B-Sure: a randomized pilot trial of an interactive web-based decision support aid versus usual care in average-risk breast cancer patients considering contralateral prophylactic mastectomy

Abstract

The use of contralateral prophylactic mastectomy (CPM) is increasing among breast cancer patients who are at average or “sporadic” risk for contralateral breast cancer. Because CPM provides no survival benefit for these patients, it is not medically recommended for them. Decision support aids may facilitate more informed, higher quality CPM decision. The purpose of this study was to evaluate the feasibility and acceptability of B-Sure, an online decision support aid to facilitate informed decisions regarding CPM, and to compare the impact of B-Sure in increasing CPM knowledge, reducing decisional conflict, and increasing preparedness to make the CPM decision among breast cancer patients at sporadic risk who are considering CPM. Ninety-three patients with unilateral, nonhereditary breast cancer considering CPM completed a baseline survey, were randomized to receive B-Sure or Usual care, and completed a 4-week follow-up survey assessing decisional conflict, preparedness to make the CPM decision, and CPM knowledge as well as self-efficacy, perceived risk, worry, CPM motivations, and the surgical decision. Study participation was high. B-Sure was viewed by almost 80% of the participants and was evaluated positively. At follow-up, patients assigned to B-Sure reported significantly higher clarity regarding the personal values relevant to the CPM decision and higher knowledge about CPM. B-Sure had smaller effects on other aspects of decisional conflict. B-Sure improved CPM knowledge and reduced decisional conflict. Patients considering CPM may benefit from an online decision support aid, but may be sensitive to approaches that they perceive as biased against CPM.

Implications

Practice: Sporadic risk breast cancer patients considering CPM have a relatively low level of knowledge about the CPM procedure and its benefits and risks. An online decision support aid, B-Sure, increased patient knowledge about the CPM procedure, perceived risks associated with the procedure, and lack of producing a survival benefit. B-Sure also clarified personal values in terms of what was most important to the patient.

Policy: Oncology settings and hospital policy makers should consider allocating resources to support the implementation of online decision support interventions like B-Sure for sporadic risk breast cancer patients considering CPM.

Research: Future research might benefit from a large-scale randomized controlled clinical trial evaluating B-Sure with a more heterogeneous sample that includes more minorities, patients with lower education and less income, patients seen in community-based oncology settings or target patients who are undecided about this surgery.

The rate of contralateral prophylactic mastectomy (CPM) has risen exponentially in the past decade [1–5]. The Society of Surgical Oncology’s position statement endorses CPM among women carrying BRCA 1–2 mutation, patients with high-risk lesions, patients with a prior unilateral breast cancer, patients who have difficulty following contralateral breast screening guidelines, or for cosmetic reasons [6,7]. Because CPM provides no survival benefit for women who are at average, or “sporadic” risk for contralateral breast cancer (CBC), it is not recommended for these patients. Despite this recommendation, up to two-thirds of women who choose CPM have sporadic breast cancer and are not in the high-risk category [8,9].

Research has identified a number of reasons patients select CPM, which include the desire to lower contralateral breast cancer risk, improve survival, improve breast symmetry, restore peace of mind, reduce worry about cancer, and avoid future cancer surveillance [10–13]. Women with sporadic breast cancer may also select CPM because they do not receive balanced information about the benefits and risks of CPM [12]. Decisions about surgical options are compressed into the first few weeks after diagnostic biopsy [14], which may contribute to pressure to make treatment decisions quickly and acceptance of more aggressive treatments regardless of their relative benefit [14].

An optimal medical decision is informed, consistent with personal values, acted upon, and results in high satisfaction [15,16]. Patients considering CPM may benefit from decision support in the form of aids to facilitate optimal decision-making. Decision support aids are used to supplement practitioners’ counseling and are designed to facilitate understanding of options, help patients weigh advantages and disadvantages of options, increase patients’ awareness of the personal importance attached to the benefits and risks of each option, and encourage patients to engage with their health care providers in deciding which option to pursue [17]. In the context of CPM decisions, decision support could address cancer worry, provide alternative ways of coping with worry, and provide education about actual risk versus perceived risk. Ultimately, improved decision-making quality might contribute to greater long-term decisional satisfaction [18]. In the oncology context, decision support has been shown to facilitate decisions about breast cancer risk reduction among women considering BRCA genetic testing [19] and breast cancer surgery [20,21].

To date, there has been only one evaluation of a decision support aid for CPM [15]. This decision support aid provided information about the tumor characteristics completed by the surgeon, surgical choices, surgical treatment goals, a list of the roles of the people involved in the decision-making process, estimates of outcomes relevant to each of the goals to allow for comparison between surgical options, and a page where the surgeon selected the surgery chosen. This study was not a randomized clinical trial and included women at hereditary risk.

In this study, we report on the development, feasibility, and pilot testing of a web-based, interactive decision aid called “B-Sure” for women with unilateral, nonhereditary breast cancer considering CPM. B-Sure address key elements of decision support such as clarifying risks and benefits of options, described nonmedical motivations for CPM, and described the experiences of patients who both chose and did not choose CPM. The study had two aims. The first aim was to evaluate the feasibility and acceptability of B-Sure in a small randomized clinical trial among breast cancer patients considering CPM. Feasibility was measured as study enrollment and retention. Acceptability was assessed by use and evaluations of B-Sure. The second aim was to assess the impact of B-Sure versus usual care (UC) on the primary outcomes of CPM knowledge, preparedness to make a decision, decisional conflict, and on secondary outcomes of self-efficacy, perceived risk for contralateral and metastatic cancer, worry, motivations for CPM, and the final decision about CPM.

METHODS

Development and content of B-Sure decision aid

B-Sure was developed over a 6-month period of time in weekly meetings between the study investigators and developers and programmers at ITX Corporation (Rochester, NY), a technology company who designs online education and consumer-facing applications. Content was guided by the Ottawa Framework for Informed Decision Making [22], which suggests that decision support should enhance knowledge and willingness to consider options, clarify values and motivations, facilitate discussion with doctors, family, and friends, improve coping ability, and assist in developing skills to make and implement the decision. Patient experiences were incorporated into the content by using quotes taken from interviews about the CPM decision with eleven patients who either selected or did not select CPM. Pictures of bilateral and unilateral mastectomies as well as different reconstruction options were provided by surgeons. Before the trial was launched, patients provided feedback on B-Sure content and ease of navigation. Changes were made before going live. B-Sure had five modules and content is shown in Table 1.

Table 1.

B-Sure content

Module	Content
Module 1	Orientation and introduction by surgical oncologist on videotapeB-Sure online navigation instructionsBasic glossary of key medical terms in B-Sure
Module 2	Basic education about average versus high riskWhen CPM is recommended and when it is not recommendedQuestions and answers regarding CPM
Module 3	Information about surgical options
	Timelines for lumpectomy, mastectomy, and bilateral mastectomy comparing length of procedure, average hospital stay, length of recovery time, risks, and complications from each procedure
	Reconstruction options
	Graphics depicting incisions for lumpectomy, unilateral and bilateral skin-sparing mastectomy, nipple-sparing mastectomy, mastectomy without reconstruction
	Patient photograph examples of reconstruction outcomes
	Common questions and answers about surgery risks and outcome
	True–false quiz with correct answers provided
Module 4	Nonmedical considerations in the CPM decision
	Patients rate their motivations:
	Worry about cancer coming back
	Worry about having a mammogram or regular breast cancer screening
	Additional surgery, longer recovery time, greater surgical complications of CPM
	Cosmetic appearance of your breasts
	Description of possible coping strategies and perspectives for each motivation by patients who selected and did not select CPM via audiotaped narratives
	Coping strategies discussed include:
	Focusing on CPM facts
	Making a plan to reduce your cancer risk by taking good care of your health
	Viewing mammograms as a way of engaging in self-care
Module 5	Review and summary of key points
	Can be printed and e-mailed to participant

CPM contralateral prophylactic mastectomy.

Participants and procedures

Eligible patients were: women 18 years of age or older with a diagnosis of Stage 0-3A breast cancer without hereditary breast cancer who were considering CPM and consulting a breast cancer surgeon at Memorial Sloan Kettering Cancer Center, NY, NY or at Massachusetts General Hospital, Boston, MA. The Tyrer Cuzick [23] risk model was used to calculate risk for a hereditary cancer syndrome if there was uncertainty about risk. Patients were required to speak English, have home internet access, and be able to provide informed consent. Patients were identified by surgeons or study staff and then approached after the appointment with the breast surgeon when surgical options were discussed. Eligible participants were provided written or electronic consent and completed a survey in clinic or online. Participants were paid $25 for the baseline and $25 for the follow-up survey.

The surgeon provided potential participants a study overview and obtained permission for staff to follow up. Participants signed a written informed consent document approved by the Institutional Review Board at each site and were either given a paper baseline in clinic or sent a link to the online survey via e-mail. After the consent and survey were received, participants were randomized to B-Sure or UC. Both sites followed the same randomization procedures. B-Sure participants received written and/or e-mailed instructions for accessing the B-Sure site and a telephone support number. UC participants received standard care which entailed a discussion with the breast surgeon about surgical options and referral to a plastic surgeon for a discussion about breast reconstruction. Participants in both B-Sure and UC received a follow-up survey 2–4 weeks after surgery.

Measures

Participant age, marital status, race/ethnicity, marital status, employment status, and health insurance status were collected on the baseline survey. Eastern Cooperative Oncology Group (ECOG) performance status [24] and date of diagnosis were collected from medical chart.

Primary outcomes were CPM knowledge, decisional conflict, and preparedness. Knowledge was measured by a 10-item multiple-choice scale developed for this study by two authors (L.K., B.L.S.). Items assessed understanding of the definition of CPM, surgical recovery time and risks/side effects, whether or not CPM improves survival, and whether CPM reduced the risk for disease progression. Preparedness for the CPM decision was assessed using a 16-item scale adapted from the Ottawa Preparation for Decision Making scale modified to address the CPM decision [17]. Items evaluated the amount of and satisfaction with information about CPM. Internal consistency was 0.92 and 0.85 at baseline and follow-up. Decisional Conflict was measured using the Ottawa Decisional Conflict scale [25] and has 16 items on five subscales: Uncertainty, Uninformed, Unclear values, Unsupported, and Ineffective decision making. Participants rated the CPM decision. Higher scores indicate more uncertainty. Internal consistencies at baseline and follow-up, respectively, were: Uncertainty, alpha = 0.91, 0.86; Uninformed, alpha = 0.88, 0.83; Unclear values, alpha = 0.90, 0.82; Unsupported, alpha = 0.95, 0.91; ineffective decision making, alpha = 0.95, 0.91. Confidence in the decision made (“How confident are you that the decision you will ultimately make/made?”) was rated on a scale from 0 (not confident at all) to 10 (extremely confident).

Secondary outcomes were self-efficacy, perceived risk, worry, reasons for considering CPM, and the CPM choice at follow-up. Self-efficacy was a 3-item measure of confidence in the ability to manage worries and uncertainty about a possible recurrence of breast cancer, concerns about future surveillance, and worries about undergoing future surveillance. Internal consistency was 0.70 at baseline and 0.72 at follow-up. Perceived risk was measured by two items. One item was: “100 women with early breast cancer are treated with a single mastectomy or lumpectomy and radiation, about how many will develop breast cancer in the “other breast” in the 5 years after treatment? ___ women out of 100.” The second item assessed risk for chest wall recurrence after bilateral mastectomy using the same scale. Worry was measured by a single item that asked how worried the participant was about having another form of breast cancer on a four-point Likert scale. Higher scores indicate more worry. Reasons for CPM were measured by an 11-item scale developed after a review of the qualitative and quantitative literature on motivations for CPM [12,13], and interviews with women who chose or did not choose CPM (Author, unpublished data). Internal consistency was 0.73 at baseline and 0.86 at follow-up. Higher scores indicate more reasons. Surgery obtained (CPM or not) was abstracted from medical chart at follow-up.

Baseline CPM intention was a proposed treatment moderator and was assessed using a single item, “How interested are you in having CPM at this time?” B-Sure evaluation consisted of 19 items that included ease of navigation, helpfulness of program components, degree to which B-Sure presented balanced information, helped with the decision, clarified the pros and cons of CPM, and contained new information. Alpha for the scale was 0.91. The User Interface Satisfaction scale [26] has 27 items and four subscales: usefulness, ease of use, ease of learning, and satisfaction. Internal consistency for the subscales was good (alpha = 0.85–0.93). Open-ended questions also asked about participants’ opinions about the positive and negative aspects of B-Sure. B-Sure use was assessed by average time spent in B-Sure and pages viewed.

Statistical analysis

For the outcomes analysis, because there was some missing data (n = 10) at follow-up, we imputed missing values from the observed data using multiple imputation with 50 imputed samples using SPSS Version 22. All reported statistics are based on the pooled results from those 50 imputed samples. Multiple regression was used to test for differences in outcomes as a function of the treatment. In addition to including a coded variable denoting treatment condition, regression models included the participant’s baseline score on the outcome measure, age, ethnicity (White/not Hispanic vs. others), and education (postgraduate education vs. others) as covariates. Given that this was a small pilot study sample, our focus is primarily on measures of effect size rather than statistical significance.

RESULTS

Study sample

A total of 93 women were enrolled, with 46 women randomly assigned to B-Sure and 47 women randomly assigned to in Usual Care. The CONSORT is shown in Fig. 1. Patient demographics are shown in Table 2.

Fig 1.

Study CONSORT. BL baseline; FU follow-up; UC usual care. One patient dropped before follow-up. One patient became ineligible before follow-up.

Table 2.

Sample characteristics

Variable	B-Sure	(n = 46)	UC	(n = 47)
	n	%	n	%
Race
Asian	2	4.3	3	6.4
Black, Hispanic	0	0	1	2.1
Black, non-Hispanic	2	4.3	3	6.4
White, Hispanic	1	2.2	2	4.3
White, non-Hispanic	38	82.6	33	70.2
American Indian/Alaskan Native	1	2.2	0	0
Two or more races	2	4.3	1	2.1
Other	0	0	4	8.5
Ethnicity
Hispanic	3	6.5	4	8.5
Education
High School	2	4.3	3	6.4
Some college	8	17.4	6	12.8
Tech or trade school	1	2.2	1	2.1
College graduate	15	32.6	17	36.2
Some graduate work	2	4.3	3	6.4
Graduate degree	18	39.1	17	36.2
Study site
MSKCC	36	78.3	37	78.7
MGH	10	21.7	10	21.3
Marital Status
Married	39	84.8	36	76.6
Single	5	10.9	7	14.9
Divorced/widowed	3	10.9	4	8.5
Insurance (yes)	45	97.8	45	95.7
Employment status
Full time	28	60.9	23	48.9
Part time	8	17.4	9	19.1
On leave	5	10.9	4	8.5
Does not work	5	10.9	11	23.4
Income
<$60,000	7	15.2	10	20.7
$60,000–$139,999	17	40.0	17	36.2
$140,000–$179,999	5	10.9	6	12.8
>$180,000	14	30.5	10	21.3
Missing	3	6.5	4	8.3
Age (M and SD)	47.5	8.4	45.5	8.4
Baseline decision made (yes)	36	78.3	41	87.2

MGH Massachusetts General Hospital; MSKCC Memorial Sloan Kettering Cancer Center.

Feasibility

Of the 128 eligible patients approached, 93 participated, 27 refused, and 8 consented but did not complete the baseline before surgery or withdrew from the study. The acceptance rate was 72.7%. The most common reasons for refusal were having: already made a decision (n = 7) and feeling overwhelmed/no time (n = 6). A comparison of available data from the 93 participants with the 35 decliner suggested that the acceptance rate was significantly higher at one site (95.2%) than the other site (73%) (Chi-square = 4.83, p < .05). There were no differences between participants and refusers with regard to age, race/ethnicity, or ECOG performance status. As shown in Fig. 1, completion of the follow-up survey was 89.2%. Completion rate was lower in B-Sure: 84.8% of participants in B-Sure completed the follow-up and 93.6% of participants in UC completed the follow-up.

Acceptability

B-Sure use calculated from program tracking data indicated that 73.9% (n = 34) viewed B-Sure before surgery, 6.5% (n = 3) viewed B-Sure after surgery, and 19.6% (n = 9) did not view B-Sure. Among the 37 participants who opened B-Sure, average time spent in it was 52.68 min (range = 2–107). Approximately 80% viewed 75% or more of the 88 B-Sure pages (median = all 88 pages).

B-Sure participants rated it positively, with an average of 5.63 on a seven-point scale. Slightly more than half of the sample reported that the overall message of B-Sure was that the patient should not have CPM. Approximately 61% reported that the material was presented in a balanced way, and 54.6% reported B-Sure was helpful in making the CPM decision. The majority (87.8%) reported it was helpful in facilitating a better understanding of CPM’s pros and cons. User interface ratings of B-sure, as shown in Table 3, participants rated the ease of use and ease of learning higher than usefulness and satisfaction.

Table 3.

B-Sure evaluation

Variable	M	SD
Overall Evaluation	5.63	0.77
User Interface
Usefulness	4.92	1.42
Ease of use	6.03	0.76
Ease of learning	6.21	0.96
Satisfaction	5.56	1.52

Qualitative feedback was gathered in terms of what patients liked and did not like. Positive feedback included, “The video and personal testimonies and photos of before and after were excellent because it makes it more real. You do not feel as alone in this and you have a better understanding of what you can expect” and “It was easy to use.” Negative feedback included, “Slightly favored keeping the healthy breast,” and “It was trying to convince me not to have CPM. Need more testimonials of women who decided to have it.”

Preliminary efficacy

Table 4 presents the estimated marginal means, standard errors, and effect sizes for the primary and secondary outcomes. The strongest treatment effect emerged for CPM knowledge, with an effect size that was moderate in size. B-Sure participants were significantly more knowledgeable about CPM than women in UC. The difference in CPM knowledge between the two groups was statistically significant, with t (86) = 2.59, p = .01. The other difference between the two conditions emerged for the Unclear values Decision Conflict subscale, t (86) = 2.06, p = .04. B-Sure participants reported lower Unclarity about values (i.e., more clarity) at follow-up, d = .44. Table 4 also shows that B-Sure had small to moderate effects on three other decision conflict subscales: Feeling uninformed, Unsupported, and Ineffective decision-making, and on confidence in the decision. The effect size for Uncertainty and Preparedness to make the decision was small. Finally, in terms of surgery choice, 71.7% of women in the UC arm chose CPM and 78.5% in B-Sure chose CPM. This difference was not significant (Chi-square = 0.55). It should be noted that there was missing data on six patients enrolled B-Sure who had their surgery elsewhere.

Table 4.

Estimated marginal means, standard errors, and effect size measures for usual care and B-Sure decision aid conditions

Outcome	Usual care			B-Sure
	M	SE	M	SE	d
Primary Outcomes
CPM Knowledge	51.33	3.24	62.47	3.40	−.56
Preparedness	3.42	0.08	3.46	0.09	−.08
Decisional Conflict scales
Uncertainty	25.51	3.23	26.38	3.50	−.04
Informed	17.13	1.81	13.43	1.96	.33
Values Clarity	20.76	1.90	15.50	2.05	.44
Support	16.33	2.33	12.48	2.51	.27
Effective Decision	12.46	1.77	9.65	1.90	.25
Confidence in Decision	8.50	0.33	9.10	0.35	−.30
Secondary Outcomes
Efficacy to manage worry	3.85	0.14	4.00	0.16	−.16
Worry	2.84	0.21	2.91	0.24	−.05
CPM Motivations	2.73	0.11	2.77	0.12	−.05
Risk for CBC after unilateral mastectomy and radiation	5.10	1.98	6.44	1.97	−.13
Risk for chest wall recurrence after CPM	12.12	2.26	11.95	2.52	.01

CBC contralateral breast cancer; CPM contralateral prophylactic mastectomy. Higher scores on decision conflict scales indicate more conflict. Means are estimated marginal means controlling for the outcome variable measured at baseline, age, ethnicity (White/not Hispanic versus others), and education (Post graduate education versus others). Cohen’s ds were calculated using a t-to-d transformation based on the pooled values of t from the pooled results based on 50 samples of imputed data. Positive Cohen’s d indicates more positive outcomes in the B-sure condition.

There was little evidence that the treatment had an impact on the secondary outcomes of self-efficacy, perceived risk, worry, and motivations for CPM. We also tested whether age or initial CPM intentions moderated the effects of B-Sure. Results indicated that there were no significant interactions, and thus B-Sure did not have a stronger impact among younger patients or patients with lower CPM intentions.

DISCUSSION

B-Sure was developed to facilitate a more well-informed and higher quality decision among average-risk breast cancer patients considering CPM. B-Sure provided information to increase knowledge about CPM and recurrence risk and clarify risks and benefits of CPM to make expectations more realistic. Because anxiety about recurrence and future screening was a motivating factor, content included alternative self-management information to manage worry about cancer recurrence and future surveillance. An online delivery modality was selected to allow for improved options for dissemination and reach for future research.

The first aim was to evaluate B-Sure’s feasibility and acceptability. Our results suggest that feasibility was high. Participation in the trial was 72.1% and the survey return rate was nearly 90%. B-Sure was viewed by almost 80% of participants. Participants felt B-Sure was easy to navigate and that it was easy to learn material. Usefulness and satisfaction were lower than we anticipated.

The second aim was to evaluate the preliminary impact of B-Sure versus UC. B-Sure achieved a key goal of a decision support aid: It increased knowledge about the CPM procedure, its risks, and its lack of producing a survival benefit. Average knowledge about CPM increased from 48% to 63% in the B-Sure arm and from 45% to 51% in the UC arm. The effect on knowledge is consistent with the one prior decision support pilot intervention study [15]. B-Sure clarified patients’ personal values in terms of what was most important to the patient. The effect sizes for these two outcomes were in the small to medium range, with larger effects on CPM knowledge and on the clarity of personal values. Effects of B-Sure versus UC on other components of decisional conflict such as uncertainty, being uninformed, feeling unsupported, as well as confidence in the decision were in the small to moderate magnitude range, indicating some preliminary evidence of B-Sure’s decision support efficacy. Effect sizes for preparedness to make a decision and uncertainty were small, suggesting that it did not impact these aspects of the decision process.

We also evaluated the impact of B-Sure on known correlates of decisional conflict in the Ottawa Decision Support Framework, including self-efficacy, perceived risk for CBC and recurrence, worry, and CPM motivations. There was no evidence of an impact on these outcomes. It was relatively surprising that B-Sure did not impact participants’ ability to manage worry and uncertainty about possible recurrence of breast cancer or future surveillance and perceived risk for contralateral breast cancer, because B-Sure contained content addressing worry about recurrence and surveillance under the content addressing CPM motivations. However, the intervention content could be enhanced by highlighting it more and by making it more personalized and comprehensive. For example, patients could be asked to rate their worry about confidence in managing worries about recurrence and/or future surveillance, choose possible strategies to cope with these worries or propose their own, select barriers to implementing these coping strategies, and ways to address these barriers.

Compared with UC, B-Sure did not ultimately have an impact on the decision made. Since the goal of decision support aids is not to influence the decision made, but rather to facilitate a more informed decision that is consistent with the person’s values and goals, these findings are not disappointing from a patient care perspective. However, it should be noted that the participants as a whole were highly motivated to choose CPM (Baseline CPM intention, M = 5.5 on a seven-point scale). Thus, it is possible that participants were already highly interested in CPM and unlikely to change their decision.

In summary, an online decision support aid was viewed and evaluated positively by women at sporadic risk for breast cancer who were considering CPM. Preliminary efficacy data indicates that the aid clarified patients’ values relevant to this decision and improved their knowledge about CPM. Effect sizes for B-Sure indicated that it had a small to moderate impact on other aspects of decisional conflict such as decision uncertainty, being uninformed, feeling unsupported in this decision. It is important to point out that participants made the CPM decision with relatively low levels of knowledge about the CPM procedure and its risks. Average knowledge scores at baseline were relatively low across both study arms (M = 46% correct). Even with the emphasis on education in B-Sure, average knowledge scores at follow-up were not extremely high (67%). These findings suggest there may be limitations of the application of the Ottawa Decision Framework in CPM decision making. Knowledge is considered an important prerequisite for informed decision making in this framework but it did not appear to play a primary role in the CPM decision. One possible explanation is that personal values as well as affective factors may play a more important role. It has been well documented that decision-making is not solely a rational process and that emotions influence behaviors and decisions [27,28]. In the context of decisions regarding CPM, both immediate and anticipated emotions—particularly fear and anxiety—can interfere with the processing of facts and influence decisions and lead to the selection of aggressive treatments [29]. Fear and anxiety can divert attention to the most frightening aspects of a situation and can lead to what is known as affective forecasting [28]. Affective forecasting occurs when people predict how they will feel and respond in the future [30]. For example, people may project their current anxiety onto the future and anticipate long-term persistent anxiety. People can also underestimate their ability to adapt to negative events [31]. Although not well studied in the CPM context, patients may decide to have CPM because they believe they will experience persistent fear of another cancer [32,33]. It may be important for the B-Sure intervention to address affective influences more directly and thoroughly by devoting more content to this topic. Although affective influences were incorporated into the nonmedical considerations module, it is possible that participants did not review this material and/or reviewed the material quickly. Future enhancements to B-Sure might benefit from a separate module on affective considerations that allows patients to become more aware of the link between their cancer-related worries, thoughts, and behaviors (e.g., choosing surgery), and facilitate more attention to and processing of current and anticipated emotional reactions to breast cancer. Thorough emotional processing may help patients manage worries and develop confidence in the ability to manage current and future worries without seeking CPM. A second possible explanation may be that the factual information may need to be enhanced and/or presented in a simpler way to facilitate better understanding.

Feedback from the B-Sure evaluation suggest that patients considering CPM may be sensitive to approaches that they perceive as biased against CPM and may desire to hear more about experiences of patients who chose CPM. Enhanced content about values and how CPM may or may not fit with values may improve the impact on decisional outcomes. Future studies would benefit from a larger randomized trial and including a more heterogenous sample that includes more minorities, patients with lower education and less income, and patients seen in community-based oncology settings. Future enhancements to B-Sure may include a separate module and more content focusing on affective influences on the CPM decision, information regarding affective forecasting influences decisions, and enhanced content about managing worries about cancer recurrence. Finally, it may be important to target patients who are undecided about this surgery, who may benefit more from decision support.

We included the nine participants who reported not viewing B-Sure in our analyses of the decision aid efficacy. Analyses dropping those nine participants yielded highly similar results on all but one variable. The exception that the effect size for the CPM knowledge, which was moderate in size for the full sample (d = .56) was considerably larger (d = .82) when those patients were excluded.

Compliance with Ethical Standards

Funding sources: This study was funded by an NIH R21 grant (CA187643) to Sharon Manne and Laurie Kirstein.

Conflicts of Interest: The authors declare that they have no conflicts of interest.

Human and Animal Rights: All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

Informed Consent: Informed consent was obtained from all individual participants included in the study.