their condition and its treatment. However, little research has examined whether
patients understand this information and are prepared to participate in their own treatment
decisions. Research that examined patient preparedness through the use of knowledge
measures and decision aids reported that there was room for improvement on the part of
patients. This finding is noteworthy. If patients are to be involved in making important
and complex treatment decisions, it is imperative that they understand the information
they receive and the decisions they face.
One way to improve patients’ abilities to make informed treatment decisions is to
develop educational interventions to prepare them for these decisions. However, reliable
and valid instruments measuring patients’ knowledge about breast cancer and their ability
to make wise treatment decisions should first be developed. These instruments are needed
to evaluate the effectiveness of future educational interventions, as well as to determine
patients’ present fitness for making their own treatment decisions.
Two instruments were designed and evaluated as part of the current research. The
first instrument was the refinement of a previously developed measure of breast cancer
knowledge and the second was a new measure of treatment decision-making that utilized
hypothetical scenarios of patients seeking treatment for breast cancer. Data was collected
from two groups of participants, breast cancer patients and a non-patient sample. It was
hypothesized that patients would outperform the non-patient sample on 1) the breast
cancer knowledge instrument, 2) the treatment decision scale (for both surgical and
adjuvant treatment decisions) and 3) the support for explanations of why treatments were
3
or were not appropriate. It was further hypothesized that greater knowledge about breast
cancer would be predictive of higher quality treatment decisions, as would patient status,
age and information use. The research should result in measures that can be used in the
future to assess both patient readiness for decision making and the effectiveness of
educational materials designed to improve knowledge and understanding of breast cancer
treatment decisions.
BACKGROUND
Physicians recognize that there are many reasons to include patients in their
treatment decisions. What comprises the most appropriate treatment is often a subjective
decision, and a great amount of uncertainty exists concerning the effectiveness of certain
treatments and the medical outcomes for any given patient (Hughes, 1993; Yates &
Patalano, 1999). Knowledge of a patient’s history, feelings, thoughts, and goals are
important aspects in choosing the correct treatment (Bartholome, 1992; Broadstock &
Michie, 2000; Charles, Gafni, & Whelan, 1999). Additionally, physicians do not always
make objective decisions based solely on medical factors (Auerbach, 2000; Bartholome,
1992; Chagpar et al., 2006; Keleman III, Poulton, Swartz, & Jatoi, 2001). Factors such as
special training, an inability to adopt new views, and geographic location affect the
decisions that physicians make. Even when the patient ultimately chooses her own
treatment, the physician has a great influence over the choices that are made. The
influence of the physician’s opinion was reported by some researchers to be four times
greater than even the patient’s preference (Sepucha, Ozanne, Silvia, Partridge, Mulley Jr.,
2007). Other researchers reported that when offered a physician’s recommendation about
treatment, over 90% of breast cancer patients followed the recommendation (Johnson,
4
Roberts, Cox, Reintgen, Levine, & Parsons, 1996; Katz, Lantz, & Zemencuk, 2001;
Kotwall, Maxwell, Covington, Churchill, Smith, & Covan, 1996).
In the treatment of early stage breast cancer, often the most important factor in
determining what treatment to select is the personal preferences of the patient, which vary
depending on the individual’s life circumstances and personal goals regarding treatment
outcomes (Barry, 1999). Some patients are more concerned with their physical
appearance, while others are concerned with the amount of time they will have to be
away from work, and still others are concerned about minimizing the chance for
recurrence. Because patients differ in regard to their preferences for treatment outcomes,
it is beneficial for patients to be involved, for doctors cannot be expected to know and
represent each patient’s preference. However, it is not only a patient’s preference that is
important. A patient’s preference needs to be based on facts, not misconceptions or
incomplete information. If the goal is to have patients take on a more participatory role in
their treatment decision making, then it is important to understand what level of
involvement patients wish to have and whether they have adequate knowledge and
understanding of breast cancer treatment to assume their desired role.
Desire for Participation
Mixed results exist about the amount of participation that breast cancer patients
wanted in the medical decision making process. Some researchers found that patients
wanted to be actively involved, including taking responsibility for the treatment decisions
that were made (Mastaglia & Kristjanson, 2001; Ward, Heidrich, and Wolberg, 1989).
However, the more consistent findings supported the claim that most patients desired a
passive, or at most, a collaborative role in the treatment decision making process (Beaver,
5
Luker, Owens, Leinster, & Degner, 1996; Bilodeau & Degner, 1996; Bruera, Willey,
Palmer & Rosales, 2002; Degner et al., 1997; Degner & Sloan, 1992; Hack, Degner &
Dyck, 1994). Different methods were used to elicit patients’ preferences and these
methods may have influenced the results.
Asking patients to retrospectively report their desired involvement was one
method for assessing their desire to participate in treatment decisions. When preferences
were elicited in this manner, the results suggested that patients desired a more active
participatory role in the decision making process. In one study, where patients rated a
single statement, 50% of patients wanted the treatment decision to be fully her own and
the other 50% indicated they would like to share the decision with someone else, most
frequently the physician or a family member (Ward et al., 1989). None of the patients
indicated that they believed their treatment decisions should be made without their
involvement. Johnson and colleagues (1996) also reported that a majority of breast cancer
patients (80%) wanted a participatory role in their treatment decisions, though the
researchers did not distinguish between active or collaborative roles.
Mastaglia & Kristjanson (2001) found similar results, with 54% of breast cancer
patients indicating that they wanted their surgical decision left entirely up to them, while
36% of participants wanted someone else involved in the decision. Only 8% of patients
indicated that they did not want to be involved in the decision making process. When
distinguishing between active and collaborative roles, other researchers found that
approximately one-fourth of breast cancer patients preferred an active role, two-thirds
preferred a collaborative role, and the remaining patients (about 10%) preferred a passive
role (Bruera et al., 2002; Keating, Guadagnoli, Landrum, Borbas, & Weeks, 2002).
6
Reports such as these, utilizing retrospective techniques to gauge patient desire for
involvement in the decision making process, indicated that overall, patients desired some
level of participation in their treatment. Patients wanted to make their own treatment
decisions, or at the very least, to have a voice in the process, working in collaboration
with their physician. However, due to differences in the questions and possible responses,
it is difficult to make comparisons across studies.
Another common method of examining patient preferences for involvement in the
treatment decision making process used a card sort task (Beaver et al., 1996; Bilodeau &
Degner, 1996; Degner & Sloan, 1992; Degner et al., 1997; Hack et al., 1994). Degner and
Sloan developed the Degner Control Preference Scale to ascertain patients’ preferences
for involvement in their medical decisions. The card sort task consisted of five cards with
treatment involvement preferences written on them. Statements ranged from the patient
taking full control of the decision to the physician having full control of the treatment
decision. The five statements for the card sort technique were:
A) I prefer to make the final selection about which treatment I will receive
B) I prefer to make the final selection of my treatment after seriously
considering my doctor’s opinion
C) I prefer that my doctor and I share responsibility for deciding which
treatment is best for me
D) I prefer that my doctor makes the final decision about which treatment
will be used, but seriously considers my opinion
E) I prefer to leave all decisions regarding my treatment to my doctor
7
Participants were presented the cards two at a time and selected the statement from
each pair that best reflected their desired role in the treatment decision making process.
Participants looked at all possible combinations of cards. This process allowed
participants to make small mistakes while still reaching a correct reflection of their
attitudes towards decisional control. Researchers believed the card sort process more
accurately represented the true preferences of participants (Degner & Sloan, 1992). Based
on the card sort technique, participants were placed in one of three categories of
treatment decision making preferences: active, collaborative, or passive.
When breast cancer patients’ preferences for participation in their treatment
decisions were elicited using the card sorting technique, most cancer patients reported
they desired a passive role in their treatment. Over 50% of patients reported a preference
for a passive role, approximately 30% wanted a collaborative role, and the remaining
patients (12-20%) indicated a desire for an active role (Beaver et al., 1996; Degner &
Sloan, 1992). Hack and colleagues (1994) also used to card sort technique with a small
sample of breast cancer patients and found a majority preferred a collaborative role
(57%). Desire for active and passive roles were split, with 23% preferring an active role
and 20% preferring a passive role.
However, when non-patients or women with a benign breast disease (versus
breast cancer) performed the same card sorting task, preferences were more in line with
the retrospective reports (Beaver et al., 1996; Degner & Sloan, 1992). Preferences shifted
from a majority desiring passive roles to a majority desiring collaborative or active roles.
Further, 64% of the non-patient sample preferred the active role, while 45% patients with
the benign breast disease preferred the collaborative role. This was perhaps an indication
8
that there was something about receiving a breast cancer diagnosis that led patients to
want a less active role in the treatment decision making process, though more research is
necessary to confirm this.
A further consideration was the degree to which patients achieved their desired
roles in the decision making process. The match between desired and actual roles was
important because congruence between preferences for involvement and actual
involvement was an indicator of satisfaction with treatment (Keating et al., 2002; Lantz et
al., 2005). It was not important whether patients were asked to take a less active role than
desired or a more active role than desired. Both groups exhibited lower satisfaction with
their surgical treatment and the treatment process in general, as well as more decisional
regret, as compared to patients who participated to the level they desired.
There was mixed evidence for the degree to which patients were able to achieve
their desired roles in the treatment decision making process. Some researchers reported
levels of congruence over 80% (Mastaglia & Kristjanson, 2001; Petrisek, Laliberte,
Allen, & Mor, 1997; Ward et al., 1989). Hawley and colleagues (2007) reported slightly
less congruence, with two-thirds of their sample achieving their desired level of
decisional involvement. Still, other researchers found agreement between desired and
actual participation roles to be less than 50% (Bilodeau & Degner, 1996; Degner et al.,
1997; Keating et al., 2002). For those patients who did not achieve their desired level of
participation, two results were common. The first result was that a majority of patients
had a more passive role than desired (Bilodeau & Degner, 1996; Degner et al., 1997).
Other researchers reported a fairly even split between more and less participation than
patients desired (Hawley et al., 2007; Keating et al., 2002).
9
There are several reasons why preferred and achieved roles did not always
coincide. First, patients may have initially preferred an active role, but then discovered
that they had actual or perceived deficits in understanding the information and decisions
presented, prohibiting them from participating to the degree to which they desired
(Charles, Gafni, & Whelan, 1997). Most often, patients did not feel adequately informed
to make their treatment decisions, particularly the complex decisions that accompany a
breast cancer diagnosis (Sutherland, Llewellyn-Thomas, Lockwood, Tritchler, & Till,
1989). Second, patients need physicians to be willing to allow them to actively (or
passively) participate. Poor agreement between physicians and patients regarding the
appropriate level of control for patients has been reported (Bruera et al., 2002; Strull, Lo
& Charles, 1984). Even patient characteristics, such as age and level of education
influenced the realization of decision making preferences (Charles, et al.). Researchers
found that older participants routinely said that patients were not qualified to make
complex medical decisions (Cassileth, Zupkis, Sutton-Smith & March, 1980). It is also
important to note that patients who did not participate at all had lower levels of
satisfaction with their surgical treatments and the decision making process (Lantz
et al., 2005).
Some relationships were established between patient characteristics and the
amount of involvement patients actually assumed in the treatment decision making
process, as well as the amount of involvement that they preferred. The most common
finding was that age was related to both preferred and actual involvement. Younger
patients took more active roles and older patients assumed more passive roles (Bilodeau
& Degner, 1996). This was possibly evidence for the cohort effects mentioned by Charles
10
and colleagues (1997). Older breast cancer patients were more likely to report that there
was not a decision to be made or that treatment decisions were made for them, indicating
they did not assume an active role in the process (Meyer, Russo & Talbot, 1995; Pierce,
1993). When looking at the roles that patients preferred, as opposed to the roles they
actually achieved, age was again a significant predictor of preferred levels of control in
the decision making process. Older women were significantly more likely to prefer a
passive role than were their younger counterparts (Beaver et al., 1996; Bruera et al.,
2002; Degner & Sloan, 1992; Degner et al., 1997; Petrisek et al., 1997). A second
demographic variable related to involvement was education; there was evidence that
women who had more education preferred a more active role in the decision making
process (Beaver et al.; Degner et al.; Meyer et al.).
Though patients were expected to take active roles in their treatment decisions,
the most consistent finding was that the active role was the least preferred role for
patients. Instead, patients preferred a passive, or at most, a collaborative role. While there
were likely many reasons that patients did not desire to be in charge of their treatment
decisions, perhaps the greatest barrier was access to adequate information. Researchers
speculated that patients with more knowledge and better decision making abilities would
prefer more active roles, while those who knew less and felt inadequate in their decision
making skills may have preferred passive roles (Beaver et al., 1996; Sutherland et al.,
1989). However, no research has attempted to measure patient preparedness or ability to
make treatment decisions. One goal of the current research is to provide the instruments
that are necessary to address patients’ treatment decision making ability.
11
Desire for Information
Researchers hypothesized that patients who preferred active roles would desire more
information about their disease and its treatment. Hack and colleagues (1994) found some
support for this hypothesis; patients classified as active were more likely to prefer
detailed information, including all treatment options and risks associated with each.
However, Hack and colleagues also reported that all patients wanted all available
information about the possible side effects, regardless of their desire to participate in the
decision making process. This second finding was more typical of the preferences
reported in the literature; most patients wanted as much information as possible,
regardless of their desire to be involved in their treatment decisions (Bilodeau & Degner,
1996; Luker, Beaver, Leinster, Owens, Degner, & Sloan, 1995; Petrisek et al., 1997).
Less is known about what information patients believed was the most important to
have when making treatment decisions. Many researchers used a list of nine information
needs and asked breast cancer patients to make paired comparisons to reach a final
ranking of what patients believe were the most important information needs (Bilodeau &
Degner, 1996; Degner et al., 1997; Luker et al., 1995; Luker, Beaver, Leinster, & Owens,
1996). Information categories were presented two at a time and participants indicated
which information item they would want first. The following are the nine categories of
informational needs that were identified most often in the literature:
1. Advancement of disease
2. Likelihood of cure
3. How treatment will affect ability to perform social activities
4. Impact of disease on family and friends
12
5. Issues related to caring for one’s self at home
6. How sexuality will be affected
7. Information about various treatments, including advantages and
disadvantages of each
8. Risk to family members of developing breast cancer
9. Information about side effects.
The results from this line of research were fairly consistent. Likelihood of cure,
disease stage, and treatment options were the three most important categories of
information for most women (Bilodeau & Degner, 1996; Degner et al., 1997; Luker et al.,
1995; Luker et al., 1996). It is important to note that a large amount of the information
was rated as very important by most breast cancer patients. Further, the desire for
information was present regardless of the patient’s desire to actually make treatment
decisions. Other research conducted by Petrisek and colleagues (1997) identified 15
information items important to breast cancer patients. The three most important pieces of
information were a doctor’s recommendation, making sure the cancer does not return,
and family opinion.
Some notable findings were that sexuality appeared at the bottom of the list of
information needs across studies (Bilodeau & Degner, 1996; Katz et al., 2001; Luker et
al., 1995; Luker et al., 1996), though sexuality and physical attractiveness were shown to
be significantly more important to younger women (Degner et al., 1997; Luker et al.;
Petrisek et al., 1997). Another interesting finding was that in a follow-up study where
patients were a mean of 21 months from diagnosis, risk to family members replaced
treatment options as the more important topic, possibly reflecting the changing
13
information needs of patients at different point in the treatment process (Luker et
al., 1996). This distinction concerning the changing information needs of breast cancer
patients was also addressed by Mallinger, Griggs, and Shields (2004) who found that
patients were significantly more satisfied with treatment information than with
survivorship information.
When patients selected information using an interactive CD-Rom decision aid,
they selected a mean of 21 pieces of information (Molenaar et al., 2007). The most
commonly selected menu was the breast conserving therapy menu; 99% of patients
selected information related to breast conserving therapy. The second most popular menu
was the mastectomy menu; 79% of patients selected information from that category. The
least selected menus were information from fellow patients (26% selected) and
information about reconstruction (21% selected). Younger patients and those with more
education selected significantly more items than their older and less educated
counterparts. This was consistent with previous reports of demographic variables related
to patients’ desires to participate in their own treatment decisions, wherein younger and
more educated patients desired more active roles (Beaver et al., 1996; Bilodeau &
Degner, 1996; Degner et al., 1997; Meyer et al., 1995).
Other age differences were found with regard to desire for information.
Researchers reported that older women rated information about the effects of breast
cancer treatment on their social life as significantly more important than younger women
(Luker et al., 1995) and rated information about self care as more important (Bilodeau &
Degner, 1996; Degner et al., 1997). Older patients were significantly less concerned
about recurrence of their breast cancer than younger patients (Petrisek et al., 1997) and
14
were significantly more likely to desire a physician’s opinion about which treatment to
undergo, though a majority of all patients wanted a physician’s recommendation
(Johnson et al., 1996).
The most consistent finding regarding desire for information was that patients
overwhelmingly desired as much information as possible about their disease and its
treatment, though they did not always receive all of the available information. The
repercussions this has on patient decision making and satisfaction has not been examined
in great detail. Further research is needed to provide a more precise picture of how much
information breast cancer patients want, what specific pieces of information they want,
and how they use that information to make their treatment decisions.
Treatment Decisions
Researchers also examined breast cancer treatments selection and what factors
were related to the selection of those treatments. Results from fifteen studies that reported
the percentage of breast cancer patients who selected either mastectomy or breast
conserving treatment and factors associated with their choice were included in this review
(Chagpar et al., 2006; Hughes, 1993; Katz et al., 2001; Kotwall et al., 1996; Lantz et al.,
2005; Mastaglia & Kristjanson, 2001; Morris & Ingram, 1988; Muss et al., 1992; Nold,
Beamer, Helmer, & McBoyle, 2000; Petrisek et al., 1997; Sepucha et al., 2007; Staradub,
Hsieh, Clauson, Langerman, Rademaker, & Morrow, 2002; Ward et al., 1989;
Williamson, Jones, and Ingram, 1999; Wolberg, Tanner, Romsaas, Trump & Malec,
1987). Rates for mastectomy ranged from 15% undergoing a mastectomy (Staradub et
al.) to 82% receiving a mastectomy (Kotwall et al.). It is important to note that there is no
ideal percentage of patients who should receive a mastectomy versus breast conserving
15
treatment; instead the treatment should meet the medical needs of the patient, reflect the
goals of the patient, and the patient should be satisfied with the treatment decision.
For patients who selected breast conserving treatment (typically with radiation) as
their surgical treatment one factor that was an important predictor was the knowledge that
a lumpectomy plus radiation had the same survival rate as a mastectomy (Katz et al.,
2001; Mastaglia & Kristjanson, 2001; Ward et al., 1989). Other factors cited as
influencing the treatment selected were concern about losing a breast (Morris & Ingham,
1988; Ward et al.) and wanting a less drastic surgery (Morris & Ingham). Women who
selected a mastectomy cited a concern about radiation (Morris & Ingham, 1998; Sepucha
et al., 2007; Ward et al., 1989), reconstruction being possible (Ward et al.), and wanting
to get rid of entire disease area (Katz et al., 2001; Morris & Ingham) as important factors
influencing their decisions. Another important factor cited for choosing a mastectomy
was a desire to be done with surgery/treatment (Katz et al.; Kotwall et al., 1996).
Additionally, Kotwall and colleagues reported that many patients who selected a
mastectomy believed that the surgery increased their chance of survival and reduced their
risk of recurrence versus breast conserving treatment. While this may have been an
influencing factor, the knowledge it was based on was partially incorrect, as mastectomy
and lumpectomy plus radiation have nearly identical survival rates.
Differences in treatment selection were also examined with regard to patient
characteristics. The most common finding reported by researchers was that patients who
selected breast conserving treatments were younger than patients who chose a
mastectomy (Chagpar et al., 2006; Mastaglia & Kristjanson, 2001; Morris & Ingham,
1988; Muss et al., 1992; Staradub et al., 2002; Ward et al., 1989; Wolberg et al., 1987). A
16
small number of researchers found results in the opposite direction; patients who had a
lumpectomy were significantly older than patients who chose a mastectomy (Hughes,
1993; Lantz et al., 2005). Finally, Petrisek and colleagues (1997) reported that
approximately two thirds of patients in their study chose a lumpectomy, with no age
differences found.
Other patient characteristics were examined with relation to treatment chosen.
These findings were neither consistent, nor strong. Patients with lower income were
significantly more likely to choose a mastectomy (Hughes, 1993). Lower education was
associated with selection of mastectomy (Muss et al., 1992). A common finding was that
patients with larger tumors were more likely to receive a mastectomy as their surgical
treatment (Chagpar et al., 2006; Kelemen III, et al., 2001; Kotwall, 1996; Muss et al.,
1992). Often, a mastectomy was the recommended treatment for larger (or multiple)
tumors. Mastaglia and Kristjanson (2001) found that breast cancer patients living in rural
areas were significantly more likely to have a mastectomy that their urban counterparts.
This is understandable, as patients who lived in rural areas had to travel further for daily
radiation treatment, so they may have been more likely to select a mastectomy.
Unfortunately, little research examined the underlying rationales for treatment selection
and this is an area that the current research is focusing on, to provide a clearer
understanding of how well participants understand the information and the decisions that
they face.
Age was also significantly related to the selection of adjuvant treatments (Muss et
al., 1992; Petrisek et al., 1997; Williamson, et al., 1999). Older patients were significantly
less likely to receive chemotherapy or radiation and significantly more likely to receive
17
Tamoxifen (hormone therapy) than younger patients, supporting one hypothesis that
younger women were more concerned about survival and older women were more
concerned with quality of life (Williamson et al.). However, Petrisek and colleagues
asserted that this age difference was because younger patients were less likely to have
cancers that were estrogen receptor positive, thus Tamoxifen would not have been an
indicated treatment. The area of adjuvant treatment selection needs further empirical
study, as most research focused solely on surgical decisions and not on the subsequent
adjuvant decisions, which are very much an important area of breast cancer treatment.
The current research moves beyond examining only surgical decisions and also explores
the quality of adjuvant decisions, an area often ignored by researchers in the past.
Information and Knowledge
Age and education were related to surgical treatment selection. However, it was
not known if this was due to younger and more educated women seeking out more
information and having increased knowledge about breast cancer, which in turn affected
the decisions that they made. Research that supported this idea showed only one third to
one half of patients knew that mastectomy and lumpectomy plus radiation had equivalent
survival rates (Fagerlin et al., 2006; Katz et al, 2001; Sepucha, et al., 2007). This lack of
knowledge influenced the decisions that patients made. Patients who knew that the two
procedures had equal survival rates were more likely to choose the less invasive
procedure (Katz et al., 2001; Mastaglia & Kristjanson, 2001; Ward et al., 1989).
Hughes (1993) examined the role of information in treatment selection for breast
cancer and found that two months after surgery, patients had very poor recall of
information presented to them before their procedures. For the group that underwent a
18
breast conserving lumpectomy, the number of items recalled ranged from zero to four,
out of a possible 13 items. Fifty percent of participants in this group had no recall of
information previously presented to them. For the mastectomy group, also out of a
possible 13 items, the recall range was zero to three items, with 66% having no recall of
previously presented information (Hughes). Multiple studies showed that patients
received incomplete information and then understood and retained very little of it. This
lack of understanding can have an adverse effect on patients’ willingness to participate in
their treatment decisions (Siminoff, 1989).
Multiple instruments were developed in an attempt to measure what women knew
about breast cancer. Many of these measures focused on specific areas of breast cancer,
ignoring the overall body of knowledge. The most common focus of the knowledge
instruments was pre-diagnosis knowledge, including susceptibility, screening, and
detection (McCance, Mooney, Smith, & Field, 1990; Ondrusek, Warner, & Goel, 1999;
Stager, 1993; Stillman, 1977). Other instruments focused on treatment-related
knowledge; one instrument focused on Ductal Carcinoma In Situ, which is a specific type
of breast cancer (Bluman, Borstelmann, Rimer, Iglehart, & Winer, 2001). Other focuses
were treatment options (Ward & Griffin, 1990) and knowledge about risks, symptoms,
and treatment options (Vaeth, 1993). One major problem with many of these instruments
was that most participants who completed the instruments in these studies were not breast
cancer patients. Researchers did find that increased knowledge was related to both
increased age and increased education (Bluman et al.; Vaeth; Ward & Griffin). Vaeth
also reported that patients had higher total scores, as well as higher scores on the
treatment and symptoms subscales than non-patients.
19
One way in which researchers examined treatment decision making was through
the use of low cost, easy to administer decision aids, called decision boards. An example
of a decision board was one designed to help breast cancer patients choose between
breast conserving treatment and mastectomy (Whelan et al., 2004). The board contained
four sections of information for the surgical treatment decision: treatment choice, side
effects, results of treatment choice for the breast, and results of the treatment choice for
survival. Information from clinical trials (effects of treatment, survival, quality of life) is
presented to patients in a sequential manner and patients are encouraged to ask questions
as the information is presented. The decision board took approximately 21 minutes to
administer and did not increase the overall length of the medical consultation. Patients
were presented with a take-home version of the decision board.
Patients in the decision board group had significantly higher knowledge scores
than the control group. The decision board group also had lower decisional conflict
scores and higher satisfaction with the decision making after the initial consultation,
though no differences existed at six or twelve months. There were also differences in the
treatment selected. Patients who received the decision board were significantly more
likely to select breast conserving treatment than the control group (94% versus 76%).
Finally, those in the decision board group were more likely to believe they had a choice
over their treatment than the control group (87% versus 69%) (Whelan et al., 2004).
Other examples of decision boards were those for adjuvant treatment for breast
cancer, including chemotherapy and radiation to the breast (Levine, Gafni, Markham, &
MacFarlane, 1992; Whelan et al., 1995). The information presented included possible
treatments, the quality of life that could reasonably be expected from undergoing
20
treatment and not undergoing treatment, and the possible outcomes for both the treatment
and non-treatment option. Again, the decision boards were easy to administer, taking
approximately 15 minutes, and patients received a take-home version.
Overall, patients were satisfied with the adjuvant decision boards; over 97% of
patients found them easy to use, 80% reported that the decision boards helped them think
of questions to ask the physician or nurse, and over two-thirds of patients showed the
take-home version of the decision board to someone else. Over 80% of patients reported
that the instrument helped them make their adjuvant treatment decisions. Additionally,
the use of the decision aid was found to increase patients’ perceptions of choice over their
treatment versus those who only consulted with the physician (Levine et al., 1992;
Whelan et al., 1995).
The research in the area of decision aids has demonstrated that patient decision
aids are easy to administer and receive positive feedback from patients and physicians.
Chapman, Elstein, and Hughes (1995) measured knowledge before and after the
implementation of a decision aid and found that knowledge scores did improve after
receiving a decision board as an educational intervention. Other researchers have not
examined this explicitly by assessing knowledge. However, patients using decision aids
made different decisions than those who did not receive the decision aid (typically
selecting a less invasive treatment), demonstrating that these aids had an impact on
decisions that were made (Chapman et al.; Kasper, Mulley & Wennberg, 1992; Whelan et
al., 2004).
An important area of research in breast cancer decision making that needs more
attention is evaluating and improving patient knowledge and decision making ability.
21
Increasing patient competence should enable patients to make more informed decisions
about their treatment. Decision boards are likely an effective method to both relay
information to patients and assess their understanding of that information before they
make their treatment decisions. However, knowing what information patients need in
order to make informed treatment decisions is an important step that must occur before
educational interventions are designed and implemented.
Conclusion
Though the research was inconclusive as to whether breast cancer patients desired
active involvement in their treatment decisions, it is clear that patients wanted to be
involved in terms of access to information. Much of the research in the area of breast
cancer treatment found considerable inconsistency in the information that patients
received, as well as gaps in the information that was recalled by patients. These
disparities can lead to lack of involvement and result in less satisfaction with the
treatment process. Further research is needed to determine whether patients are prepared
to participate in the complex treatment decisions that they face after receiving a breast
cancer diagnosis.
STATEMENT OF RESEARCH PURPOSE
Previous research suggests that women who have been diagnosed with breast
cancer have significantly greater knowledge about breast cancer than the non-patient
sample (Brown, 2003; Vaeth, 1993). However, it is not known whether increased
knowledge translates to better treatment decisions. Two instruments were developed and
used to evaluate the relationships between patient status, breast cancer knowledge and
treatment decision quality. Due to their exposure to breast cancer specific information
22
during the treatment process, it is hypothesized that women who have been diagnosed
with breast cancer will outperform the non-patient sample on 1) breast cancer knowledge,
2) surgical and adjuvant treatment selection, and 3) the support of correct rationales for
both surgical and adjuvant treatment selections.
This research further aims to examine the relationships between the independent
variables of patient status, age, breast cancer knowledge, and information use and the
dependant variables of surgical treatment recommendations (ST), surgical rationale
support (SR), adjuvant treatment recommendations (AT) and adjuvant rationale support
(AR). It is expected that more breast cancer knowledge, a previous diagnosis of breast
cancer, and increased age will lead to better decisions and support of better rationales.
Finally, the predictive ability of information use will be examined for both treatment
decisions and support of explanatory statements. This is primarily an exploratory
examination. It is expected that participants who rate medical information as more
important will make better decisions and support more sound rationales than those
participants who rate situational information as important. Beyond that, it is unclear what,
if any, relationships will emerge between information importance and decision quality.
23
Chapter 2: Method
PARTICIPANTS
Participants were recruited from online message boards at two websites, one that
focuses on hereditary breast cancers, the other an online, community-based support
group. Non-patients were recruited from a community-based message board that
advertised community activities and volunteer opportunities. One hundred and fifty
participants began the survey; eight of those participants completed only the demographic
information. Participants with usable data consisted of 142 women. Fifty-three had a
previous diagnosis of breast cancer, while the remaining 89 women had no personal
experience with breast cancer. The mean age for the patient group was 47.6 years
(SD = 9.7 years). The non-patient group was significantly younger, with a mean age of
39.2 years (SD = 10.9 years), t = -4.69, p < .01. The majority of both groups were white,
married, and had at least a college degree. Complete participant characteristics are
included in Table 1.
Forty-three percent of patients were diagnosed with breast cancer less than 2 years
ago. The remaining patients were evenly split, with 28% diagnosed between two and five
years ago, and the 28% diagnosed more than five years ago. The most common stage of
cancer reported was Stage II (38%), followed by Stage I (28%), and Stage 0, non-
invasive breast cancer (25%). Patients were also asked to report the treatments they
underwent. Twenty-two percent of patients reported that they received both a
lumpectomy and a mastectomy and since no attempt was made to record recurrences,
the percentages are difficult to interpret. Table 2 presents a summary of complete
patient characteristics.
24
Table 1
Participant Characteristics
Patients (n=53) Non-Patients (n=89)
Mean Age (SD) 47.6 (9.7) 39.2 (10.9)
Race/Ethnicity
Caucasian 94.3% 77.5%
Hispanic 3.8% 2.2%
African American 0.0% 6.7%
Asian 0.0% 10.1%
Other 1.9% 3.4%
Education
Less than High School 0.0% 1.1%
High School Diploma 7.5% 7.9%
Some College 17.0% 32.6%
College Degree 45.3% 37.1%
Graduate School 30.2% 21.3%
Marital Status
Single 11.3% 32.6%
Married/Living as Married 75.5% 56.2%
Divorced/Separated 11.3% 11.2%
Widowed 1.9% 0.0%

