Dissertation
The influence of preexisting mental illness, associated medication regimen factors, and access to care on breast cancer endocrine therapy adherence
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Spring 2020
DOI: 10.17077/etd.005283
Abstract
Oral endocrine therapy is a guideline-recommended treatment for estrogen receptor positive (ER+) breast cancer which reduces recurrence and mortality risk. Despite the serious nature of breast cancer, many patients are non-adherent to treatment. To improve cancer outcomes, efforts to identify populations at-risk and factors associated with non-adherence are warranted. This dissertation aimed to: (1) examine the prevalence of preexisting mental illness and its influence on breast cancer endocrine therapy adherence, (2) examine whether mental illness modifies the effects of adherence risk factors including prior medication (regimen complexity, number of medications and pharmacy visits) and financial burden, and (3) determine if primary care and mental health professional access and area socioeconomic factors are associated with endocrine therapy non-adherence and whether these factors disproportionately affect patients living with mental illness.
Using the Surveillance, Epidemiology, and End Results cancer database linked to Medicare claims (SEER-Medicare), we selected women aged 68 and older at the time of their stage I-IV ER+ breast cancer diagnosis in 2007 to 2013. Included patients had continuous enrollment in fee-for-service Medicare Parts A and B for three or more years prior to cancer and 18 months or more after, and Part D for four months or more before and 18 or more after. Mental illness diagnoses were obtained from Medicare claims 36 months before cancer. Medication regimen factors in the four months before cancer included the count of unique medications used, pharmacy visits, and an administration complexity score subscale of the Medication Regimen Complexity Index. Access to care was measured using primary care and mental Health Professional Shortage Area (HPSA) assessments in a patient’s home county at the time of cancer diagnosis. Endocrine therapy outcomes included initiation and discontinuation (Cox regression), and adherence (longitudinal linear regression) as measured by the Proportion of Days Covered.
Prior to breast cancer, 24.9% of patients had mental illness diagnoses, with unipolar depression (11.0%) and anxiety (9.5%) disorders most prevalent. Twenty percent of all women never initiated endocrine therapy and 25.7% of users discontinued. Continuous users for one year or longer had an 84% mean adherence rate. Patients with bipolar and psychotic disorders [HR 0.93 (95% CI 0.87-0.99)] and dementia disorders [HR 0.90 (95% CI 0.83-0.98)] were less likely to initiate. Those with anxiety [HR 1.21 (95% CI 1.09-1.34)] and substance use [HR 1.43 (95% CI 1.26-1.64)] disorders were more likely to discontinue. Those with substance use disorders had 2.3% lower average adherence rates.
Patients who more frequently used medication and visited pharmacies in the four months prior to cancer (“4+ Meds & 2+ Visits”) were more likely to initiate [HR 1.47 (95% CI 1.33-1.63)] and adhere [+6.0% (95% CI 4.3-7.6)] and less likely to discontinue [HR 0.48 (95% CI 0.39-0.59)], relative to those who did not. In a focused analysis of patients with bipolar and psychotic disorders – who are less likely to initiate endocrine therapy – these protective factors were more common (they used more medications and visited pharmacies more often), but were nevertheless not prognostic of better initiation of maintenance of endocrine therapy in this subgroup.
After adjustment for access to oncology care and patient characteristics, patients living in counties with the lowest primary care access were less likely to initiate endocrine therapy [highest shortage vs least HR 0.92 (95% CI 0.86-0.97)] but there was no association with daily adherence or discontinuation. Mental health shortage level was not associated with initiation, daily adherence, or discontinuation, overall or within subgroups with particular mental illness. However, patients who lived in counties not evaluated for mental health shortage were less likely to discontinue, including those with bipolar and psychotic disorders [discontinuation HR 0.35 (95% CI 0.17-0.73)], substance use [HR 0.48 (95% CI 0.24-0.95)], and anxiety disorders [HR 0.56 (95% CI 0.35-0.90)]. Counties not evaluated for shortage had significantly more oncologists per capita, a lower proportion living below the federal poverty level, and higher median income, all suggestive of generally high access to health care.
Suboptimal adherence and preexisting mental illness are common among Medicare-enrolled women, with certain illnesses associated with endocrine therapy non-initiation, discontinuation, and non-adherence. Prior experience taking medication and visiting pharmacies appeared protective for patients in general, but patients with bipolar and psychotic disorders (the mental illness we examined more closely) did not receive a significant benefit. Breast cancer patients living in counties containing high shortage primary care HPSAs were less likely to initiate, but women with bipolar and psychotic disorders, anxiety, and substance use disorders had better outcomes if living in higher access areas not evaluated for mental HPSA designations. These findings suggest that patient-level and policy interventions designed to improve breast cancer endocrine therapy adherence for average breast cancer patients may not be as effective for patients with mental illness who may benefit from different approaches emphasizing the coordination of mental health, primary care and cancer care.
Details
- Title: Subtitle
- The influence of preexisting mental illness, associated medication regimen factors, and access to care on breast cancer endocrine therapy adherence
- Creators
- Cole Haskins
- Contributors
- Elizabeth Chrischilles (Advisor)Ryan Carnahan (Committee Member)Jess Fiedorowicz (Committee Member)Bradley McDowell (Committee Member)Brian Smith (Committee Member)Robert Wallace (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Epidemiology
- Date degree season
- Spring 2020
- DOI
- 10.17077/etd.005283
- Publisher
- University of Iowa
- Number of pages
- xv, 159 pages
- Copyright
- Copyright 2020 Cole Haskins
- Grants
- Language
- English
- Description illustrations
- color illustrations
- Description bibliographic
- Includes bibliographical references (pages 151-159).
- Public Abstract (ETD)
Endocrine therapies are used to treat the most common type of breast cancer and can prolong life, but many patients do not consistently use these long-term oral medications (non-adherence). To identify patients less likely to adhere, we used a nationally representative cohort of Medicare enrolled women aged 68 or older diagnosed with breast cancer and examined their medical claims to identify preexisting mental illness and effects on non-adherence, and additional risk factors including: characteristics of their prior medication regimens, and primary care and mental health professional shortage areas in patient home counties.
Nearly 25% of patients had diagnosed mental illness in the three years before cancer, with depression (11.0%) and anxiety (9.5%) disorders being common. Eighty percent of patients started endocrine therapy. On average, active users took their medication 84% of the time, but 25.7% prematurely discontinued. Women with bipolar and psychotic disorders, dementia, anxiety, and substance use disorders were less likely to adhere to treatment. Patients with frequent medication use and pharmacy visits before cancer were more likely to adhere, but this was not true for those with bipolar and psychotic disorders. Patients with bipolar and psychotic disorders, anxiety, and substance use disorders living in counties never evaluated for mental health shortage designations (which had more oncologists per capita, fewer people living below the federal poverty level, and higher median incomes) were more likely to continuously use therapy, but patients without mental illness were less likely to start treatment if living in areas with high primary care shortages. Our findings can be used to identify patients at-risk for endocrine therapy non-adherence and develop interventions to improve adherence and therefore breast cancer survival.
- Academic Unit
- Epidemiology
- Record Identifier
- 9983966299102771
Metrics
20 File views/ downloads
58 Record Views