An informed decision? Breast cancer patients and their knowledge about treatment

Transcription

1 Patient Education and Counseling 64 (2006) An informed decision? Breast cancer patients and their knowledge about treatment Angela Fagerlin a,b,c, *, Indu Lakhani b, Paula M. Lantz d, Nancy K. Janz e, Monica Morrow f, Kendra Schwartz g, Dennis Deapen h, Barbara Salem b, Lihua Liu h, Steven J. Katz a,b,d a VA Health Services Research & Development Center for Practice Management and Outcomes Research, VA Ann Arbor Healthcare System, Ann Arbor, MI, United States b Department of Internal Medicine, University of Michigan, United States c Center for Behavioral and Decision Sciences in Medicine, Ann Arbor, MI, United States d Department of Health Management and Policy, University of Michigan, Ann Arbor, MI, United States e Department of Health Behavior and Health Education, University of Michigan, Ann Arbor, MI, United States f Department of Surgical Oncology, Fox Chase Cancer Center, Philadelphia, PA, United States g Department of Family Medicine and Karmanos Cancer Institute, Wayne State University, Detroit, MI, United States h Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA, United States Received 11 August 2005; received in revised form 10 March 2006; accepted 10 March 2006 Abstract Objective: Although involving women in breast cancer treatment decisions is advocated, there is little understanding of whether women have the information they need to make informed decisions. The objective of the current study was to evaluate women s knowledge of survival and recurrence rates for mastectomy and breast conserving surgery (BCS) and the factors associated with this knowledge. Methods: We used a population-based sample of women diagnosed with breast cancer in metropolitan Los Angeles and Detroit between December 2001 and January All women with ductal carcinoma in situ and a random sample of women with invasive disease were selected (N = 2382), of which 1844 participated (77.4%). All participants were mailed surveys. The main outcome measures were knowledge of survival and recurrence rates by surgical treatment type. Results: Only 16% of women knew that recurrence rates were different for mastectomy and BCS, and 48% knew that the survival rates were equivalent across treatment. Knowledge about survival and recurrence was improved by exposure to the Internet and health pamphlets ( p < 0.01). Women who had a female (versus male) surgeon, and/or a surgeon who explained both treatments (rather than just one treatment) demonstrated higher survival knowledge ( p < 0.01). The majority of women had inadequate knowledge with which to make informed decisions about breast cancer surgical treatment. Conclusion: Previous explanations for poor knowledge, such as irrelevance of knowledge to decision making and lack of access to information, were not shown to be plausible explanations for the low levels of knowledge observed in this sample. Practice implications: These results suggest a need for fundamental changes in patient education to ensure that women are able to make informed decisions about their breast cancer treatment. These changes may include an increase in the use of decision aids and in decreasing the speed at which treatment decisions are made. Published by Elsevier Ireland Ltd. Keywords: Informed decision making; Breast cancer; Treatment decisions; Knowledge 1. Introduction * Corresponding author. Tel.: ; fax: address: fagerlin@umich.edu (A. Fagerlin). Very few would challenge the belief that patients are typically entitled to make their own medical decisions. The rationale underlying this belief is multifaceted. First, the /$ see front matter. Published by Elsevier Ireland Ltd. doi: /j.pec

2 304 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) authority to make important decisions in one s life is part of what it means to be a free, self-governing human being. Second, it is presumed that the majority of the decisions patients make for themselves are better than the decisions others can make for them. Third, there is some evidence that patients who make their own decisions tend to be more satisfied with their health care [1 4], are more adherent to treatment recommendations [5], and have a better quality of life than those who do not make their own decisions [6 9]. The principle that patients ought to make their own decisions has been invoked with particular frequency and conviction with regard to the choice of breast cancer surgical treatment options, as this is a decision that encompasses what is often referred to as preference-sensitive care [10,11]. The reasons for this are two-fold. First, concerns have been raised that mastectomy is an over-treatment, that it is overused, and that many women may not receive the treatment they prefer (i.e., may prefer BCS over mastectomy, but receive mastectomy) [12,13]. Second, there is evidence that women want to be involved in making their own treatment decisions. In a recent study, 24% of breast cancer patients stated that they wanted to take sole responsibility for deciding which surgical treatment to undergo, while 64% wanted to make the decision in conjunction with their surgeon [14]. These data suggest that while this decision is one in which many women choose to participate in, either by making their treatment decisions on their own or in conjunction with their physician, other women may choose to not be the primary decision maker. Previous research has shown that not all patients want to make their own medical decisions and that patients often have valid reasons for choosing to not participate in the decision making process [15]. Women s choices about surgical treatment for breast cancer have received enormous attention as a result of concerns that have been raised regarding lack of patient involvement in decision making [16]. Indeed, 20 states have enacted laws which require physicians to inform women about both mastectomy and breast conserving surgery (BCS) [17 19]. Furthermore, there are few significant treatment choices for which patients will have better access to information than they do for breast cancer. Randomized controlled trials have yielded definitive information about the relative benefits of the two surgical options [20 24]. A great deal of effort has been devoted to developing educational materials for patients. Breast cancer surgery and other treatments have been widely discussed in the media. Many women making the decision are likely to know other women who have already made the same decision. Given each of these conditions, breast cancer treatment decision making is an ideal case to investigate notions of informed decision making. If, under these conditions, patients are unable to make informed decisions, we ought to be concerned about how well patients are able to use available information to make any important decisions regarding their health. A diagnosis of breast cancer plunges a woman into a journey filled with decisions that until recently were routinely left to surgeons, but are now more often placed in the hands of patients [25]. Yet, we have little understanding of how well informed women are when they are asked to make these decisions. What do patients know about the trade-offs between mastectomy compared to BCS with radiation with regard to survival and recurrence of disease? What socio-demographic factors are correlated with this knowledge? Where do women get their information about treatment and how does the source of information affect their knowledge? To address these questions we analyzed data from a population-based survey of breast cancer survivors recently diagnosed in the metropolitan areas of Detroit and Los Angeles. More specifically, our goal was to examine women s knowledge regarding the risks and benefits of mastectomy and BCS and to determine what factors were related to this knowledge. 2. Methods 2.1. Study population Women who were 79 years and younger (no lower age limit) and who were diagnosed with either ductal carcinoma in situ (DCIS) or invasive disease as identified by the Surveillance, Epidemiology, and End Results (SEER) cancer registries of the greater metropolitan areas of Detroit and Los Angeles during a 14-month period from December 2001 to January 2003 were eligible for the study. SEER registries are population-based. All cases of cancer confirmed by pathology laboratories in the catchment area are entered into the registry system Database and sampling We prospectively selected all cases of DCIS (to assure adequate representation) and a random sample of invasive cases (over-sampling African American women to 25% and 33% of accrued cases in Los Angeles and Detroit, respectively) each month into the preliminary study sample (N = 2647). Eligible subjects had a primary diagnosis of breast cancer, underwent a definitive surgical procedure, resided in the catchment area of the SEER site, and were able to complete a questionnaire in either English or Spanish. All Asian women and all U.S.-born women younger than 50 years of age who were diagnosed with invasive disease in Los Angeles during our study period were excluded because they were already being enrolled in other studies. Women with a diagnosis of lobular carcinoma in situ were also excluded because the natural history of and recommended treatment for this disease are different than for DCIS.

3 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) Using initial pathology reports, we selected all cases of DCIS and an approximate 20% random sample of invasive breast cancer cases. Ninety percent of all accrued cases were eligible for the study (N = 2382). The survey was completed by 77.4% of eligible patients (N =1844). We restricted the analyses in this paper to patients with DCIS and stage 1 invasive disease (N = 1215) because randomized clinical trials have clearly shown that disease recurrence is higher for women who have received BCS and radiation compared to those who received mastectomy [26,27] Data collection and management Physicians were notified of our intent to contact patients. Patients then were sent an introductory letter and a follow-up telephone call which assessed eligibility. A questionnaire and a US$ 10 grocery gift certificate were mailed to all eligible women who agreed to participate and to those that could not be reached by phone (approximately 14% of potential respondents). The Dillman survey method, which involves follow-up reminders and subsequent mailings to non-respondents at periodic intervals, was employed to encourage participant response [28,29]. SEER clinical data (tumor size, node status, regional and distant extension, histologic grade, and treatment information) from hospital-based sources were merged with survey data for 98.2% of cases. The study protocol was approved by the Institutional Review Boards of the University of Michigan, the University of Southern California, and Wayne State University Measures Knowledge about survival and recurrence Two true/false questions assessed women s knowledge about survival and recurrence rates for mastectomy versus BCS: (1) the chance of being alive 5 years after surgery is the same for mastectomy as it is for lumpectomy with radiation therapy (true) and (2) the chances that breast disease will come back after treatment are the same for mastectomy as they are for lumpectomy with radiation therapy (false). Response options were true, false, and don t know. These questions were designed specifically for this study and were found to be reliable (Cronbach s a =0.74) Sources of information Women indicated whether they consulted any of 12 sources of health information (listed in Table 2). Of these 12 sources, 3 asked about personal relationships (i.e., spouse/ partner, relatives, and friends), 4 had their roots in mass media (i.e., exposure to television/movies, news source, magazines, and books about breast cancer), and 2 were educational/ research-based (i.e., informational videos/cd-roms and scientific journals/research papers). The remaining sources included health pamphlets, the Internet, and places of worship. There was no space provided for participants to list other sources of information. Within each category (personal relationship, mass media, and educational/researchbased), data were coded to reflect whether participants viewed none, one, or more than one of each source. Finally, using an open-ended question, we asked women to indicate the source of information (from the list provided) they found most helpful during their breast cancer surgery decision making Factors influencing treatment decisions Using a four-point Likert scale (ranging from did not influence decision to greatly influenced decision ), women rated the extent to which two issues (worry about the disease coming back and desire to reduce the chances of disease coming back) influenced their surgical treatment decision. In addition, an open-ended question asked women to Describe the most important factor that influenced your decision about whether to have a mastectomy or lumpectomy. This question was asked at the very end of the survey (and 22 pages after the two recurrence questions) Surgeon interaction Women s satisfaction with their surgeon was measured by six questions using 5-point Likert scale responses that ranged from strongly disagree to strongly agree. These questions were: (1) My surgeon explained everything I wanted to know about my breast disease, (2) My surgeon listened to everything I had to say, (3) My surgeon really cared about me as a person. I was not just part of his/her job, (4) My surgeon treated me with a great deal of respect, (5) My surgeon calmed my worries about my breast disease, and (6) My surgeon was very kind and considerate of my feelings. Participants were classified as having high or low satisfaction using a median split (cut-off was 4.33). Women indicated whether their surgeon discussed both mastectomy and BCS with them, or if only one treatment was described. Women also reported the gender of their surgeon Surgical treatment Women indicated the type of treatment they received (BCS or mastectomy) Stage Summary cancer stage was classified using the American Joint Committee on Cancer Tumor, Node, Metastasis Staging System for breast cancer [30]. Tumor classification, tumor size, and histologic grade were derived from the SEER clinical data Demographics Standard demographics were assessed (i.e., age at diagnosis, ethnicity, and education). Ethnicity was coded as White, African American, or other, while

4 306 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) educational attainment was categorized as: (1) having less than a high school education, (2) high school graduate, (3) having attended some college, and (4) college graduate and beyond. Data were derived from participant self-report. Missing information in self-report was updated using SEER data for the following measures: age at diagnosis, ethnicity, response time from treatment to their completion of the survey, primary surgical treatment, and stage Data analysis We first described the population characteristics. We then compared proportions of women who answered the recurrence and survival questions accurately by demographic characteristics, number and types of information sources, provider characteristics (gender and physician communication), and surgical treatment received. We used logistic regression to examine the independent associations of these covariates with knowledge accuracy. Two dichotomous dependent variables were created based on the two knowledge questions (coded as 1 for correct answers and 0 for incorrect or don t know answers). We first regressed the knowledge accuracy variables on patient age, education, ethnicity groups, and stage. We then examined the independent association of the use of various sources of information with women s knowledge by adding additional covariates to the model. Specifically, we tested the independent effect of each category of information source by adding three sets of variables indicating personal relationships, mass media, and education/ research-based materials. Each set contained two dummy variables indicating whether the respondent used exactly one source or two or more sources from within the category. We also included dummy variables indicating whether the respondent used any Internet source and any pamphletrelated sources. Significance of group variables was determined using Wald tests. We also independently examined the role of the Internet and health pamphlets on knowledge. We tested the role of the patient s treatment by regressing the knowledge accuracy variables on treatment received. We evaluated provider factors by regressing the knowledge accuracy variables on surgeon gender, type of information provided by the surgeon, and surgeon satisfaction, after controlling for patient age, education, ethnicity groups, and stage. Finally, we evaluated whether women thought that concerns about recurrence and survival were important to their treatment decision making. We used logistic regression to calculate the proportion of women with incorrect answers to the two knowledge questions across respondent groups with different levels of concern about disease recurrence risk controlling for demographic factors. All analyses were evaluated for second order interactions and point estimates were adjusted for design effects by using a sample weight that accounted for differential selection by stage, ethnicity, and non-response. 3. Results 3.1. Sample characteristics Table 1 shows the weighted and unweighted characteristics of the study participants (subsequent statistics presented in text are the weighted results). The sample consisted of 1215 women who had been diagnosed with either stage 0 (DCIS) or 1 breast cancer during the time period of December 2001 January 2003 (mean time between diagnosis and completion of survey was 8.3 months). The women were, on average, 60.7 years old (range = 28 79); 76.7% were White and 15.7% were African American. Approximately one-third of the study participants reported having a high school degree or less, and slightly more than one-quarter had at least a college education. Furthermore, 33.4% of the women were diagnosed with DCIS and 66.6% with stage 1 breast cancer. The majority of women underwent BCS (74.0%); an additional 12.4% underwent both a mastectomy and Table 1 Characteristics of study sample (N = 1215) Unweighted (%) Weighted (%) Ethnicity White Black Other Education Less than high school High school graduate Some college College graduate and beyond Age at diagnosis (years) M = 59.8 (10.93) [range = 28 79] Provider gender Female Male AJCC stage 0 (DCIS) Treatment received BCS Mastectomy with reconstruction Mastectomy no reconstruction Type of information provided by surgeon Described both BCS and mastectomy Described only BCS Described only mastectomy Physician recommended Physician recommended BCS Physician recommended mastectomy Satisfaction with surgeon Low satisfaction High satisfaction M = 60.7 (9.91) [range = 28 79]

5 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) reconstructive surgery, while 13.5% had a mastectomy only. The majority of women (66%) believed they had played a role in the decision about which treatment to undergo Knowledge of breast cancer treatment Women s knowledge regarding survival and recurrence was assessed using two true/false questions. The survival question, The chances of being alive 5 years after surgery are the same for mastectomy as they are for lumpectomy with radiation therapy, was correctly identified as true by 48.5% of the women (with an additional 44.3% responding don t know ). Moreover, for the recurrence question, The chances that breast disease will come back after treatment are the same for mastectomy as they are for BCS with radiation therapy, only 18.7% of women correctly answered false (46.9% responded that they did not know the answer). Knowledge differed across ethnicity with 51.5% of White women answering the survival question correctly compared to 24.2% of African Americans ( p < after controlling for age, education, and stage). No ethnic differences were observed among responses to the recurrence knowledge question. Fig. 1 shows that respondents who reported having higher education levels demonstrated more accurate knowledge ( p < 0.03). For example, whereas 55.7% of the college graduates answered the survival question correctly, only 24.6% of women who had less than a high school education were able to provide the correct response (after controlling for age, race, and stage) How do women obtain their information? We asked women to indicate the sources of information they used to learn about breast cancer and their breast cancer treatment options. As shown in Table 2, the most commonly Fig. 1. Survival and recurrence knowledge across different levels of educational attainment. Proportions adjusted for age, stage, and ethnicity (N = 1215). Wald test: x 2 = 1.50, p > 0.20 for recurrence knowledge. Wald test: x 2 = 33.50, p < for survival knowledge. Table 2 Sources of information used by women (N = 1215) Percent of women who used source Health pamphlets or brochures 82.5 Friends 64.8 Books 59.6 Relatives 59.3 Scientific journal/research papers 51.7 Magazines 51.2 Internet (world wide web) 48.0 News source: newspapers, TV, radio 43.4 Spouse or partner 34.7 Television shows or popular movies 33.2 Place of worship 21.9 Information videos or CD-ROMs 13.4 used sources of information were health pamphlets and/or brochures, as 82.5% of women reported that they had consulted these materials. Friends (64.8%) and books (59.6%) were also frequently cited as a source of information. Slightly less than half of the women reported that they had searched the Internet for breast cancer information. Only 8% of women indicated that they did not consult any of the sources. In an open-ended question, women reported that books, health pamphlets, and the Internet were the most helpful sources. As expected, women also received information about breast cancer treatment from their surgeons. Nearly 70% of the women reported that their surgeon described both BCS and mastectomy to them, while 23.2% indicated that their surgeon informed them of BCS only, and 2.7% of the women reported that they learned only about mastectomy from their surgeon (see Table 1) Factors associated with accuracy on knowledge questions We next examined women s knowledge across three categories of information sources (personal relationships, mass media, and education/research-based materials) while controlling for age, education, ethnicity, stage, and other information source categories. We also separately examined the impact of the Internet and health pamphlets/brochures on women s knowledge. Fig. 2 shows that knowledge about survival rates was greater among women who used the Internet ( p < 0.02), health pamphlets ( p < 0.002), and education/research-based sources ( p = 0.06) to obtain information. Furthermore, knowledge of recurrence was increased among women who used more than one education/ research-based source ( p < 0.003; see Fig. 3). Women who had a mastectomy had greater knowledge regarding recurrence than did those women who had BCS (29.3% accurate versus 14.9%, Wald x 2 = 36.6, p < 0.001). In contrast, women who had undergone BCS were more knowledgeable about survival than were those who had received a mastectomy (52.2% versus 38.22%, Wald x 2 = 106.0, p < 0.001).

6 308 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) Fig. 2. The impact of the number of informational sources on patient knowledge of survival. Proportions adjusted for age, education, stage, and ethnicity (N = 1215). Wald test: x 2 = 5.5, p < 0.02 for Internet; Wald test: x 2 = 9.8, p < for health pamphlets. As described in Table 3, results from logistic regression analysis indicate that, after controlling for age, ethnicity, education, and stage, surgeon gender was significantly associated with women s knowledge about recurrence ( p < 0.03), with patients of female surgeons possessing greater knowledge of recurrence than patients who reported having a male surgeon. In addition, women s knowledge about survival was found to be related to the type(s) of treatment described by their surgeon ( p < 0.02). Surgeon satisfaction was not related to women s recurrence or survival knowledge Were concerns about recurrence and survival important factors in women s treatment decisions? Given the lack of knowledge about the survival and recurrence rates across treatment options, we examined whether women considered these issues important in their treatment decision making. If knowledge about these issues was not salient in the surgical decision, the findings would be less troubling. However, we found that 50.7% of women indicated that their treatment decision was greatly Fig. 3. The impact of the number of informational sources on patient knowledge of recurrence. Proportions adjusted for age, education, stage, and ethnicity (N = 1215). Wald test: x 2 = 8.7, p < 0.01 for education/ research. Table 3 Effect of patient provider interactions on recurrence and survival knowledge Odds ratio Confidence interval (95%) Recurrence knowledge Surgeon gender Female 1.00 * Male Surgery treatments discussed BCS and mastectomy 1.00 BCS only 0.58 * Mastectomy only Don t remember Wald test x 2 = 7.10, p = Satisfaction with surgeon Low satisfaction 1.00 High satisfaction Survival knowledge Surgeon gender Female 1.00 Male Surgery treatments discussed BCS and mastectomy 1.00 BCS only Mastectomy only 0.34 * Don t remember Wald test x 2 = 9.7, p < 0.03 Satisfaction with surgeon Low satisfaction 1.00 High satisfaction Analyses adjusted for age, education, stage, and ethnicity. * p < influenced by their desire to reduce the chance that their breast disease would come back and 44.8% of women claimed that their decision was greatly influenced by their worry about their breast disease coming back. Yet, 77.4% and 77.6% (after controlling for age, education, ethnicity, and stage) of these women did not know that the recurrence rates are indeed unequal for mastectomy and BCS. However, women who indicated that their treatment decision was greatly influenced by their desire to reduce the chance that their disease would come back or their worry about the breast disease coming back were more accurate on the recurrence question than women whose decision was not greatly influenced by their decision (t =2.2 and 2.0, p < 0.04, after adjusting for age, education, and ethnicity). Of the women who indicated, in an open-ended survey question, that recurrence was the most influential factor in their treatment decision making, only 20.8% answered the recurrence question correctly. Furthermore, of the women who stated that their concern about survival most influenced their treatment decision, less than half (41.5%) knew that survival rates do not differ between mastectomy and BCS with radiation therapy.

7 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) Discussion and conclusion 4.1. Discussion This study revealed that the majority of breast cancer patients in two large population-based samples had levels of knowledge that are not commensurate with making informed decisions about their breast cancer surgical treatment. Less than half of our sample knew that the survival rates for mastectomy and BCS with radiation are identical. Additionally, fewer than one in five study participants knew that recurrence rates differ across local treatment. Other analyses (reported elsewhere) revealed that the women in our study also had very low knowledge about other treatments such as breast reconstruction [31]. Ours is one of the few studies to address these issues in a breast cancer context. If ours were the only study that reported such poor knowledge about such important treatment components, perhaps it would not raise significant questions. However, over many years a considerable number of studies have reached similar conclusions [32 34]. Our results are disturbing since they indicate that patients remain poorly informed even after many years of legal and ethical insistence on informed consent. The current study s findings suggest that patients lack of knowledge may not be due to easily fixable problems. First, it may be that the information that patients are tested on is not the information that is most influential in their decision making. This is likely not the case, though, since many women in our study reported that either survival or recurrence was the most important factor in their treatment decision making process and most of these women did not have accurate knowledge about these treatment issues. A second potential explanation is that patients lack of access to information may cause poor outcomes, such as psychosocial health [35]. The current results do not support this hypothesis as patients reported having had exposure to a relatively large variety of information sources. In fact, nearly three-quarters of the women recalled being told about both mastectomy and BCS by their surgeons; but even among these women, knowledge was low. In addition, many patients reported consulting other notable information sources, such as health pamphlets, the Internet, and journal articles. Only 8.0% never consulted any of the 12 sources of information (although they may have consulted sources other than those listed). Of course, it is possible that even though women consulted numerous sources, the sources did not contain the information that was measured in the current study. Other explanations for women s lack of knowledge are more plausible. First, some of the sources women consulted may have provided inaccurate information. In fact, the second and fourth most commonly used sources were friends and family members, many of whom may have provided women with anecdotal and/or erroneous information. This is an important consideration, as anecdotal information has been found to have a disproportionate amount of influence on people s treatment decisions [36,37]. Additionally, it is possible that the information was incomplete and did not contain information regarding recurrence and survival rates. Another explanation is the possibility that cognitive barriers related to numeracy and literacy may inhibit processing of risk-related information. There is compelling evidence that a substantial proportion of patients have difficulty understanding basic risk information about diseases and treatments [38]. Yet, many materials developed for patients (e.g., decision aids and health pamphlets) are written at a college reading level, and in a way that is often not conducive for most patients to truly understand the risks and benefits of treatment [39]. A third potential explanation for this observed lack of knowledge is the presence of emotional barriers that inhibit patients who are dealing with a potentially life threatening illness from absorbing treatment information. For example, one-third of the women in our study said that the fear that they felt upon receiving the diagnosis of cancer made it difficult for them to make a good decision. It is possible that this fear inhibited women from hearing and remembering the information given to them by their provider. In fact, previous research has found that high levels of anxiety and emotion can lead to decreased knowledge and/or recall of information [40 42]. A fourth possible explanation is that patients simply do not have enough time to make an informed decision. Respondents reported making their treatment decisions very quickly; in fact, half chose a surgical treatment during their first consultation with their surgeon. This would suggest that many patients select a treatment so soon after they are presented with their options that they could not have had time to process the information they had received and effectively use it in making their final treatment decision. Cognitive representations may have played an important role in women s understanding of benefits of mastectomy and BCS. For instance, women s schemas ( the active organization of past reactions, or of past experiences that guide reconstruction or memory for information [43]) regarding the risks and benefits of mastectomy and BCS may have impaired their ability to assimilate the numerical information into their knowledge base. For example, if women believe that BCS is as effective as mastectomy, they may not incorporate the fact that mastectomy has a lower recurrence rate than BCS into their schema of the two treatments. This would thus result in inaccurate knowledge when subsequently tested. Finally, women may have had better knowledge at the time they made their treatment decision, but in the interim had forgotten or not retained information that was salient months earlier. However, secondary analyses revealed that no association existed between patient knowledge and the duration from surgical treatment date to completion of the patient questionnaire.

8 310 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) Several potential limitations merit comment. Although this is a large, population-based study, it was limited to two large metropolitan areas, excluded Asian women, and only included women with stage 0 (DCIS) and 1 breast cancer. Thus, study findings may not be generalizable to women who reside in other geographic regions, to Asian women, and/or to women with more advanced breast cancer. In addition, the current study focused on a limited range of knowledge issues and used a limited number of questions. Finally, this was a retrospective cross-sectional study and associations which were observed between women s knowledge at the time of survey administration and key patient and provider characteristics may not reflect associations which may have been present at the time the surgical treatment decision was actually made Conclusion Our findings challenge current thinking about the role of cognition in treatment decisions and how to improve it, and suggest that it is time for researchers to further explore some critical questions: How much information is it reasonable to expect patients to understand, analyze, and use, especially during times of stress? What changes can be made to facilitate and improve physician patient communication? To what extent can changes in the ways patients are instructed in the facts of their disease and treatment options improve patient decision making? And how should we help patients participate in the decisions that affect their lives given whatever limitations in learning that this research revealed? Until these questions are addressed, true informed decision making will not be a clinical reality Practice implications Our study results may help to inform interventions for improving patient knowledge of breast cancer treatment. The conventional solution has traditionally relied on decision aids, which are interventions designed to help people make specific and deliberative choices among options by providing (at the minimum) information on the options and outcomes relevant to the person s health status [44]. The stress and anxiety inherent in a potentially life-threatening and life-changing decision such as breast cancer treatment makes decision aids especially valuable, as it may be difficult for patients to focus on and remember all of the detailed treatment information provided by their surgeon. Several challenges should be addressed to develop the most effective cancer treatment decision aids. First, the tools patients receive must be designed to address the counterintuitive notion that breast cancer surgery options yield the same survival but different local recurrence risks. Second, the tools must be written at low literacy and numeracy levels, as well as in multiple languages. Patients with low literacy and numeracy skills and/or who do not speak English as a first language are at a significant disadvantage in attempting to make critical medical decisions [45]. However, it is critically important to understand that designing decision aids for non-english speakers is not merely about translating the words, but also incorporating cultural beliefs and sensitivities. A less conventional solution lies in the need to slow down the treatment decision making process. While breast cancer is indeed a very serious disease, the emotion-driven perception that it must be treated emergently may not be justified. It is critical for physicians to communicate to patients that they have time to explore various information sources and to make an informed treatment decision. This is counter to popular lore, and as such this idea may initially encounter resistance from patients. However, for women who want to be involved in the treatment decisions, it is crucial that they are reassured that they have the time necessary to evaluate their treatment options. Finally, allowing patients to record office visits and use the tapes to help remember the information provided by their physician may help improve recall. In fact, a comprehensive literature review demonstrated that providing audiotapes to patients can help improve information recall (although more research needs to investigate this question) [46]. Other research has shown that patients prefer audiotapes to summary letters and that they use the tapes and share them with friends and family [47]. Our study is now one of many to conclude that many patients make medical decisions with inadequate and even incorrect information. In breast cancer this finding is particularly noteworthy as it comes after many years of efforts to improve the quality of information available to breast cancer patients and to promote patient involvement in the surgical treatment decision process. We must find improved methods, either using decision aids or by improving patient physician discussions, for communicating the risks and benefits of breast cancer treatment to women. Until we provide women with the information they need and can understand, we are at risk of having thousands of women making uninformed and potentially harmful medical decisions about their breast cancer treatment. Acknowledgements Portions of this work were presented at the annual meeting of the Society for Medical Decision Making in October This work was funded by a grant from the National Cancer Institute (RO1 CA8837-A1) to the University of Michigan. This project has been funded in part with Federal funds from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services, under Contract Nos. N01-PC and NO1-PC Dr. Fagerlin is supported by an MREP Early Career Award from the U.S. Department of Veterans Affairs. The collection of cancer incidence data used in this

9 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) publication was supported by the California Department of Health Services as part of the statewide cancer reporting program mandated by California Health and Safety Code Section The funding agreements ensured the authors independence in designing and conducting the study, interpreting the data, and publishing the report. References [1] Brody DS. The patient s role in clinical decision-making. Ann Intern Med 1980;93: [2] Kinmonth AL, Woodcock A, Griffin S, Spiegal N, Campbell MJ. Randomised controlled trial of patient centred care of diabetes in general practice: impact on current wellbeing and future disease risk. The Diabetes Care from Diagnosis Research Team. Brit Med J 1998;317: [3] Moyer A, Salovey P. Patient participation in treatment decision making and the psychological consequences of breast cancer surgery. Wom Hlth: Res Gender Behav Policy 1998;4: [4] Weiss SM, Wengert Jr PA, Martinez EM, Sewall W, Kopp E. Patient satisfaction with decision-making for breast cancer therapy. Ann Surg Oncol 1996;3: [5] Haynes RB, McKibbon KA, Kanani R. Systematic review of randomised trials of interventions to assist patients to follow prescriptions for medications. Lancet 1996;348: [6] Stewart MA. Effective physician patient communication and health outcomes: a review. Can Med Ass J 1995;152: [7] Fallowfield L. Offering choice of surgical treatment to women with breast cancer. Patient Ed Couns 1997;30: [8] Street Jr RL, Voigt B. Patient participation in deciding breast cancer treatment and subsequent quality of life. Med Decis Making 1997;17: [9] Barry M, Cherkin D, Chang Y, Fowler Jr F, Skates S. A randomized trial of a multimedia shared decision-making program for men facing a treatment decision for benign prostatic hyperplasia. Dis Manage Clin Outcomes 1997;1:5 14. [10] Wennberg JE. Promoting disease management in Medicare. Testimony before the Subcommittee on Health of the House Committee on Ways and Means; [11] Wennberg JE. Unwarranted variations in healthcare delivery: implications for academic medical centres. Brit Med J 2002;325: [12] 13th Annual Coggeshall Lecture. University of Chicago; [13] Institute of