25
Table 2
Patient Characteristics
Time Since Diagnosis
Less than 2 years 43.4%
Between 2 and 5 years 28.3%
More than 5 years 28.3%
Stage of Cancer
Stage 0 (DCIS or LCIS) 24.5%
Stage 1 28.3%
Stage 2 37.7%
Stage 3 7.5%
Stage 4 0.0%
Don’t Know 1.9%
Treatment
Lumpectomy 54.7%
Mastectomy 67.9%
Reconstruction 54.7%
Radiation 60.4%
Chemotherapy 56.6%
Hormonal Therapy 43.4%
Lymph Node Dissection 64.2%
Mastectomy of Unaffected Breast 37.7%
Other 11.3%
26
Table 2 (continued)

Patient Characteristics

Level of Participation
I made the final decision on treatment 30.2%
My doctor and I collaborated on final decision 67.9%
My doctor made the final decision on treatment 1.9%
Note. Treatment percentages exceeded 100% because participants were instructed to select all treatments
that they underwent.
In addition to reporting diagnostic and treatment related information, all
participants were queried as to other experiences they had with breast cancer, such as
friends’ and/or family members’ experiences, as well as any professional experiences.
Overall, 72% of participants reported knowing someone with breast cancer, with mother
and aunt as the two most commonly reported family members. When looking at only
breast cancer patients’ responses, 91% reported knowing someone with breast cancer,
again with mother and aunt as the two most frequent family members mentioned. Sixteen
percent of participants reported professional experience related to breast cancer. The most
commonly reported experiences were nursing, patient advocates, and research positions.
Table 3 presents a summary of this information.
MATERIALS
Demographic Measure
Participants completed a demographic questionnaire that requested basic personal
information such as age, race/ethnicity, and education level, as well as information
specific to breast cancer experiences. Participants indicated whether they had ever been
27
Table 3
Other Experiences with Breast Cancer
Overall (n=142) Patients (n=53) Non-Patients (n=89)
Know someone with
breast cancer?

Yes 71.8% 90.6% 60.7%
No 28.2% 9.4% 39.3%
Of those who said yes,
person they indicated

Mother 39.2% 35.4% 42.6%
Sister 17.6% 22.9% 13.0%
Aunt 39.2% 43.8% 35.2%
Daughter 0.9% 2.1% 0.0%
Grandmother 16.7% 16.7% 16.7%
Cousin 12.7% 16.7% 9.3%
Friend 42.1% 50.0% 33.3%
Acquaintance 41.1% 58.3% 27.8%
Other 17.6% 18.8% 16.7%
Professional Experience
Yes 15.5% 11.3% 18.0%
No 84.5% 88.7% 82.0%
Note. Those who selected “other” indicated in-laws, distant relatives, and great grandmothers.
Professional experience most frequently included nursing, patient advocates, and research.
28
diagnosed with breast cancer, and if so, described their disease history, including time
since diagnosis and medical treatments they received. In addition to reporting diagnostic
and treatment related information, all participants were queried as to other experiences
they had with breast cancer, such as friends’ and/or family members’ experiences, as well
as any professional experiences. The complete demographic survey is located in
Appendix A.
Breast Cancer Knowledge Measure
The breast cancer knowledge measure consisted of 25 multiple choice and seven
true/false items. Items were written with regard to four broad areas of knowledge
believed to be central concepts in breast cancer. These four areas were diagnosis,
treatment options, treatment outcomes, and physiology. Test items were created based on
materials from three main sources. The National Cancer Institute (NCI) has several
publications commonly given to patients, free upon request. Two of these publications
(NIH No. 00-1556 and NIH No. 98-4251) were used in the current research to develop
the knowledge scale (NIH, 1998; NIH 2000). Additionally, the state of California’s
Department of Health Services developed a booklet, which is given to every patient
diagnosed with breast cancer in the state of California (1999). The information from this
source was also utilized in the creation of test items. Test items were designed to be
challenging, but also to cover information that breast cancer patients would likely
encounter and which would be relevant to their treatment decision making efforts.
A breast surgeon and a medical oncologist (GL & CR), both who specialized in
the diagnosis and treatment of breast cancer, reviewed the items to establish content
validity. For the evaluation process, the items were broken down into the categories in
29
which we believed they belonged and placed into a format that allowed the physicians to
easily evaluate the items. The physicians rated each item in three areas: 1) quality and
correctness of the item, 2) whether the item belonged in the category it had been placed
in, and 3) how relevant it was for a patient who was making decisions about breast cancer
to know the information it sampled. The two physicians were not required to reach a
consensus with one another regarding item quality.
Revisions were made to the original draft of the knowledge instrument in order to
incorporate the suggestions of both physicians, including removing items, rewording
items (or the possible responses) and adding new items suggested by the physicians.
Additionally, after initial data collection and analysis in 2003, revisions were made,
which included removing and/or revising items that lacked variability or were too easy or
too difficult. Additional items were added that the experts believed would round out the
relevant knowledge that breast cancer patients should have to be adequately informed
about their disease and its treatment. The complete knowledge measure is in Appendix B.
Items on the current version of the knowledge instrument were examined to ensure
adequate difficulty and discrimination. Item difficulty was calculated as the mean score
of each dichotomously scored item. Possible difficulty ranged from zero to one, with zero
indicating that no one answered the item correctly and one indicating that everyone
answered the item correctly. Actual item difficulties from the knowledge measure ranged
from .36 to .92. The ability of the item to discriminate between participants of high and
low ability was calculated using the point biserial correlation. In this case, total
knowledge score, not patient status, was used to determine participants’ ability. Point
30
biserial correlations of items ranged from .004 to .688. Table 4 presents the item analysis
of the knowledge measure.
Table 4
Item Analysis of Knowledge Measure
Item Difficulty Point Biserial Correlation
K1 .68 .27**
K2 .80 .41**
K3 .70 .51**
K4 .88 .42**
K5 .57 .25*
K6 .76 .63**
K7 .36 .34**
K8 .60 .53**
K9 .52 .22*
K10 .73 .50**
K11 .82 .59**
K12 .46 .07 (Removed)
K13 .62 .34**
K14 .56 .42**
K15 .76 .47**
K16 .58 .22*
K17 .53 .09 (Removed)
31
Table 4, Continued

Item Analysis of Knowledge Measure

Item Difficulty Point Biserial Correlation
K18 .56 .69**
K19 .76 .20
K20 .64 .16 (Removed)
K21 .72 .36**
K22 .59 .59**
K23 .68 .67**
K24 .69 .57**
K25 .77 .59**
K26 .87 .33**
K27 .82 .41**
K28 .92 .32**
K29 .72 .02 (Removed)
K30 .90 .29*
K31 .48 .09 (Removed)
K32 .87 .00 (Removed)
Note. Items marked (removed) were not included in further analyses.
* p < .01. ** p < .001.
The following criterion was used to evaluate items. Point biserial correlation
values below .20 were considered poor and items needed to be eliminated or completely
revised. Items with point biserial correlations between .20 and .29 were considered fair,
32
some revisions should be considered. Items with correlations between .30 and .39 were
considered good, with little or no revisions necessary. Items with correlations above .40
were considered excellent (Crocker & Algina, 1986). Though these are general
guidelines, Nunnally and Bernstein (1994) state that the final decision to retain or delete
an item must rest on the judgment of the test developers; it cannot be based on statistical
factors alone. The item analyses for many tests are secondary to the content validity of
the scale, which for the knowledge instrument, was established by experts in the field of
breast cancer treatment.
Items from the knowledge instrument were retained or deleted based on the above
criteria. Six items were removed in total, though with revisions, they could be included in
a future version of the instrument. Reliability of the knowledge measure was estimated by
calculating Cronbach’s alpha. The reliability estimate of the full scale was .78. With the
six items that were poor discriminators removed, alpha increased to .83. Finally, a factor
analysis was conducted on the breast cancer knowledge measure. Neither the four-factor
pattern that was proposed (diagnosis, treatment options, treatment outcomes, physiology),
nor any other factor pattern emerged that explained the underlying structure of the
measure. Therefore, one total score for correct responses was computed and used in
further analyses.
Breast Cancer Decision Quality Measure
The breast cancer decision quality measure was designed to assess participants’
ability to 1) choose the best treatments, 2) use relevant information to make treatment
choices, and 3) provide support for sound conceptual explanations of why some treatment
choices were better than others. The measurement instrument used realistic hypothetical
33
patient scenarios as a testing environment to query research participants as to what they
believed were the best treatments, the most important pieces of information, and the
correct causal explanations. The measure consisted of two pairs of scenarios; with one
surgical and one adjuvant scenario for the same hypothetical patient paired together to
create a realistic treatment case. Pairs of scenarios were counterbalanced to reduce
practice and/or fatigue effects.
The scenarios were designed in collaboration with a medical oncologist and a
breast surgeon (CC & DH), both of whom specialized in breast cancer treatment and
research. The medical collaborators were essential to the development of accurate and
relevant scenarios. In order to make the scenarios as realistic as possible, the physicians
provided both the medical and the personal factors that are commonly observed in patient
consultations. Scenarios were designed to represent the most common diagnostic
characteristics, as these would cover information and knowledge that most breast cancer
patients would be expected to encounter. The decision quality measure is located in
Appendix C.
Treatment Selection
Participants’ ability to select the best treatments was measured with a 10-point
scale that required participants to rate the strength of their recommendation for a list of
possible treatments (both surgical and adjuvant). Possible surgical treatments ranged
from taking no action to a contralateral prophylactic mastectomy (where the unaffected
breast is removed). A rating of zero indicated that the treatment was not recommended at
all, a rating of four to five indicated the treatment was moderately recommended, and a
rating of nine indicated that the treatment was strongly recommended. Participants’
34
scores were compared to the scores provided by the two breast cancer specialists to
calculate deviation scores that were used in the data analysis. For example, if one
treatment recommendation was “do nothing at this time, wait to see if the tumor gets
larger” and the physicians gave that treatment a score of four, a participant’s score would
be the absolute value of four minus the participant’s rating. Therefore, if a participant
rated that option seven, their deviation would be three. The best possible deviation score
would be zero, indicating the participant was in perfect agreement with the physicians’
recommendations.
Scores were then added together within each category. For each surgical
treatment, there were 11 options that participants were asked to rate. Each participant’s
score for surgical treatment was the sum of the absolute values of the deviations for each
of 11 surgical treatment options. The same procedure was used to calculate a score for
adjuvant treatment selection. One adjuvant scenario had five treatments to rate, while the
other had nine treatments. Possible adjuvant treatments ranged from taking no further
action to a combination of radiation, chemotherapy, and hormone therapy. After
calculating deviation scores, the surgical treatment deviation scores were added together
and the adjuvant treatment deviation scores were added together, providing two final
treatment deviation scores (ST and AT). The possible range of deviation scores depended
on both the number of options available and the physicians’ responses to each item. The
range of possible scores for surgical treatment recommendations was 0 – 195. The range
of possible scores for adjuvant treatment recommendations was 0 – 118. Again, it is
important to note that for the treatment decision scenarios, the lower the score, the better,
35
more accurate the participants’ ratings were, when compared to the physicians’
responses.
Reasoned Statements
For each treatment scenario, between six and ten rationales were written that
clearly explained correct treatments and also reflected faulty reasoning that could be
made to justify incorrect decisions. Participants again used a 10-point scale to indicate the
extent to which they agreed or disagreed with each explanation. A rating of zero
indicated the participant strongly disagreed with the rationale provided by the statement,
a score of four to five indicated the participant neither agreed nor disagreed with the
rationale provided, and a score of nine indicated that the participant strongly agreed with
the rationale. Participants’ scores were again compared to the scores provided by the two
breast cancer specialists; deviation scores were calculated and used in the data analysis.
One possible response was “Jane should not have any form of breast surgery
because of surgical risks involved.” If the physicians’ rating of this rationale was a four,
each participant’s deviation score would be the absolute value of four minus the
participant’s response. So, if a participant’s rating was a nine, their deviation score for
that item would be five. Again, lower scores indicated less deviation from physician
responses, indicating better support of correct rationales. The best possible deviation
score a participant could receive was zero, indicating perfect agreement with the
physicians’ ratings. For each scenario, a surgical and an adjuvant rationale deviation
score was computed by adding together the participants’ deviations for each item on the
scale. After calculating deviation scores, the surgical rationale scores were combined and
the adjuvant rationale scores were combined, resulting in two final deviation scores for
36
the rationales (SR and AR). The possible range of deviation scores depended on both the
number of options available and the physicians’ responses to each item. For the surgical
rationales, the possible range was 0 – 166. For the adjuvant rationales, the possible range
was 0-103.
Information Use
The information use section of the decision quality measure presented participants
with a list of each piece of information that was presented in the scenario. Participants
were asked to rate how important the information was to them, when making their
treatment recommendations. Participants used a 10-point scale to rate the importance of
each piece of information. A rating of zero indicated that the piece of information had no
importance in their treatment recommendations, a score of four to five indicated
moderate importance, and a score of nine indicated the information had great importance
when making treatment recommendations. The information section was not scored in the
same manner as the treatment decision and rationale subscales, as even the physicians’
responses showed considerable variation.
Instead, an exploratory factor analysis was conducted to determine the underlying
structure of the scale. Principal components analysis was chosen as the exploratory
technique and varimax rotation was used to extract the underlying factors. Factor
solutions were selected based on eigenvalues greater than one. Item loadings greater than
.32 were initially selected for interpretation. If items loaded less than .45 on a factor, they
were considered for deletion, as that is the cutoff considered “fair” by Comrey and Lee
(1992). Items were also considered for deletion if they loaded greater than .45 on more
than one factor. Reliability of the factors was estimated by calculating Cronbach’s alpha.
37
Additionally, mean factor scores were computed by adding together the scores from the
items that loaded on each factor and calculating their mean. These mean factor scores
were used in further analyses.
Participation Level
A single question asked participants how willing they would be to make the
treatment decision, if they were the patient portrayed in the scenario. This item was asked
after each treatment decision (two surgical and two adjuvant decisions). There were three
options participants selected from: 1) prefer to make the treatment decision completely on
my own, 2) prefer to make the treatment decision in collaboration with physician, and 3)
prefer to leave this decision regarding treatment to the physician. Additionally, on the
demographic survey, patients were asked what level of participation they actually
assumed in their own breast cancer treatment decisions.
PROCEDURE
The demographic, knowledge, and decision quality instruments were all
completed online. Participants were recruited from online message boards. A link in the
message advertising the study took participants to an information sheet that explained the
research, as well as their rights as research participants. Participants were entered into a
lottery with a 1 in 50 chance of winning $100. Potential participants were informed that
they did not need to complete the survey to be eligible for the lottery. After reading the
information sheet and supplying an email address to be notified at if they were selected
for the lottery, participants were taken directly to the demographic page. The
demographic questionnaire was presented upfront, so that if participants did not complete
all of the materials, differences between those who completed and those who did not
38
could be examined, based on demographic variables. The eight participants who
completed only the demographic measure did not differ from those who completed all of
the materials, based on any of the participant characteristics. After completing the
demographic measure, participants completed the breast cancer knowledge instrument.
Finally, participants completed the decision quality instrument. Upon completing both
pairs of scenarios, participants were directed to a final page that thanked them for their
participation and again supplied the contact information for the researchers in case
participants had any questions or comments concerning the research. All materials took
approximately one hour to complete and participants were required to complete them in
one sitting. Participants were told that we were interested in their unaided knowledge and
ability and were instructed to not use any outside sources to assist them with completing
the materials. However, it is impossible to know whether participants accessed any
additional information, and if so, how that might have affected the results.
Data collection via the internet had several advantages for the current research.
The first was that it allowed for more data to be collected in a shorter period of time.
Second, the internet allowed for easier recruitment of a specialized sample, in this case
breast cancer patients. As I did not have access to a continuous patient sample, in person
recruitment would have been through handing out flyers and in person data collection
would have required participants to arrange an hour block of time and travel to USC to
complete the measures in person. I believe this would have greatly reduced the interest in
completing the survey. Previous research demonstrated that data collected over the
internet is comparable to data collected by traditional paper and pencil methods
(Birnbaum, 2001).
39
Because these measures were administered over the internet, a convenience
sample was utilized and I had only weak control over the characteristics of the
participants in my study. This weak control was partially countered by careful placement
of advertisements to recruit a wide range of participants, both patients and non patients.
Extensive demographic information was collected, allowing for a thorough description of
participants. However, caution must be exercised when making generalizations to
population as a whole.
DATA ANALYSIS
Incomplete Data
Instead of treating incomplete data with a simple technique such as listwise
deletion, pairwise deletion, or mean substitution, the expectation maximization (EM)
algorithm was utilized. This technique employed an iterative process to estimate both the
parameters and the incomplete data. First, the parameters were estimated based on the
data that was present. Then, missing values were estimated based on those parameters.
Parameters were then estimated again, as were values for the missing data. This process
repeated until a stable solution was reached, at which point the parameters were used to
make final estimates of the missing values (Howell, 2007; Roth, 1994; Schafer, 1997).
The EM algorithm was less biased than listwise deletion and mean substitution and more
accurate than pairwise deletion (Roth). The EM algorithm added error to the estimated
variances and covariances based on the residual variance from the regression (Howell).
The EM was computed using the freely available program NORM (Schafer, 1999).
40
Chapter 3: Results
MEAN DIFFERENCES IN KNOWLEDGE AND DECISION QUALITY
A series of independent samples t-tests were conducted to determine if there were
mean differences on the knowledge and decision quality scales between patients and non-
patients. Patients had a mean score of 22.40 (SD = 2.16) on the knowledge instrument,
versus a mean score of 15.75 (SD = 4.52) for non-patients, t = -11.78, p < .001. Further,
patients performed significantly better than non-patients on each of the four decision
outcome measures. Patients had a mean score of 48.79 (SD = 18.85) for the surgical
treatment recommendations, while non-patients had a mean score of 74.04 (SD = 19.86),
t = 7.47, p < .001. For surgical rationale support, patients had a mean score of 20.98
(SD = 11.79) and non-patients had a mean score of 41.12 (SD = 18.87), t = 7.83, p < .001.
For the adjuvant treatment recommendations, patient had a mean score of 31.74
(SD = 14.54) and non-patients had a mean score of 49.52 (SD = 14.58), t = 7.04, p < .001.
Finally, for adjuvant rationale support, patients had a mean score of 12.98 (SD = 8.97)
while non-patients had a mean score of 31.89 (SD = 14.99), t = 9.40, p < .001. Table 5
presents the means, standard deviations, and associated statistics.
The treatments that the physicians gave a top rating to were also examined to
determine whether patients and non-patients differed in the ratings they assigned to these
treatments. For the surgical treatments, physicians gave a score of nine to three options:
lumpectomy plus lymph node surgery (followed by 6 weeks of daily radiation),
mastectomy of the affected breast plus lymph node surgery, and mastectomy of the
affected breast plus lymph node surgery, plus reconstructive surgery. Patients and non-
patients differed significantly only on their ratings of this last option for both surgical

41
Table 5
Mean Differences between Patients and Non-Patients on Breast Cancer Knowledge and Decision Quality Scores
Overall (n = 142) Patients (n = 53) Non-Patients (n = 89)
Range M SD Range M SD Range M SD t df p η
2

Knowledge 6-26 18.23 4.99 16-25 22.40 2.16 6-26 15.75 4.52 -11.78134.8.000 0.51
ST 1-122 64.6222.96 1-83 48.7918.85 2-122 74.0419.86 7.47140.0.000 0.28
SR 2-95 33.61 19.21 2-48 20.98 11.79 7-95 41.12 18.87 7.83 139.6 .000 0.31
AT 2-93 42.8816.88 2-67 31.7414.54 16-93 49.5214.58 7.04140.0.000 0.26
AR 2-72 24.83 15.94 2-38 12.98 8.97 2-72 31.89 14.99 9.40 140.0 .000 0.39

ST: Surgical Treatment Deviation Score; SR: Surgical Rationale Deviation Score;
AT: Adjuvant Treatment Deviation Score; AR: Adjuvant Rational Deviation Score

42
scenarios. For scenario one, patients had a mean rating of 5.82 (SD = 3.60) and non-
patients had a mean rating of 4.38 (SD = 3.34), t = -2.13, p < .05. For scenario two,
patients had a mean score of 7.48 (SD = 2.13) and non-patients had a mean score of 5.49
(SD = 3.15), t = -3.71, p < .001. None of the other top rated treatment options, either
surgical or adjuvant, had significant differences between patient and non-patients.
EXPLORATORY FACTOR ANALYSIS OF INFORMATION USE SCALES
Principal component analysis of the surgical treatment information use scale
resulted in five factors, accounting for 65.4% of the variance. The first factor, accounting
for 26.1% of the variance was labeled “Cancer Concerns.” Five items initially loaded on
this factor, though one item with a loading of only .33 was subsequently removed from
analysis. The reliability estimate of this factor was .84. The second factor, accounting for
13.3% of the variance was labeled “Situational Characteristics.” Three items loaded on
this factor, and the reliability estimate was .81. Factor three was labeled “Individual
Characteristics” and accounted for 12.9% of the variance. Three items loaded on this
factor, with a reliability estimate of .73. Factor four was labeled “Medical” and accounted
for 6.9% of the variance. Four items loaded on this factor, and the reliability estimate was
.64. The final factor was labeled “Personal Concerns” and accounted for 6.1% of the
variance. While only two items loaded on this factor, the loadings were high and a
sufficient amount of variance was accounted for, though the reliability estimate was
lower, only .62. Table 6 presents a description of the factor structure of the surgical
information use, including the items that loaded on each factor. Table 7 presents
descriptive statistics of the surgical information factor scores.

43
The most important factor to participants when making surgical decisions was the
“Medical” factor (M = 8.10, SD = 0.91). The least important factor was the “Situational
Characteristics” factor (M = 2.30, SD = 1.94). With regard to mean factor scores, patients
and non-patients did not differ significantly from one another on any of the five surgical
factor scores.
Table 6
Factor Structure of Surgical Information Use Items
Factor and Variables Factor
Loading
Reliability
Estimate
Factor 1: Cancer Concerns
Concern about dying
Concern about recurrence in same breast
Concern about recurrence in other breast
Concern about recurrence outside the breast

.72
.76
.81
.79
.84
Factor 2: Situational Characteristics
Occupation
Relationship status
Living in an urban area

.86
.75
.84
.81
Factor 3: Individual Characteristics
Age
Menopausal status
Childbearing status

.68
.87
.74
.73

44
Table 6, Continued
Factor Structure of Surgical Information Use Items
Factor and Variables Factor
Loading
Reliability
Estimate
Factor 4: Medical
Size of cancer
Location of cancer
Type of cancer (invasive/non-invasive)
Results from biopsy

.79
.59
.80
.55
.64
Factor 5: Personal Concerns
Concern with health
Concern with appearance

.62
.75
.62

Principal component analysis of the adjuvant treatment scale resulted in six
factors, accounting for 73.4% of the variance. The first factor was labeled “Concerns”
and contained six items, accounting for 26.4% of the variance. The reliability estimate for
this factor was .85. The second factor accounted for 16.3% of the variance and was
labeled “Situational Characteristics.” Three items loaded on this factor and the reliability
estimate was .82. Factor three was labeled “Medical” and accounted for 10.4% of the
variance. Three items loaded on this factor and the reliability estimate was .70. Factor
four was labeled “Individual Characteristics” and accounted for 8.8% of the variance.
Three items loaded on this factor and the reliability estimate was .74. The fourth factor
was labeled “Cancer Characteristics” and accounted for 6.1% of the variance. Two items
loaded on this factor and the reliability estimate was .82. The final factor was labeled

45
Table 7
Mean Differences between Patients and Non-Patients on Surgical Information Factor Scores
Overall (n =142) Patients (n = 53) Non-Pts (n = 89)
Factor M SD M SD M SD t df p
Factor 1: Cancer Concerns 6.53 1.69 6.83 1.66 6.55 1.71 -0.96 140.0 .338
Factor 2: Situational Character. 2.30 1.94 2.02 2.02 2.46 1.88 1.32 140.0 .191
Factor 3: Individual Character. 5.67 1.85 5.35 1.94 5.86 1.79 1.59 140.0 .114
Factor 4: Medical 8.10 0.91 8.07 0.77 8.11 0.98 0.25 140.0 .804
Factor 5: Personal Concerns 6.25 1.76 6.16 1.81 6.30 1.74 0.47 140.0 .643

46
“Invasiveness of Cancer” and also contained two items. This factor accounted for
5.4% of the variance and the reliability estimate was .79. Table 8 presents a description of
the items that loaded on each factor for adjuvant information use. Table 9 presents the
descriptive statistics of the adjuvant information use factor scores.
The most important information to participants making adjuvant treatment
decisions was the information contained in factor six, labeled “Invasiveness of Cancer”
(M = 8.50, SD = 0.92). As was the case with the surgical information, the least important
information for participants was the “Situational Characteristics” factor (M = 2.37,
SD = 2.04). Patients and non-patients differed significantly from one another on two out
of six factor scores. For the Cancer Characteristics factor, patients had a mean score of
8.63 (SD = 0.71) and non-patients had a mean score of 7.74 (SD = 1.67), t = -4.97,
p < .001. For the Invasiveness of Cancer factor, patients had a mean score of 8.81
(SD = 0.34) and non-patients had a mean score of 8.32 (SD = 1.10), t = -3.87, p < .001.
PREDICTING DECISION QUALITY
A series of regression analyses were conducted to determine the amount of variance
that was explained for each dependant variable, based on the independent variables
entered. The four dependant variables were deviation scores for surgical treatment
recommendations (ST), surgical rationale support (SR), adjuvant treatment
recommendation (AT), and adjuvant rationale support (AR). Independent variables
included patient status, age, and breast cancer knowledge. Additionally, the scores from
each of the five surgical information factors that emerged from the factor analysis were
entered as independent variables for the two surgical outcomes, while the scores from

47
Table 8
Factor Structure of Adjuvant Information Use Items
Factor and Variables Factor
Loading
Reliability
Estimate
Factor 1: Concerns
Personal concern with health
Personal concern with appearance
Concern about dying
Concern about recurrence in same breast
Concern about recurrence in other breast
Concern about recurrence outside breast

.66
.53
.91
.84
.71
.71
.85
Factor 2: Situational Characteristics
Occupation
Relationship status
Living in an urban area

.90
.67
.88
.82
Factor 3: Medical
Location of cancer
Size of lump
Grade of cancer

.76
.80
.68
.70
Factor 4: Individual Characteristics
Age
Menopausal status
Childbearing status

.52
.79
.81
.74

48
Table 8
Factor Structure of Adjuvant Information Use Items

Factor and Variables Factor
Loading
Reliability
Estimate
Factor 5: Cancer Characteristics
Estrogen and progesterone receptor status
Her-2/neu receptor status

.82
.82
.82
Factor 6: Invasiveness of Cancer
Type of cancer (invasive/non-invasive)
Results of lymph node surgery

.73
.82
.79

each of the six adjuvant information factors were entered as independent variables for the
two adjuvant outcomes.
Analyses were performed using SPSS REGRESSION. Regression analyses for
each dependant variable are presented separately below. For each regression analysis,
unstandardized regression coefficients (B) and intercepts, standardized regression
coefficients ( β), and the squared semi-partial correlations (sr
i
2
) are reported. Note that
some regression coefficients are negative indicating that higher scores on those variables
are related to better responses on the decision scale, as lower deviation scores on the
decision quality measures indicate better decisions. In addition to the regression analyses,
correlations between age, patient status, knowledge, each information use factor score
and the dependant variables were computed. Simple correlations are presented in Table
10 (surgical) and Table 11 (adjuvant).

49
Table 9
Mean Differences between Patients and Non-Patients on Adjuvant Information Factor Scores
Overall (n =142) Patients (n = 53) Non-Pts (n = 89)
Factor M SD M SD M SD t df p
Factor 1: Concerns 6.48 1.69 6.51 1.83 6.47 1.61 -0.14 140.0 .889
Factor 2: Situational Character. 2.37 2.04 2.01 1.96 2.59 2.06 1.66 140.0 .099
Factor 3: Medical 7.68 1.31 7.63 1.16 7.71 1.40 0.34 140.0 .736
Factor 4: Individual Character. 5.50 2.01 5.35 2.06 5.59 1.98 0.67 140.0 .507
Factor 5: Cancer Character. 8.00 1.47 8.63 0.71 7.74 1.67 -4.97 129.1 .000
Factor 6: Invasiveness of Cancer 8.50 0.92 8.81 0.34 8.32 1.10 -3.87 113.5 .000

50
Table 10
Correlations between Surgical Outcome Variables and Predictor Variables
ST SR Knowl Pt Stat Age SF1 SF2 SF3 SF4 SF5
ST 1.0 .76**-.64**-.53**-.48** -.08 .26** .24** -.05 -.14
SR 1.0 -.65** -.51** -.39** -.00 .31** .38** .09 -.03
Knowl 1.0 .65** .44* .03 -.18* -.11 -.09 .04
Pt Stat 1.0 .37** .08 -.11 -.13 -.02 -.04
Age 1.0 .17* .02 -.02 .19* .17*
SF1 1.0 .28** .23** .31** .56**
SF2 1.0 .37** .06 .34**
SF3 1.0 .36** .27**
SF4 1.0 .20*
SF5 1.0
Note. * p < .05, ** p < .01

ST: Surgical Treatment Deviation Score; SR: Surgical Rationale Deviation Score; Knowl: Knowledge Score; Pt Stat: Patient Status
SF1: Cancer Concerns; SF2: Situational Characteristics; SF3: Individual Characteristics; SF4: Medical; SF5: Personal Concerns

51
Table 11

Correlations between Adjuvant Outcome Variables and Predictor Variables
AT AR Knowl Pt Stat Age AF1 AF2 AF3 AF4 AF5 AF6
AT 1.0 .79** -.64** -.51** -.41** -.15 .33** .16 .17* -.25** -.31**
AR 1.0 -.76** -.58** -.39** .11 .37** .13 .17* -.21* -.43**
Knowl 1.0 .65** .44* -.02 -.19* -.17 -.06 .33** .31**
Pt Stat 1.0 .37** .01 -.14 -.03 -.06 .33** .26**
Age 1.0 .09 .04 .14 -.03 .34** .39**
AF1 1.0 .41** .28** .35** .15 .10
AF2 1.0 .11 .40**-.02 -.15
AF3 1.0 .42** .18* .47**
AF4 1.0 .22** .15
AF5 1.0 .48**
AF6 1.0
Note. * p < .05, ** p < .01

AT: Adjuvant Treatment Deviation Score; AR: Adjuvant Rationale Deviation Score; Knowl: Knowledge Score; Pt Stat: Patient Status
AF1: Concerns; AF2: Situational Characteristics; AF3: Medical; AF4: Individual Characteristics; AF5: Cancer Characteristics; AF6: Invasiveness of
Cancer
52
All regression assumptions were examined and met. The residual plots were
examined and it was determined that the assumptions of linearity, normality, and
homoscedasticity were met. Multicollinearity was assessed with the Variance Inflation
Factor (VIF) and values were all well below four, indicating that multicollinearity was
not a problem. Finally, Mahalabobis distances were examined and it was determined that
outliers were not problematic. As all of the regression assumptions were examined and
met, data transformations were not necessary.
Surgical Treatment Deviation
The simultaneous (enter) regression method was used, and a significant model
emerged (F
8, 133
= 21.13, p < .001). Four variables were significant predictors of surgical
treatment decision deviations; their unique variability was .18. The significant predictors
were knowledge (sr
i
2
= .08), patient status (sr
i
2
= .02), the situational characteristics
factor (sr
i
2
= .02), and the individual characteristics factor (sr
i
2
= .06). All independent
variables in combination contributed an additional .38 in shared variability. The total R
2

was .56 (.53 adjusted). Table 12 presents the unstandardized regression coefficients (B)
and intercept, the standardized regression coefficients ( β), and the squared semi-partial
correlations (sr
i
2
).

53
Table 12
Summary of Simultaneous Regression Analyses for Variables Predicting Surgical
Treatment Decision Quality (N = 142)
Variable B SE B β sr
i
2

Constant 108.97 13.50
Age -0.25 0.14 -0.12 0.01
Patient Status -8.30 3.67 -0.18* 0.02
Knowledge Score -1.89 0.38 -0.41** 0.08
Cancer Concerns (F1) 0.01 0.12 0.01 0.00
Situational Characteristics (F2) 0.31 0.13 0.16* 0.02
Individual Characteristics (F3) 0.58 0.14 0.28** 0.06
Medical (F4) -0.04 0.21 -0.01 0.00
Personal Concerns (F5) -0.41 0.24 -0.12 0.01
Note. R
2
= .56 (adjusted R
2
= .53), F
8, 133
= 21.13 (p < .001); * p < .05, ** p < .01
Surgical Rationale Deviation
The simultaneous (enter) regression method was used, and a significant model
emerged (F
8, 133
= 20.15, p < .001). Five variables were significant predictors of surgical
rationale deviations; their unique variability was .36. The significant predictors were
knowledge (sr
i
2
= .18), age (sr
i
2
= .04), the situational characteristics factor
(sr
i
2
= .04), the individual characteristics factor (sr
i
2
= .06), and the personal concerns
factor (sr
i
2
= .05). All independent variables in combination contributed an additional .18
in shared variability. The total R
2
was .55 (.52 adjusted). Table 13 presents the
unstandardized regression coefficients (B) and intercept, the standardized regression

54
coefficients ( β), and the squared semi-partial correlations (sr
i
2
).
Table 13
Summary of Simultaneous Regression Analyses for Variables Predicting Surgical
Assessments of Reasoning (N = 142)
Variable B SE B β sr
i
2

Constant 99.02 11.45
Age -0.28 0.12 -0.16* 0.04
Patient Status -4.98 3.11 -0.13 0.02
Knowledge Score -1.75 0.32 -0.45** 0.18
Cancer Concerns (F1) 0.03 0.11 0.03 0.00
Situational Characteristics (F2) 0.26 0.11 0.16* 0.04
Individual Characteristics (F3) 0.33 0.12 0.19** 0.06
Medical (F4) -0.29 0.18 -0.11 0.02
Personal Concerns (F5) -0.52 0.20 -0.19** 0.05
Note. R
2
= .55 (adjusted R
2
= .52), F
8, 133
= 20.15 (p < .001); * p < .05, ** p < .01
Adjuvant Treatment Deviation
The simultaneous (enter) regression method was used, and a significant model
emerged (F
9, 132
= 15.77, p < .001). Three variables were significant predictors of
adjuvant treatment decision deviations; their unique variability was .20. The significant
predictors were knowledge (sr
i
2
= .11), age (sr
i
2
= .04), and the situational characteristics
factor (sr
i
2
= .04). All independent variables in combination contributed an additional .32
in shared variability. The total R
2
was .52 (.49 adjusted). Table 14 presents the
unstandardized regression coefficients (B) and intercept, the standardized regression

55
coefficients ( β), and the squared semi-partial correlations (sr
i
2
).
Table 14
Summary of Simultaneous Regression Analyses for Variables Predicting Adjuvant
Treatment Decision Quality (N = 142)
Variable B SE B β sr
i
2

Constant 83.95 10.32
Age -0.26 0.11 -0.18* 0.04
Patient Status -5.11 2.83 -0.15 0.02
Knowledge Score -1.25 0.31 -0.37** 0.11
Concerns (F1) 0.05 0.06 0.06 0.01
Situational Characteristics (F2) 0.25 0.10 0.18* 0.04
Medical (F3) 0.31 0.18 0.15 0.02
Individual Characteristics (F4) 0.00 0.11 0.00 0.00
Cancer Characteristics (F5) 0.04 0.21 0.02 0.00
Cancer Invasiveness (F6) -0.69 0.39 -0.15 0.02
Note. R
2
= .52 (adjusted R
2
= .49), F
8, 133
= 14.10 (p < .001);* p < .05, ** p < .01
Adjuvant Rationale Deviation
The simultaneous (enter) regression method was used, and a significant model
emerged (F
9, 132
= 36.01, p < .001). Six variables were significant predictors of adjuvant
rationale deviations; their unique variability was .67. The significant predictors were
knowledge (sr
i
2
= .31), patient status (sr
i
2
= .04), the situational characteristics factor
(sr
i
2
= .07), the medical factor (sr
i
2
= .03), the cancer characteristics factor (sr
i
2
= .06),
and the cancer invasiveness factor (sr
i
2
= .15). All independent variables in combination

56
only contributed an additional .04 in shared variability. The total R
2
was .71
(.69 adjusted). Table 15 presents the unstandardized regression coefficients (B) and
intercept, the standardized regression coefficients ( β), and the squared semi-partial
correlations (sr
i
2
).
Table 15
Summary of Simultaneous Regression Analyses for Variables Predicting Adjuvant
Assessments of Reasoning (N = 142)
Variable B SE B β sr
i
2

Constant 83.33 7.55
Age -0.07 0.81 -0.05 0.01
Patient Status -4.86 2.07 -0.15* 0.04
Knowledge Score -1.73 0.22 -0.54** 0.31
Concerns (F1) -0.01 0.04 -0.01 0.00
Situational Characteristics (F2) 0.24 0.08 0.19** 0.07
Medical (F3) 0.28 0.13 0.14* 0.03
Individual Characteristics (F4) 0.02 0.08 0.02 0.00
Cancer Characteristics (F5) 0.43 0.16 0.16** 0.06
Cancer Invasiveness (F6) -1.40 0.28 -0.32** 0.15
Note. R
2
= .71 (adjusted R
2
= .69), F
8, 133
= 32.23 (p < .001); * p < .05, ** p < .01
Separate Regression Analyses for Patients and Non-Patients
Separate regression analyses were also conducted for the patient and non-patient
groups to determine whether the predictive relationships were the same. Correlations
were strongest when the two groups were combined. When the groups were separated,

57
the non-patients had stronger correlations between knowledge and each of the four
decision outcomes than did the patient group. With one exception, all correlations were
significant (at the .05 level) for both groups: for the patient group, the correlation
between knowledge and surgical treatment recommendations was not significant.
Regression analyses were also examined for the separate groups and the regressions were
more likely to be significant for the non-patient group (all four were significant) than the
non-patient group (only 1 was significant at the p < .05 level; an additional regression had
a p-value of .056). The unique contributions of the individual predictors were also
examined, and knowledge was the largest predictor for the four decision quality outcome
variables, and this was the case for both patients and non-patients. Separating the groups
resulted in a significant loss of power and it was determined that the initial data analysis
method of combining the groups was the most appropriate method.
PARTICIPATION LEVEL
The level of participation that participants (both patients and non-patients)
indicated they would have desired, if they were the patients in the scenarios, was assessed
for the two surgical and the two adjuvant treatment decisions. The most common level of
participation that participant indicated they preferred was the collaborative role; 85.3% to
88.7% of participants selected this role (varied slightly across scenarios). Between 7.5%
and 11.9% of participants indicated that they preferred the active role with the final
treatment decision left completely up to them. Across the four treatment decisions, no
patients indicated that the physician should make the treatment decision without patient
involvement, while only 1.9%-2.8% of non-patients indicated that they would prefer to
leave the treatment decision up to the physician. Significant differences were found for

58
preferred participation level between patients and non-patients for both surgical scenarios
( χ
2
= 6.55, p < .05; χ
2
= 9.07, p < .05). Significant differences were not found for
preferred participation level between patients and non-patients for the adjuvant treatment
scenarios. Patients were also asked to report the level of participation that they actually
achieved in their own treatment. The most common role was the collaborative role
(67.9%) followed by the active role (30.2%). Only 1.9% of patients indicated that they
had a passive role in their treatment participation.
59
Chapter 4: Discussion
Previous research established that patients knew more about breast cancer than
their non-patient counterparts (Brown, 2003; Vaeth, 1993). However, the more important
question of whether increased knowledge led to better, more informed decisions has
never been examined. If patients are expected to be truly active participants in their
treatment decisions, it is important to understand if they have the ability to participate.
Therefore, one goal of this research was to develop and assess instruments that measured
not only breast cancer knowledge, but also participants’ decision making ability and their
comprehension of the causal explanations that supported those decisions. An additional
goal was to better understand the role of information use in treatment decision making.
MEAN DIFFERENCES IN KNOWLEDGE AND DECISION QUALITY
Building upon previous research (Brown, 2003), the breast cancer knowledge
measure was revised and administered to the patient and non-patient groups. It was
expected that patients would outperform the non-patient group, as patients have been
through treatment and have likely learned a great deal about breast cancer from their
physicians and nurses, as well as from other patients and written materials. The first
hypothesis, that patients would have higher breast cancer knowledge scores than their
non-patient counterparts, was supported. Out of a possible score of 26, patients had a
mean score of 22.4 (SD = 2.2) while non-patients had a mean score of 15.8 (SD = 4.5).
Patients also had a higher minimum score than the non-patients (16 versus 6),
demonstrating that even the lowest scoring patients had higher knowledge scores than the
mean of the non-patient group.

60
The knowledge instrument was designed to cover four broad areas of breast
cancer knowledge and to be comprehensive and representative of what patients need to
know to make treatment decisions. However, the factor analysis that was performed on
the knowledge measure did not yield the proposed factor structure based on the four
content areas proposed (diagnosis, treatment options, treatment outcomes, and
physiology). It was not surprising that an underlying structure did not emerge, as patients
are typically exposed to medical information in an unorganized manner (McGuire, 1996).
The psychometric analyses presented previously demonstrate that the breast cancer
knowledge instrument is both a valid and reliable instrument.
This is currently where most of the previous research ends, with a demonstration
that instruments are reliable and valid and patients have more knowledge than their non-
patient counterparts (Brown, 2003; Vaeth, 1993). However, the more interesting question
that the present research addresses is: does increased breast cancer knowledge translate
into better treatment decisions? This question was examined in the second hypothesis,
which asserted that patients would make better treatment decisions than their non-patient
counterparts. This hypothesis was clearly supported by the detailed analyses presented in
the results section. Patients scored significantly better than non-patients on both the
surgical treatment decision scale and the adjuvant treatment decision scale. Keeping in
mind that lower deviation scores corresponded with better decisions, for the surgical
treatment decisions, patients had a mean score of 48.8 (SD = 18.9) while non-patients had
a mean score of 74.0 (SD = 19.9). The mean score of the patients was 25.2 points lower,
or 34% better than non-patients. For the adjuvant treatment decisions, patients had a
mean score of 31.7 (SD = 14.5) and the non-patients had a mean score of 49.5

61
(SD = 14.6). This was a mean difference of 17.8, with patients performing 36% better
than non-patients.
When examining only the treatments that the physicians gave the top rating of
nine, only the surgical treatment option of mastectomy plus lymph node surgery plus
reconstruction resulted in significant differences between patients and non-patients. This
was the case for this option in both scenarios. It is not clear why patients rated this option
as significantly better than non-patients. Perhaps it is the case that women who have been
through treatment view reconstruction as more important than women who have not
experienced breast cancer, though further research would be necessary to confirm this.
The third hypothesis was that patients would exhibit better comprehension than
the non-patient group for the causal reasons supporting treatment decisions. This
hypothesis was also supported for both the surgical decision rationales and the adjuvant
decision rationales. Again, note that lower scores corresponded with better responses. For
the surgical decision rationale support, patients had a mean score of 21.0 (SD = 11.8)
while non-patients had a mean score of 41.1 (SD = 18.9). The difference in mean scores
was 21.2, with patients performing 51% better than non-patients. Additionally, the
difference in standard deviations indicates that patients not only scored better on the
surgical rationales, they were also more homogeneous in their responses than the non-
patients. Furthermore, patients outperformed the non-patient group for the adjuvant
decision rationale support. Patients had a mean score of 13.0 (SD = 9.0) and non-patients
had a mean score of 31.9 (SD = 15.0); a difference of 18.9. This was the largest
difference, as a percentage, with patients scoring 59% better than non-patients. Again, in
addition to patients supporting more sound rationales, they were also more

62
homogeneous in their responses, as reflected by the smaller deviation around the mean.
The more homogeneous responses given by patients are not surprising, in that even those
patients who had the lowest causal understanding about what treatments were better than
others were likely exposed to large amounts of information during their consultations
with physicians and nurses and while undergoing treatment.
The patient group scored significantly better than the non-patient group on all four
treatment-related decision outcomes. The largest mean difference between groups was for
the surgical treatment scores, while the smallest mean difference was for adjuvant
treatment scores; however this was an artifact of the differences in possible ranges for the
various measures. When looking at the percentage differences between patients and non-
patients as a percentage, the smaller differences were for the mean treatment decision
deviation scores, both surgical (34%) and adjuvant (36%), while the mean scores for the
rationale support showed larger differences, 51% for surgical rationales and 59% for
adjuvant rationales. This is perhaps related to breast cancer knowledge. Knowledge is a
larger predictor for both surgical and adjuvant rationale support, than for the
corresponding treatment decisions. As patients have higher knowledge scores, it follows
that the differences between patients and non-patients would be larger when knowledge
plays more of a role.
PREDICTING DECISION QUALITY
In addition to demonstrating that the knowledge and decision quality instruments
discriminated between patients and non-patients, regression analyses were conducted to
determine which variables predicted decision quality. An examination of the regression
analyses of the separate groups (patients and non-patients) revealed similar patterns,

63
with a significant loss of power, particularly for the patient sample which was smaller
and had more restricted ranges on all scales. Therefore, the regressions with the two
groups combined (and patient status entered as a predictor) were used as the main
analyses.
For the regression analyses, knowledge was entered as an independent variable, as
were age and patient status. Additionally, the factor scores for the structures that emerged
from the information use scales were entered into the regression equations as predictors.
Specifically, the five surgical information use factor scores were entered as predictors of
the two surgical dependant variables (ST and SR), while the six adjuvant information use
factor scores were used as predictors of the two adjuvant dependant variables (AT and
AR).
In all four regression analyses, knowledge was the largest predictor of decision
quality, which was the expected outcome. Patient status, the independent variable
expected to be the next largest predictor of decision quality was only a significant
predictor of the surgical treatment recommendations and adjuvant rationale support,
while age was only a significant predictor for surgical rationale support and adjuvant
treatment recommendations. Despite the large, significant correlations between patient
status and age with each of the four outcome variables, in both cases, only patient status
or age was significant, never both in the same regression equation. There is a relationship
between age, patient status, and knowledge. Patients were significantly older than non-
patients and had more breast cancer knowledge. Further, age and breast cancer
knowledge were significantly correlated to one another. However, when you enter all

64
three variables into the regression equation, it is knowledge that accounts for the most
variance, as it is the most important predictor, not patient status or age alone.
A common finding emerged for each of the four dependent variables with regard
to the information use factor scores. The situational characteristics factor score was a
significant predictor in each regression (this was a common factor for both surgical and
adjuvant information use). The situational characteristics factor consisted of three
information items; occupation, relationship status, and living in an urban area. The items
that loaded on this factor had the three lowest mean scores, indicating their relative lack
of importance to participants when making treatment decisions. There was a significant,
positive correlation between scores on this factor and each of the four outcome variables;
higher scores on this factor were significantly correlated with higher deviation scores
(poorer decisions). As the information contained in this factor was rated by participants
as the least important, it logically follows that a relationship exists between rating this
information as important and making poor decisions and supporting unsound rationales.
With regard to the surgical outcomes, the individual characteristics factor also
emerged as a significant predictor for both treatment selection and rationale support. This
factor consisted of three items: age, menopausal status, and childbearing status. There
were significant positive correlations between scores on this factor and surgical treatment
recommendations and surgical rationale support; again, participants who had higher
scores on this factor had higher deviation scores, indicating worse decisions. One factor
that was expected to predict decision quality, but was not predictive of either surgical
treatment recommendations or surgical rationale support, was the medical factor. The
items that loaded on the medical factor of the surgical information use scale had the

65
highest mean scores, indicating participants thought this information was the most
important to have when making surgical treatment decisions. However, this factor was
not predictive of surgical treatment recommendations or surgical rationale support. The
likely explanation is a ceiling effect, whereas the majority of participants rated items that
loaded on this factor as very important in their decision making, thereby eliminating its
predictive ability.
When examining the two adjuvant outcome variables, the only common
information use predictor was the situational characteristics factor, which was previously
discussed as a predictor of poor performance for all four dependant variables. No other
factor scores were predictive of adjuvant treatment recommendations. Three additional
information use factor scores were predictive of adjuvant rationale support: medical,
cancer characteristics, and invasiveness. However, only the scores on the invasiveness
factor were predictive of better treatment decisions. Higher scores on the medical and
cancer characteristics factors were predictive of worse treatment decisions. The adjuvant
rationales had the largest amount of variance explained (R
2
= .71), most of it unique
variability. Adjuvant rationale support also had the largest difference between patients
and non-patients, with patients performing 59% better than non-patients.
One possible explanation for why patients performed better on the adjuvant
rationale scale is that they are likely more involved in their adjuvant treatments. This
involvement may include traveling to medical centers for repeated treatments (such as
radiation or chemotherapy) or involvement in self-care to reduce the common side effects
of many of the adjuvant treatments. As non-patients also performed best on the adjuvant
rationale scale, an additional explanation is that many adjuvant treatments, such as

66
radiation and chemotherapy, are common across different types of cancer. Therefore, the
likelihood of being exposed to the information is greater, even if the participants do not
have any direct or indirect experience with breast cancer.
It is of import that almost all of the significant relationships for the information
use factors were in the direction of information use being a predictor of poor decision
quality. That is, participants who paid attention to the less important information
performed worse on the instruments. For the surgical and adjuvant treatment decisions,
all of the significant predictors were in the direction of information use being a poor
predictor of decision quality. For rationale support, both surgical and adjuvant, five of the
seven significant predictors were also in the direction of information use being a
predictor of worse support for causal rationales.
Further, on the adjuvant rationale scale, two information factors that had high
mean importance scores were also predictive of worse decisions. For the cancer
characteristics factor, a likely explanation was suppression, the cancer characteristics
factor had a significant negative correlation with adjuvant rationale support, but a
significant positive β. This was not the case for the medical factor, which was positively
(non-significantly) correlated with adjuvant rationale support and also had a significant
positive β. It is clear that participants believed that the information contained in the
medical factor was important for making their treatment decisions. It is unclear why those
high ratings were not predictive of better support for decisions. It is possible that
participants recognized the information as important, yet they were unable to use that
information properly when rating the rationales for making adjuvant treatment decisions.
The overall picture from the information use results showed that participants who paid

67
attention to the wrong information tended to make worse treatment decisions. This
finding should be valuable for designing educational materials.
INFORMATION USE
Previous research that examined information use typically asked participants to
rate the importance of broad categories of information, such as “information about side
effects” and “information about various treatments” (Bilodeau & Degner, 1996; Degner
et al., 1997; Luker et al., 1995; Luker et al., 1996). In the current research, an attempt was
made to move beyond the broad categories and hone in on the specific pieces of
information that a patient would consider and ask participants to indicate how important
that information was, when making their treatment decisions. Previous researchers
reported that likelihood of cure, disease stage, and information about treatment options
were the most important categories of information, while affect on sexuality was the least
important category of information (Bilodeau & Degner; Degner et al.; Luker et al.).
Although the information presented in the current research was not the same, the
results were similar to those reported by other researchers. For the surgical scenarios, the
three most important pieces of information were results from biopsy, type of cancer, and
size of cancer, while the three most important pieces of information for the adjuvant
scenarios were type of cancer, results from lymph node surgery, and grade of cancer.
These findings supported the general categories of information that patients reported as
important in previous research, in that likelihood of cure and disease stage are both
categories of information that would be ascertained from type and size of cancer, as well
as from results from biopsy and lymph node dissection. Additionally, the least important
pieces of information for both surgical and adjuvant scenarios were occupation, living

68
in an urban area, and relationship status. This was also in support of previous research
that showed that personal information was rated as less important than medical
information (Bilodeau & Degner, 1996; Degner et al., 1997; Luker et al., 1995; Luker et
al., 1996). While previous research mostly asked patients what information they thought
was important, this research asked participants to indicate the importance of the
information, when making treatment recommendations. The results from this research
support general information needs reported by other researchers. The current research
also offers insight into the information used and decisions made by participants,
specifically, that paying attention to the wrong information can lead to worse decisions.
PARTICIPATION LEVEL
An additional item was presented to participants to determine the desired level of
participation that patients and non-patients indicated they would like to have, if they were
making the decisions presented in the scenarios. Based on previous literature (Beaver et
al., 1996; Bilodeau & Degner, 1996; Degner et al., 1997; Degner & Sloan, 1992; Hack et
al, 1994), it was expected that participants would desire a collaborative role, followed by
a passive role, and the least preferred role would be the active role. Across the four
treatment decisions (two surgical and two adjuvant), the most commonly desired role was
the collaborative role, with 85.3% to 88.7% of participants indicating they preferred the
collaborative role. The second most preferred role was the active role, with 7.5% to
11.9% of participants choosing that role. The passive role was only selected by 1.9% to
2.8% of participants, all of them non-patients. There were significant differences between
patient and non-patients for their desired role in making surgical treatments, though not
for the adjuvant decisions.

69
The actual participation reported by the patients differed from their preferred
participation for the scenario treatment decisions. Between 80% and 89.4% of patients
reported they preferred the collaborative role in the scenarios, while the remainder of
patients reported they desired the active role. However, 67.9% of patients reported their
actual participation in their own treatment decision making was reflective of the
collaborative role, followed by 30.2% of patients who reported they assumed an active
role in their own treatment. This differs from previous research, which typically reported
that patients assumed less active roles in their own treatment than in hypothetical
scenarios (Beaver et al., 1996; Degner & Sloan, 1992).
Due to the small percentages of participants who selected either active or passive
roles, group differences in knowledge and treatment were not examined with regard to
preferred participation level. The participation level was presented to participants with
three options: prefer to make the decision completely on my own, prefer to make the
decision in collaboration with a physician, and prefer to leave the decision regarding
treatment to the physician. In the future, those choices should be expanded to include
more alternatives, which might lead to more variation in responses. This would allow
researchers to determine if those participants who prefer more active roles in their
treatment decisions do in fact have more ability to make better treatment decisions. Lack
of ability, or perceived ability, was an often cited reason that patients may not desire or
assume an active role in their own treatment decisions (Beaver et al., 1996; Charles et al.,
1997; Sutherland et al., 1989) and research is necessary to confirm this claim.
70
IMPORTANCE OF THE RESEARCH
Many of the previous attempts to measure patients’ breast cancer treatment
decision making relied on measures that did not fully capture the complexity of those
decisions. Knowledge instruments typically had a very narrow scope, often focusing on
specific types of breast cancer or specific areas of knowledge, such as diagnostic
information. Much of the previous research exploring breast cancer treatment decision
making examined either desire to participate in treatment decisions or desire for
information, with little effort to determine if patients were prepared for their decision
making responsibilities. Research that moved beyond desire to participate and desire for
information was mostly descriptive in nature; again, with little effort to ascertain the
quality of the treatment decisions that patients made.
The current research sought to fill in some of the gaps in the literature by
measuring both breast cancer knowledge and treatment decision making ability. The
knowledge instrument developed for this research encompassed four broad areas of
breast cancer knowledge, incorporating items that breast cancer specialists believed
patients needed to know when making treatment decisions. The decision quality measure
attempted to ascertain participants’ preparedness to make treatment decisions. To this
end, participants not only selected treatments for patient cases presented in hypothetical
scenarios, but they also judged the quality of statements that explained those decisions.
This process allowed for an objective measure of participants’ understanding of the
strengths or weaknesses of various treatment options. The unique contribution of this
research was this development of reliable and valid instruments that measured breast
cancer knowledge, treatment decision quality, and understanding of rationales for

71
selecting various treatment options. The relationships between these outcomes provided
new information about patient preparedness for participating in their own treatment
decisions.
Results from this research showed that increased breast cancer knowledge
translated into better treatment decisions. This is an important finding, as previous
research has not directly assessed this relationship. Further, based on the results of the
regression analyses, it was concluded that participants who paid attention to less
important information tended to make worse treatment decisions, though a relationship
did not appear between using information rated as more important and making better
treatment decisions. Based on these findings, a description of an optimal treatment
decision maker would be someone who has a lot of technical knowledge, in this case
about breast cancer, and is not easily distracted by irrelevant information.
Educational interventions should be designed that help patients not only gain
knowledge about breast cancer and treatments, but also teach them what information to
pay attention to, and perhaps just as importantly, what information to ignore.
Psychological research has long demonstrated that people have limits in attention and
working memory abilities (Miller, 1956). If people pay attention to a limited amount of
information, it makes sense that they should pay attention to the best, most relevant
information. Experts in the field of breast cancer (physicians and nurses) should assist
patients in determining what information is most relevant to the decisions they are
making, allowing them to ignore less important information. In addition to providing
valuable information about the quality of treatment decisions and the design of future
educational interventions, this research also provides a template for conducting similar

72
research in other medical areas where patients are expected to take an active involvement
in their treatment decisions.
LIMITATIONS OF THE RESEARCH
Both the knowledge measure and the decision quality measures were valid and
reliable instruments of breast cancer knowledge and decision making ability. However,
some limitations exist. The scenarios were designed with the assistance of two breast
cancer specialists (a medical oncologist and a breast surgeon) who supplied both the
medical and personal information that is typically presented in medical consultations with
patients. However, when designing the scenarios, we sought to completely remove the
physician’s opinion from the scenarios (i.e., there was no opinion offered by a physician
as to what course of treatment would be best). While this is not completely realistic,
previous research has clearly demonstrated the importance of the physician’s opinion to
patients (Johnson et al., 1996; Katz et al., 2001; Kotwall et al., 1996; Sepucha et al,
2007). Offering a physician’s opinion would have certainly influenced the decisions that
participants made, and the goal of the research was to measure participant’s unaided
treatment decisions.
Additionally, it was not feasible to represent every breast cancer case in the
decision scenarios because of the time and effort limits that participants were willing to
invest in research. This research was an important first step in developing and evaluating
instruments that may prove useful in clinical settings. The scenarios used in this research
were designed to be representative of the most common cases of breast cancer, in terms
of stage of cancer, lymph node involvement, and personal characteristics of the patient. It
is important to note that many different scenarios could have been used instead. It

73
would be useful to conduct research examining patients’ actual decision making and
information use patterns. However, the goal of this research was to develop an objective
measure of treatment decision making that could be used to make comparisons among
different groups of participants who were presented with identical information and
decisions. Additionally, the use of an objective measure allowed for the evaluation of the
role of breast cancer knowledge and information use in making those decisions.
The second limitation to this study was the utilization of a relatively small,
convenience sample. Several data collection methods were considered. The first was to
have physicians recruit consecutive patients. However, the two physicians who assisted
with the development of the scenarios were unable to assist with patient recruitment.
Permission was sought to recruit patients from waiting rooms at USC/Norris
Comprehensive Cancer Center; however, that permission was never granted.
Additionally, we contacted the California Breast Cancer Registry to recruit a
consecutively diagnosed sample of patients. The only patient sample available through
this method was a sample of non-Hispanic white cases diagnosed in the first part of 2005.
As this sample was not going to be a racially/ethnically diverse sample, we decided to
recruit participants via the internet, utilizing online messaging boards. This recruitment
method may have resulted in participants who were more interested in, or more
knowledgeable about, breast cancer choosing to participate. This may have occurred even
without the use of convenience samples. Further, had the non-patient population been
more interested in and knowledgeable about breast cancer, this should have worked
against finding large mean differences between the groups.

74
The majority of participants in the current research were white and highly
educated. However, patients and non-patients were similar in terms of race/ethnicity and
education. The major differences found in the current research are unlikely to be due to
comparing very different groups of people across the patient and non-patient divide.
Given the strength of the findings, we are optimistic that these results will hold for
participants from other demographic groups. However, until data is collected with a
larger, more representative, sample, broad generalization must be limited.
Further, if the instruments are administered to a larger sample of patients and non-
patients, more sophisticated data analytic techniques, such as structural equation
modeling, can be utilized. Subsets of knowledge, based on the four proposed categories
of the knowledge instrument could also be examined to determine if some knowledge is
more important for patients to know, when they are making treatment decisions. This
should provide for a more complete understanding of the relationships between
participant characteristics, knowledge, information use, decision quality, and rationale
support.
CONCLUSIONS
Despite the limitations to the current research, it was the first attempt to measure
the quality of patient decisions regarding breast cancer treatment. This research moved
beyond merely measuring breast cancer knowledge or only reporting treatment selection.
The current research also assessed the quality of breast cancer treatment decisions and the
comprehension of the underlying explanations for why those treatments were appropriate.
The instruments measured the deviations from physician recommendations, allowing for
the evaluation of the quality of patients’ treatment decisions. Additionally, by focusing

75
on the information rated by participants as important, determinations were made as to
whether they were paying attention to the wrong information, while missing the crucial
information that is necessary to make informed treatment decisions.
This research should prove useful in the development and assessment of
educational interventions that can aid patients in making informed breast cancer
treatment decisions. Feedback received by participants in this research was
overwhelmingly positive. Participants reported that the instruments were challenging,
though not too difficult as to dissuade them from completing them. These instruments, or
revised versions, could also serve a useful role in clinical settings by helping physicians
(or nurses) ascertain whether their patients are adequately informed and ready for active
involvement in their breast cancer treatment. Finally, this research can serve as a template
for measuring patient preparedness in other medical areas where patients are encouraged
to actively participate in their treatment decisions.

76
References
Auerbach, S. M. (2000). Should patients have control over their won health care?
Empirical evidence and research issues. Annals of Behavioral Medicine, 22,
246– 259.

Barry, M. J. (1999). Involving patients in medical decisions: How can physicians do
better? Journal of the American Medical Association, 282, 2356–2357.

Bartholome, W. G. (1992). A revolution in understanding: How ethics has transformed
health care decision making. Quality Review Bulletin, 18, 6–11.

Beaver, K., Luker, K. A., Owens, R. G., Leinster, S. J., & Degner, L. F. (1996).
Treatment decision making in women newly diagnosed with breast cancer. Cancer
Nursing, 19, 8–19.

Bilodeau, B. A. & Degner, L. F. (1996). Information needs, sources of information, and
decisional roles in women with breast cancer. Oncology Nursing Forum, 23,
691–696.

Birnbaum, M. H. (2001). Introduction to behavioral research on the internet. Upper
Saddle River, NJ: Prentice-Hall, Inc.

Bluman, L. G., Borstelmann, N. A., Rimer, B. K., Iglehart, J. D., & Winer, E. P. (2001).
Knowledge, satisfaction, and perceived cancer risk among women diagnosed with
ductal carcinoma in situ. Journal of Women’s Health & Gender Based Medicine,
10, 589–598.

Broadstock, M. & Michie, S. (2000). Processes of patient decision making: Theoretical
and methodological issues. Psychology and Health, 15, 191–204.

Brown, C. E. (2003). The development of a scale to measure breast cancer knowledge.
Unpublished manuscript, University of Southern California, Los Angeles.

Bruera, E., Willey, J. S., Palmer, J. L. & Rosales, M. (2002). Treatment decisions for
breast carcinoma: Patient preferences and physician perceptions. Cancer, 94,
2076–2080.

California Department of Health Services (1999). A woman’s guide to breast cancer
diagnosis and treatment (4
th
ed). [Brochure]. Sacramento, CA: Breast Cancer Early
Detection Program.

Cassileth, B. R., Zupkis, R. V., Sutton-Smith, K. & March, V. (1980). Information and
participation preferences among cancer patients. Annals of Internal Medicine, 92,
832–836.

77
Chagpar, A. B., Studts J. L., Scoggins, C. R., Martin, R. C. G., II, Carlson, D. J., Laidley,
A. L., et al., (2006). Factors associated with surgical options for breast carcinoma.
Cancer, 106, 1462–1466.

Chapman, G. B., Elstein, A. S., & Hughes, K. K. (1995). Effects of patient education on
decisions about breast cancer treatments: A preliminary report. Medical Decision
Making, 15, 231–239.

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 &
Medicine, 44, 681–692.

Charles, C. Gafni, A., & Whelan, T. (1999). Decision-making in the physician–patient
encounter: Revisiting the shared treatment decision making model. Social Science
& Medicine, 49, 651–661.

Comrey, A. L. & Lee, H. B. (1992). A First Course in Factor Analysis. Hillsdale, NJ:
Lawrence Erlbaum Associates, Publishers.

Crocker, L. & Algina, J. (1986). Introduction to Classical and Modern Test Theory. Fort
Worth, TX: Holt, Rinehart, and Winston, Inc.

Degner, L. F. & Sloan, J. A. (1992). Decision making during serious illness: What role do
patients really want to play? Journal of Clinical Epidemiology, 45, 941–950.

Degner, L. F., Kristjanson, L. J., Bowman, D., Sloan, J. A., Carriere, K. C., O’Neil, J., et
al. (1997). Information needs and decisional preferences in women with breast
cancer. Journal of the American Medical Association, 277, 1485–1492.

Fagerlin, A., Lakhani, I., Lantz, P. M., Janz, N. K., Morrow, M., Schwartz, K., et al.,
(2006). An informed decision? Breast cancer patients and their knowledge about
treatment. Patient Education and Counseling, 64, 303–312.

Fisher, B., Anderson, S., Bryant, J., Margolese, R. G., Deutsch, M., Fisher, E. R., et al.
(2002). Twenty-year follow up of a randomized study comparing total mastectomy,
lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast
cancer. New England Journal of Medicine, 347, 1233–1241.

Hack, T. F., Degner, L. F., & Dyck, D. G. (1994). Relationship between preferences for
decisional control and illness information among women with breast cancer: A
quantitative and qualitative analysis. Social Science & Medicine, 39, 279–289.

Hawley, S. T., Pantz, P. M., Janz, N. K., Salem, B., Morrow, M., Schwartz, K., et al.
(2007). Factors associated with patient involvement in surgical treatment decision
making for breast cancer. Patient Education and Counseling, 65, 387–395.

78
Howell, D. C. (2007). The analysis of missing data. In W. Outhwaite & S. Turner (Eds.),
The Sage Handbook of Social Science Methodology (pp. 208–224). London: Sage.

Hughes, K. K. (1993). Decision making by patients with breast cancer: The role of
information in treatment selection. Oncology Nursing Forum, 20, 623–628.

Johnson, J. D., Roberts, C. S., Cox, C. E., Reintgen, D. S., Levine, J. S. & Parsons, M.
(1996). Breast cancer patients’ personality style, age, and treatment decision
making. Journal of Surgical Oncology, 63, 183–186.

Kasper, J. F., Mulley, A. G., & Wennberg, J. E. (1992). Developing shared decision-
making programs to improve the quality of health care. Quality Review Bulletin, 18,
183–190.

Katz, S. J., Lantz, P. M., & Zemencuk, J. K. (2001). Correlates of surgical treatment type
for women with noninvasive breast cancer. Journal of Women’s Health & Gender
Based Medicine, 10, 659–670.

Keating, N. L., Guadagnoli, E., Landrum, M. B., Borbas, C., & Weeks, J. C. (2002).
Treatment decision making in early-stage breast cancer: Should surgeons match
patients’ desired level of involvement? Journal of Clinical Oncology, 20,
1473–1479.

Keleman, J. J., III, Poulton, T., Swartz, M. T., & Jatoi, I. (2001). Surgical treatment of
early-stage breast cancer in the Department of Defense healthcare system. Journal
of the American College of Surgeons, 192, 293–297.

Kotwall, C. A., Maxwell, J. G., Covington, D. L., Churchill, P., Smith, S. E., & Covan, E.
K. (1996). Clinicopathologic factors and patient perceptions associated with
surgical breast-conserving treatment. Annals of Surgical Oncology, 3, 169–175.

Lantz, P. M., Janz, N. K., Fagerlin, A., Schwartz, K., Liu, L., Lakhani, I., et al. (2005).
Satisfaction with surgery outcomes and the decision process in a population-based
sample of women with breast cancer. Health Services Research, 40, 745–767.

Levine, M. N., Gafni, A., Markham, B., & MacFarlane, D. (1992). A bedside decision
instrument to elicit a patient’s preference concerning adjuvant chemotherapy for
breast cancer. Annals of Internal Medicine, 117, 53–58.

Luker, K. A., Beaver, K., Leinster, S. J., Owens, R. G., Degner, L. F., & Sloan, J. A.
(1995). The information needs of women newly diagnosed with breast cancer.
Journal of Advanced Nursing, 22, 134–141.

Luker, K. A., Beaver, K., Leinster, S. J., & Owens, R. G. (1996). Information needs and
sources of information for women with breast cancer: A follow-up study. Journal of
Advanced Nursing, 23, 487–495.

79

Mallinger, J. B., Griggs, J. J., & Shields, C. G. (2005). Patient-centered care and breast
cancer survivors’ satisfaction with information. Patient Education and Counseling,
57, 342–349.

Mastaglia, B. & Kristjanson, L. J. (2001). Factors influencing women’s decisions for
choice of surgery for Stage I and Stage II breast cancer in Western Australia.
Journal of Advanced Nursing, 35, 836–847.

McCance, K. L., Mooney, K. H., Smith, K. R., & Field, R. (1990). Validity and reliability
of a breast cancer knowledge test. American Journal of Preventative Medicine, 6,
93–98.

McGuire, L. C. (1996). Remembering what the doctor said: Organization and adults’
memory for medical information. Experimental Aging Research, 22, 403–428.

Meyer, B. J. F., Russo, C. & Talbot, A. (1995). Discourse comprehension and problem
solving: Decisions about the treatment of breast cancer by women across the life
span. Psychology and Aging, 10, 84–103.

Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our
capacity for processing information. The Psychological Review, 63, 81–97.

Molenaar, S., Sprangers, M., Oort, F., Rutgers, E., Luiten, E., Mulder, J., et al. (2007).
Exploring the black box of a decision aid: What information do patients select from
an interactive cd-rom on treatment options in breast cancer? Patient Education and
Counseling, 65, 122–130.

Morris, J. & Ingham, R. (1988). Choice of surgery for early breast cancer: Psychosocial
considerations. Social Science & Medicine, 27, 1257–1262.

Muss, H. B., Hunter, C. P., Wesley, M., Correa, P., Chen, V. W., Greenberg, R. S. et al.,
(1992). Treatment plans for black and white women with stage II node-positive
breast cancer. Cancer, 70, 2460–2467.

National Institutes of Health (2000). What you need to know about breast cancer. (NIH
Publication No. 00-1556). Bethesda, MD: National Cancer Institute.

National Institutes of Health (1998). Understanding breast cancer treatment: a guide for
patients. (NIH Publication No. 98-4251). Bethesda, MD: National Cancer Institute.

Nold, R. J., Beamer, R. L., Helmer, S. D., & McBoyle, M. F. (2000). Factors influencing
a woman’s choice to undergo breast-conserving surgery versus modified radical
mastectomy. American Journal of Surgery, 180, 413–418.

80
Nunnally, J. C. & Bernstein, I. H. (1994). Psychometric Theory. New York: McGraw-
Hill, Inc.

Ondrusek, N., Warner, E., & Goel, V. (1999). Development of a knowledge scale about
breast cancer and heredity (BCHK). Breast Cancer Research and Treatment, 53,
69–75.

Petrisek, A. C., Laliberte, L. L., Allen, S. M. & Mor, V. (1997). The treatment decision-
making process: Age differences in a sample of women recently diagnosed with
nonrecurrent, early-stage breast cancer. The Gerontologist, 37, 598–608.

Pierce, P. F. (1993). Deciding on breast cancer treatment: A description of decision
behavior. Nursing Research, 42, 22–28.

Roth, P. L. (1994). Missing data: A conceptual review for applied psychologists.
Personnel Psychology, 47, 537–560.

Schafer, J. L. (1997). Analysis of incomplete multivariate data. London: Chapman &
Hall.

Schafer, J. L. (1999). Software for multiple imputation. Retrieved April 7, 2009, from
http://www.stat.psu.edu/~jls/misoftwa.html

Sepucha, K., Ozanne, E., Silvia, K., Partridge, A., & Mulley, A. G., Jr. (2007). An
approach to measuring the quality of breast cancer decisions. Patient Education and
Counseling, 65, 261–269.

Siminoff, L. A. (1989). Cancer patient and physician communication: Progress and
continuing problems. Annals of Behavioral Medicine, 11, 108–112.

Stager, J. L. (1993). The comprehensive breast cancer knowledge test: Validity and
reliability. Journal of Advanced Nursing, 18, 1133–1140.

Staradub, V. L., Hsieh, Y., Clauson, J., Langerman, A., Rademaker, A. W., & Morrow,
M. (2002). Factors that influence surgical choices in women with breast carcinoma.
Cancer, 95, 1185–1190.

Stillman, M. J. (1977). Women’s health beliefs about breast cancer and breast self-
examination. Nursing Research, 26, 121–127.

Strull, W. M., Lo, B., & Charles, G. (1984). Do patients want to participate in medical
decision making? Journal of the American Medical Association, 252, 2990–2994.

Sutherland, H. J., Llewellyn-Thomas, H. A., Lockwood, G. A., Tritchler, D. L., & Till, J.
E. (1989). Cancer patients: Their desire for information and participation in
treatment decisions. Journal of the Royal Society of Medicine, 82, 260–263.

81

Vaeth, P. A. (1993). Women’s knowledge about breast cancer: dimensions of knowledge
and scale development. American Journal of Clinical Oncology, 16, 446–454.

Ward, S. & Griffin, J. (1990). Developing a test of knowledge of surgical options for
breast cancer. Cancer Nursing, 13, 191–196.

Ward, S., Heidrich, S. & Wolberg, W. (1989). Factors women take into account when
deciding upon type of surgery for breast cancer. Cancer Nursing, 12, 344–351.

Whelan, T. J., Levine. M. N., Gafni, A., Lukka, H., Mohide, E. A., Patel, M., et al.
(1995). Breast irradiation postlumpectomy: Development and evaluation of a
decision instrument. Journal of Clinical Oncology, 13, 847–853.

Whelan, T. J., Levine, M., Willan, A., Gafni, A., Sanders, K., Mirsky, D., et al. (2004).
Effect of a decision aid on knowledge and treatment decision making for breast
cancer surgery. Journal of the American Medical Association, 292, 435–441.

Williamson, G. M., Jones, D. J. & Ingram, L. A. (1999). Medical and psychosocial
predictors of breast cancer treatment decisions. In D. C. Park, R. W. Morrell, & K.
Shifren (Eds.), Processing of Medical Information in Aging Patients (pp. 69–91).
Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

Wolberg, W. H., Tanner, M. A., Romsaas, E. P., Trump, D. L. & Malec, J. F. (1987).
Factors influencing options in primary breast cancer treatment. Journal of Clinical
Oncology, 5, 68–74.

Yates, J. F. & Patalano, A. L. (1999). Decision making and aging. In D. C. Park, R. W.
Morrell, & K. Shifren (Eds.), Processing of Medical Information in Aging Patients
(31–54). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

82
Appendix A
Demographic & Professional Information
Please answer each of the questions below. All of the information you provide will
remain completely confidential. To ensure confidentiality, your name is not requested.
Only an assigned participant number will appear with your data. The answers to the
following questions will not be linked to you personally and will be stored in a password
protected database.

1. How did you hear about this research?

2. What is your age? _____ years.

3. Please select your racial/ethnic background.
_____ Caucasian
_____ Hispanic
_____ Native American
_____ African American
_____ Asian
_____ Other, Please Specify:______________________________

4. Please select your highest level of education.
_____ Less than a high school education
_____ High School
_____ Some College
_____ College Graduate
_____ Graduate School

5. Please select your current marital status.
_____ Single
_____ Married or Living as Married
_____ Divorced or Separated
_____ Widowed

6. Have you ever been diagnosed with breast cancer?
_____ No, Please continue to Item #7
_____ Yes, Please continue to next items (6a-6d)

6a. When were you diagnosed with breast cancer? _________________________

83
6b. What was your stage of breast cancer? (Please report only your first diagnosis.)
_____ Stage 0 (DCIS or LCIS)
_____ Stage I
_____ Stage II
_____ Stage III
_____ Stage IV
_____ I do not know this information

6c. What did your treatment consist of? Please select all that apply.
_____ Lumpectomy
_____ Mastectomy
_____ Reconstructive Surgery
_____ Radiation
_____ Chemotherapy
_____ Hormonal Treatment, such as Tamoxifen
_____ Lymph Node Evaluation
_____ Mastectomy of the Unaffected Breast
_____ Other, Please Specify: ________________________________
_____ I do not know this information

6d. How involved do you feel you were in your treatment decision making process?
_____ I made the final decisions regarding my treatment
_____ My physician(s) and I collaborated to make treatment decisions
_____ My physician(s) made the decisions concerning my treatment

7. Do you have friends or family who have been diagnosed with breast cancer?
_____ No, Please continue to item #8
_____ Yes, Please continue to next item. #7a

7a. Please specify who was diagnosed. Select all that apply.
_____ Mother
_____ Sister
_____ Aunt
_____ Daughter
_____ Close Friend
_____ Acquaintance
_____ Other, please specify: ___________________________

8. Do you have any professional or occupational experience with breast cancer? This can
include diagnosis, treatment, research, or any other professional or educational
experience.
_____ No, Please continue to next page.
_____ Yes, Please continue to next item, #8a

8a. Please describe the details of this experience, including the length of time (dates)
spent working in the area.

84
Appendix B
Breast Cancer Knowledge Measure
For each of the following items, please select the correct answer. If you do not know the
answer, please make you best guess. Please do not use any other sources of information to
find the correct answers; we are interested in what you know at this point in time.
1. Cancer that is no larger than 2 cm (about 1 inch) and has not spread to the axillary lymph
nodes is characteristic of what stage of breast cancer?
a. Stage I
b. Stage II
c. Stage III
d. None of the Above
2. Which of the following is a local therapy to the breast that requires treatment 5 days a
week for 6 weeks and is used to prevent cancer from recurring in the breast?
a. Chemotherapy
b. Tamoxifen
c. Radiation Therapy
d. Hormonal Therapy
3. Most breast cancers start in what type of cells in the breast?
a. Fat Cells
b. Duct Lining Cells
c. Lobular Cells
d. Muscle Cells
4. A surgery that removes the tumor and a small amount of normal tissue around it is
called?
a. Lumpectomy
b. Incisional Biopsy
c. Stereotactic Biopsy
d. Modified Mastectomy
5. In women over 40, mammograms fail to detect about how many cases of breast cancer?
a. Less than 1%
b. 10-15%
c. 20-25%
d. 30-35%
6. Which of the following is the most frequent side effect of radiation?
a. Red and dry skin in the treated area
b. Muscle stiffness in the arm and shoulder
c. Hair loss
d. Nausea

85
7. Although tumors can occur anywhere in the breast, where do they most frequently appear?
a. Lower-outer portion (away from center of chest)
b. Lower-inner portion (towards center of chest)
c. Upper-outer portion
d. Upper-inner portion
8. A women’s chance of developing breast cancer in her lifetime is__________.
a. 1 in 4 (25 %)
b. 1 in 8 (12.8 %)
c. 1 in 16 (6.25%)
d. 1 in 32 (3.1%)
9. Approximately what percent of breast lumps that are biopsied turn out to be breast cancer?
a. 10%
b. 20%
c. 40%
d. 80%
10. Over-expression of the HER-2 gene is an indicator of what aspect of breast cancer?
a. Lower rate of recurrence of breast cancer
b. More aggressive form of breast cancer
c. Very early stage of breast cancer
d. Locally advanced breast cancer
11. Lymphedema is a condition where which of the following is likely to occur?
a. Severe bruising occurs on the affected breast
b. Fluid build-up occurs in the arm
c. Numbness occurs in the chest
d. Skin of the breast becomes red and irritated

12. What is neoadjuvant therapy?
a. A therapy that reduces side effects of other treatments
b. A therapy used to shrink a tumor before surgery
c. A therapy that helps prevent cancer recurrences
d. None of the above
13. When breast cancer recurs after surgical treatment of a lumpectomy, where is the most
common site of the recurrence?
a. Bones
b. Lungs
c. Opposite Breast
d. Original Site of the Tumor

86
14. Which of the following statements is not correct?
a. Reconstruction can be performed using tissue from the body.
b. Reconstruction can be performed using saline or silicone implants.
c. Reconstruction cannot be started at the same time the mastectomy takes place.
d. Reconstructive surgery will create a breast that looks different than the natural
breast.

15. Which of the following is the one factor that is the most important in determining the risk
of metastasis?
a. Family History
b. Lymph node involvement
c. Estrogen receptor status
d. Number of tumors in breast
16. The use of chemotherapy after surgery for breast cancer is decided by what factor(s)?
a. Menopausal Status
b. Lymph Node Involvement
c. Estrogen Receptors
d. All of the above
17. When is a breast conserving surgery not recommended?
a. When tumors are widely separated in more than one area of the breast
b. When there is a prior radiation to the breast
c. During the first or second trimester of pregnancy
d. All of the above
18. A procedure that removes the first lymph node(s) that the cancer is likely to spread to is
called?
a. Axillary Lymph Node Dissection
b. Conventional Lymph Node Dissection
c. Sentinel Lymph Node Dissection
d. Core Biopsy of Axillary Lymph Nodes
19. Which of the following is not a possible risk factor for increased chance of developing
breast cancer?
a. Having children at an early age
b. Early onset of menstruation
c. Late age of menopause onset
d. Family history of breast cancer

87
20. Hormone therapy for treating breast cancer in post-menopausal women can work by what
mechanism?
a. Block the effect of estrogen on the cancer cell
b. Prevent production of estrogen in the body
c. By turning off the estrogen receptors in the cancer
d. All of the above
21. Which of the following is a possible side effect of Tamoxifen use?
a. Increased risk of developing endometrial cancer
b. Hot flashes
c. Blood clots in the veins
d. All of the Above
22. When used to treat early stage breast cancer, Mastectomy, as compared to Lumpectomy
plus Radiation has ________________.
a. A lower survival rate
b. An about equal survival rate
c. A 50% higher survival rate
d. A 100% higher survival rate
23. Ductal Carcinoma In Situ (DCIS) is what type of breast cancer?
a. Inflammatory
b. Non-Invasive
c. Aggressive
d. Metastatic

24. Which of the following is the most important factor for determining the use of hormone
therapy for breast cancer?
a. Age
b. Tumor Aggressiveness
c. Estrogen and Progesterone receptors
d. Menstrual Status
25. Stage IV breast cancer is also referred to as __________ cancer.
a. Inflammatory
b. Invasive
c. Locally Advanced
d. Metastatic

26. Radiation is typically performed after a lumpectomy to reduce the risk of recurrence.
a. True
b. False

88
27. Lymph node surgery is performed to determine whether cancer has spread beyond the
breast.
a. True
b. False

28. Once breast cancer occurs in one breast, it almost always occurs in the opposite breast.
a. True
b. False

29. Lymph node surgery is necessary for all cases of breast cancer, regardless of whether the
cancer is invasive or non-invasive.
a. True
b. False

30. There are effective treatments to combat many of the side effects of chemotherapy.
a. True
b. False

31. Cancer that has metastasized has returned after initial treatment.
a. True
b. False

32. The overall survival rate for Stage I breast cancer is greater than 95%.
a. True
b. False

89
Appendix C
Jane’s Surgical Treatment Scenario
General Instructions

Please complete the following scenarios to the best of your ability. Please understand that
completing these scenarios will be difficult for all participants, including those with
previous breast cancer experience. With the results from this research we hope to find out
where deficits in decision making occur so that we can design educational tools to help
future breast cancer patients make informed treatment decisions.

For the case presented below, we want you to be the patient's medical consultant. Your
job is to carefully review the information and use your knowledge of breast cancer to
make treatment recommendations. First we will present the information for the case, and
then we will list possible treatments. You will be asked to use 10-point rating scales to
express your recommendations for each treatment and the importance of information you
considered in reaching your recommendations. You will also be asked to evaluate
explanations of why some treatment choices may be better than others for treating this
patient.

Description of Jane’s Case

Jane is a 45-year-old married woman who lives in a major metropolitan area and works
full-time as a legal secretary. She has no children and she is pre-menopausal, but neither
she nor her husband desire to have children now or in the future. Jane is health conscious
and works very hard at staying physically fit and physically attractive.

Jane recently detected a small lump in her right breast while performing a breast self-
examination. While she has no prior screening mammograms for comparison, she did
obtain both a mammogram and sonogram to help evaluate the lump she detected. These
procedures confirmed that Jane had a 2 cm solid nodule in the upper outer portion of her
right breast that was suspicious for cancer. She followed these tests with a core needle
biopsy, which confirmed she has invasive breast cancer.

Jane is terrified of her cancer diagnosis, terrified that the cancer could kill her or return
after treatment. Her physician discussed multiple treatments with her, but did not offer a
recommendation about what treatment Jane should select.

90
Treatment Recommendation
Based on the information given about Jane and your knowledge of breast cancer, use the
following 10-point scale to indicate how strongly you would recommend each of the
treatments listed below. Please write the number from the following rating scale that
describes your recommendation in the corresponding space of the "Strength of
Recommendation" column. You should provide a number, ranging from zero to nine
for each treatment option. If you would recommend two or more treatments equally, you
should give them the same rating. Please note, several treatment options below include
lumpectomy or mastectomy, but the combinations of treatments included with each
option differ. You might find it useful to read through all treatments before beginning
your recommendations.
For your information, we are presenting brief definitions of some of the treatment options.
Lumpectomy: Surgical procedure to remove a tumor and a small amount of normal tissue.
Mastectomy: Surgical procedure to remove the whole breast that contains cancer.
Radiation: The use of high energy x-rays or other types of radiation to kill cancer cells.
Lymph Node Surgery: Removal of some of the lymph node under the arm for biopsy.
Reconstructive Surgery: Surgery to rebuild the breast after a mastectomy.
0 1 2 3 4 5 6 7 8 9
Not Moderately Strongly
Recommend Recommended Recommended
Treatment Types Strength of
Recommendation
1. Do nothing at the present; wait to see if the lump grows larger.
2. Lumpectomy only.
3. Lumpectomy plus Lymph Node Surgery.
4. Lumpectomy followed by 6 weeks of daily radiation treatments.
5. Lumpectomy plus Lymph Node Surgery, followed by 6 weeks of
daily radiation

6. Mastectomy of the affected breast only.
7. Mastectomy of the affected breast plus Lymph Node Surgery.
8. Mastectomy of the affected breast plus Reconstructive Surgery.
9. Mastectomy of the affected breast plus Lymph Node Surgery plus
Reconstructive Surgery.

10. Mastectomy of the unaffected breast (Contralateral Prophylactic
Mastectomy) with or without Reconstructive Surgery.

11. Alternative Treatments (e.g. herbal remedies & diet modification).

91
If you were the woman in the scenario, how willing would you be to make this decision?
Please select the statement that best represents your willingness to make this treatment
decision.
_____ I would prefer to make the final treatment decision completely on my own.
_____ I would prefer to make the treatment decision in collaboration with a physician.
_____ I would prefer to leave this decision regarding my treatment to the doctor.

92
Information Importance

Listed below is the information that was included in our description of Jane's case. Please
use the 10-point scale immediately below to indicate how much importance each piece of
information played in your treatment recommendations for Jane. Write the scale number
you choose for each piece of information in the "Rated Importance" column next to that
item. You should provide a number, ranging from zero to nine for each piece of
information.

0 1 2 3 4 5 6 7 8 9
No Moderate Great
Importance Importance Importance
Information Type Rated
Importance
1. Age.
2. Occupation
3. Relationship status.
4. Living in an urban area.

5. Menopausal status.
6. Child bearing status and plans.
7. Personal concern with health.
8. Personal concern with appearance.

9. Size of the cancer.
10. Location of the cancer.
11. Type of cancer (invasive versus non-invasive).
12. Result of mammogram and sonogram.
13. Result of biopsy.
14. Concern about dying from breast cancer.
15. Concern about breast cancer recurrence in same breast.
16. Concern about breast cancer developing in other breast.

17. Concern about cancer recurring outside of the breast, such as in
the lungs, liver, or bones.

93
Reasons Behind Recommendations

Below are statements (both true and false) of reasons for recommending some treatments,
and not recommending other treatments. Please use the 10-point scale below to indicate the
extent to which you agree (or disagree) with the reasoning behind each statement. Write the
scale number you choose for each Reasoned Statement in the "Strength of Agreement"
column next to that item. You should provide a number, ranging from zero to nine for
each statement.

0 1 2 3 4 5 6 7 8 9
Strong Moderate Strong
Disagreement Agreement Agreement
Reasoned Statements Strength of
Agreement
1. Jane should have a lumpectomy followed by radiation because
in terms of survival, it will be as effective as a mastectomy for
the size and location of her cancer. It will also minimize the
disfigurement of her breast.

2. Jane should have a mastectomy because it is the only effective
treatment for the size and location of her breast cancer. The
mastectomy will improve her chances of survival compared to
a lumpectomy and radiation.

3. Jane should have a mastectomy of both breasts, because once
breast cancer occurs in one breast, it almost always appears in
the other breast. Having both breasts removed at once will
minimize cost and inconvenience to Jane and increase her
chance of surviving breast cancer.

4. Jane should have a lumpectomy, but need not undergo radiation
treatment given the size and location of her cancer. Radiation
will not affect the rate of recurrence.

5. Jane should wait at least six months before undergoing any
surgical treatment for her small breast cancer. Often the
immune system is able to cure these cancers on its own and
there is little risk to Jane at this early stage.

6. Jane should have a mastectomy because it will require less time
off work and inconvenience to her family life because she can
avoid daily radiation treatments for six weeks.

7. If Jane chooses mastectomy, she can have reconstructive
surgery to handle her desire to maintain physical appearance of
her breast.

8. Jane should not have any form of breast surgery because of
surgical risks involved.

9. Regardless of treatment chosen, Jane should also have her
physician perform lymph node surgery, as this will inform her
about whether her cancer has spread beyond the breast.

10. Jane need not undergo lymph node surgery because of the small
size of her cancer; it is highly unlikely that the cancer would
have spread beyond the breast.

94
Jane’s Adjuvant Treatment Scenario
The paragraphs below present the original information about Jane, as well as some new
information about her case. For the new information, we have selected a surgical
treatment for Jane. This may or may not have been the treatment you selected in the
previous portion. It is not the only correct treatment option. We selected this treatment to
keep the information consistent for all participants. Please do not change your previous
answers to correspond with the information presented below. Please complete the
remaining portion of Jane's scenario based on the information you already have about
Jane, as well as the new information about her case presented below.
Original Description of Jane's Case
Jane is a 45-year-old married woman who lives in a major metropolitan area and works
full-time as a legal secretary. She has no children and she is pre-menopausal, but neither
she nor her husband desire to have children now or in the future. Jane is health conscious
and works very hard at staying physically fit and physically attractive.
Jane recently detected a small lump in her right breast while performing a breast self-
examination. While she has no prior screening mammograms for comparison, she did
obtain both a mammogram and sonogram to help evaluate the lump she detected. These
procedures confirmed that Jane had a 2 cm solid nodule in the upper outer portion of her
right breast that was suspicious for cancer. She followed these tests with a core needle
biopsy, which confirmed she has invasive breast cancer.
Jane is terrified of her cancer diagnosis, terrified that the cancer could kill her or return
after treatment. Her physician discussed multiple treatments with her, but did not offer a
recommendation about what treatment Jane should select.
New Information about Jane's Case
Jane ultimately decided to have her surgeon perform a lumpectomy to remove the lump
in her right breast and to follow the surgery with six weeks of radiation treatment. Her
surgeon referred her to a medical oncologist to discuss further treatment options. The
following information was obtained from Jane's pathology report.
The lump removed was a 2.0 centimeters, high grade, invasive (ductal) breast cancer. A
clear margin was obtained. Estrogen and Progesterone receptor levels were both negative,
while the Her-2/neu receptor was positive. An axillary lymph node dissection was also
performed during the surgery and the results showed 3 positive lymph nodes. The
oncologist presented Jane with the adjuvant treatment options listed below, but again did
not offer Jane a recommendation about what, if any additional treatment would be
optimal.

95
Treatment Recommendation
Based on the information given about Jane and your knowledge of breast cancer, use the
following 10-point scale to indicate how strongly you would recommend each of the
treatments listed below. Please write the number from the following rating scale that
describes your recommendation in the corresponding space of the "Strength of
Recommendation" column. You should provide a number, ranging from zero to nine
for each treatment option. If you would recommend two or more treatments equally, you
should give the same rating. You might find it useful to read through all treatments before
beginning your recommendations.
For your information, we are presenting brief definitions of some of the treatment options.
Chemotherapy: The use of drugs to stop the growth of cancer cells, either by killing the cells or
by stopping the cells from dividing.
Hormonal Therapy: A treatment that removes hormones or blocks their action and stops cancer
cells from growing.
0 1 2 3 4 5 6 7 8 9
Not Moderately Strongly
Recommend Recommended Recommended
Treatment Types Strength of
Recommendation
1. No adjuvant treatment, stop after surgery and radiation.

2. Chemotherapy, in addition to surgery and radiation.
3. Hormone therapy (such as Tamoxifen) only, after surgery and
radiation.

4. Both chemotherapy and hormonal therapy in addition to surgery
and radiation.

5. Alternative Treatments (e.g. herbal remedies & diet modification).

If you were the woman in the scenario, how willing would you be to make this decision?
Please select the statement that best represents your willingness to make this treatment
decision.
_____ I would prefer to make the final treatment decision completely on my own.
_____ I would prefer to make the treatment decision in collaboration with a physician.
_____ I would prefer to leave this decision regarding my treatment to the doctor.

96
Information Importance

Listed below is the information that was included in our description of Jane's case. Please
use the 10-point scale immediately below to indicate how much importance each piece of
information played in your treatment recommendations for Jane. Write the scale number
you choose for each piece of information in the "Rated Importance" column next to that
item. You should provide a number, ranging from zero to nine for each piece of
information.

0 1 2 3 4 5 6 7 8 9
No Moderate Great
Importance Importance Importance
Information Type Rated
Importance
1. Age.
2. Occupation
3. Relationship status.
4. Living in an urban area.

5. Menopausal status.
6. Child bearing status and plans.
7. Personal concern with health.
8. Personal concern with appearance.

9. Location of the cancer.
10. Size of the lump removed.
11. The grade of the cancer.
12. Type of cancer (invasive versus non-invasive).
13. Estrogen & Progesterone receptor status.
14. Her-2/neu receptor status.
15. Results of lymph node surgery.
16. Concern about dying from breast cancer.

97
Information Type Rated
Importance
17. Concern about breast cancer recurrence in same breast.
18. Concern about breast cancer developing in other breast.
19. Concern about cancer recurring outside of the breast, such as in the
lungs, liver, or bones.

98
Reasons Behind Recommendations

Below are statements (both true and false) of reasons for strongly recommending some
treatments, and not recommending other treatments. Please use the 10-point scale below to
indicate the extent to which you agree (or disagree) with the reasoning behind each
statement. Write the scale number you choose for each Reasoned Statement in the
"Strength of Agreement" column next to that item. You should provide a number, ranging
from zero to nine for each statement.

0 1 2 3 4 5 6 7 8 9
Strong Moderate Strong
Disagreement Agreement Agreement
Reasoned Statements Strength of
Agreement
1. Jane need not undergo any adjuvant treatments because the positive
lymph nodes have no bearing on recurrence. Further, the high grade
cancer and negative Estrogen and Progestin receptor status indicate
the cancer is not life threatening.

2. Jane’s stage of breast cancer indicates she should receive
aggressive adjuvant treatment consisting of chemotherapy and
hormonal therapy, in addition to her radiation treatments.

3. Jane need not undergo hormonal therapy for her breast cancer, as
the Estrogen and Progestin receptors are both negative. However,
the size of the cancer and the lymph node involvement indicate she
should receive chemotherapy.

4. Because Jane is pre-menopausal, she should receive hormonal
therapy to reduce the risk of recurrence. However, chemotherapy is
not necessary because of Jane’s excellent overall health and the
small risk of recurrence.

5. No adjuvant treatment should be administered because there are no
known impacts on survival from breast cancer. The risks outweigh
any possible benefits.

6. Alternative treatments should be used instead because they are
known to be just as effective as traditional adjuvant regimens, but
allow Jane to avoid the risks and toxicities associated with other
treatments.

99
Elizabeth’s Surgical Treatment Scenario
For the case presented below, we want you to be the patient's medical consultant. Your
job is to carefully review the information and use your knowledge of breast cancer to
make a treatment recommendation. First we will present the information for the case, and
then we will list possible treatments. You will be asked to use 10-point rating scales to
express your recommendations for each treatment and the importance of information you
considered in reaching your recommendations. You will also be asked to evaluate
explanations of why some treatment choices may be better than others for treating this
patient.
Description of Elizabeth's Case (Final Scenario)
Elizabeth is a 60-year-old woman who lives in a major metropolitan area and works full-
time as an elementary school teacher. She is married and has four grown children.
Elizabeth is post-menopausal. Elizabeth is in good health and works out regularly; both
her physical health and attractiveness are important to her.
A lump was recently detected in Elizabeth's right breast while her physician was
performing a routine physical examination. Elizabeth has multiple prior screening
mammograms for comparison, and she obtained a mammogram to help evaluate the lump
that was detected. These procedures confirmed that Elizabeth had an approximately 5 cm
solid nodule in the upper outer portion of her right breast that was suspicious of cancer.
She followed these tests with a core needle biopsy, which confirmed she has invasive
breast cancer.
Elizabeth is terrified of her cancer diagnosis, terrified that the cancer could kill her or
return after treatment. Her physician discussed multiple treatments with her, but did not
offer a recommendation about what treatment Elizabeth should select.

100
Treatment Recommendation
Based on the information given about Elizabeth and your knowledge of breast cancer, use
the following 10-point scale to indicate how strongly you would recommend each of the
treatments listed below. Please write the number from the following rating scale that
describes your recommendation in the corresponding space of the "Strength of
Recommendation" column. You should provide a number, ranging from zero to nine
for each treatment option. If you would recommend two or more treatments equally, you
should give them the same rating. Please note, several treatment options below include
lumpectomy or mastectomy, but the combinations of treatments included with each
option differ. You might find it useful to read through all treatments before beginning
your recommendations.
For your information, we are presenting brief definitions of some of the treatment options.
Lumpectomy: Surgical procedure to remove a tumor and a small amount of normal tissue.
Mastectomy: Surgical procedure to remove the whole breast that contains cancer.
Radiation: The use of high energy x-rays or other types of radiation to kill cancer cells.
Lymph Node Surgery: Removal of some of the lymph node under the arm for biopsy.
Reconstructive Surgery: Surgery to rebuild the breast after a mastectomy.
0 1 2 3 4 5 6 7 8 9
Not Moderately Strongly
Recommend Recommended Recommended
Treatment Types Strength of
Recommendation
1. Do nothing at the present; wait to see if the lump grows larger.
2. Lumpectomy only.
3. Lumpectomy plus Lymph Node Surgery.
4. Lumpectomy followed by 6 weeks of daily radiation treatments.
5. Lumpectomy plus Lymph Node Surgery, followed by 6 weeks of
daily radiation

6. Mastectomy of the affected breast only.
7. Mastectomy of the affected breast plus Lymph Node Surgery.
8. Mastectomy of the affected breast plus Reconstructive Surgery.
9. Mastectomy of the affected breast plus Lymph Node Surgery plus
Reconstructive Surgery.

10. Mastectomy of the unaffected breast (Contralateral Prophylactic
Mastectomy) with or without Reconstructive Surgery.

11. Alternative Treatments (e.g. herbal remedies & diet modification).

101
If you were the woman in the scenario, how willing would you be to make this decision?
Please select the statement that best represents your willingness to make this treatment
decision.
_____ I would prefer to make the final treatment decision completely on my own.
_____ I would prefer to make the treatment decision in collaboration with a physician.
_____ I would prefer to leave this decision regarding my treatment to the doctor.

102
Information Importance

Listed below is the information that was included in our description of Elizabeth's case.
Please use the 10-point scale immediately below to indicate how much importance each
piece of information played in your treatment recommendations for Elizabeth. Write the
scale number you choose for each piece of information in the "Rated Importance" column
next to that item. You should provide a number, ranging from zero to nine for each piece
of information.

0 1 2 3 4 5 6 7 8 9
No Moderate Great
Importance Importance Importance
Information Type Rated
Importance
1. Age.
2. Occupation
3. Relationship status.
4. Living in an urban area.

5. Menopausal status.
6. Child bearing status and plans.
7. Personal concern with health.
8. Personal concern with appearance.

9. Size of the cancer.
10. Location of the cancer.
11. Type of cancer (invasive versus non-invasive).
12. Result of mammogram and sonogram.
13. Result of biopsy.
14. Concern about dying from breast cancer.
15. Concern about breast cancer recurrence in same breast.
16. Concern about breast cancer developing in other breast.
17. Concern about cancer recurring outside of the breast, such as in the
lungs, liver, or bones.

103
Reasons Behind Recommendations

Below are statements (both true and false) of reasons for strongly recommending some
treatments, and not recommending other treatments. Please use the 10-point scale below to
indicate the extent to which you agree (or disagree) with the reasoning behind each
statement. Write the scale number you choose for each Reasoned Statement in the
"Strength of Agreement" column next to that item. You should provide a number, ranging
from zero to nine for each statement.

0 1 2 3 4 5 6 7 8 9
Strong Moderate Strong
Disagreement Agreement Agreement
Reasoned Statements Strength of
Agreement
1. Elizabeth should have a lumpectomy followed by radiation because
in terms of survival, it will be as effective as a mastectomy for the
size and location of her cancer. It will also minimize the
disfigurement of her breast.

2. Elizabeth should have a mastectomy because it is an effective
treatment and will be less inconvenient to her schedule because she
can avoid radiation treatments for six weeks.

3. Elizabeth should have a mastectomy of both breasts, because once
breast cancer occurs in one breast, it almost always appears in the
other breast. Having both breasts removed at once will minimize
cost and inconvenience to Elizabeth and increase her chance of
surviving breast cancer.

4. Elizabeth should have a lumpectomy, but need not undergo
radiation treatment given the size and location of her cancer.
Radiation will not affect the rate of recurrence.

5. Elizabeth should wait at least six months before undergoing any
surgical treatment for her breast cancer. Often the immune system
is able to cure these cancers on its own and there is little risk to
Elizabeth at this stage.

6. If Elizabeth chooses mastectomy, she can have reconstructive
surgery to handle her desire to maintain physical appearance of her
breast.

7. Elizabeth should not have any form of breast surgery because of
surgical risks involved.

8. Regardless of treatment chosen, Elizabeth should also have her
physician perform lymph node surgery, as this will inform her
about whether her cancer has spread beyond the breast.

9. Elizabeth need not undergo lymph node surgery because of the size
of her cancer; it is highly unlikely that the cancer would have
spread beyond the breast.

104
Elizabeth’s Adjuvant Treatment Scenario
The paragraphs below present the original information about Elizabeth, as well as some
new information about her case. For the new information, we have selected a surgical
treatment for Elizabeth. This may or may not have been the treatment you selected in the
previous portion. It is not the only correct treatment option. We selected this treatment to
keep the information consistent for all participants. Please do not change your previous
answers to correspond with the information presented below. Please complete the
remaining portion of Elizabeth's scenario based on the information you already have
about Elizabeth, as well as the new information about her case presented below.
Original Description of Elizabeth's Case
Elizabeth is a 60-year-old woman who lives in a major metropolitan area and works full-
time as an elementary school teacher. She is married and has four grown children.
Elizabeth is post-menopausal. Elizabeth is in good health and works out regularly; both
her physical health and attractiveness are important to her.
A lump was recently detected in Elizabeth's right breast while her physician was
performing a routine physical examination. Elizabeth has multiple prior screening
mammograms for comparison, and she obtained a mammogram to help evaluate the lump
that was detected. These procedures confirmed that Elizabeth had an approximately 5 cm
solid nodule in the upper outer portion of her right breast that was suspicious of cancer.
She followed these tests with a core needle biopsy, which confirmed she has invasive
breast cancer.
Elizabeth is terrified of her cancer diagnosis, terrified that the cancer could kill her or
return after treatment. Her physician discussed multiple treatments with her, but did not
offer a recommendation about what treatment Elizabeth should select.
New Information about Elizabeth's Case
Elizabeth ultimately decided to have her surgeon perform a mastectomy of her right
breast. Her surgeon referred her to a medical oncologist to discuss further treatment
options. The following information was obtained from Elizabeth’s pathology report.
The lump removed was a 5.0 centimeters, intermediate grade, invasive (ductal) breast
cancer. Estrogen and Progesterone receptor levels were both positive, while the Her-
2/neu receptor was negative. An axillary lymph node dissection was also performed
during the surgery and the results showed 4 positive lymph nodes. The oncologist
presented Elizabeth with the adjuvant treatment options listed below, but again did not
offer Elizabeth a recommendation about what, if any, additional treatment would be
optimal.

105
Treatment Recommendation
Based on the information given about Elizabeth and your knowledge of breast cancer, use
the following 10-point scale to indicate how strongly you would recommend each of the
treatments listed below. Please write the number from the following rating scale that
describes your recommendation in the corresponding space of the "Strength of
Recommendation" column. You should provide a number, ranging from zero to nine
for each treatment option. If you would recommend two or more treatments equally, you
should give them the same rating. You might find it useful to read through all treatments
before beginning your recommendations.
For your information, we are presenting brief definitions of some of the treatment options.
Chemotherapy: The use of drugs to stop the growth of cancer cells, either by killing the cells or
by stopping the cells from dividing.
Hormonal Therapy: A treatment that removes hormones or blocks their action and stops cancer
cells from growing.
0 1 2 3 4 5 6 7 8 9
Not Moderately Strongly
Recommend Recommended Recommended
Treatment Types Strength of
Recommendation
1. No adjuvant treatment, stop after surgery.

2. Only chemotherapy, after surgery.
3. Only hormone therapy (such as Tamoxifen), after surgery.
4. Only radiation to the breast, after surgery.
5. Both chemotherapy and hormone therapy in addition to surgery.
6. Both radiation and chemotherapy, in addition to surgery.
7. Both radiation and hormone therapy, in addition to surgery.
8. Radiation to the breast, chemotherapy, and hormonal therapy, in
addition to surgery.

9. Alternative Treatments (e.g. herbal remedies & diet modification).
If you were the woman in the scenario, how willing would you be to make this decision?
Please select the statement that best represents your willingness to make this treatment
decision.
_____ I would prefer to make the final treatment decision completely on my own.
_____ I would prefer to make the treatment decision in collaboration with a physician.
_____ I would prefer to leave this decision regarding my treatment to the doctor.

106
Information Importance

Listed below is the information that was included in our description of Elizabeth's case.
Please use the 10-point scale immediately below to indicate how much importance each
piece of information played in your treatment recommendations for Elizabeth. Write the
scale number you choose for each piece of information in the "Rated Importance" column
next to that item. You should provide a number, ranging from zero to nine for each piece
of information.

0 1 2 3 4 5 6 7 8 9
No Moderate Great
Importance Importance Importance
Information Type Rated
Importance
1. Age.
2. Occupation.
3. Relationship status.
4. Living in an urban area.

5. Menopausal status.
6. Child bearing status and plans.
7. Personal concern with health.
8. Personal concern with appearance.

9. Location of the cancer.
10. Size of the lump removed.
11. The grade of the cancer.
12. Type of cancer (invasive versus non-invasive).
13. Estrogen & Progesterone receptor status.
14. Her-2/neu receptor status.
15. Results of lymph node surgery.
16. Concern about dying from breast cancer.

107

Information Type Rated
Importance
17. Concern about breast cancer recurrence in same breast.
18. Concern about breast cancer developing in other breast.
19. Concern about cancer recurring outside of the breast, such as in the
lungs, liver, or bones.

108
Reasons Behind Recommendations

Below are statements (both true and false) of reasons for strongly recommending some
treatments, and not recommending other treatments. Please use the 10-point scale below to
indicate the extent to which you agree (or disagree) with the reasoning behind each
statement. Write the scale number you choose for each Reasoned Statement in the
"Strength of Agreement" column next to that item. You should provide a number, ranging
from zero to nine for each statement.

0 1 2 3 4 5 6 7 8 9
Strong Moderate Strong
Disagreement Agreement Agreement
Reasoned Statements Strength of
Agreement
1. Elizabeth need not undergo any adjuvant treatments because the
positive lymph nodes have no bearing on recurrence. Further, her
surgical treatment of a mastectomy indicates that no further
adjuvant treatment is necessary.

2. Because of the size and lymph node involvement of Elizabeth’s
breast cancer, she should undergo chemotherapy. Further, she
should receive hormone therapy because of the positive Estrogen
and Progestin receptors.

3. Elizabeth need not undergo hormonal therapy for her breast cancer,
as the Estrogen and Progestin receptors are both positive. However,
the size of the cancer and the lymph node involvement indicate she
should receive chemotherapy.

4. Because Elizabeth is post-menopausal, she should receive
hormonal therapy to reduce the risk of recurrence. However,
chemotherapy is not necessary because of Elizabeth’s excellent
overall health and the small risk of recurrence.

5. No adjuvant treatment should be administered because there are no
known impacts on survival from breast cancer. The risks outweigh
any possible benefits.

6. Alternative treatments should be used instead because they are
known to be just as effective as traditional adjuvant regimens, but
allow Elizabeth to avoid the risks and toxicities associated with
other treatments.

7. Elizabeth should receive radiation to the breast five days a week for
six weeks in order to ensure that the cancer is no longer a threat to
her.

8. Elizabeth should receive radiation to the breast, as well as
chemotherapy and hormone therapy. The combination of these
treatments will eliminate any risk of recurrence.

Abstract (if available)

Abstract Research in the area of breast cancer treatment decision making has typically focused separately on two issues: 1) what patients know about breast cancer and 2) what surgical treatments they undergo. The current research seeks to examine the link between knowledge and treatment decision making. To this end, two instruments were developed. The first instrument was a measure of breast cancer knowledge and the second was a measure of treatment decision making ability that assessed decision quality and causal reasoning skills. Participants were 142 women (53 breast cancer patients, 89 nonpatients). Patients performed significantly better than non-patients on the knowledge instrument and each of 4 decision quality outcomes. Results from regression analyses showed multiple predictors of decision quality, including breast cancer knowledge, patient status, age, and various information use factor scores. Results from this research provide a clearer picture of what participants know about breast cancer, the quality of their decisions, the importance of information and how information use is related to the decisions that participants make. In the future, these instruments can be used to determine patients' readiness to make their own breast cancer treatment decisions. Additionally, the knowledge gained from this research can be used to design educational interventions that teach patients what they need to know to make informed treatment decisions.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Health-related quality of life correlates of the treatment decision making process of newly diagnosed prostate cancer patients

PDF

Discussion during treatment decision-making predicts emotional adjustment in prostate cancer patients

PDF

Evaluating cancer treatments with health economics methods

PDF

Association between informed decision-making and mental health-related quality of life in long term prostate cancer survivors

PDF

Mining an ePR system using a treatment plan navigator for radiation toxicity to evaluate proton therapy treatment protocol for prostate cancer

PDF

Molecular mechanisms of chemoresistance in breast cancer

PDF

Deconstructing the psychological components of emotional decision making and their relation to the suicide continuum

PDF

Examining the applications of data-driven decision making on classroom instruction

PDF

A two-level analysis of foreign policy decision making: an empirical investigation of the case of China-Taiwan

PDF

Arm lymphedema in a multi-ethnic cohort of female breast cancer survivors

PDF

Low-income, minority cancer patients who drop out of depression treatment

PDF

How urban school superintendents effectively use data-driven decision making to improve student achievement

PDF

The interpersonal effect of emotion in decision-making and social dilemmas

PDF

A school's use of data-driven decision making to affect gifted students' learning

PDF

A comparison of improvements in sight reading pitch accuracy in beginning instrumentalists after treatment

PDF

Affective decision-making predictive of Chinese adolescents drinking behaviors: a longitudinal study

PDF

Social support, constraints, and protective buffering in prostate cancer patients and their partners

PDF

High intensity interval training for breast cancer patients receiving anthracycline-based chemotherapy

PDF

Screening and association testing of coding variation in steroid hormone coactivator and corepressor genes in relationship with breast cancer risk in multiple populations

PDF

Evaluating novel chemotherapeutic strategies in colorectal and gastric cancer: the role of histone deacetylase inhibitors and human epidermal receptor family inhibitors

Asset Metadata

Creator Brown, Carol E. (author)

Core Title An empirical evaluation of breast cancer treatment decision making

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 06/10/2009

Defense Date 03/30/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag breast cancer,Decision making,information use,Knowledge,OAI-PMH Harvest,treatment selection

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Walsh, David A. (committee chair), John, Richard S. (committee member), Read, Stephen J. (committee member), Richardson, Jean L. (committee member)

Creator Email carol.brown.usc@gmail.com,ceb@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2284

Unique identifier UC1412944

Identifier etd-Brown-2897 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-235792 (legacy record id),usctheses-m2284 (legacy record id)

Legacy Identifier etd-Brown-2897.pdf

Dmrecord 235792

Document Type Dissertation

Rights Brown, Carol E.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu