Abstract
Background:
Adjuvant endocrine therapy (AET) is a life-saving medication for patients with hormone-sensitive breast cancer, yet many struggle with adherence, warranting behavioral intervention. In our recent trial, participation in a group cognitive behavioral intervention (STRIDE) for symptom management and adherence was associated with improvements in symptom distress, coping, quality-of-life, and mood. We now explore whether baseline patient- and medication-specific factors—that may be modifiable by clinician-led discussions— moderated the effect of STRIDE on adherence rates.
Methods:
From 10/2019 to 6/2021, 100 patients with early-stage breast cancer reporting AET-related distress were enrolled and randomly assigned to STRIDE or a medication monitoring control group (MM). All patients stored their AET in electronic pill bottles to track objective adherence. Patients also self-reported their adherence on the Medication Adherence Report Scale (MARS-5) and their perceptions of AET on the Cancer Therapy Satisfaction Questionnaire (CTSQ) at baseline. We conducted hierarchical linear modeling to test moderators of intervention effects on objective adherence rates. We report the ‘time X group X moderator’ effects.
Results:
Among patients reporting greater perceived difficulties with AET adherence at baseline, STRIDE participants had higher adherence rates over time, compared to MM (b=−13.80, SE=4.56, p<.01). Patients with greater expectations of therapeutic benefit from AET also had improved adherence rates if they were assigned to STRIDE, versus MM (b=0.25, SE=0.10, p=.01). Patients who perceived taking AET as convenient and had been taking their AET for less time had higher adherence rates in STRIDE, versus MM.
Conclusions:
The current study identified patient- and medication-specific factors that may augment AET adherence interventions and may be modifiable through clinician-led discussions, such as perceptions of adherence problems, therapeutic efficacy, and convenience of AET.
Keywords: adjuvant endocrine therapy, adherence, breast cancer, STRIDE, treatment perceptions
INTRODUCTION
Adjuvant endocrine therapy (AET) is a central component of treatment for patients diagnosed with hormone-sensitive breast cancer, the most common form of breast cancer in the United States.1 AET, including tamoxifen and aromatase inhibitors, is prescribed daily for 5–10 years after initial treatment to reduce risk of recurrence.2 Nevertheless, common side effects of AET—including musculoskeletal symptoms, hot flashes, sexual dysfunction, and mood changes—make adherence to these medications challenging.3,4 Indeed, previous longitudinal studies have found that 30–60% of patients with breast cancer either take the medication less frequently than prescribed or discontinue AET altogether,5–7 increasing their risk for breast cancer recurrence and mortality.8,9 Breast oncology clinicians report difficulty optimizing adherence to AET for patients struggling with more severe side effects.10 AET is typically prescribed when primary treatment ends and as patients transition from active treatment to routine surveillance. During the majority of AET receipt, patients tend to interact less frequently with their oncology team; therefore, optimizing initial conversations about the benefits and importance of AET may be most influential for long-term adherence.
Understanding factors that influence adherence to AET requires a multifaceted lens. Recent systematic reviews have organized predictors of nonadherence into five categories, as proposed by the World Health Organization:11,12 socioeconomic factors, condition/disease-related factors, medication-related factors, healthcare provider/system-related factors, and patient-related factors. Essential modifiable factors that are both patient-related and provider-related are patients’ perceptions of the medication itself and their own medication-taking behaviors, including their views of how essential, beneficial, and challenging AET may be.13–15 For example, a study by Lee and Min (2018)16 found that a patient’s beliefs about the necessity of AET for breast cancer recurrence—which is influenced by clinician-delivered education as well as patient-clinician conversations and rapport17—explained the largest proportion of variance in adherence compared to psychological and sociodemographic/medical factors, suggesting a more significant role for breast cancer clinicians in promoting AET adherence. The multiple factors promoting adherence, including those driven by clinician conversations, may be modifiable through behavioral intervention.
Burgeoning interventions have targeted AET adherence, given the aforementioned challenges and relevance to survival.18 A 2023 systematic review and meta-analysis suggests that while certain behavioral intervention components contribute to changes in AET adherence, there is an overall lack of evidence for improvement in AET adherence following behavioral intervention at this time.16,19 Our group previously developed a digital mobile application (‘CORA app’) for oral chemotherapy symptom management and adherence for patients with solid tumors.20 In a randomized controlled trial (RCT), we demonstrated that among patients with adherence problems at baseline, those assigned to the CORA app had better adherence compared to usual care.21 We recently completed a RCT of a cognitive behavioral, therapist-delivered, telehealth intervention (STRIDE) targeting patients’ self-management of AET symptoms and adherence to AET, compared to a medication monitoring control group (MM).22–24 STRIDE was associated with greater self-management of AET symptoms, lower AET symptom distress, better ability to cope with distress, fewer mood symptoms, and better QOL, compared to the MM control group.23 The STRIDE group overall exhibited slightly higher adherence rates after intervention, though these did not reach significance when compared to the control group. Given prior evidence, the current study sought to identify whether specific patient- and medication-specific factors moderated the intervention effect on objective adherence rates for patients in the STRIDE intervention versus the MM control. Specifically, we examined the role of patients’ initial perceived difficulty with AET adherence, time since starting AET, and beliefs about therapeutic efficacy and convenience as factors that may influence the intervention effect on adherence and may be, in part, modifiable through clinician-led conversations and education.
METHODS
Study Design
The current study is an exploratory analysis following a RCT testing STRIDE versus MM (ClinicalTrials.gov identifier: NCT03837496). We recruited and enrolled patients at the Massachusetts General Hospital Cancer Center in Boston, MA and three community affiliates between October 2019 and June 2021. The study was reviewed and approved by the Dana-Farber/Harvard Cancer Center Institutional Review Board.
Participants
Eligible participants included English-speaking, females (≥21 years old) with a diagnosis of early-stage (stage 0-IIIb) hormone-sensitive breast cancer who had completed treatment and were within one week to 36 months of starting AET. Patients were screened and deemed eligible if they endorsed AET distress (≥4) on at least one of three questions adapted from the National Comprehensive Cancer Network distress thermometer (range = 0–10).25 Full eligibility and screening criteria are listed in Table 1 and have been published previously.24
Table 1.
Eligibility criteria
| Inclusion Criteria | Exclusion Criteria |
|---|---|
| • Female • Age 21 or older • Diagnosis of early stage (stage 0-IIIb), hormone-sensitive breast cancer • Within 1 week to 36 months of starting adjuvant endocrine therapy • Ability to read and respond in English • Eastern Cooperative Oncology Group performance status ≥2 • Currently taking adjuvant endocrine therapy (i.e., if took a recent break, has taken within the past 2 weeks) • Completed primary treatment (i.e., chemotherapy, surgery and/or radiation) for early-stage breast cancer • Indicates a score ≥4 on one of the three National Comprehensive Cancer Network adapted distress thermometer study screening questions25 |
• Uncontrolled psychosis, active suicidal ideation, or psychiatric hospitalization within the past year • Cognitive impairment that prohibits participation in the study • Enrollment in a different clinical trial for breast cancer • Current participation in formal group psychotherapy or other psychosocial intervention |
Study Procedures
Study staff reviewed the electronic health records (EHR) of patients with upcoming oncology appointments. Potentially eligible patients were approached with permission from their oncologist. Interested and eligible patients provided informed consent, completed a baseline psychosocial assessment, and received a Medication Event Monitoring System (MEMS) Cap26 and bottle to store their AET. All patients used the MEMS Cap for a one-week baseline period, after which they were randomly assigned 1:1 to either the STRIDE intervention or the MM control group via a computer-generated randomization scheme. Randomization was stratified by participants’ baseline distress as measured by the Hospital Anxiety and Depression Scale subscales (high ≥8 vs low <8).23,27 All patients used the MEMS Cap for the duration of the study and completed follow-up assessments at 12- and 24-weeks post-baseline. In the current analysis, we report data from the baseline and 12-week follow-up only. The study protocol has been published previously.24
Intervention Arms
STRIDE Intervention.
Patients randomized to STRIDE received six weekly, small-group, 1-hour sessions over Zoom with a trained clinical psychologist. Following the six sessions, patients received two individual 20-minute booster sessions via telephone to review skills and discuss challenges. As published in-depth elsewhere, 22 the intervention incorporated cognitive behavioral skills such as cognitive restructuring, behavioral activation, relaxation training, and coping to improve adherence and mood. The intervention also addressed risks and benefits of AET, barriers to adherence, and problem-solving skills.
Medication Monitoring (MM) Control.
Participants in the control group received follow-up oncology care as usual and stored their AET medication in the MEMS Cap, which did not provide feedback or reminders about medication-taking.
Measures
Sociodemographic and Medical Variables.
Participants self-reported their age, gender, race, ethnicity, relationship status, and current level of education. We collected clinical and treatment characteristics from the EHR, including their recruitment site, disease stage, node status, hormone receptor status, treatment received, type of AET, and date of AET initiation, which was also tested as a baseline moderator.
Primary Outcome
Objective Adherence.
MEMS Caps electronically recorded the date and time the bottle was opened, as a proxy for taking medication.26 Prior to randomization, participants completed a 1-week trial of the MEMS Caps, giving participants time to adjust to the new device. They did not receive any feedback regarding their adherence data during the study. Data were aggregated across the study period to produce monthly adherence percentages (doses taken of doses prescribed). To account for any discrepancies, patients tracked doses of medication taken outside of their MEMS Cap in a study-provided medication diary. Physician-prescribed drug holidays were noted at the conclusion of the study and excluded as to not penalize the patient for adhering to the regimen as prescribed.
Potential Baseline Moderators
Perceived Difficulty to AET Regimen.
The Medication Adherence Report Scale (MARS-5) is a 5-item scale assessing patients’ perceived difficulties with AET adherence behaviors, including pausing, forgetting, or altering the dose (1 = Always to 5 = Never).28 Lower scores denote greater difficulties with AET adherence behaviors.
AET Perceptions.
Perceptions of AET were evaluated using the Cancer Therapy Satisfaction Questionnaire (CTSQ).29 This 16-item measure assessed participants’ preferences and beliefs across four domains: AET expectations, satisfaction with AET, AET side effects, and convenience of taking AET. The AET adherence subscale consists of two items and was excluded from the current study due to low variability and conceptual overlap with the MARS-5. Patients rated statements on a scale of Always (5) to Never (1). Higher scores indicated greater AET satisfaction respective to each domain.
Statistical Plan
In these secondary analyses, we adjusted models for level of baseline distress via the Hospital Anxiety and Depression Scale27 (high distress ≥8; low distress <8), which was used to stratify participants for randomization at baseline, and receipt of ovarian suppression to account for baseline group differences.
We employed hierarchical linear modeling using R Software, version 4.3.2 (2023-10-31), to test 3-way, cross-level interaction effects of unique, continuous moderators (between-person), group assignment (between-person), and time (months, within-person) on adherence trajectories (i.e., MEMS-Cap monthly adherence rates across months one, two, and three). Significant interaction effects indicated differences in adherence over time by group condition and at higher versus lower levels of each moderator. We tested random time effects and used a likelihood ratio test (LRT) to assess if model fit improved and if so, were retained. Interaction terms with a p-value of less than .10 were probed for simple slopes, an approach used to examine trends and modify the type I error rate within underpowered interaction analyses.30,31 Related analyses from this trial have been published separately32 and do not report on any of the tested moderators from the current study.
RESULTS
Sample Composition
As illustrated in Table 2, the study sample (N=100) mainly consisted of non-Hispanic, White (87%) women with stage I breast cancer (77%). Patients ranged from 31 to 81 years old (M=56.1, SD=10.94) and had been prescribed AET for 18 months on average (SD=8.64). Sixty percent of patients were taking an aromatase inhibitor. Objective adherence data was available for 83 of the 100 patients who returned their MEMS Caps following the study period. Monthly adherence rates fluctuated between 88–91%, and no differences existed between groups.23 As previously reported from this trial,32 an LRT revealed that random effects significantly improve model fit (χ2=6.97(2), p=.03) and thus, were retained for all models tested. The interclass correlation was .53, indicating that 53% of adherence variance was explained by between-patient differences. Full model results may be found in Table 3.
Table 2.
Sociodemographic, Clinical, and Treatment Characteristics at Baseline
| STRIDE Intervention N = 50 |
Medication Monitoring N = 50 |
Full Sample N = 100 |
|
|---|---|---|---|
| Mean (Standard Deviation) | |||
|
| |||
| Age (years; range = 31–81) | 57.2 (10.6) | 54.9 (11.2) | 56.1 (10.9) |
| Months on AET (AET start to enrollment) | 17.70 (8.87) | 18.13 (8.49) | 17.91 (8.64) |
|
| |||
| N (%) | |||
|
| |||
| Gender | |||
| Women | 50 (100) | 50 (100) | 100 (100) |
| Race | |||
| White | 47 (94) | 44 (88) | 91 (91) |
| Asian | 0 (0) | 4 (8) | 4 (4) |
| Black or African American | 1 (0) | 0 (0) | 1 (1) |
| Other | 1 (2) | 2 (4) | 3 (3) |
| Not Reported | 1 (2) | 0 (0) | 1 (1) |
| Ethnicity | |||
| Hispanic or Latino/a | 1 (2) | 2 (4) | 3 (3) |
| Not Hispanic or Latino/a | 47 (94) | 47 (94) | 94 (94) |
| Not Reported | 2 (4) | 1 (2) | 3 (3) |
| Education | |||
| Advanced Professional Degree | 3 (6) | 6 (12) | 9 (9) |
| Master’s Degree | 16 (32) | 16 (32) | 32 (32) |
| College Graduate | 19 (38) | 15 (30) | 34 (34) |
| Some college/Technical School | 9 (18) | 7 (14) | 16 (16) |
| High school graduate/GED | 3 (6) | 5 (10) | 8 (8) |
| 11th grade or less | 0 (0) | 1 (2) | 1 (1) |
| Relationship Status | |||
| Married/Cohabitating | 38 (76) | 35 (70) | 73 (73) |
| Non-cohabitating relationship | 1 (2) | 2 (4) | 3 (3) |
| Single, never married | 5 (10) | 4 (8) | 9 (9) |
| Divorced/separated | 6 (12) | 6 (12) | 12 (12) |
| Loss of long-term partner/widowed | 0 (0) | 3 (6) | 3 (3) |
| Employment Status | |||
| Full-time/part-time work or student | 33 (66) | 29 (58) | 62 (62) |
| Caring for home or family | 4 (8) | 6 (12) | 10 (10) |
| Unemployed | 1 (2) | 3 (6) | 4 (4) |
| Not working due to illness/disability | 0 (0) | 1 (2) | 1 (1) |
| Retired | 9 (18) | 9 (18) | 18 (18) |
| Other or missing | 3 (6) | 2 (4) | 5 (5) |
| Income | |||
| $25,000–$49,999 | 2 (4) | 5 (10) | 7 (7) |
| $50,000–$99,999 | 11 (22) | 11 (22) | 22 (22) |
| $100,000–$149,999 | 9 (18) | 9 (18) | 18 (18) |
| > $150,000 | 27 (54) | 23 (46) | 50 (50) |
| Declined to respond | 1 (2) | 2 (4) | 3 (3) |
| Breast Cancer Stage | |||
| Stage 0 | 5 (10) | 3 (6) | 8 (8) |
| Stage I | 36 (72) | 41 (82) | 77 (77) |
| Stage II | 6 (12) | 5 (10) | 11 (11) |
| Stage III | 3 (6) | 1 (2) | 4 (4) |
| Type of AET | |||
| Aromatase Inhibitor | 28 (56) | 32 (64) | 60 (60) |
| Tamoxifen | 22 (44) | 18 (36) | 40 (40) |
| Primary Treatment Type | |||
| Surgery only | 10 (20) | 13 (26) | 23 (23) |
| Radiation only | 1 (2) | 0 (0) | 1 (1) |
| Surgery & Radiation | 24 (48) | 23 (46) | 47 (47) |
| Surgery & Chemotherapy | 2 (4) | 6 (12) | 8 (8) |
| Surgery, Chemotherapy, & Radiation | 13 (26) | 8 (16) | 21 (21) |
| Node Status | |||
| Node Positive | 9 (18) | 12 (24) | 21 (21) |
| Node Negative | 35 (70) | 32 (64) | 67 (67) |
| Not Reported | 6 (12) | 6 (12) | 12 (12) |
| Menopausal Status | |||
| Pre- or Peri-menopausal | 18 (36) | 18 (36) | 36 (36) |
| Post-menopausal | 25 (50) | 23 (46) | 48 (48) |
| Not Reported | 7 (14) | 9 (18) | 16 (16) |
| HER2/neu Status | |||
| HER2/neu Positive | 5 (10) | 8 (16) | 13 (13) |
| HER2/neu Negative | 40 (80) | 40 (80) | 80 (80) |
| Not Reported | 5 (10) | 2 (4) | 7 (7) |
| Ovarian Suppression | |||
| Receiving Ovarian Suppression | 8 (16) | 20 (40) | 28 (28) |
| Not Receiving Ovarian Suppression | 42 (84) | 30 (60) | 72 (72) |
Note:
AET = Adjuvant Endocrine Therapy
Table 3.
Model Estimates
| Parameters | Interaction model: Time since AET Initiation | Interaction model: CTSQ Convenience | Interaction model: CTSQ Side Effects | Interaction model: CTSQ Expectations | Interaction model: CTSQ Satisfaction | Interaction model: MARS-5 |
|---|---|---|---|---|---|---|
| Regression coefficients (fixed effects) | ||||||
| Intercept | 103.58 (9.06)*** | 66.15 (13.40) | 97.61 (14.03)*** | 79.41 (9.37)*** | 78.48 (19.34)*** | 54.65 (46.88) |
| Time | −11.95 (3.64)** | 4.30 (5.49) | −3.87 (5.97) | 2.94 (3.92) | −5.01 (8.12) | −52.59 (19.36)** |
| Distress | −2.69 (3.42) | −2.60 (3.37) | 02.43 (3.58) | −2.85 (3.46) | −2.54 (3.43) | −5.58 (3.06)+ |
| Ovarian Suppression | 0.49 (4.12) | 3.37 (4.30) | 0.56 (4.19) | 0.52 (4.15) | 1.32 (4.11) | 3.85 (3.71) |
| Group | −6.14 (12.14) | 33.53 (16.60)* | −5.11 (17.53) | 29.92 (12.23)* | 39.04 (27.01) | −11.52 (68.78) |
| Moderator | −0.64 (0.42) | 0.60 (0.31)+ | −0.15 (0.33) | 0.24 (0.17) | 0.21 (0.30) | 1.59 (2.00) |
| Group X Time | 9.05 (5.00)+ | −11.96 (7.10)+ | 6.66 (7.50) | −11.55 (5.19)* | −13.07 (11.41) | 60.46 (28.66)* |
| Group X Moderator | 0.55 (0.61) | −0.71 (0.40) | 0.24 (0.41) | −0.53 (0.23)* | −0.54 (0.42) | 0.68 (2.91) |
| Time X Moderator | 0.54 (0.18)** | −0.15 (0.13) | 0.05 (0.14) | −0.09 (0.07) | 0.05 (0.13) | 2.18 (0.83)* |
| Group X Time X Moderator | −0.43 (0.25)+ | 0.33 (0.17)+ | −0.15 (0.18) | 0.25 (0.10)* | 0.21 (0.18) | −2.55 (1.21)* |
| Residual variance (random effects) | ||||||
| Residual | 124.89 | 124.88 | 126.31 | 124.89 | 124.89 | 124.90 |
| Intercept | 245.37 | 233.19 | 260.35 | 224.98 | 247.93 | 253.40 |
| Time | 29.93 | 36.55 | 40.55 | 31.94 | 35.29 | 32.70 |
Note: Estimate (standard error).
= p<.10
= p<.05
= p<.01
= p<.001.
Subjective Adherence Difficulty
The 3-way interaction effect of time by group by perceived adherence difficulties on objective adherence rate reached significance (B=−2.55, SE=1.21, p=.038), suggesting that as baseline difficulties with adherence increased, the improvement in objective adherence in STRIDE compared to the MM control increased. See Figure 1.
Figure 1.
Improved Adherence to AET Over Time for Patients in STRIDE with Perceived Adherence Difficulties at Baseline
Note: Though analyzed continuously via moderation models, adherence differences among patients with greater adherence difficulties are visualized here, split at the median score on the MARS-5 (<24).
Patient Perceptions of AET
Two of the four CTSQ domains reached significance when tested for moderation effects. We observed a significant 3-way interaction effect of time by group by CTSQ Expectations subscale on monthly adherence (B=0.25, SE=0.10, p=.01), such that as patients’ baseline expectations of therapeutic benefit of AET increased, the improvement in adherence in STRIDE compared to MM control increased. Additionally, the 3-way interaction effect of time by group by CTSQ Convenience approached significance (B=0.33, SE=0.17, p=.06) per our a priori p = 0.10 for probing interaction effects. Specifically, as patients’ baseline perceptions of AET as convenient increased, the improvement in adherence in STRIDE compared to MM control increased. Models testing the 3-way interaction effects of the CTSQ Side Effects (B=−0.15, SE=0.18, p=.40) and Satisfaction (B=0.21, SE=0.18, p=.24) did not reach significance.
Time Since AET Initiation
A 3-way interaction effect of time by group by time since AET initiation reached significance (B=−0.43, SE=0.25, p=.089), per our a priori p = 0.10 for probing interaction effects. Patients who had been receiving AET for fewer months at the time they enrolled in the study had slightly better adherence rates across the 3 months if they were assigned to STRIDE, compared to those in the MM control.
DISCUSSION
In the current study, we examined moderators of a cognitive behavioral, therapist-delivered, telehealth intervention on patients’ objective AET adherence rates. Though we previously reported no overall group differences in adherence rates between the STRIDE intervention and MM control groups,23 results of the current study suggest that a subgroup of participants did, in fact, increase their adherence during participation in STRIDE. Namely, among those who reported greater perceived difficulty adhering to their AET regimen at baseline, those who had higher expectations of the therapeutic benefit of AET, those who believed AET is a convenient treatment option, and those who had started the medication more recently, the participants assigned to STRIDE demonstrated better adherence rates over the study period. Though the current study was exploratory, these findings highlight moderating effects that may improve a patient’s response to behavioral intervention for adherence. Moreover, given many of the identified moderators are clinical in nature, our findings underscore the essential role of breast cancer clinicians in optimizing intervention effects, including clinician-led discussions with patients about their perceived difficulties with AET, education regarding the therapeutic efficacy of AET, and strategies to simplify adherence within their daily routine. Lastly, because patients earlier in their AET treatment course were more likely to experience improvements in objective adherence, our findings suggest that this may be a critical period for effective behavioral intervention.
Given the paucity of evidence-based behavioral interventions for increasing adherence to AET for patients with breast cancer, the results of the current study are promising.19 This study builds on a previous examination of the moderating effects of our mobile CORA app for patients taking oral chemotherapy, in which patients with worse adherence at baseline exhibited better adherence in the CORA app intervention versus usual care.21 In addition, this study builds on our previous work highlighting patient-level demographic and mood-related factors moderating the effects of STRIDE; specifically, older patients, those with fewer symptoms of anxiety and depression at baseline, and those with lower AET symptom distress at baseline experienced improvements in adherence following STRIDE, suggesting behavioral interventions may be more beneficial for those who exhibit a less severe psychosocial symptom profile at baseline.32 Altogether, this work highlights the multifaceted factors at play in increasing adherence through behavioral intervention.
Our findings also align with previous models of AET adherence, including studies that have identified perceptions of AET and time since AET initiation as important contributors to adherence.13,16,33–37 Additionally, our results mirror a moderating effect of the REACH intervention—a low-touch, values-based, educational intervention developed to increase AET adherence in patients with early-stage breast cancer.38 In this randomized pilot study, positive AET attitudes and better oncologist-patient communication at baseline predicted greater adherence in those assigned to REACH compared to control, 38 bolstering the idea that 1) patients’ attitudes of AET contribute to the efficacy of a behavioral intervention, and 2) clinician-led communication is an important mechanism of addressing patients’ concerns and maximizing intervention effects. Interestingly, in the current study, patients’ perceptions of side effects did not significantly moderate intervention effects on adherence, though previous research on this relationship remains mixed.39
Breast cancer clinicians may be able to readily identify these subgroups of patients in clinical settings in need of AET support, providing an opportunity for them to assess patients’ beliefs about AET and intervene when appropriate. Previous research suggests that bidirectional conversations employing a non-judgmental, person-centered approach may lead to increased motivation and feelings of empowerment.17,40 Clinicians may assess each patient’s personal reasons for taking AET to better gauge their expectations of the medication. Our findings suggest that clinicians may be more successful when intervening earlier in their AET treatment course, perhaps because it is challenging to initially adjust to AET or because some patients may be maximally open to learning and implementing behavioral strategies at the start of their treatment. Clinicians may also support patients reporting difficulties by strategizing ways to support their adherence, underscoring AET’s convenience, and providing education on both the therapeutic benefit and necessity of the medication.41 Through these deliberate conversations, clinicians have an opportunity to help patients see the overall value of AET and correct any misinformation about its efficacy or side effects. While these aforementioned strategies are essential components of the STRIDE intervention, our results suggest a role for clinicians to engage in these conversations as well.
Future studies should use these exploratory findings in their development and/or refinement of behavioral interventions for adherence. Specifically, the results suggest that future trials should continue to use patients’ perceived difficulties with adherence as a part of the screening criteria for inclusion, as these patients may experience the most benefit. Moreover, future trials of STRIDE may focus on those who are earlier in their AET initiation, as these individuals may be more responsive to improving their objective adherence through intervention, making this a critical point in time to intervene. Relatedly, a more in-depth understanding of barriers to adherence in patients in later years of their regimen may be important for refining and optimizing the intervention. Lastly, trials should continue to strive to enroll patients regardless of their perceptions of AET (e.g., therapeutic benefit and convenience), given that they may still be struggling with adhering to their regimens at baseline for other reasons. Considering these patient- and medication-specific factors in behavioral intervention design may produce more efficacious treatments for this ubiquitous obstacle in breast cancer care that is directly tied to clinical outcomes.
The current study was an exploratory analysis of potential moderating effects of the STRIDE intervention and not without limitations. First, these exploratory findings are hypothesis-generating and, therefore, warrant replication in a larger study (e.g., future RCT of STRIDE). Next, the current study found that the longer a patient has been taking AET, the less modifiable their adherence behaviors were; however, there are currently no clinical thresholds that characterize a patient as “early” to AET treatment. In addition, the sample was fairly homogenous in that participants were predominantly White, non-Hispanic, partnered, highly educated, and receiving care from an urban academic medical center. Previous research suggests that demographic characteristics and socioeconomic status may affect adherence,11 highlighting the need to explore the intervention and these specific moderating relationships in a more diverse and representative sample. In addition, most patients demonstrated high levels of objective adherence (88–91%), and thus, a ceiling effect may be at play. Relatedly, approximately 83% of participants returned their MEM Cap bottles, which may bias results. There is also the possibility that the use of MEMS Cap alone impacted participants’ adherence behaviors, given the participants knew they were being monitored (i.e., Hawthorne effect); however, our study included a one-week baseline trial period to give patients time to adjust to the novelty of the MEMS caps as a way to minimize the effect of this new tracking device on adherence. The MEMS Cap also did not provide any reinforcement or feedback to participants. While the MEMS Cap is a gold standard measure of objective adherence,42 future studies may supplement this measure with other data collection methods, as no adherence measure to date is without flaws. Nevertheless, the results of the current study suggest that patients’ subjective understanding of their adherence may be an important inclusion criterion for a behavioral intervention such as STRIDE.
Conclusions
The current study highlights that patients who report worse adherence may benefit from a behavioral intervention, and that clinician-led conversations regarding AET efficacy may be important to optimize intervention effects. Overall, these findings should inform future behavioral interventions for improving adherence to AET, including future iterations of STRIDE. In addition, the findings suggest an opportunity for providers to assess patients’ beliefs and expectations about AET and intervene when appropriate.
CONTEXT SUMMARY.
Key Objective
How might specific patient- and medication-related factors moderate the effect of a brief behavioral symptom management and adherence intervention (‘STRIDE’) on adherence to adjuvant endocrine therapy (AET) for patients with early-stage breast cancer?
Knowledge Generated
Among patients with breast cancer taking AET participating in the STRIDE behavioral intervention, those who had difficulty with AET adherence, expected greater therapeutic benefit, perceived taking AET to be convenient, and who had started AET more recently exhibited higher objective adherence rates over time, measured by an electronic pill device, compared to patients assigned to the control group.
Relevance
The findings highlight patient characteristics and opportunities for clinician-patient discussions about AET expectations and beliefs that may bolster adherence alongside a behavioral intervention. In addition, adherence behaviors may be more modifiable for patients earlier in their AET regimen.
Funding:
This study was supported by NCI Career Development Award (K07CA211107; JMJ). KDW was also supported by a Dana-Farber Harvard Cancer Center T32 Training in Oncology Population Sciences Fellowship (T32CA092203).
Footnotes
Disclaimers: None
Presentation History: The results of the current study were presented at the American Society for Clinical Oncology 2023 Annual Meeting.
REFERENCES
- 1.Howlader N, Altekruse SF, Li CI, et al. US incidence of breast cancer subtypes defined by joint hormone receptor and HER2 status. J Natl Cancer Inst 2014;106(5). doi: 10.1093/jnci/dju055 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Burstein HJ, Lacchetti C, Anderson H, et al. Adjuvant endocrine therapy for women with hormone receptor–positive breast cancer: ASCO clinical practice guideline focused update. Journal of Clinical Oncology. 2019;37(5). doi: 10.1200/JCO.18.01160 [DOI] [PubMed] [Google Scholar]
- 3.Vaz-Luis I, Francis PA, Di Meglio A, Stearns V. Challenges in Adjuvant Therapy for Premenopausal Women Diagnosed With Luminal Breast Cancers. American Society of Clinical Oncology Educational Book. 2021;(41). doi: 10.1200/edbk_320595 [DOI] [PubMed] [Google Scholar]
- 4.Nabieva N, Fehm T, Häberle L, et al. Influence of side-effects on early therapy persistence with letrozole in post-menopausal patients with early breast cancer: Results of the prospective EvAluate-TM study. Eur J Cancer. 2018;96. doi: 10.1016/j.ejca.2018.03.020 [DOI] [PubMed] [Google Scholar]
- 5.Van Herk-Sukel MPP, Van De Poll-Franse LV, Voogd AC, Nieuwenhuijzen GAP, Coebergh JWW, Herings RMC. Half of breast cancer patients discontinue tamoxifen and any endocrine treatment before the end of the recommended treatment period of 5 years: A population-based analysis. Breast Cancer Res Treat 2010;122(3). doi: 10.1007/s10549-009-0724-3 [DOI] [PubMed] [Google Scholar]
- 6.Nekhlyudov L, Li L, Ross-Degnan D, Wagner AK. Five-year patterns of adjuvant hormonal therapy use, persistence, and adherence among insured women with early-stage breast cancer. Breast Cancer Res Treat 2011;130(2). doi: 10.1007/s10549-011-1703-z [DOI] [PubMed] [Google Scholar]
- 7.Murphy CC, Bartholomew LK, Carpentier MY, Bluethmann SM, Vernon SW. Adherence to adjuvant hormonal therapy among breast cancer survivors in clinical practice: A systematic review. Breast Cancer Res Treat 2012;134(2). doi: 10.1007/s10549-012-2114-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hershman DL, Shao T, Kushi LH, et al. Early discontinuation and non-adherence to adjuvant hormonal therapy are associated with increased mortality in women with breast cancer. Breast Cancer Res Treat 2011;126(2). doi: 10.1007/s10549-010-1132-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Haque R, Ahmed SA, Fisher A, et al. Effectiveness of aromatase inhibitors and tamoxifen in reducing subsequent breast cancer. Cancer Med 2012;1(3). doi: 10.1002/cam4.37 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wheeler SB, Roberts MC, Bloom D, et al. Oncology providers’ perspectives on endocrine therapy prescribing and management. Patient Prefer Adherence. 2016;10. doi: 10.2147/PPA.S95594 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Yussof I, Mohd Tahir NA, Hatah E, Mohamed Shah N. Factors influencing five-year adherence to adjuvant endocrine therapy in breast cancer patients: A systematic review. Breast. 2022;62. doi: 10.1016/j.breast.2022.01.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Peh KQE, Kwan YH, Goh H, et al. An Adaptable Framework for Factors Contributing to Medication Adherence: Results from a Systematic Review of 102 Conceptual Frameworks. J Gen Intern Med 2021;36(9). doi: 10.1007/s11606-021-06648-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Walker HE, Rosenberg SM, Stanton AL, Petrie KJ, Partridge AH. Perceptions, Attributions, and Emotions Toward Endocrine Therapy in Young Women with Breast Cancer. J Adolesc Young Adult Oncol 2016;5(1). doi: 10.1089/jayao.2015.0051 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Donevant S, Heiney SP, Wineglass C, Schooley B, Singh A, Sheng J. Perceptions of endocrine therapy in African-American breast cancer survivors: Mixed methods study. JMIR Form Res 2021;5(6). doi: 10.2196/23884 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Todd A, Waldron C, McGeagh L, et al. Identifying determinants of adherence to adjuvant endocrine therapy following breast cancer: A systematic review of reviews. Cancer Med Published online January 19, 2024. doi: 10.1002/cam4.6937 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lee JY, Min YH. Relationships between determinants of adjuvant endocrine therapy adherence in breast cancer. BMC Womens Health. 2018;18(1). doi: 10.1186/s12905-018-0522-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Eraso Y Oncologists’ perspectives on adherence/non-adherence to adjuvant endocrine therapy and management strategies in women with breast cancer. Patient Prefer Adherence. 2019;13. doi: 10.2147/PPA.S211939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bright EE, Finkelstein LB, Nealis MS, et al. A Systematic Review and Meta-Analysis of Interventions to Promote Adjuvant Endocrine Therapy Adherence Among Breast Cancer Survivors. Journal of Clinical Oncology. 2023;41(28). doi: 10.1200/JCO.23.00697 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Finitsis DJ, Vose BA, Mahalak JG, Salner AL. Interventions to promote adherence to endocrine therapy among breast cancer survivors: A meta-analysis. Psychooncology. 2019;28(2):255–263. doi: 10.1002/pon.4959 [DOI] [PubMed] [Google Scholar]
- 20.Fishbein JN, Nisotel LE, MacDonald JJ, et al. Mobile application to promote adherence to oral chemotherapy and symptom management: A protocol for design and development. JMIR Res Protoc 2017;6(4). doi: 10.2196/resprot.6198 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Greer JA, Jacobs JM, Pensak N, et al. Randomized trial of a smartphone mobile app to improve symptoms and adherence to oral therapy for cancer. JNCCN Journal of the National Comprehensive Cancer Network. 2020;18(2). doi: 10.6004/jnccn.2019.7371 [DOI] [PubMed] [Google Scholar]
- 22.Jacobs JM, Walsh EA, Rapoport CS, et al. Development and Refinement of a Telehealth Intervention for Symptom Management, Distress, and Adherence to Adjuvant Endocrine Therapy after Breast Cancer. J Clin Psychol Med Settings. 2021;28(3). doi: 10.1007/s10880-020-09750-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Jacobs JM, Post K, Massad K, et al. A telehealth intervention for symptom management, distress, and adherence to adjuvant endocrine therapy: A randomized controlled trial. Cancer. 2022;128(19). doi: 10.1002/cncr.34409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Jacobs JM, Rapoport CS, Horenstein A, et al. Study protocol for a randomised controlled feasibility trial of a virtual intervention (STRIDE) for symptom management, distress and adherence to adjuvant endocrine therapy after breast cancer. BMJ Open. 2021;11(1). doi: 10.1136/bmjopen-2020-041626 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Dabrowski M, Boucher K, Ward JH, et al. Clinical experience with the NCCN distress thermometer in breast cancer patients. JNCCN Journal of the National Comprehensive Cancer Network. 2007;5(1). doi: 10.6004/jnccn.2007.0011 [DOI] [PubMed] [Google Scholar]
- 26.Farmer KC. Methods for measuring and monitoring medication regimen adherence in clinical trials and clinical practice. Clin Ther 1999;21(6). doi: 10.1016/S0149-2918(99)80026-5 [DOI] [PubMed] [Google Scholar]
- 27.Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67(6):361–370. [DOI] [PubMed] [Google Scholar]
- 28.Chan AHY, Horne R, Hankins M, Chisari C. The Medication Adherence Report Scale: A measurement tool for eliciting patients’ reports of nonadherence. Br J Clin Pharmacol 2020;86(7). doi: 10.1111/bcp.14193 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Trask PC, Tellefsen C, Espindle D, Getter C, Hsu MA. Psychometric validation of the cancer therapy satisfaction questionnaire. Value in Health. 2008;11(4). doi: 10.1111/j.1524-4733.2007.00310.x [DOI] [PubMed] [Google Scholar]
- 30.Wang R, Ware JH. Detecting Moderator Effects Using Subgroup Analyses. Prevention Science. 2013;14(2). doi: 10.1007/s11121-011-0221-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Marshall SW. Power for tests of interaction: Effect of raising the Type I error rate. Epidemiologic Perspectives and Innovations. 2007;4. doi: 10.1186/1742-5573-4-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Walsh EA, Post K, Massad K, et al. Identification of patient subgroups who benefit from a behavioral intervention to improve adjuvant endocrine therapy adherence: a randomized-controlled trial. Breast Cancer Res Treat Published online 2024. doi: 10.1007/s10549-023-07228-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.De Geest S, Sabaté E. Adherence to long-term therapies: Evidence for action. European Journal of Cardiovascular Nursing. 2003;2(4). doi: 10.1016/S1474-5151(03)00091-4 [DOI] [PubMed] [Google Scholar]
- 34.Toivonen KI, Williamson TM, Carlson LE, Walker LM, Campbell TS. Potentially modifiable factors associated with adherence to adjuvant endocrine therapy among breast cancer survivors: A systematic review. Cancers (Basel). 2021;13(1). doi: 10.3390/cancers13010107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Bright EE, Petrie KJ, Partridge AH, Stanton AL. Barriers to and facilitative processes of endocrine therapy adherence among women with breast cancer. Breast Cancer Res Treat 2016;158(2). doi: 10.1007/s10549-016-3871-3 [DOI] [PubMed] [Google Scholar]
- 36.Stanton AL, Petrie KJ, Partridge AH. Contributors to nonadherence and nonpersistence with endocrine therapy in breast cancer survivors recruited from an online research registry. Breast Cancer Res Treat 2014;145(2). doi: 10.1007/s10549-014-2961-3 [DOI] [PubMed] [Google Scholar]
- 37.Jacob Arriola KR, Mason TA, Bannon KA, et al. Modifiable risk factors for adherence to adjuvant endocrine therapy among breast cancer patients. Patient Educ Couns 2014;95(1). doi: 10.1016/j.pec.2013.12.019 [DOI] [PubMed] [Google Scholar]
- 38.Arch JJ, Crespi CM, Levin ME, et al. Randomized Controlled Pilot Trial of a Low-Touch Remotely-Delivered Values Intervention to Promote Adherence to Adjuvant Endocrine Therapy Among Breast Cancer Survivors. Annals of Behavioral Medicine. 2022;56(8). doi: 10.1093/abm/kaab118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Fleming L, Agnew S, Peddie N, Crawford M, Dixon D, MacPherson I. The impact of medication side effects on adherence and persistence to hormone therapy in breast cancer survivors: A quantitative systematic review. Breast. 2022;64. doi: 10.1016/j.breast.2022.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Gagné M, Lauzier S, Lemay M, et al. Women with breast cancer’s perceptions of nurse-led telephone-based motivational interviewing consultations to enhance adherence to adjuvant endocrine therapy: a qualitative study. Supportive Care in Cancer. 2022;30(6). doi: 10.1007/s00520-021-06692-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Horne R, Chapman SCE, Parham R, Freemantle N, Forbes A, Cooper V. Understanding patients’ adherence-related Beliefs about Medicines prescribed for long-term conditions: A meta-analytic review of the Necessity-Concerns Framework. PLoS One. 2013;8(12). doi: 10.1371/journal.pone.0080633 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Shah KK, Touchette DR, Marrs JC. Research and scholarly methods: Measuring medication adherence. JACCP Journal of the American College of Clinical Pharmacy. 2023;6(4). doi: 10.1002/jac5.1771 [DOI] [Google Scholar]