Medicine. Protecting data privacy in health services research National Academy Press; [14] Keating NL, Guadagnoli E, Landrum MB, Borbas C, Weeks JC. Treatment decision making in early-stage breast cancer: should surgeons match patients desired level of involvement? J Clin Oncol 2002;20: [15] Schneider CE. The practice of autonomy: patients, doctors, and medical decisions Oxford University Press; [16] Institute of Medicine and Commission on Life Sciences. Ensuring quality cancer care Washington, DC: National Cancer Policy Board, Institute of Medicine Report and National Research Council, National Academy Press; [17] Lantz PV, Zemencuk JK, Katz SJ. Is mastectomy overused? A call for an expanded research agenda. Hlth Serv Res 2002;37: [18] Montini T. Resist and redirect: physicians respond to breast cancer informed consent legislation. Wom Hlth 1997;26: [19] Nayfield SG, Bongiovanni GC, Alciati MH, Fischer RA, Bergner L. Statutory requirements for disclosure of breast cancer treatment alternatives. J Natl Cancer I 1994;86: [20] National Institutes of Health. Treatment of early-stage breast cancer. NIH consensus statement online, vol. 8(6) ed.; June 18 21, [21] Anon. Mastectomy or lumpectomy? The choice of operation for clinical stages I and II breast cancer. The Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer. Canadian Association of Radiation Oncologists. Can Med Ass J 1998;158(Suppl. 3):S [22] Morrow M, Strom EA, Bassett LW, Dershaw DD, Fowble B, Giuliano A, Harris JR, O Malley F, Schnitt SJ, Singletary SE, Winchester DP. Standard for breast conservation therapy in the management of invasive breast carcinoma. CA-Cancer J Clin 2002;52: [23] Jacobson B, Bowling A. Involving the public: practical and ethical issues. Brit Med Bull 1995;51: [24] Jacobson JA, Danforth DN, Cowan KH, d Angelo T, Steinberg SM, Pierce L, Lippman ME, Lichter AS, Glatstein E, Okunieff P. Ten-year results of a comparison of conservation with mastectomy in the treatment of stage I and II breast cancer. New Engl J Med 1995; 332: [25] Lerner BH. The breast cancer wars: hope, fear, and the pursuit of a cure in twentieth-century America New York: Oxford University Press; [26] Veronesi U, Cascinelli N, Mariani L, Greco M, Saccozzi R, Luini A, Aguilar M, Marubini E. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. New Engl J Med 2002;347: [27] van Dongen JA, Voogd AC, Fentiman IS, Legrand C, Sylvester RJ, Tong D, van der Schueren E, Helle PA, van Zijl K, Bartelink H. Longterm results of a randomized trial comparing breast-conserving therapy with mastectomy: European Organization for Research and Treatment of Cancer trial. J Natl Cancer I 2000;92: [28] Anema MG, Brown BE. Increasing survey responses using the total design method. J Contin Educ Nurs 1995;26: [29] Dillman DA. Mail and telephone surveys: the total design method New York: John Wiley & Sons, Inc.; [30] American Joint Committee on Cancer. AJCC cancer staging manual, 5th ed., Philadelphia: Lippincott-Raven Publishers; [31] Morrow M, Mujahid M, Lantz PM, Janz NK, Fagerlin A, Schwartz K, Liu L, Deapen D, Salem B, Lakhani I, Katz J. Patterns Correlates of breast reconstruction: results from a population-based study. Cancer 2005;104(11): [32] Coward DD. Lymphedema prevention and management knowledge in women treated for breast cancer. Oncol Nurs Forum 1999;26: [33] Alexander M, Gordon NP, Davis CC, Chen RS. Patient knowledge and awareness of hypertension is suboptimal: results from a large health maintenance organization. J Clin Hypertens 2003;5: [34] Roche-Nagle G, Chambers F, Nanra J, Bouchier-Hayes D, Young S. Evaluation of patient knowledge regarding oral coagulants. Irish Med J 2003;96: [35] Arora NK, Johnson P, Gustafson D, McTavish F, Hawkins RP, Pingree S. Barriers to information access, perceived health competence, and psychosocial health outcomes: test of a mediation model in a breast cancer sample. Patient Educ Couns 2002;47: [36] Ubel PA, Jepson C, Baron J. The inclusion of patient testimonials in decision aids: effects on treatment choices. Med Decis Making 2001;21:60 8. [37] Institute of Medicine, Board on Neuroscience and Behavioral Health. In: Nielsen-Bohlman L, Panzer AM, Kindig DA, editors. Health literacy: a prescription to end confusion. Washington, DC: Committee on Health Literacy; [38] Committee on Health Literacy. In: Nielsen-Bohlman L, Panzer AM, Kindig DA, editors. Health literacy: a prescription to end confusion. Washington,. DC: Board of Neuroscience and Behavioral Health; 2004 [39] Fagerlin A, Rovner DR, Jentoft C, Wei JT, Holmes-Rovner M. Patient education materials about the treatment of early-stage prostate cancer: a critical review. Ann Intern Med 2004;140: [40] Maguire P, Faulkner A. Communicate with cancer patients: 1. Handling bad news and difficult questions. Brit Med J 1988;297: [41] Jedlicka-Kohler I, Gotz M, Eichler I. Parents recollection of the initial communication of the diagnosis of cystic fibrosis. Pediatrics 1996;97:204 9.

10 312 A. Fagerlin et al. / Patient Education and Counseling 64 (2006) [42] Sorg BA, Whitney P. The effect of trait anxiety and situational stress on working memory capacity. J Res Pers 1992;26: [43] Bartlett FC. Remembering Cambridge: Cambridge University Press; [44] O Connor AM, Stacey D, Entwistle V, Llewellyn-Thomas H, Rovner D, Holmes-Rovner M, Tait V, Tetroe J, Fiset V, Barry M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2003;CD [45] Gazmararian JA, Williams MV, Peel J, Baker DW. Health literacy and knowledge of chronic disease. Patient Educ Couns 2003;51: [46] McClement SE, Hack TF. Audio-taping the oncology treatment consultation: a literature review. Patient Educ Couns 1999;36: [47] Tattersall MH, Butow PN, Griffin AM, Dunn SM. The take-home message: patients prefer consultation audiotapes to summary letters. J Clin Oncol 1994;12: