Skip to main content

Patient Preferences and Treatment Adherence Among Women Diagnosed with Metastatic Breast Cancer

Original Research
October 2014 Vol 7, No 7 - Clinical
Download PDF
Abstract

BACKGROUND: Given the various profiles (eg, oral vs intravenous administration, risk of hot flashes vs fatigue) of treatment options (eg, endocrine therapy, chemotherapy) for metastatic breast cancer (mBC), how patients value these attributes of their medications has implications on making treatment decisions and on adherence.

OBJECTIVES: To understand how patients trade off medication side effects with improved effectiveness and/or quality of life, to provide estimates of nonadherence among women with mBC, and to quantify the association of medication nonadherence with health outcomes.

METHODS: The study was a cross-sectional, Internet-based survey of 181 women diagnosed with mBC who were recruited from cancer-specific online panels (response rate, 7%). Treatment information, demographics, nonadherent behaviors, and quality of life assessed by the Functional Assessment of Cancer Therapy-Breast (FACT-B) were collected in the survey, and each respondent completed a choice-based conjoint exercise to assess patient preferences. The patients’ preferences were analyzed using hierarchical Bayesian logistic regression models, and the association between the number of nonadherent behaviors and the health outcomes was analyzed using general linear models.

RESULTS: The mean age of the patient sample was 52.2 years (standard deviation, ±9.1), with 93.9% of participants being non-Hispanic white. Results from the conjoint model indicated that effectiveness (overall survival) was of primary importance to patients, followed by side effects—notably alopecia, fatigue, neutropenia, motor neuropathy, and nausea/vomiting—and finally, dosing regimen. In all, 34.8% of survey respondents either discontinued their treatment or were nonadherent to their treatment regimen. Among those who have ever used oral chemotherapy (N = 95; 52.5%) and those currently using oral chemotherapy (N = 44; 24.3%), the number of nonadherent behaviors was significantly associated with a decrease in functional well-being (b [unstandardized regression coefficient] = –2.01 for patients who had ever used a targeted therapy and b = –3.14 for current users of a targeted therapy), FACT-General total score (b = –4.30 and b = –7.37, respectively), FACT-B total score (b = –3.93 and b = –6.11, respectively), and FACT trial outcome index (b = –5.22 and b = –8.63, respectively; all P <.05).

CONCLUSIONS: Patients were willing to accept substantial additional risks from side effects for gains in overall survival. Approximately 33% of women with mBC reported engaging in nonadherent behaviors. Because forgetfulness and adverse events were among the most frequent reasons for nonadherence, these results suggest that less complex treatment regimens, as well as regimens with less toxic profiles, may be associated with improvements in adherence and, subsequently, could correspond to perceptible patient benefits.

Am Health Drug Benefits.
2014;7(7):386-396
www.AHDBonline.com

Received July 7, 2014
Accepted in final form September 16, 2014

Disclosures are at end of text

Breast cancer is the most common cancer diagnosed among women in the United States, with an estimated 232,000 new cases diagnosed annually.1 It is also the second most deadly cancer, accounting for nearly 40,000 deaths annually.1 The 5-year survival rates for early-stage breast cancer is between 84% for regional disease (ie, contained within the breast and lymph nodes) and 99% for localized disease (ie, contained within the breast); the survival rate drops to 24% in more advanced stages of the disease.2 Metastatic breast cancer (mBC) is defined as breast cancer that has spread to other parts of the body. Whereas less than 10% of women are initially diagnosed with mBC, approximately 33% of women who are treated for early-stage disease will progress to mBC.2,3 The majority of breast cancer metastases affect the lymph nodes, followed by bone, liver, and lung.4,5

At present, mBC is incurable, and treatment is focused on arresting the disease and extending patient survival, as well as promoting quality of life (QOL) and ensuring adequate symptom management.3,4 Treatment strategies may include endocrine therapy (eg, exemestane); targeted therapies, such as anti-HER2 agents (eg, trastuzumab); and chemotherapy (eg, capecitabine).6 As a result of the currently incurable nature of mBC and the side-effect profile associated with various forms of therapy, investigators have explored the potential role of patient preferences in decision-making regarding their treatment goals and desired outcomes.7-11 For example, aggressive treatment may maximize the duration of survival but may also be associated with significant and burdensome side effects that impair QOL.3,4

Using the responses of 102 patients with breast cancer, Beusterien and colleagues recently revealed distinct preferences across side-effect profiles and a willingness to undergo more difficult treatment regimens to reduce the risk of more severe symptoms.7 Moreover, another study of 121 patients with breast cancer demonstrated the importance of a relatively minor survival benefit (as little as 3 months) for the perceived value of chemotherapy among patients with breast cancer, despite an increased risk for treatment toxicities.8 Further studies have shown a willingness of patients to trade a minor increase in disease recurrence risk for more convenient treatment regimens,9 as well as examining patient preferences for follow-up care10 and the ability for preferences to predict the eventual use of chemotherapy.11

To our knowledge, no study to date has examined patient preferences using a conjoint method in women with mBC. In part, this may be because it is difficult to recruit this patient population for research survey purposes. Several studies have been conducted that have focused on patients diagnosed with early-stage breast cancer.12,13 McQuellon and colleagues surveyed women diagnosed with early-stage disease to assess their preferences for the hypothetical treatment of mBC.14 Although there was a wide range of preference profiles among the women who were surveyed, they were once again consistently willing to trade the risk of major side effects and toxicities for a modest survival benefit.14 However, because that study was initiated nearly 20 years ago, treatment advances since the time of that study were not included in that analysis.

The primary objective of the current study is to provide an examination of contemporary treatments and to provide data on the treatment preferences of women with mBC to understand how these patients trade off side effects with increases in effectiveness and/or QOL.

Patient preferences may have implications not only for treatment decision-making but also potentially for treatment adherence and follow-up care.15 If patient preferences and prescribed treatment regimens are misaligned, women diagnosed with breast cancer may become nonadherent, which could have implications for symptom management and for survival.15 To date, much of the research in this domain has focused on adherence to adjuvant therapy in patients with early-stage disease.16-18 A secondary objective of this study is to provide real-world evidence of nonadherence among women with mBC and to quantify the association with nonadherent behavior and health outcomes. We focused on adherence among patients receiving oral chemotherapy, because these agents are increasing in availability, and they represent a frequently self-administered treatment (as opposed to intravenous treatment, which is often not self-administered), and are thus more susceptible to nonadherent behaviors.


Table 1

Methods
Patient Sample and Procedures
Qualitative interviews. An initial qualitative study was conducted to inform the participants of the survey. A total of 10 telephone interviews were conducted using a structured discussion guide with women who were diagnosed with mBC. Patients were invited to participate through cancer-specific online panels (eg, the Find A Cure Panel), which recruit members from cancer-oriented nonprofit Internet communities for the purpose of participating in research projects.

To be eligible to participate in the study, women had to report that they met the criteria, including being diagnosed with mBC, aged ≥18 years, and be proficient in the English language. Patients who never received treatment with a taxane (eg, paclitaxel, docetaxel) or who did not have health insurance, were covered by Medicaid, or who did not know their form of health insurance were excluded. These latter criteria were included to ensure sufficient treatment experience to put our hypothetical treatments into context. For research questions that were unrelated to the current study, the study focused on patients with private insurance types. Interviews lasted approximately 30 minutes and focused on the patient’s experience with previous treatments, as well as the frequency, severity, and tolerability of side effects.

Participant survey. The present study was a cross- sectional, Internet-based survey of 181 women who were diagnosed with mBC. The inclusion and exclusion criteria of the women surveyed were identical to the qualitative participants. Potential participants were also invited from cancer-specific online panels, the same source used in the qualitative research. A total of 2500 panelists were e-mailed invitations to participate in the current study (response rate, 7%); those who clicked the invitation link were directed to the statement of informed consent, and, if consent was given, they were then directed to screening questions to determine eligibility; 28 respondents were ineligible: 21 because they had never received a taxane and 7 because of unknown reasons or for having Medicaid insurance coverage.

All participants who completed the survey were compensated with a $100 donation made in their name to a nonprofit charitable organization of their choice. For both the qualitative interview and the participant survey, all respondents provided informed consent and had their responses kept confidential. The protocol was approved by an Institutional Review Board (Essex IRB; Lebanon, NJ).

Survey Measures
Demographics and health characteristics. All participants provided information about their age, race/ethnicity, marital status, college education, annual household income, employment, insurance type, and out-of-pocket (OOP) costs. Health-related information was also provided, such as height and weight, which was then converted to a body mass index category of underweight (<18.5 kg/m2), normal weight (18.5 to <25 kg/m2), overweight (25 to <30 kg/m2), obese (≥30 kg/m2), or missing (for those participants who chose not to provide their weight). A family history of breast cancer, years diagnosed, specialty of diagnosing physician, stage at diagnosis, and months in metastatic stage were also reported.

Treatment history and adherence. The participants’ treatment modality experience was defined by surgery, intravenous chemotherapy, oral chemotherapy, radiation therapy, hormone therapy, clinical trial medication, and palliative care. The participants reported their experience in terms of previous treatment and current treatment. All women were asked whether they had been nonadherent to or had discontinued any of their previous treatments. For the participants who responded affirmatively for nonadherence, specific reasons for nonadherence or discontinuation were provided, and respondents were asked to indicate whether they engaged in the behavior.

Quality of life. The Functional Assessment of Cancer Therapy-Breast (FACT-B) was also included as a health-related QOL (HR-QOL) measurement instrument.19 The FACT-B includes the 7 subscales from the Functional Assessment of Cancer Therapy-General (FACT-G)—physical well-being, social well-being, family well-being, relationship with doctor, emotional well-being, functional well-being, trial outcome index—and the total score for the FACT-G. The FACT-B also includes a separate breast cancer subscale and a total score (FACT-B total score) that combines all of the subscales. For all of the subscales and total scores, higher values reflect better HR-QOL.

Stated preferences and choice task. Participants rated the importance of various attributes (from 1, indicating “extremely important,” to 5, indicating “not at all important”) and completed a choice-based conjoint task consisting of 7 choice scenarios, each containing 2 profiles of hypothetical treatments. Participants then selected the treatment they preferred from the 2 choices based on the information in the profile; participants were told to assume that all other attributes of the medications that were not explicitly mentioned in the profiles were identical (Table 1).

For each profile of a hypothetical treatment in the choice scenarios, 11 different attributes were comprised of 8 safety attributes—including alopecia, motor neuropathy, myalgia/arthralgia, nausea/vomiting, fatigue, neutropenia, mucositis/stomatitis, and diarrhea—1 effectiveness attribute, 1 dosing regimen attribute, and 1 QOL attribute. The actual levels of each attribute varied with every hypothetical profile and across all choice scenarios. These attributes were identified from the preliminary qualitative research described earlier as well as an examination of past literature.

Given the number of unique attribute-level combinations (ie, profiles), a fractional factorial balanced incomplete block design was used. Kuhfeld’s SAS macros (SAS Institute; Cary, NC)20 were used to derive a set of orthogonal arrays that were 99.9% D-efficient with respect to parameter variance. The resulting instruments had orthogonal profiles and balanced attribute levels.

Statistical Analysis
The choice tasks were analyzed using hierarchical Bayesian logistic regression models, whereas the regression models were parameterized using effects coding. The resulting model coefficients are viewed as part-worth utilities that also served as inputs to the relative importance analysis. To understand the relative importance of the attributes, deviations of the part-worth utilities from the overall expected value were calculated to create sums of squares for each individual attribute; for each attribute, the regression coefficients for each level of that attribute were squared and were summed together. The resulting sums of squares were divided by the attribute-specific degrees of freedom to generate a mean sum of squares (MSS) for each attribute. The relative importance of each attribute was calculated by dividing the MSS for that attribute by the sum of all MSS values for all attributes. The attributes with higher relative importance have a disproportionately larger MSS than other attributes, which is a result of large coefficients of the individual levels. The individual-level hierarchical Bayesian modeling was conducted using Sawtooth’s CBC/HB v4.6.4 (Sawtooth Software; Orem, UT).


Table 2

Because they were significantly related to adherence and associated with QOL, general linear models controlling for age, race/ethnicity, education, and body mass index were used to estimate the relationship between the number of nonadherent behaviors and QOL measured with the FACT-B (unstandardized regression coefficients [ie, b] are provided, indicating the effect on the dependent variable [ie, QOL] with a 1-unit increase in the independent variable [ie, nonadherent behaviors]). The effect of adherence was examined separately among patients who had ever received an oral chemotherapy agent (N = 95) and those who were currently receiving an oral chemotherapy agent (N = 44).


Table 3

Results
Demographics and Health History
The mean age of the patient sample was 52.2 years (standard deviation [SD], 9.1) with 93.9% of participants reporting being non-Hispanic white. The patients were generally of high socioeconomic status, with 71.8% of participants reporting having a college degree, 47.5% reporting an annual household income of ≥$75,000, and 76.8% reporting that their insurance covers all of their treatments for breast cancer (Table 2).

The total monthly OOP costs per patient were approximately $303 (SD, $785) for treatments related to breast cancer and approximately $107 (SD, $200) for physician visits related to breast cancer (Table 3).
The OOP costs, particularly for the treatment of breast cancer, were higher among patients who were non-Hispanic white, were college educated, had a higher household income, and were not enrolled in a patient assistance program. Women in the Midwest United States (28.18%) reported the highest cancer-related OOP costs (ie, $586.96).

The patients had extensive previous treatment experience across treatment modalities (100% intravenous chemotherapy, 82.3% surgery, 77.4% hormone therapy, 71.3% radiation therapy, and 52.5% oral chemotherapy). The majority of participants (90.6%) were currently receiving treatment, with the most common treatment modalities being hormone therapy (47.5%), intravenous chemotherapy (42%), and oral chemotherapy (24.3%).

Patient Preferences
When asked directly, the most important attributes of treatments for participants were related to effectiveness, followed by side effects. Cost-related attributes were the least important (Table 4).
Results from the conjoint model reaffirmed the primary importance of effectiveness, as indicated by overall survival, followed by side effects (the most notable of which were alopecia, fatigue, neutropenia, motor neuropathy, and nausea/vomiting), and, finally, dosing regimen (Table 5, Figure). For example, the logit column in Table 5 represents the strength (and direction) of the relationship between the presence of an attribute and selecting that treatment. Survival of 3 months (5.24), alopecia of 0% (1.70), and fatigue of 0% (1.19) had the highest positive logits, suggesting the strongest relationship with the probability of selection.

Post-hoc analyses investigated whether these preferences would vary by treatment experience (ie, rounds of chemotherapy). Our findings suggest that patient preference was remarkably consistent and did not vary by treatment experience. In our analyses, all attributes were rank ordered identically, and relative importance values of any given subgroup were within 2% to 3% of any other subgroup. For example, the relative importance of effectiveness attribute was highest for having received more than 6 rounds of chemotherapy (34.51%) and lowest for having received less than 2 rounds of chemotherapy (32.45%).


Table 4


Table 5

Adherence
A total of 63 (34.8%) survey participants either discontinued their treatment or were nonadherent to their treatment regimen. Patients who had ever received hormone therapy (37.9%) reported the greatest level of nonadherence, followed by patients who had ever received an oral chemotherapy agent (36.8%). Across all treatment modalities, forgetfulness (41.3%) and intolerance of side effects (36.5%) were the most common reasons for nonadherence among patients who reported their nonadherence or discontinuation (Table 6).

Among the patients who had ever received an oral chemotherapy agent (N = 95), the number of nonadherent behaviors (mean, 1.57) was used to predict HR-QOL (Table 7). Analogous models were also conducted among patients who were currently receiving a chemotherapy agent (N = 44). The number of nonadherent behaviors was significantly associated with a decrease in functional well-being (b = –2.01 for patients who had ever used a targeted therapy and b = –3.14 for current users of a targeted therapy), FACT-G total score (b = –4.30 and b = –7.37, respectively), FACT-B total score (b = –3.93 and b = –6.11, respectively), and FACT trial outcome index (b = –5.22 and b = –8.63, respectively; all P <.05).

Discussion
The primary objective of this study was to assess patient preferences for the treatment of mBC, because most of the existing literature focused on preferences among women with early-stage breast cancer. Although our results were generally comparable with the results of patients with early-stage disease,7,8 we found that treatment effectiveness was rated as the most important attribute among women with mBC, nearly twice as much as alopecia, and more than 3 times more important than other side effects. The rank order of important attributes for a treatment is similar to the findings by Beusterien and colleagues8; however, the current study estimated a greater preference for treatment effectiveness and the avoidance of alopecia, and a lower preference for regimen and diarrhea. These findings help clarify the patient perspective of treatments for mBC, which, if aligned with prescribing patterns, may maximize treatment satisfaction and adherence.

A secondary objective of this study was to assess the level of nonadherence for women with mBC who are receiving an oral chemotherapy agent and to establish the relationship between nonadherence and QOL. Approximately 33% of women with mBC reported engaging in medication nonadherent behavior. This rate is consistent with previous reviews of treatment adherence in patients with breast cancer.21 However, given the fairly affluent and engaged patient population, it can be hypothesized that adherence levels in the overall population with mBC are even lower. Of note, even after adjusting for established predictors of QOL, the nonadherent behaviors of women receiving an oral chemotherapy agent were associated with significant decrements in health status.


Figure


Table 6

Forgetfulness and adverse events were among the most common reasons for nonadherence, and these results suggest that a less complex treatment regimen and regimens with less toxic profiles may be associated with improvements in adherence, and subsequently could correspond to perceptible patient benefits. Other potential interventions may include greater education about potential side effects (which could mitigate the influence of side effects on nonadherence) and memory aids (eg, reminder systems). Further research is necessary to understand these potential interventions and their effectiveness in this patient population.


Table 7

Targeted therapies have become increasingly important in the treatment of many cancers, including breast cancer.21 Such therapeutic agents are often taken orally and can often have less severe side effects than traditional chemotherapy regimens, which can improve patient outcomes.22 Although these advantages can increase adherence—because patients prefer oral therapies to those administered intravenously—medication adherence is complex. The complexity of adherence can be influenced by a multitude of patient- and care-based factors, including the prohibitive costs of targeted treatments for some patients.21

Research examining adherence to targeted oral medications in the context of breast cancer has shown that approximately 20% of patients are nonadherent, depending on the way in which adherence was measured.22-24 Furthermore, in a study of adherence to lapatinib, Kartashov and colleagues reported that nonadherence is associated with increased provider visits and suggested an increase in the cost of care.23

Further research is needed to enhance our understanding of the factors influencing adherence rates in the treatment of breast cancer, and to develop effective interventions in line with advances in the pharmaceutical industry.

A literature search did not uncover any epidemiology study that could be useful to compare the general US population of patients receiving treatment for mBC against the sample in this current study.

Limitations
The results of the current study should be examined within the context of its limitations. All data, including treatments and adherence information, were self-reported and could have been subject to recall biases and other self-presentation effects. The conjoint task is a simplifying hypothetical exercise to estimate how patients value certain attributes of treatments, and may differ with the inclusion of other relevant attributes or levels. For example, in actual decision-making, the true benefit or risks that will be experienced (eg, there is a probability of certain survival outcomes, but not a guaranteed 3-month survival) cannot be known with certainty. In addition, differences in analytical methods and attributes can also make it difficult to directly compare preference results from one study to another.

Of note, the relationship between nonadherence and QOL was presented in a single direction as nonadherence predicting QOL. It is possible that the relationship could be bidirectional (eg, QOL influencing adherence) or that there could be additional third variables that are not included in the current study that could explain the relationships observed here. In addition, detailed information on disease progression, treatment chronology, side effects experienced, and dosing adjustments was unavailable, so the context for the results on adherence is incomplete. For example, it is not known what the mechanism was for the relationship between nonadherence and health outcomes (eg, nonadherence leading to faster disease progression leading to poorer QOL).

Finally, the current study used a convenience sample of women with mBC. These women were generally of high socioeconomic status, and, because of their participation in online oncology panels, may be more engaged and/or knowledgeable about their condition and treatments, which may influence their preferences, adherence, and health outcomes.

Conclusion
The current study suggests that the treatment preferences of women diagnosed with mBC, an understudied group in this context, are similar to the preferences of women diagnosed with early-stage disease. Among the patient sample surveyed, treatment effectiveness was of primary importance, an effect that did not appear influenced by other study variables. Further analysis revealed that approximately 33% of the women surveyed reported some degree of treatment nonadherence, which, in turn, was associated with impairment of QOL among patients receiving an oral chemotherapy agent. These results have important implications for treatment planning in the context of mBC, as well as reinforce the need to consider medication adherence in promoting QOL and desired treatment outcomes.

Acknowledgments
The authors would like to thank Duncan Brown, PhD, for his contribution to the survey design and recruitment of respondents. Dr Brown was a full-time employee of Kantar Health at the time of the study. The authors would also like to thank Errol Philip, PhD, for his assistance with the literature review and editorial comments on the manuscript. Dr Philip is a paid consultant to Kantar Health.

Funding Source
This study was funded by Eisai Inc.

Author Disclosure Statement
Dr DiBonaventura’s institution received research support from Eisai Inc; Dr Copher is an employee of Eisai Inc; Mr Basurto is a consultant to and received research support from Eisai; Dr Faria is an employee of Eisai; and Ms Lorenzo’s institution received research support from Eisai Inc.

Dr DiBonaventura is Vice President, Health Outcomes, Health Outcomes Practice, Kantar Health, New York, NY; Dr Copher is Associate Director, Health Economics and Outcomes Research, Eisai Inc, Woodcliff Lake, NJ; Mr Basurto is Senior Associate Methodologist, Kantar Health, New York, NY; Dr Faria is Director, Health Economics and Outcomes Research, Eisai Inc, Woodcliff Lake, NJ; and Ms Lorenzo is Senior Director of Research, Kantar Health, New York, NY.

References
1. American Cancer Society. Cancer facts & figures 2014. 2014. www.cancer.org/
acs/groups/content/@research/documents/webcontent/acspc-042151.pdf. Accessed March 14, 2014.
2. American Cancer Society. Breast cancer facts and figures 2013-2014. 2013. www.cancer.org/acs/groups/content/@research/documents/document/acspc-042725.pdf. Accessed March 14, 2014.
3. Cardoso F, Fallowfield L, Costa A, et al; for the ESMO Guidelines Working Group. Locally recurrent or metastatic breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2011;22(suppl 6):vi25-vi30.
4. Beaumont T, Leadbeater M. Treatment and care of patients with metastatic breast cancer. Nurs Stand. 2011;25:49-56.
5. Gao S, Barber B, Schabert V, Ferrufino C. Tumor hormone/HER2 receptor status and pharmacologic treatment of metastatic breast cancer in Western Europe. Curr Med Res Opin. 2012;28:1111-1118.
6. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines): breast cancer. Version 3.2014. April 1, 2014. www.nccn.org/professionals/physician_gls/pdf/breast.pdf. Accessed September 26, 2014.
7. Beusterien K, Grinspan J, Kuchuk I, et al. Use of conjoint analysis to assess breast cancer patient preferences for chemotherapy side effects. Oncologist. 2014;19:127-134.
8. Beusterien K, Grinspan J, Tencer T, et al. Patient preferences for chemotherapies used in breast cancer. Int J Womens Health. 2012;4:279-287.
9. Alvarado MD, Conolly J, Park C, et al. Patient preferences regarding intraoperative versus external beam radiotherapy following breast-conserving surgery. Breast Cancer Res Treat. 2014;143:135-140.
10. Kimman ML, Dellaert BG, Boersma LJ, et al. Follow-up after treatment for breast cancer: one strategy fits all? An investigation of patient preferences using a discrete choice experiment. Acta Oncol. 2010;49:328-337.
11. Mandelblatt JS, Sheppard VB, Hurria A, et al; for the Cancer Leukemia Group B. Breast cancer adjuvant chemotherapy decisions in older women: the role of patient preference and interactions with physicians. J Clin Oncol. 2010;28:3146-3153.
12. Simes RJ, Coates AS. Patient preferences for adjuvant chemotherapy of early breast cancer: how much benefit is needed? J Natl Cancer Inst Monogr. 2001;30:146-152.
13. Fallowfield L, McGurk R, Dixon M. Same gain, less pain: potential patient preferences for adjuvant treatment in premenopausal women with early breast cancer. Eur J Cancer. 2004;40:2403-2410.
14. McQuellon RP, Muss HB, Hoffman SL, et al. Patient preferences for treatment of metastatic breast cancer: a study of women with early-stage breast cancer. J Clin Oncol. 1995;13:858-868.
15. Magai C, Consedine N, Neugut AI, Hershman DL. Common psychosocial factors underlying breast cancer screening and breast cancer treatment adherence: a conceptual review and synthesis. J Womens Health (Larchmt). 2007;16:11-23.
16. Banning M. Adherence to adjuvant therapy in post-menopausal breast cancer patients: a review. Eur J Cancer Care (Engl). 2012;21:10-19.
17. Murphy CC, Bartholomew LK, Carpentier MY, et al. Adherence to adjuvant hormonal therapy among breast cancer survivors in clinical practice: a systematic review. Breast Cancer Res Treat. 2012;134:459-478.
18. Mayer EL, Partridge AH, Harris LN, et al. Tolerability of and adherence to combination oral therapy with gefitinib and capecitabine in metastatic breast cancer. Breast Cancer Res Treat. 2009;117:615-623.
19. Bonomi AE, Cella DF, Hahn EA, et al. Multilingual translation of the Functional Assessment of Cancer Therapy (FACT) quality of life measurement system. Qual Life Res. 1996;5:309-320.
20. Kuhfeld WF. Marketing Research Methods in SAS: Experimental Design, Choice, Conjoint, and Graphical Techniques. SAS 9.2 Edition, MR-2010. Cary, NC: SAS
Institute Inc; 2010. http://support.sas.com/techsup/technote/mr2010.pdf. Accessed April 15, 2014.
21. Geynisman DM, Wickersham KE. Adherence to targeted oral anticancer medications. Discov Med. 2013;15:231-241.
22. Weingart SN, Brown E, Bach PB, et al. NCCN Task Force Report: oral chemotherapy. J Natl Compr Canc Netw. 2008;6(suppl 3):S1-S14.
23. Kartashov A, Delea TE, Sharma PP. Retrospective study of predictors and consequences of nonadherence with lapatinib (LAP) in women with metastatic breast cancer (MBC) who were previously treated with trastuzumab. J Clin Oncol. 2012;30(15 suppl). Abstract e11067.
24. Addeo R, Vincenzi B, Riccardi F, et al. Multicenter observational study on adherence and acceptance of lapatinib treatment in patients with HER2+ metastatic breast cancer. J Clin Oncol. 2011;29(15 suppl). Abstract e11102.

Stakeholder Perspective
Predicting Behavior: It’s a Matter of Preference

In the well-known quip that has been attributed to individuals as diverse as the quantum physicist Niels Bohr and the extraordinary baseball manager Casey Stengel, both men are purported to have said, “I never make predictions, especially about the future.” Predictions, especially when they come to the ability and/or desire to adhere to a medical regimen of care, are highly dependent on a number of factors.

And so it is in the article by DiBonaventura and colleagues in this issue of American Health and Drug Benefits.1 Not long ago, the paternalistic aspect of medical care consisted of the physician prescribing a course of medical care that was incumbent on the patient to follow. Societal norms stipulated such behavior as a function of the asymmetric knowledge relationship between a physician and a patient. As patients became more empowered, they demanded a greater role in determining their own care, and we saw the genesis of the era of “shared decision-making.” As Mazur discusses in his book on shared decision-making, clarifying the notion of this type of decision-making includes, among other things, taking into account the risk preferences of patients.2 It is relatively easy for physicians and patients to believe that they can and will conform to certain therapies when they expect a relatively easy trade-off of side-effect tolerance versus gains in survival time. However, experience does not always mirror expectation.

One reason for this lack of congruence between plan and result may lie in behavioral economics. In their seminal article on prospect theory, Kahneman and Tversky posited and demonstrated that people make decisions based on the potential value of losses and gains rather than on the final outcome.3 In the article by DiBonaventura and colleagues, women with metastatic breast cancer were willing to trade increases in side effects for increases in life expectancy.1 Of course, one could also argue that prospect theory notes how people underweight the likelihood of high probability events (eg, side effects from medical treatment) while overvaluing the likelihood of lower probability events (eg, cure in overall survival). This may also explain some of the findings encountered in the current study by DiBonaventura and colleagues.

PAYERS: Payers understand these findings all too well. In addition to being a HEDIS (Healthcare Effectiveness Data and Information Set) measure, medication adherence is an area of great interest, because of its overall impact on cost and outcomes. It is well-documented that nonadherence to treatment, whether pharmacologic or otherwise, results in increased costs and worsened health. Many payers have thus developed specialty analytic units to study and address the conundrum why patients are nonadherent to therapy.

PATIENTS: People often make decisions without fully understanding the subconscious reasons why they have chosen to act as they do. If individual patients perhaps had more time or more inclination to truly share their real perspectives, it is very possible that there would be more adherence to treatment, or that more appropriate therapy decisions would be made in a shared manner in the first place. Of course, the key to this occurring is to study why it is not currently happening. As DiBonaventura and colleagues note, perhaps a decrease in the complexity of therapy may help in that endeavor.1

Although we can certainly all hope for more understanding of patients’ preferences and their role in successful care outcomes, what we cannot do, however, is predict that this will actually happen.

1. DiBonaventura MdC, Copher R, Basurto E, et al. Patient preferences and treatment adherence among women diagnosed with metastatic breast cancer. Am Health Drug Benefits. 2014;7:386-396.
2. Mazur DJ. Shared Decision Making in the Physician-Patient Relationship. Tampa, FL: American College of Physician Executives; 2001:103.
3. Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica. 1979;47:263-291.

1. DiBonaventura MdC, Copher R, Basurto E, et al. Patient preferences and treatment adherence among women diagnosed with metastatic breast cancer. Am Health Drug Benefits. 2014;7:386-396.
2. Mazur DJ. Shared Decision Making in the Physician-Patient Relationship. Tampa, FL: American College of Physician Executives; 2001:103.
3. Kahneman D, Tversky A. Prospect theory: an analysis of decision under risk. Econometrica. 1979;47:263-291.

Related Items
Changes in Antipsychotic Medication Use Among Medicare Patients in a Nursing Home, 2010 to 2015
Michele Berrios, Bruce S. Pyenson, FSA, MAAA, Kyle Pérez, MPH, Heidi C. Waters, PhD
Web Exclusives published on November 10, 2023 in Original Research, Clinical
The Hidden Inferno: Burn Pit Exposure in the Military and Its Potential Links to Cancer
Claire Szewczyk
Web Exclusives published on October 20, 2023 in Clinical
Real-World Treatment Patterns and Healthcare Costs Among Patients with FL with Early Treatment Failure of First-Line Chemoimmunotherapy
Lori A. Leslie, MD, Bruno Emond, MSc, Marie-Hélène Lafeuille, MA, Maude Vermette-Laforme, BSc, Patrick Lefebvre, MA, Qing Huang, PhD, MHS
September 2022 Vol 15, No 3 published on September 27, 2022 in Clinical, Original Research
The Quality of Care and Economic Burden of COPD in the United States: Considerations for Managing Patients and Improving Outcomes
David L. Larsen, RN, MHA, Hitesh Gandhi, MBBS, Michael Pollack, MS, Norbert Feigler, MD, Sushma Patel, PharmD, Robert A. Wise, MD
June 2022 Vol 15, No 2 published on June 23, 2022 in Clinical, Review Article
Migration of Hospital Total Hip and Knee Arthroplasty Procedures to an Ambulatory Surgery Center Setting and Postsurgical Opioid Use: A Private Practice Experience
James Van Horne, MD, Alaine Van Horne, BS, Nick Liao, MS, Victoria Romo-LeTourneau, PharmD
March 2022 Vol 15, No 1 - Online Only published on March 23, 2022 in Original Research, Clinical
Last modified: August 30, 2021