Abstract
Objective:
To quantify preferences for aspects of contralateral prophylactic mastectomy(CPM) decision-making process among key stakeholders.
Background:
Despite increasing numbers of women with unilateral breast cancer undergoing CPM, quantitative evidence of all stakeholder preferences regarding CPM is lacking, particularly for healthy volunteers. Conjoint analysis, a marketing tool, can be used to quantify tradeoffs surrounding CPM.
Methods:
Healthy volunteers(HV), women-with-cancer(WwCa), surgical oncologists(SO), and plastic surgeons(PS) were surveyed with the same conjoint simulation exercise. Respondents chose between either single(SM) or double(DM) mastectomy under varying recurrence and complication rates, surveillance, and symmetry conditions. Hierarchical Bayesian models calculated partworth utilities and importance scores.
Results:
Overall, 1244 respondents participated. The top three important factors for all stakeholders were surgical complication rates following DM, type of surgery(SM vs. DM) independent of other variables, and 10-year future contralateral cancer risk following SM. HV and surgeons placed greatest importance on high rates of surgical complications following DM. WwCa preferred DM regardless of complication risk or low rates of a 10-year future cancer episode following SM. SO strongly preferred SM and were more accepting of future cancer risk of 3% or 10% than other stakeholders. Symmetry and need for surveillance were least important factors for all stakeholders.
Conclusion:
The threshold of acceptability for future cancer episodes and risk tolerance for complications varies by stakeholder with a profound influence upon WwCA. The current findings suggest room for improved provider and patient alignment through behavioral techniques, such as framing, meanwhile highlighting changes in risk perception following a breast cancer diagnosis.
Keywords: mastectomy, prophylactic, decision, preference
Graphical Abstract
Introduction
Despite studies showing a lack of survival benefit among low-risk breast cancer patients who undergo double mastectomy, rates of contralateral prophylactic mastectomy (CPM) are increasing in the United States [1-13]. CPM is associated with a greater complication rate and loss of sensation in the uninvolved breast [11, 14-17]. However, CPM has potential non-medical benefits such as reduced worry about contralateral breast cancer episodes and absent need for future breast imaging. Improved reconstructive symmetry, especially with implants, [17-29] is also a factor in the rise in bilateral mastectomies [10, 14].
For patients without a strong family history or genetic predisposition, where the risk of a subsequent contralateral breast cancer remains very low, physicians may view CPM as medically unnecessary and feel uneasy about performing CPM [30]. This dilemma generates a tension between the adage of “do no harm” and the patient-centered model of healthcare currently advocated for by the Institute of Medicine [29, 31, 32]. Ideally, the choice for CPM is made through a shared decision-making process and involves several tradeoffs for patients and practitioners. Benefits of the shared decision-making process for patients include, but are not limited to, increased knowledge, more accurate risk perception, decisions consistent with patients’ values, and reduced, internal decisional conflict. However, differing opinions among relevant stakeholders about aspects of CPM may lead to sub-optimal shared decision-making. In particular, the importance placed by each stakeholder on the pros and cons of CPM is critical yet remains unknown. Importantly, the viewpoint of healthy volunteers on CPM is largely missing from the scientific literature and has only been explored qualitatively [33]. The opinion of healthy volunteers on CPM is relevant as it has been documented that preferences, attitudes or behaviors towards a condition may change after diagnosis and the viewpoint from unaffected individuals provides an important perspective about healthcare resource utilization.
Existing studies do not rank or discriminate the relative importance of factors weighed in CPM decision-making. Descriptive surveys and qualitative interviews have offered insight into the important issues but have not quantitatively measured the relative weight of each issue, nor the potential tradeoffs that patients and providers consider. In market research, conjoint analysis is a commonly used and rigorous quantitative tool, to rank and measure the relative value of preferences for attributes of a product [34-38]. To elicit preferences, conjoint analysis forces the user to make tradeoffs between competing factors, thereby simulating actual, real-world choices. The purpose of the current study is to have key stakeholders for CPM participate in the same online conjoint exercise in order to quantify their preferences in decision-making for CPM. The hypothesis is that each stakeholder group will have varying preferences for aspects of CPM. Enhanced knowledge of preferences could facilitate shared decision-making between patients and doctors through better alignment of values, leading to improved health outcomes. Moreover, the viewpoint of healthy volunteers is novel, providing an alternative perspective on this issue from future potential stakeholders.
METHODS:
Conjoint Study Design:
An evidence-based literature review was performed to understand important factors that contribute to CPM utilization [26, 39] and served to define the attributes for an alternative-specific design conjoint exercise. Final attributes of interest were: type of surgery (single [SM] versus double mastectomy [DM]), postoperative surgical complication rates following either SM or DM, 10-year future contralateral cancer risk following either SM or DM, frequency of follow-up surveillance imaging, and chest symmetry (Table 1), which demonstrates Attributes and Levels Used in Conjoint Analysis). Clinical factors associated with CPM such as preoperative MRI, prior biopsy, and attempts at breast conservation were not included in the exercise as they are not downstream effects of choosing a SM or DM. The choices for a conjoint exercise can be administered in either an unlabeled (i.e., blinded) or labeled manner; however, the current study required a labeled approach because certain combinations of attributes and levels are not realistic, for example, the need for surveillance imaging after DM.
Table 1:
Attributes and Levels Used in Conjoint Analysis
| Attribute | Levels |
|---|---|
| Type of Surgery | Single Mastectomy Double Mastectomy |
| Postoperative Surgical Complication Rates Following Single Mastectomy | 5% 10% 20% |
| Postoperative Surgical Complication Rates Following Double Mastectomy | 15% 30% 50% |
| 10-year Future Contralateral Cancer Risk Following Single Mastectomy | 3% 10% 15% |
| 10-year Future Cancer Risk Following Double Mastectomy | 0.3% 1% 3% |
| Frequency of Follow-Up Surveillance Imaging | 6 months 12 months None |
| Chest Symmetry Following Single Mastectomy | Somewhat Symmetric Asymmetric |
| Chest Symmetry Following Double Mastectomy | Highly Symmetric Somewhat Symmetric |
A choice-based-conjoint technique that included these attributes was administered to all stakeholder groups. This involved constructing profiles for either a SM or DM with a random combination of varying levels for each attribute (Table 1). Ten different scenarios were generated by the software for each participant to answer, containing the attributes with varying levels. For example, participants chose the scenario they preferred the most (SM or DM) after comparing the attributes and levels between them (Figure 1).
Figure 1:
Two Example Scenarios of the Conjoint Analysis Simulation Exercise
Participants/Stakeholder Groups:
Four main stakeholder groups were created: healthy volunteers, women with breast cancer (WwCA), surgical oncologists, and plastic surgeons. Inclusion criteria for WwCA were women 18-75 years old with a history of unilateral breast cancer who underwent either SM or DM in the past 6-24 months. Women who underwent lumpectomy only were excluded. Inclusion criteria for healthy volunteers were women with no personal history of cancer or genetic predisposition for breast cancer. Upon approval, both WwCA and healthy volunteers were recruited using the Dr. Susan Love Research Foundation’s Love Army of Women® registry (AoW). A second group of healthy volunteers, not associated with breast disease, was recruited from Amazon Mechanical Turk (mTurk), an online crowdsourcing marketplace. All respondents were asked questions about age, race/ethnicity, and cancer genetic predisposition. WwCA were additionally asked about their cancer stage, tumor histology, prior imaging with breast MRI, prior attempts at breast conserving surgery, receipt of post-mastectomy radiotherapy, and type of mastectomy and reconstruction.
Plastic surgeons and surgical oncologists were contacted by email using the American Society of Plastic Surgeons and Society of Surgical Oncology membership directory. An email provided study details with a link to the Sawtooth website. All surgeons were asked questions about age, race/ethnicity, fellowship training, practice setting, practice length, and breast case volume per year.
Study participants were directed to the web-based online conjoint exercise experiment, hosted on the Sawtooth Software (Lighthouse studio 9.2.0.) website and server [40]. Recruitment schedules were as follows: AoW was from June-September 2018, mTurk was from April-May 2018, Society of Surgical Oncology from August-September 2018, and American Society for Plastic Surgeons from May-July 2018. mTurk workers were compensated for completing each conjoint exercise.
Statistical Analysis:
Sawtooth Software was used to perform the conjoint analysis. Patients’ responses were converted into a dependent variable with each scenario as the independent variable. The more an attribute was preferred, the higher its importance score, relative to the other attributes. All importance scores add up to 100%. Bayesian hierarchical modeling was used to assign relative values (partworth utilities) for each level of an attribute with all partworth utilities adding up to 0[41]. Partworth utilities provide a quantitative measure of preference for levels of an attribute, where positive values demonstrate an inclination while negative values demonstrate an aversion.
Because a labeled approach was utilized in the current study, the type of surgery (SM vs. DM) was also considered an attribute associated with its own utility score. This is known as the alternative-specific constant and captures the effect of SM or DM not explicitly measured by the other attributes. Said differently, the type of surgery can be interpreted as the influence of the label itself, akin to a “brand”. The data was stored and collected from Sawtooth Software and analyzed at this institution using a combination of their proprietary software and SPSS 23.0 (IBM SPSS Statistics for Windows, Version 23.0 Armonk, NY). All data was stored on a secure network as per this institution’s guidelines and accessible only to the personnel listed in this study.
RESULTS:
Stakeholder Demographics:
In total the study had 1244 volunteer participants. There were 692 healthy volunteers, 224 WwCA, 133 surgical oncologists and 195 plastic surgeons who completed the online conjoint exercise. Healthy volunteers who participated via mTurk were significantly younger than AoW participants which included both healthy volunteers and WwCA (Table 2). The majority of participants were white. Amongst WwCA, 10% had a genetic predisposition, 14.3% had prior attempts at breast conservation, 63.8% underwent DM (36.2% SM), and 68.3% underwent reconstruction with the majority receiving prosthetic reconstruction (65.4%). Amongst surgeons (Table 3), there was a greater proportion of female oncologists whereas a greater proportion of plastic surgeons were male. The majority of physician participants were fellowship trained and encompassed a variety of practice settings. Both groups had a median of practice length of more than 10 years and a high volume of breast cases per year.
Table 2:
Demographics of Healthy Volunteers and Women with a History of Cancer
| Characteristics | mTurk Healthy Volunteers (n = 198) |
AoW Healthy Volunteers (n = 494) |
WwCA (n = 224) |
p value* |
|---|---|---|---|---|
| Age, median years (range) | 36 (18–71) | 56 (25–81) | 51 (27–76) | < 0.001 |
| Race or Ethnicity, n (%) | < 0.001 | |||
| White | 152 (76.8) | 465 (94.1) | 200 (89.3) | |
| Black | 18 (9.1) | 5 (1.0) | 4 (1.8) | |
| Hispanic/Latino | 11 (5.6) | 11 (2.2) | 13 (5.8) | |
| Asian | 14 (7.1) | 6 (1.2) | 3 (1.3) | |
| Other/Mixed | 3 (1.5) | 7 (1.4) | 4 (1.8) | |
| Cancer Stage at Diagnosis, n (%) | -- | |||
| I | -- | -- | 34 (15.2) | |
| II | -- | -- | 71 (31.7) | |
| III | -- | -- | 76 (33.9) | |
| IV | -- | -- | 37 (16.5) | |
| Unknown | -- | -- | 6 (2.7) | |
| Genetic predisposition, n (%) | -- | |||
| Yes | 0 | 0 | 24 (10.7) | |
| No | 128 (64.6) | 337 (68.2) | 183 (81.7) | |
| Unknown | 68 (34.3) | 157 (31.8) | 17 (7.6) | |
| Previous Biopsy: False Negative, n (%) | -- | |||
| Yes | -- | -- | 53 (23.7) | |
| No | -- | -- | 169 (75.4) | |
| Unknown | -- | -- | 2 (0.9) | |
| Previous MRI, n (%) | -- | |||
| Yes | -- | -- | 173 (77.2) | |
| No | -- | -- | 49 (21.9) | |
| Unknown | -- | -- | 2 (0.9) | |
| Previous BCS, n (%) | -- | |||
| Yes | -- | -- | 32 (14.3) | |
| No | -- | -- | 192 (85.7) | |
| Type of Mastectomy, n (%) | -- | |||
| Single | -- | -- | 81 (36.2) | |
| Double | -- | -- | 143 (63.8) | |
| PMRT, n (%) | -- | |||
| Yes | -- | -- | 71 (31.7) | |
| No | -- | -- | 153 (68.3) | |
| Reconstruction, n (%) | -- | |||
| Yes | -- | -- | 153 (68.3) | |
| No | -- | -- | 71 (31.7) | |
| Reconstruction Method, n (%) ** | -- | |||
| Implant-based | -- | -- | 100 (65.4) | |
| Tissue-based | -- | -- | 40 (26.1) | |
| Other | -- | -- | 13 (8.5) |
Abbreviations: mTurk Amazon Turk; AoW Army of Women; WwCA Women with Cancer; BCS Breast Conserving Therapy; PMRT Post Mastectomy Radiation Therapy
3-way p value: categorical variables calculated with Chi-Square or Fisher's Exact Test, continuous variables calculated with ANOVA
Percentages based on total reconstructions (total n = 153)
Table 3:
Demographics of Surgical Oncologists and Plastic Surgeons
| Characteristics | Surgical Oncologists (n = 133) |
Plastic surgeons (n = 195) |
p value* |
|---|---|---|---|
| Age, median years (range) | 47 (29–77) | 51 (34–82) | 0.135 |
| Gender, n (%) | < 0.001 | ||
| Male | 49 (36.8) | 156 (80) | |
| Female | 84 (63.2) | 39 (20) | |
| Race or Ethnicity, n (%) | 0.1899 | ||
| White | 105 (78.9) | 160 (82.1) | |
| Black | 6 (4.5) | 2 (1.0) | |
| Hispanic/Latino | 4 (3.0) | 2 (1.0) | |
| Asian | 11 (8.3) | 21 (10.8) | |
| Other/mixed | 7 (5.3) | 10 (5.1) | |
| Fellowship training, n (%) | - | ||
| Yes (breast fellowship[s]) | 108 (81.2) | 129 (66.2) | |
| No | 24 (18.0) | 63 (32.3) | |
| Other (non-breast fellowship[s]) | 1 (0.8) | 3 (1.5) | |
| Practice setting, n (%) | < 0.001 | ||
| Academic practice | 64 (48.1) | 25 (12.8) | |
| Employed physician | 39 (29.3) | 29 (14.9) | |
| Group practice | 16 (12.0) | 58 (29.7) | |
| Military/Veteran’s Affairs | 4 (3.0) | 2 (1.0) | |
| Solo practice | 10 (7.5) | 81 (41.5) | |
| Practice length, median years (range) | 12 (1-50) | 17 (2–46) | 0.024 |
| Breast case volume per year, n (%) | |||
| [≤20] | 13 (9.8) | 74 (37.9) | |
| [21-50] | 12 (17.3) | 69 (35.4) | |
| [>50] | 97 (72.9) | 52 (26.7) |
2-way p value: categorical variables calculated with Chi-Square or Fisher's Exact Test, continuous variables calculated with Student t-test
Importance Scores:
For WwCA the most important attribute in the decision-making process was the type of surgery, SM versus DM (importance score 33.6%) independent of other variables, followed by the increased risk of surgical complications after a DM (importance score 21.7%; Table 4). In contrast, among healthy volunteers, plastic surgeons, and surgical oncologists, risk of surgical complications following DM was considered the most important factor (importance score range 28.0%-29.7%). The top three most important factors influencing decision-making for all stakeholder group were surgical complications following a DM, type of surgery (SM vs. DM) and 10-year future contralateral cancer risk following a SM. The 10-year future cancer risk following a DM was rated highest by WwCA (importance score 9.6%) compared to other stakeholder groups (importance score 5.7-7.4%). Surgical oncologists and plastic surgeons along with mTurk healthy volunteers prioritized symmetry (importance score: 6.4%-8.8%) relatively more than AoW healthy volunteers and WwCA who placed greater importance on 10-year future cancer risk following a DM (importance score: 7.4-9.6%).
Table 4:
Importance Scores for each Attribute by Stakeholder
| Attributes | mTurk Healthy Volunteers (n = 198) |
AoW Healthy Volunteers (n = 494) |
WwCA (n = 224) |
Surgical Oncologists (n = 133) |
Plastic surgeons (n = 195) |
|---|---|---|---|---|---|
| Type of Surgery | 20.00% | 25.80% | 33.60% | 22.00% | 19.20% |
| Surgical Complications | |||||
| SM | 15.70% | 13.90% | 10.50% | 15.60% | 15.00% |
| DM | 29.00% | 28.00% | 21.70% | 29.70% | 28.90% |
| 10-Year Future Cancer Risk | |||||
| SM | 20.70% | 19.10% | 18.90% | 18.30% | 19.20% |
| DM | 5.90% | 7.40% | 9.60% | 5.70% | 6.30% |
| Symmetry | 6.40% | 4.10% | 4.00% | 6.80% | 8.80% |
| Follow-up Schedule: SM | 2.40% | 1.87% | 1.65% | 1.92% | 2.70% |
Abbreviations: SM Single Mastectomy; DM Double Mastectomy; mTURK Amazon Turk; AoW Army of Women; WwCA Women with Cancer
Interpretation: Relative importance explains the impact of each attribute in the total utility of the surgical decision-making process. Importance percentages are calculated from utility values and total up to 100% for each stakeholder. Higher score indicates greater importance to respondents.
Partworth Utility Scores:
Following a DM, surgical oncologists were the most risk averse group to surgical complications (partworth: −133) compared to all other stakeholders (Figure 2). In contrast, WwCA were the least risk averse to complications following DM (partworth: −93.5). Similar preferences for avoidance of complications were seen following SM (Figure 3) with the greatest risk aversion demonstrated by surgical oncologists and the least by WwCA. Risk aversion to surgical complications following either SM or DM for healthy volunteers and plastic surgeons was more similar to surgical oncologists than WwCA (see Figures 2 and 3).
Figure 2: Surgical Complications Following a Double Mastectomy, Partworth Utilities by Stakeholder.
Abbreviations: mTurk Amazon Mechanical Turk, AoW Army of Women, WwCA with Cancer
Figure 3: Surgical Complications Following a Single Mastectomy, Partworth Utilities by Stakeholder.
Abbreviations: mTurk Amazon Mechanical Turk, AoW Army of Women, WwCA with Cancer
The threshold for 10-year future breast cancer episodes following a SM or DM differed for surgical oncologists versus all other stakeholders. Compared to surgical oncologists, all other stakeholders had a lower threshold of tolerance for future contralateral cancer episodes following a SM (Figure 4). Whereas surgical oncologists were accepting of future contralateral cancer risk of 3% or 10% following a SM, all other groups were averse to any future contralateral cancer risk above 3%. Following DM all stakeholder groups were accepting of 10-year future cancer episodes at a rate of 0.3-1%; however, WwCA were most strongly against future cancer rates of 3% (Figure 5). In general, surgical oncologists felt least strongly about future cancer episode rates (partworth utilities range: 4.1 to −4.1) following a DM than all other participants.
Figure 4: 10-Year Future Contralateral Cancer Risk Following a Single Mastectomy, Partworth Utilities by Stakeholder.
Abbreviations: mTurk Amazon Mechanical Turk, AoW Army of Women, WwCA with Cancer
Figure 5: 10-Year Future Cancer Risk Following a Double Mastectomy, Partworth Utilities by Stakeholder.
Abbreviations: mTurk Amazon Mechanical Turk, AoW Army of Women, WwCA with Cancer
Stakeholder opinion for type of surgery, SM versus DM following a breast cancer diagnosis varied greatly (Figure 6). WwCA had a strong preference for DM (partworth: 70.5) followed by plastic surgeons who also had a preference as well (partworth: 6.2). In contrast surgical oncologists felt strongly against DM (partworth: −58.3), favoring SM. Healthy volunteers did not have a strong preference for either SM or DM
Figure 6: Type of Surgery, Partworth Utilities by Stakeholder.
Abbreviations: mTurk Amazon Mechanical Turk, AoW Army of Women, WwCA with Cancer
Following a DM, plastic surgeons and surgical oncologists felt most strongly about a highly symmetric outcome (see Figure, Supplemental Digital Content 1 demonstrating Partworth Utilities by Stakeholder for Symmetry Following a Double Mastectomy). This was followed next by healthy volunteers and least by WwCA. Following SM, plastic surgeons are least accepting of asymmetry followed by patients (see Figure, Supplemental Digital Content 2 demonstrating Partworth Utilities by Stakeholder for Symmetry Following a Single Mastectomy). Surgical oncologists and healthy volunteers felt less strongly about asymmetry following SM.
Discussion:
The modern era of shared decision-making advocated by the Institute of Medicine is highlighted in cases of preference sensitive care. Examples in breast surgery include lumpectomy versus mastectomy or prosthetic versus autologous tissue reconstruction and have spawned a number of tools to improve decision quality such as decision aids [42, 43]. However, perhaps the most interesting example of preference sensitive care, is the choice for a CPM, since the organ is not affected by disease and the increased rate of complications following bilateral surgery are well documented[44]. So how do physicians reconcile patient preference with the Hippocratic oath to “First do not harm” (Latin: Primum non nocere), especially when shared decision-making has been shown to positively influence cognitive, behavioral, and health-related outcomes [45, 46]? The current study aimed to quantify the preferences of different stakeholders for CPM. Importantly, the same instrument was administered to all groups allowing for more meaningful comparisons in the analysis. Overall, the findings, parsed below, demonstrate divergent opinions regarding aspects of CPM for each stakeholder group suggesting significant room for improving education and the quality of the shared decision-making.
The top three most important factors influencing decision-making for all stakeholder groups were surgical complication rates following a DM, type of surgery (SM versus DM) irrespective of other factors, and 10-year future contralateral cancer risk following a SM. For women with a prior history of breast cancer, when all else was considered, type of surgery-DM, was the most important factor. The preference for DM may simply reflect a confirmation bias by women in the current study cohort who underwent DM in 63.8% of cases; however, the wide scale national increase in CPM suggests additional mechanisms need consideration [47, 48]. For example, the strength of implicit biases for CPM likely have increased over time due to sociocultural factors including social media connectivity, the “Angelina Jolie” effect albeit bilateral prophylactic mastectomy, or personal experiences, including friends and family with breast disease, especially in the backdrop of increasing US breast cancer incidence over the past 40 years. Moreover, while most patients faced with a cancer diagnosis are usually concerned with either local recurrence or metastatic disease, the paired nature of breasts makes the possibility of a second breast cancer episode a potential issue requiring significant deliberation. While the majority of breast cancer patients are at low risk for a subsequent metachronous contralateral breast event, choosing to undergo CPM may mitigate cancer worry and marginally minimize risk, perhaps a heuristic response to an extremely complex problem.
In contrast to former breast cancer patients, all other stakeholder groups ranked the high rate of surgical complications from a DM as the most important aspect to consider in surgical decision-making. Stakeholders, yet unaffected by a breast cancer diagnosis, weigh surgical complications more so than a predilection for a specific surgery type. It would not have been surprising if healthy volunteers from mTurk or AoW demonstrated a preference for DM based upon shared breast cancer experiences with friends or family; however, the adjustment in internal standards, or response shift, appears to occur only after disease diagnosis. Not unexpectedly, surgeons were the most risk averse group to complications following a second mastectomy likely reflecting their first-hand knowledge of the harm of surgical complications including the potential need for reoperation, poor wound healing, delays in adjuvant therapy, and reconstructive failure. Moreover, a consensus statement from the American Society of Breast Surgeons recommends that women with unilateral breast cancer who are at average risk should be discouraged from undergoing CPM, because most will not obtain a survival benefit and CPM doubles the risk of surgical complications [49].
Although there is wide variation in contralateral breast cancer risk, impacted by age, tumor receptor profile, family history, genetic predisposition, and adjuvant therapies, data from the Surveillance, Epidemiology, and End Results Program show a 0.2-0.3% annual contralateral breast cancer risk for women over age 40 with ER+ breast cancer, highlighting that the majority of sporadic breast cancer patients remain at very low risk for a contralateral event following their diagnosis. However, the current data suggest each stakeholder group interprets cancer risk differently or may have different risk perceptions. Surgical oncologists were more accepting of future breast cancer episodes following either a SM or DM than other stakeholder groups. Evaluation of partworth utilities demonstrates that a 10%, 10- year risk of developing a contralateral breast cancer following SM was acceptable to surgical oncologists but was perceived to be strongly negative by other stakeholder groups. Following DM, surgical oncologists also did not feel strongly about cancer risks ranging from 0.3-3% (partworth utilities range: 4.1 to −4.1). In contrast, even a 3% 10-year cancer risk after a DM was highly negative to all other stakeholder groups (partworth utilities range: −15.5 to −31.9) with the greatest negative value demonstrated by women previously diagnosed with breast cancer, followed by healthy volunteers. In summation, the threshold of acceptability for future cancer episodes as well as risk perception varies depending upon the stakeholder with a profound influence upon women who have experienced breast cancer. Therefore, preoperative patient education should highlight that an individual’s risk perception may have changed in the setting of a new breast cancer diagnosis [50] especially when considering a DM. In contrast, surgical oncologists likely balance potential risk-benefit ratios through tradeoffs between surgical morbidity versus observation of the uninvolved breast, modulated by a nuanced understanding of cancer risk and treatment effects.
Often the choice for a CPM is ascribed to symmetry particularly in the setting of prosthetic reconstruction. Overall symmetry was low on importance scores for all stakeholder groups and does not appear to be an important driver in the decision-making process for a SM or DM. Plastic surgeons were the stakeholder group that felt most importantly about symmetry (importance score 8.8) and believed most strongly in symmetry following a SM or DM (partworth utility 37.9 & 25.9, respectively) Interestingly, WwCA felt strongly about asymmetry following a SM (partworth utility range: 20 to −20), but much less so after a DM (partworth utility range: 6.4 to −6.4). The significance of this finding is unclear but may simply reflect knowledge of the high degree of symmetry achieved for women who undergo reconstruction for DM compared to SM, particularly with implants. This finding could vary based on method of reconstruction, although this was not assessed in the current study.
The overall study findings demonstrate that each stakeholder group has different preferences or concerns for aspects of CPM. Suggestions about how to align provider and patient differences may come from lessons learned in the business world. The field of behavioral economics and concepts like “choice architecture” has demonstrated that framing options in a particular way, even when performed without bias, can influence decisions made [48, 51, 52]. For example, the order in which treatment options are communicated as well as whether consequences are described as gains (e.g., “survival rates are higher”) or losses (e.g., “death rates are lower”) can be impactful. Following introduction of a 1-hour training session by physicians in a systemic counseling approach to patients [48], a study of men with early prostate cancer showed an increase in active surveillance by 9.1% compared with prostatectomy. Behavioral decision-making techniques for breast cancer patients could include appropriate framing which starts by encouraging patients to not ignore the harms of prophylactic mastectomy. Importantly, the goal of physician training in such techniques should be to change patient behavior, not patient beliefs. The current study helps us better understand patient interests and serves as a useful starting point for framing principles that will optimize effective physician communication.
One aspect of CPM which deserves attention is insurance payment for the procedure, especially in the US where escalating healthcare expenditures continue to outpace growth of gross domestic product. Within the US, health insurance design does not appear to be an impediment to CPM since the additional costs are absorbed or distributed amongst all members of a shared risk pool through greater premiums. As such, the individual bears no financial burden when making healthcare choices, an economics concept known as “moral hazard”. In sharp contrast, single payer healthcare systems around the world have fixed budgets with more stringent eligibility requirements and thus much lower CPM rates [53]. The inclusion of healthy volunteers as a stakeholder group in the current study may reveal attitudes of the general public about CPM for health policy makers to consider in the balance between cost containment and value (i.e. health related quality of life) in US healthcare reform. Despite some demographic differences between mTurk and AoW, preferences for various aspects of CPM were quite similar between these two groups, overall. Healthy volunteers’ indifference to SM or DM (partworth utilities range 2 to −2) combined with the aversion to complications following a DM suggests the need for further exploration about the general attitude of the public towards insurance coverage of this procedure.
This study emphasizes that all perspectives matter in the shared-decision making process around CPM. However, what supersedes these perspectives is the context of the healthcare structure. Outside of the US, economics and fixed resources underlie decision-making in countries with national healthcare structures (i.e., single-payer). Given that CPM is a contributor to increased healthcare expenditure, these countries demonstrate lower rates of CPM compared to the US[53, 54]. In contrast, the US values preference-sensitive decision making, sometimes in the absence of economic decisions. Therefore, CPM decision making in the US primarily takes place among conversations between doctors and patients.
Although novel in design and methodology, this study has some limitations. The generalizability of AoW participants’ preferences for CPM may be limited by these volunteers’ personal relationship with breast cancer. To account for this possible source of bias, this study included an additional group of healthy volunteers from mTurk with no known relationship to breast cancer. There may also be a potential bias inherent to any survey study in who chooses to respond, including surgeons and participants. The current study focused on women with mastectomy and does not consider lumpectomies, the predominant treatment for breast cancer. The downstream effects of having DM such as higher rates of complications leading to financial toxicity were not studied. This study has limited racial/ethnic diversity of healthy women and cancer survivors and did not collect data on aspects of socioeconomic status – characteristics that may impact preferences. Lastly, volunteers and WwCA were not provided specifics on how surgical complications could delay adjuvant cancer therapy. This additional knowledge may affect how these stakeholder groups weigh the importance of surgical complications in relation to other attributes.
Conclusion:
Preferences for CPM vary by stakeholder group suggesting significant room for improved shared decision-making. Women with a prior history of breast cancer may perceive SM as “not doing the most” to mitigate future cancer episodes within the context of altered risk perception. Interestingly, healthy volunteer preferences align more closely with breast surgeons than patients, which may have potential health policy implications. Although physicians have the greatest understanding of the morbidity to mortality benefit associated with CPM, they may benefit from improved communication skills, such as framing, which address patient inherent biases. Preoperative patient education should also highlight that an individual’s risk perception may change in the setting of a new breast cancer diagnosis especially when considering a DM. Most patients want to participate with their clinicians in making choices; therefore, it is incumbent on physicians to act more like coaches and ask open-ended questions in order to achieve the full meaning of patient-centered care [48].
Supplementary Material
Supplemental Digital Content 2: Symmetry Following a Single Mastectomy, Partworth Utilities by Stakeholder (Supplemental Digital Content 5.pptx)
Supplemental Digital Content 1: Symmetry Following a Double Mastectomy, Partworth Utilities by Stakeholder (Supplemental Digital Content 4.pptx)
Acknowledgement:
This research was funded in part though the NIH/NCI Cancer Center Support Grant P30 CA008748
Footnotes
Financial Disclosure: None of the authors has a financial interest in any of the products, devices, or drugs mentioned in this manuscript.
References:
- 1.Yao K, et al. , Contralateral prophylactic mastectomy and survival: report from the National Cancer Data Base, 1998-2002. Breast Cancer Res Treat, 2013. 142(3): p. 465–76. [DOI] [PubMed] [Google Scholar]
- 2.Pesce CE, et al. , Changing surgical trends in young patients with early stage breast cancer, 2003 to 2010: a report from the National Cancer Data Base. J Am Coll Surg, 2014. 219(1): p. 19–28. [DOI] [PubMed] [Google Scholar]
- 3.Portschy PR, Kuntz KM, and Tuttle TM, Survival outcomes after contralateral prophylactic mastectomy: a decision analysis. J Natl Cancer Inst, 2014. 106(8). [DOI] [PubMed] [Google Scholar]
- 4.Tuttle TM, et al. , Increasing use of contralateral prophylactic mastectomy for breast cancer patients: a trend toward more aggressive surgical treatment. J Clin Oncol, 2007. 25(33): p. 5203–9. [DOI] [PubMed] [Google Scholar]
- 5.Dragun AE, et al. , Increasing use of elective mastectomy and contralateral prophylactic surgery among breast conservation candidates: a 14-year report from a comprehensive cancer center. Am J Clin Oncol, 2013. 36(4): p. 375–80. [DOI] [PubMed] [Google Scholar]
- 6.Tuttle TM, et al. , Increasing rates of contralateral prophylactic mastectomy among patients with ductal carcinoma in situ. J Clin Oncol, 2009. 27(9): p. 1362–7. [DOI] [PubMed] [Google Scholar]
- 7.Yao K, et al. , Trends in contralateral prophylactic mastectomy for unilateral cancer: a report from the National Cancer Data Base, 1998-2007. Ann Surg Oncol, 2010. 17(10): p. 2554–62. [DOI] [PubMed] [Google Scholar]
- 8.Kummerow KL, et al. , Nationwide trends in mastectomy for early-stage breast cancer. JAMA Surg, 2015. 150(1): p. 9–16. [DOI] [PubMed] [Google Scholar]
- 9.Rutter CE, et al. , Growing Use of Mastectomy for Ductal Carcinoma-In Situ of the Breast Among Young Women in the United States. Ann Surg Oncol, 2015. 22(7): p. 2378–86. [DOI] [PubMed] [Google Scholar]
- 10.Albornoz CR, et al. , Bilateral Mastectomy versus Breast-Conserving Surgery for Early-Stage Breast Cancer: The Role of Breast Reconstruction. Plast Reconstr Surg, 2015. 135(6): p. 1518–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fayanju OM, et al. , Contralateral prophylactic mastectomy after unilateral breast cancer: a systematic review and meta-analysis. Ann Surg, 2014. 260(6): p. 1000–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kurian AW, et al. , Use of and mortality after bilateral mastectomy compared with other surgical treatments for breast cancer in California, 1998-2011. JAMA, 2014. 312(9): p. 902–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wong SM, et al. , Growing Use of Contralateral Prophylactic Mastectomy Despite no Improvement in Long-term Survival for Invasive Breast Cancer. Ann Surg, 2017. 265(3): p. 581–589. [DOI] [PubMed] [Google Scholar]
- 14.Miller ME, et al. , Operative risks associated with contralateral prophylactic mastectomy: a single institution experience. Ann Surg Oncol, 2013. 20(13): p. 4113–20. [DOI] [PubMed] [Google Scholar]
- 15.Osman F, et al. , Increased postoperative complications in bilateral mastectomy patients compared to unilateral mastectomy: an analysis of the NSQIP database. Ann Surg Oncol, 2013. 20(10): p. 3212–7. [DOI] [PubMed] [Google Scholar]
- 16.Rosenberg SM, et al. , Perceptions, knowledge, and satisfaction with contralateral prophylactic mastectomy among young women with breast cancer: a cross-sectional survey. Ann Intern Med, 2013. 159(6): p. 373–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Momoh AO, et al. , Tradeoffs Associated With Contralateral Prophylactic Mastectomy in Women Choosing Breast Reconstruction: Results of a Prospective Multicenter Cohort. Ann Surg, 2017. 266(1): p. 158–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Katz SJ and Morrow M, Contralateral prophylactic mastectomy for breast cancer: addressing peace of mind. JAMA, 2013. 310(8): p. 793–4. [DOI] [PubMed] [Google Scholar]
- 19.Rosenberg SM, et al. , Local Therapy Decision-Making and Contralateral Prophylactic Mastectomy in Young Women with Early-Stage Breast Cancer. Ann Surg Oncol, 2015. 22(12): p. 3809–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Parker PA, et al. , Prospective Study of Surgical Decision-making Processes for Contralateral Prophylactic Mastectomy in Women With Breast Cancer. Ann Surg, 2016. 263(1): p. 178–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hawley ST, et al. , Social and Clinical Determinants of Contralateral Prophylactic Mastectomy. JAMA Surg, 2014. 149(6): p. 582–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rendle KA, et al. , Redefining Risk and Benefit: Understanding the Decision to Undergo Contralateral Prophylactic Mastectomy. Qual Health Res, 2015. 25(9): p. 1251–9. [DOI] [PubMed] [Google Scholar]
- 23.Agarwal S, et al. , Defining the relationship between patient decisions to undergo breast reconstruction and contralateral prophylactic mastectomy. Plast Reconstr Surg, 2015. 135(3): p. 661–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ashfaq A, et al. , Impact of breast reconstruction on the decision to undergo contralateral prophylactic mastectomy. Ann Surg Oncol, 2014. 21(9): p. 2934–40. [DOI] [PubMed] [Google Scholar]
- 25.Beesley H, et al. , Risk, worry and cosmesis in decision-making for contralateral risk-reducing mastectomy: analysis of 60 consecutive cases in a specialist breast unit. Breast, 2013. 22(2): p. 179–184. [DOI] [PubMed] [Google Scholar]
- 26.King TA, et al. , Clinical management factors contribute to the decision for contralateral prophylactic mastectomy. J Clin Oncol, 2011. 29(16): p. 2158–64. [DOI] [PubMed] [Google Scholar]
- 27.Soran A, et al. , Decision making and factors influencing long-term satisfaction with prophylactic mastectomy in women with breast cancer. Am J Clin Oncol, 2015. 38(2): p. 179–83. [DOI] [PubMed] [Google Scholar]
- 28.Covelli AM, et al. , 'Taking control of cancer': understanding women's choice for mastectomy. Ann Surg Oncol, 2015. 22(2): p. 383–91. [DOI] [PubMed] [Google Scholar]
- 29.Balch CM and Jacobs LK, Mastectomies on the Rise for Breast Cancer: “The Tide Is Changing”. Annals of Surgical Oncology, 2009. 16(10): p. 2669–2672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Bellavance E, et al. , Surgeons' Perspectives of Contralateral Prophylactic Mastectomy. Ann Surg Oncol, 2016. 23(9): p. 2779–87. [DOI] [PubMed] [Google Scholar]
- 31.Musiello T, Bornhammar E, and Saunders C, Breast surgeons' perceptions and attitudes towards contralateral prophylactic mastectomy. ANZ Journal of Surgery, 2012. 83(7–8): p. 527–532. [DOI] [PubMed] [Google Scholar]
- 32.Covelli AM, et al. , Increasing Mastectomy Rates—The Effect of Environmental Factors on the Choice for Mastectomy: A Comparative Analysis Between Canada and the United States. Annals of Surgical Oncology, 2014. 21(10): p. 3173–3184. [DOI] [PubMed] [Google Scholar]
- 33.Hooper RC, et al. , Breast Cancer Knowledge and Decisions Made for Contralateral Prophylactic Mastectomy: A Survey of Surgeons and Women in the General Population. Plast Reconstr Surg, 2019. 143(5): p. 936e–945e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Waltzman JT, Scholz T, and Evans GR, What patients look for when choosing a plastic surgeon: an assessment of patient preference by conjoint analysis. Ann Plast Surg, 2011. 66(6): p. 643–7. [DOI] [PubMed] [Google Scholar]
- 35.Marsidi N, van den Bergh MW, and Luijendijk RW, The best marketing strategy in aesthetic plastic surgery: evaluating patients' preferences by conjoint analysis. Plast Reconstr Surg, 2014. 133(1): p. 52–7. [DOI] [PubMed] [Google Scholar]
- 36.Damen TH, et al. , Patients' preferences for breast reconstruction: a discrete choice experiment. J Plast Reconstr Aesthet Surg, 2011. 64(1): p. 75–83. [DOI] [PubMed] [Google Scholar]
- 37.Pinell-White XA, Kolegraff K, and Carlson GW, Predictors of contralateral prophylactic mastectomy and the impact on breast reconstruction. Ann Plast Surg, 2014. 72(6): p. S153–7. [DOI] [PubMed] [Google Scholar]
- 38.Bridges JF, et al. , Conjoint analysis applications in health--a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. Value Health, 2011. 14(4): p. 403–13. [DOI] [PubMed] [Google Scholar]
- 39.Marmor RA, et al. , Increase in contralateral prophylactic mastectomy conversation online unrelated to decision-making. J Surg Res, 2017. 218: p. 253–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Sawtooth Software. [cited 2019; Available from: https://www.sawtoothsoftware.com/.
- 41.Howell J CBC/HB for Beginners. Research Paper Series; 2009. [cited 2021; Available from: https://sawtoothsoftware.com/resources/technical-papers/cbc-hb-for-beginners. [Google Scholar]
- 42.Berlin NL, et al. , Feasibility and Efficacy of Decision Aids to Improve Decision Making for Postmastectomy Breast Reconstruction: A Systematic Review and Meta-analysis. Med Decis Making, 2019. 39(1): p. 5–20. [DOI] [PubMed] [Google Scholar]
- 43.Nicholas Z, et al. , A systematic review of decision aids for patients making a decision about treatment for early breast cancer. Breast, 2016. 26: p. 31–45. [DOI] [PubMed] [Google Scholar]
- 44.Wilkins EG, et al. , Complications in Postmastectomy Breast Reconstruction: One-year Outcomes of the Mastectomy Reconstruction Outcomes Consortium (MROC) Study. Ann Surg, 2018. 267(1): p. 164–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Shay LA and Lafata JE, Where is the evidence? A systematic review of shared decision making and patient outcomes. Med Decis Making, 2015. 35(1): p. 114–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Hughes TM, et al. , Association of shared decision-making on patient-reported health outcomes and healthcare utilization. Am J Surg, 2018. 216(1): p. 7–12. [DOI] [PubMed] [Google Scholar]
- 47.Panchal H, et al. , National trends in contralateral prophylactic mastectomy in women with locally advanced breast cancer. J Surg Oncol, 2019. 119(1): p. 79–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Grimmer L, et al. , Variation in Contralateral Prophylactic Mastectomy Rates According to Racial Groups in Young Women with Breast Cancer, 1998 to 2011: A Report from the National Cancer Data Base. J Am Coll Surg, 2015. 221(1): p. 187–96. [DOI] [PubMed] [Google Scholar]
- 49.Boughey JC, et al. , Contralateral Prophylactic Mastectomy (CPM) Consensus Statement from the American Society of Breast Surgeons: Data on CPM Outcomes and Risks. Annals of surgical oncology, 2016. 23(10): p. 3100–3105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Ubel PA, et al. , Misimagining the unimaginable: the disability paradox and health care decision making. Health Psychol, 2005. 24(4S): p. S57–62. [DOI] [PubMed] [Google Scholar]
- 51.Offodile AC 2nd, and Clemens MW, Decisions and Incisions: The Role of Choice Architecture in Surgical Decision Making. Aesthet Surg J, 2018. 38(5): p. 575–577. [DOI] [PubMed] [Google Scholar]
- 52.Hughes TM, et al. , Recognizing Heuristics and Bias in Clinical Decision-making. Ann Surg, 2020. 271(5): p. 813–814. [DOI] [PubMed] [Google Scholar]
- 53.Güth U, et al. , Increasing rates of contralateral prophylactic mastectomy - a trend made in USA? Eur J Surg Oncol, 2012. 38(4): p. 296–301. [DOI] [PubMed] [Google Scholar]
- 54.Boughey JC, et al. , Contralateral Prophylactic Mastectomy Consensus Statement from the American Society of Breast Surgeons: Additional Considerations and a Framework for Shared Decision Making. Ann Surg Oncol, 2016. 23(10): p. 3106–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Digital Content 2: Symmetry Following a Single Mastectomy, Partworth Utilities by Stakeholder (Supplemental Digital Content 5.pptx)
Supplemental Digital Content 1: Symmetry Following a Double Mastectomy, Partworth Utilities by Stakeholder (Supplemental Digital Content 4.pptx)