Article Tools
Original Contributions
CARE DELIVERY
Article Tools
Nonadherence to Oral Medications for Chronic Conditions in Breast Cancer Survivors
Abstract
Nonadherence to oral endocrine therapy is common among women with breast cancer (BC). Less is known about nonadherence to medications for other chronic conditions among survivors of BC.
We used the MarketScan Database to identify women older than 18 years who had nonmetastatic BC diagnosed between January 1, 2009, and December 31, 2013. Prescriptions were identified for the following six non–cancer-related conditions: hypertension, thyroid disease, hyperlipidemia, gastroesophageal reflux disease, diabetes, and osteoporosis. The study period was defined as 1 year before BC diagnosis (index date) through 1.5 years after the index date, with a 6-month washout period after the index data to control for adherence during the preoperative period and during chemotherapy if necessary. Adherence was defined as a medication possession ratio ≥ 80%. Change in adherence was defined as a 20% decrease in the medication possession ratio from the time before diagnosis compared with after treatment. Factors associated with change in adherence were evaluated in multivariable logistic models.
Among 36,149 patients diagnosed with BC, the average adherence to these medications before BC was 91.4%. However, after BC treatment, adherence decreased to 77.9% (P < .001). Looking at drugs for each condition, nonadherence ranged from 15.6% to 38% (P < .001). Factors associated with an increase in nonadherence included older age, insurance type, number of medications, and comorbid conditions.
Advances in the diagnosis and treatment of breast cancer (BC) have decreased rates of recurrence and improved survival in women with BC.1 Therefore, improving the management of common chronic conditions has been recognized as a high priority for survivors of BC. In a cohort study that used data collected by the Swedish Cancer Registry, 12.6% of 42,646 women with BC had at least one other medical condition. The risk of competing-cause mortality was 1.84 times higher than the risk of BC-specific mortality among these women with at least one other comorbid condition.1 In the United States, two studies2,3 used population-based cancer registry data and found that 24% of patients with BC had coexisting medical conditions, and among patients age ≥ 66 years, 9.8% were reported to have two or more coexisting conditions. Compared with advanced-stage BC, concurrent comorbidities had the greatest prognostic impact on patients with early-stage BC (ESBC).4,5 Previous studies have shown that the number and severity of comorbidities at the time of ESBC diagnosis were more strongly associated with noncancer mortality compared with having comorbidities with advanced-stage BC.6-10 Hence, given the relatively low cancer-specific mortality for ESBC, the management of chronic conditions plays an important role in overall survival.
Comorbid conditions, such as diabetes, hypertension, hyperlipidemia, thyroid disease, gastroesophageal reflux disease (GERD), and osteoporosis, are common among patients with BC. For instance, 14.5% of patients with BC have diabetes, according to a report from the Centers for Disease Control and Prevention,3 and a retrospective study of older survivors of ESBC found that the risk of hypertension was 1.48 times higher compared with women without cancer.11 In the elderly, particularly in postmenopausal patients with BC, management of osteoporosis is emphasized, because hormone therapy increases the potential likelihood of bone loss.12-14 Moreover, increased rates of thyroid disease and GERD have been observed among patients with BC,15-17 although the etiology for coexisting diseases and BC is unclear.18 Hence, it is of clinical importance to examine the management of these non–cancer-related conditions in survivors of BC.
Successful adherence to self-administered treatments is crucial in managing chronic conditions. Although it is known that adherence to cancer-related treatment can be suboptimal in survivors of BC,19 less is known about non–cancer-related medication adherence. In an observational study of 4,216 women diagnosed with ESBC, Calip et al20 found a significant reduction in oral diabetes medication adherence during and after BC treatment. A similar finding was found with statin use.21 However, a case-cohort study done in the Netherlands found that adherence to glucose-lowering drug treatment was unchanged after a BC diagnosis.22 Similarly, a Korean population-based study reported that adherence to antihypertensive drugs was similar in patients with BC compared with the general population.23 The aim of the current study was to evaluate medication adherence before and after a diagnosis of ESBC for six non–cancer-related chronic conditions and to identify factors associated with nonadherence.
A retrospective cohort study was conducted using the Truven Health Analytics MarketScan Database from January 1, 2009, through December 31, 2013. The database contains health care claims data on approximately 49 million employees, spouses, retirees, and their dependents, covered by more than 100 payers in various types of health care coverage, such as privately insured fee-for-service, point-of-service (POS), or capitated health plans. The health plans include health maintenance organizations (HMOs), preferred provider organizations (PPOs), POS plans, and indemnity plans.
The database contains information on demographics, health plan characteristics, and medical claims for outpatient and inpatient services. Diagnoses (in International Classification of Diseases, ninth revision, clinical modification format), procedures (in Current Procedural Terminology, fourth edition, and Healthcare Common Procedure Coding System formats), and pharmacy claims for prescription drugs are available. Pharmacy claims include the National Drug Code and the quantity of each medication dispensed. Because the MarketScan data set is de-identified and uses a confidential enrollee identifier to link patient records across data files over time, the current study was exempt from institutional review board approval.
We identified patients age 18 years or older who received a first diagnosis of BC (International Classification of Diseases, ninth revision, codes 174.x) between January 1, 2009, and December 31, 2013. We used a claims-based staging algorithm24 to identify patients with ESBC that was adapted from previously described algorithms.25-28 Markers of ESBC in the algorithm included, but were not limited to, mastectomy, lumpectomy, breast-conserving surgery, axillary lymph node dissection, and radiation therapy. Medications were categorized into six clusters according to the following chronic conditions: hypertension, hyperlipidemia, thyroid diseases, GERD, diabetes, and osteoporosis (Appendix Table A1, online only). The six chronic condition groups were not mutually exclusive; patients with more than one condition could be included in more than one cohort. Moreover, for each of the six major conditions, patients were required to have at least two prescription claims to estimate the medication adherence. Patients were not counted as new users of a treatment if they filled a prescription for any drug in the same class (or therapeutic equivalent).
The study period was defined as 1 year before BC diagnosis (index date) through 1.5 years after the index date, with a 6-month washout period after the index data to control for adherence during the preoperative period and during chemotherapy if necessary. Patients who had continuous enrollment of pharmacy and medical benefits during the 12 months before and 18 months after ESBC diagnosis were eligible for the study. Patients had to obtain a retail or mail-order prescription drug of interest both before ESBC diagnosis and after ESBC treatment. Patients with BC were excluded if they were diagnosed within the first 3 months of their enrollment date.
Medication adherence was assessed using the medication possession ratio (MPR). MPR was calculated as the ratio of the sum of the days’ supply for all medication fills divided by the number of days between the first fill and the last fill plus the days’ supply of the last fill. The pre– and post–index date MPR for each class of medications was estimated for each patient. Adherence was defined as an MPR ≥ 0.8. We further dichotomized the outcome as whether there was a significant change in medication adherence by comparing the MPR before ESBC diagnosis and after ESBC treatment. A decline in MPR of at least 20% was considered a significant change.
The following demographic characteristics were included as covariates: age, geographic region (Northeast, North Central, South, or West), type of insurance plans (comprehensive, HMO, PPO, POS with and without capitation, consumer-driven health plan, exclusive provider organization, or high-deductible health plan), number of medications at index date (defined as the total number of unique medications with prescription days supplied inclusive of patient’s index date), and the use of chemotherapy or endocrine therapy after ESBC diagnosis. The Charlson comorbidity index (CCI), a weighted index on the basis of the 1-year preindex period and all available diagnosis codes, was used to control for overall health status of the study population. The CCI was categorized as 0, 1, 2, and 3 to identify the number of comorbid conditions.
Demographic and clinical characteristics of the study population were summarized by frequency and percentage. For each chronic disease, comparison of medication adherence before and after ESBC diagnosis was described using the McNemar test. To identify risk factors associated with a decline in MPR after ESBC treatment, both bivariable and multivariable logistic regression analyses were performed. The odds ratio (OR) and 95% CI were reported. Two-sided tests with a P < .05 were used to determine statistical significance. In addition, a sensitivity analysis was performed on a control group of patients before their BC diagnosis. We selected women who had 2 years of BC-free follow-up (ie, before the incident BC diagnosis) to assess their annual adherence pattern for comparison purposes. All statistical analyses were carried out using SAS v.9.4 (SAS Institute, Cary, NC).
A total of 36,149 patients with ESBC were included in the present analysis (Table 1). The majority (52.4%) were age 65 years or older. A large proportion of the patients (42.1%) were on three or more medications at the time of ESBC diagnosis. Almost half of patients (47.6%) had PPO insurance, and approximately one quarter (23.7%) had comprehensive insurance. The majority had a comorbidity index of 0 (72.2%). The most common medications were hypertension drugs (n = 21,120, 58.4%), followed by hyperlipidemia drugs (n = 16,313, 45.1%), thyroid drugs (n = 10,673, 29.5%), GERD drugs (n = 7,326, 20.3%), osteoporosis drugs (n = 3,627, 10.0%), and diabetes drugs (n = 3,424, 9.5%).
|
| Odds Ratio (95% CI) | |||||||
|---|---|---|---|---|---|---|---|
| Characteristic | Total No. (%) | Hypertension Drugs | Cholesterol Drugs | Thyroid Drugs | GERD Drugs | Osteoporosis Drugs | DM Drugs |
| Age, years | |||||||
| 18-54 | 4,905 (13.6) | Ref | Ref | Ref | Ref | Ref | Ref |
| 55-64 | 12,317 (34.1) | 1.11 (1.00 to 1.22) | 0.98 (0.87 to 1.11) | 1.13 (1.00 to 1.27) | 1.19 (1.02 to 1.40) | 1.14 (0.86 to 1.51) | 1.15 (0.91 to 1.46) |
| ≥ 65 | 18,927 (52.4) | 1.29 (1.17 to 1.41) | 1.14 (1.02 to 1.29) | 1.24 (1.10 to 1.39) | 1.38 (1.19 to 1.60) | 1.30 (0.99 to 1.70) | 1.34 (1.07 to 1.68) |
| Region | |||||||
| Northeast | 7,205 (19.9) | Ref | Ref | Ref | Ref | Ref | Ref |
| North Central | 10,197 (28.2) | 1.07 (0.98 to 1.15) | 1.07 (0.98 to 1.17) | 1.01 (0.90 to 1.13) | 0.97 (0.85 to 1.12) | 0.97 (0.80 to 1.18) | 1.09 (0.89 to 1.34) |
| South | 12,076 (33.4) | 1.02 (0.94 to 1.10) | 0.99 (0.90 to 1.08) | 0.94 (0.84 to 1.05) | 0.87 (0.77 to 0.99) | 0.93 (0.76 to 1.13) | 1.03 (0.85 to 1.26) |
| West | 6,671 (18.5) | 1.09 (1.00 to 1.19) | 0.99 (0.89 to 1.09) | 1.02 (0.91 to 1.16) | 0.83 (0.72 to 0.97) | 0.93 (0.75 to 1.14) | 0.91 (0.73 to 1.14) |
| No. of medications at index date | |||||||
| 1 | 8,584 (23.7) | Ref | Ref | Ref | Ref | Ref | Ref |
| 2 | 12,342 (34.1) | 1.10 (1.01 to 1.19) | 0.95 (0.85 to 1.06) | 0.95 (0.84 to 1.07) | 0.94 (0.79 to 1.11) | 1.13 (0.93 to 1.38) | 1.14 (0.77 to 1.68) |
| ≥ 3 | 15,223 (42.1) | 1.15 (1.07 to 1.24) | 0.96 (0.86 to 1.06) | 1.09 (0.98 to 1.22) | 1.00 (0.86 to 1.16) | 1.20 (1.00 to 1.44) | 1.24 (0.86 to 1.78) |
| CCI | |||||||
| 0 | 26,110 (72.2) | Ref | Ref | Ref | Ref | Ref | Ref |
| 1 | 7,130 (19.7) | 1.11 (1.04 to 1.18) | 1.02 (0.95 to 1.10) | 1.04 (0.94 to 1.16) | 1.10 (0.98 to 1.23) | 1.12 (0.92 to 1.35) | 1.13 (0.95 to 1.35) |
| 2 | 1,928 (5.3) | 1.21 (1.08 to 1.34) | 1.11 (0.98 to 1.26) | 1.22 (1.02 to 1.46) | 1.24 (1.03 to 1.50) | 1.20 (0.83 to 1.73) | 1.18 (0.93 to 1.50) |
| > 2 | 981 (2.7) | 1.35 (1.17 to 1.56) | 1.25 (1.06 to 1.48) | 1.42 (1.13 to 1.80) | 0.95 (0.74 to 1.21) | 0.86 (0.52 to 1.43) | 1.34 (0.94 to 1.89) |
| Type of Insurance | |||||||
| Comprehensive | 8,572 (23.7) | Ref | Ref | Ref | Ref | Ref | Ref |
| HMO | 5,294 (14.6) | 0.88 (0.81 to 0.96) | 0.77 (0.69 to 0.85) | 0.75 (0.66 to 0.86) | 0.71 (0.60 to 0.83) | 0.89 (0.72 to 1.09) | 0.68 (0.55 to 0.85) |
| PPO | 17,198 (47.6) | 0.85 (0.80 to 0.91) | 0.88 (0.82 to 0.95) | 0.86 (0.78 to 0.95) | 0.91 (0.81 to 1.03) | 0.87 (0.74 to 1.02) | 0.84 (0.71 to 1.00) |
| Other | 5,085 (14.1) | 0.83 (0.75 to 0.91) | 0.85 (0.76 to 0.95) | 0.77 (0.68 to 0.88) | 0.84 (0.72 to 0.98) | 1.02 (0.82 to 1.28) | 0.87 (0.69 to 1.10) |
| On endocrine therapy | 5,256 (14.5) | 1.03 (0.95 to 1.11) | 1.08 (0.98 to 1.18) | 0.91 (0.82 to 1.02) | 0.95 (0.83 to 1.09) | 1.14 (0.90 to 1.43) | 0.97 (0.81 to 1.15) |
| On chemotherapy | 5,183 (14.3) | 1.03 (0.95 to 1.11) | 1.08 (0.98 to 1.18) | 0.90 (0.81 to 1.01) | 0.95 (0.83 to 1.08) | 1.14 (0.90 to 1.43) | 0.98 (0.82 to 1.18) |
Abbreviations: CCI, Charlson comorbidity index; DM, diabetes mellitus; GERD, gastroesophageal drug; HMO, health maintenance organization; PPO, preferred provider organization; Ref, reference.
The average adherence to noncancer medications in the current study before ESBC diagnosis was 91.4%, which was significantly higher compared with the average adherence to the noncancer medications after ESBC treatment (77.9%; P < .001). The largest decline in adherence was observed for the osteoporosis medications, with a decline from 83.5% to 45.5% (P < .001), followed by diabetes medications (decline from 79.9% to 53.1%; P < .001), hyperlipidemia medications (decline from 83.2% to 57.1%; P < .001), GERD medications (decline from 79.8% to 55.0%; P < .001), thyroid medications (decline from 89.7% to 69.1%; P < .001), and hypertension medications (decline from 89.6% to 74.0%; P < .001; Fig 1). Among the control group of women evaluated for 2 years before the BC diagnosis, the mean MPR by each chronic condition ranged from 0.84 to 1.00. There was no difference in mean MPR between the periods from 24 to 12 months before diagnosis and 12 to 1 month before diagnosis.
FIG 1. Comparison of adherence (medication possession ratio > 80%) to noncancer medications before and after treatment in patients with early-stage breast cancer (ESBC) in the MarketScan database, 2009 to 2013. P < .001. DM, diabetes mellitus; GERD, gastroesophageal drug.
Table 1 lists the results of bivariable logistic regression models assessing change in MPR from baseline. Nonadherence was significantly higher among patients age ≥ 65 years (hypertension drugs: OR, 1.29; 95% CI, 1.17 to 1.41; thyroid drugs: OR, 1.24; 95% CI, 1.10 to 1.39; cholesterol drugs: OR, 1.14; 95% CI, 1.02 to 1.29; GERD drugs: OR, 1.38; 95% CI, 1.19 to 1.60; diabetes drugs: OR, 1.34; 95% CI, 1.07 to 1.68) compared with patients age 18 through 54 for all noncancer medications other than osteoporosis drugs. For hypertension drugs, patients on two medications (OR, 1.10; 95% CI, 1.01 to 1.19) and three or more medications (OR, 1.15; 95% CI, 1.07 to 1.24) were more likely to have a decrease in MPR than patients who were on one medication. CCI was a statistically significant factor associated with decreased MPR, and higher CCI predicted nonadherence compared with a CCI of 0. However, insurance type was inversely associated with nonadherence. Compared with patients with a comprehensive plan, a reduced MPR was less likely in those who had an HMO (Table 1).
The results from the multivariable logistic regression models are listed in Table 2. Being older remained a significant predictor of nonadherence for hypertension drugs (≥ 65 v 18 to 54 years: OR, 1.21; 95% CI, 1.09 to 1.34) and GERD drugs (≥ 65 and 55 to 64 v 18 to 54 years: OR, 1.39; 95% CI, 1.18 to 1.63; and OR, 1.22; 95% CI, 1.04 to 1.42, respectively). A higher number of comorbid conditions increased the likelihood of MPR decrease for hypertension drugs (CCI of 1: OR, 1.08; 95% CI, 1.01 to 1.16; CCI of 2: OR, 1.16; 95% CI, 1.04 to 1.29; CCI > 2: OR, 1.28; 95% CI, 1.11 to 1.48) as well as thyroid drugs (CCI > 2: OR, 1.34; 95% CI, 1.06 to 1.70). For all medications except osteoporosis drugs, patients with HMO, PPO, and other insurance types were less likely to experience a decrease in MPR than those with comprehensive insurance.
|
| Characteristic | Odds Ratio (95% CI) | |||||
|---|---|---|---|---|---|---|
| Hypertension Drugs | Cholesterol Drugs | Thyroid Drugs | GERD Drugs | Osteoporosis Drugs | DM Drugs | |
| Age, years | ||||||
| 18-54 | Ref | Ref | Ref | Ref | Ref | Ref |
| 55-64 | 1.09 (0.99 to 1.20) | 0.99 (0.87 to 1.12) | 1.11 (0.98 to 1.25) | 1.22 (1.04 to 1.42) | 1.11 (0.83 to 1.47) | 1.12 (0.88 to 1.41) |
| ≥ 65 | 1.21 (1.09 to 1.34) | 1.12 (0.99 to 1.27) | 1.12 (0.98 to 1.27) | 1.39 (1.18 to 1.63) | 1.24 (0.93 to 1.66) | 1.25 (0.98 to 1.59) |
| Region | ||||||
| Northeast | Ref | Ref | Ref | Ref | Ref | Ref |
| North Central | 1.05 (0.97 to 1.14) | 1.04 (0.95 to 1.14) | 0.97 (0.87 to 1.09) | 0.96 (0.84 to 1.11) | 0.98 (0.80 to 1.20) | 1.08 (0.88 to 1.33) |
| South | 1.04 (0.96 to 1.13) | 0.99 (0.90 to 1.08) | 0.93 (0.83 to 1.04) | 0.86 (0.76 to 0.98) | 0.94 (0.77 to 1.15) | 1.06 (0.86 to 1.29) |
| West | 1.11 (1.02 to 1.22) | 1.03 (0.93 to 1.14) | 1.07 (0.94 to 1.21) | 0.87 (0.74 to 1.01) | 0.95 (0.76 to 1.17) | 0.99 (0.79 to 1.25) |
| No. of medications at index date | ||||||
| 1 | Ref | Ref | Ref | Ref | Ref | Ref |
| 2 | 1.06 (0.97 to 1.15) | 0.92 (0.82 to 1.03) | 0.92 (0.81 to 1.04) | 0.89 (0.75 to 1.05) | 1.10 (0.90 to 1.34) | 1.11 (0.75 to 1.63) |
| ≥ 3 | 1.07 (0.99 to 1.16) | 0.90 (0.81 to 1.00) | 1.02 (0.90 to 1.15) | 0.90 (0.76 to 1.05) | 1.13 (0.93 to 1.37) | 1.17 (0.81 to 1.69) |
| CCI | ||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref |
| 1 | 1.08 (1.01 to 1.16) | 1.02 (0.95 to 1.11) | 1.01 (0.91 to 1.12) | 1.07 (0.95 to 1.21) | 1.06 (0.87 to 1.30) | 1.13 (0.95 to 1.35) |
| 2 | 1.16 (1.04 to 1.29) | 1.11 (0.97 to 1.26) | 1.16 (0.97 to 1.40) | 1.21 (1.00 to 1.47) | 1.13 (0.78 to 1.64) | 1.18 (0.93 to 1.50) |
| > 2 | 1.28 (1.11 to 1.48) | 1.25 (1.05 to 1.48) | 1.34 (1.06 to 1.70) | 0.93 (0.72 to 1.19) | 0.80 (0.48 to 1.34) | 1.34 (0.94 to 1.91) |
| Type of insurance | ||||||
| Comprehensive | Ref | Ref | Ref | Ref | Ref | Ref |
| HMO | 0.89 (0.81 to 0.98) | 0.79 (0.70 to 0.88) | 0.75 (0.65 to 0.86) | 0.77 (0.64 to 0.92) | 0.95 (0.76 to 1.19) | 0.73 (0.57 to 0.93) |
| PPO | 0.91 (0.84 to 0.97) | 0.93 (0.86 to 1.01) | 0.89 (0.80 to 0.99) | 1.00 (0.88 to 1.14) | 0.93 (0.79 to 1.11) | 0.91 (0.75 to 1.09) |
| Other | 0.92 (0.83 to 1.02) | 0.93 (0.82 to 1.04) | 0.82 (0.71 to 0.95) | 0.95 (0.80 to 1.13) | 1.14 (0.90 to 1.45) | 0.98 (0.76 to 1.28) |
| On endocrine therapy | 0.75 (0.39 to 1.45) | 1.21 (0.52 to 2.81) | 1.48 (0.73 to 3.01) | 2.24 (0.53 to 9.39) | 1.12 (0.16 to 8.02) | 0.89 (0.76 to 1.04) |
| On chemotherapy | 1.42 (0.74 to 2.75) | 0.92 (0.39 to 2.14) | 0.63 (0.31 to 1.28) | 0.45 (0.11 to 1.88) | 1.03 (0.14 to 7.49) | 1.04 (0.86 to 1.25) |
Abbreviations: CCI, Charlson comorbidity index; DM, diabetes mellitus; GERD, gastroesophageal drug; HMO, health maintenance organization; PPO, preferred provider organization; Ref, reference.
We found that adherence to noncancer medications declined after a diagnosis of ESBC. The magnitude of the MPR change varied according to the chronic condition but was greatest for patients on medications to treat osteoporosis. Factors such as older age, insurance type, higher number of medications, and more comorbid conditions were all associated with a decline in noncancer medication adherence after BC treatment.
Consistent with our findings, a Korean study demonstrated that older patients were less likely than younger patients to adhere to their antihypertensive medication after a diagnosis of cancer.23 A systematic review conducted by Hall et al29 evaluated reasons for medication nonadherence in patients with hematologic cancer and found that older age was a contributing factor. This may be a result of poorer physical condition, less family and social support, or decreased motivation compared with younger patients.30,31 In addition, elderly patients may have difficulty remembering to take medications as a result of declines in cognitive function or difficulties obtaining transportation to refill drugs. However, the literature has not found older age to be a consistent predictor of medication nonadherence. A literature review by Balkrishnan32 concluded that age was not an important predictor of medication adherence in most studies.
Our study found that the number of medications taken at the time of BC diagnosis was associated with an increased odds of nonadherence after treatment. Consistent with our findings, a Korean population-based cohort study demonstrated that concurrent drug therapy was a also a predictor of nonadherence.23 It is possible that women are more likely to prioritize the BC-related treatments over medications for other chronic conditions. It is also possible that the out-of-pocket costs of multiple medications are a barrier to adherence.
A review examining polypharmacy in patients with advanced cancer has suggested that medication adherence is reduced with increasing numbers of prescription drugs.33 The route of drug administration, frequency of doses, and other administration requirements of noncancer drugs can be affected by BC treatment. A literature review of 21 studies reported that combining cancer and noncancer drug regimens may increase adverse drug reactions, drug-drug interactions, and hospitalization.34 daCosta DiBonaventura et al30 conducted a cross-sectional, Internet-based survey of 181 women with metastatic BC and reported that adverse events were among the most frequent reasons for medication nonadherence. Moreover, the adverse effects of chemotherapy, such as nausea, may reduce medication adherence; we found no association between medication nonadherence and treatment in the current study. Similar findings have been reported.22
BC treatment is a contributor to high out-of-pocket health care costs in some patients. Our findings demonstrate that insurance type is an important predictor of adherence decline after treatment, which may be related to different out-of-pocket costs associated with each health insurance type. Although no studies have looked at the impact of insurance on noncancer drugs in patients with BC, studies by our group and others have shown that insurance is associated with adherence to hormonal therapy in survivors of BC.35-37 For example, a retrospective cohort study using a claims database suggested that higher copayments were associated with nonadherence to aromatase inhibitors.35,38 A study that used cancer registry data showed that socioeconomically disadvantaged women had poor adherence to adjuvant endocrine therapy.36 In the general population, several retrospective studies have found that patients with comprehensive insurance packages are more likely to be adherent to medications.39-42 The association between copayment status and adherence varies by socioeconomic status, with middle- and low-income patients with a high copayment having lower adherence compared with women with higher income with a high copayment.43
Our study has several limitations. One of the limitations is the absence of a control group. To deal with the limitation, we looked at the adherence patterns for a group of women who had 2 years of BC-free follow-up (ie, before the incident BC diagnosis). Adherence did not change appreciably during that time. Second, our analysis was restricted to patients who had continuous enrollment. It is possible that patients who remained in the cohort were more likely to have stable employment and, therefore, have a better financial situation, which may lead to an overestimate of the adherence. Third, we only assessed the adherence on the basis of whether or not the prescription was filled. However, we do not know whether patients were actually adherent once they got the prescription. In addition, there may have been uncontrolled confounders that were associated with medication adherence in survivors of ESBC.
Our study has several strengths. We evaluated nonadherence across a wide spectrum of chronic medications that are frequently used among patients with ESBC. In addition, the sample size was large and representative of the US population, which enhances the generalizability of our findings. Importantly, continuous enrollment of the study population in the data set allows us to compare adherence before and after ESBC diagnosis. In addition to patients’ demographic and clinical characteristics, our claims database also had comprehensive information on patients’ insurance, which is an important predictor of nonadherence.
Given the relatively low risk of cancer-specific mortality in patients with ESBC, our findings suggest that adherence to medications for the treatment of chronic conditions should be addressed by health care providers. In a Cochrane review of interventions to improve medication adherence, there was a suggestion that solutions include the following: involving allied health care providers such as nurses and pharmacists, encouraging social support from family or peers, and using technology-based strategies such as Internet-based home monitoring, short text messages, and reminders sent to patients’ mobile phone.44 Interventions to assist with medication costs may also improve adherence. Research is needed to address the negative effects of polypharmacy and the toxicity of drug-drug interactions. In conclusion, increased attention to the management of chronic conditions is needed to improve survival outcomes in survivors of BC.
Acknowledgment
Supported by grants from the Department of Defense (Grant No. BC041320; A.I.N.), Breast Cancer Research Foundation, and the American Society of Clinical Oncology (D.L.H.). The funders did not participate in the conduct of this study.
Conception and design: Jingyan Yang, Dawn L. Hershman
Financial support: Alfred I. Neugut, Dawn L. Hershman
Administrative support: Alfred I. Neugut
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml.
No relationship to disclose
Stock or Other Ownership: Stemline Therapeutics
Consulting or Advisory Role: Pfizer, TEVA Pharmaceuticals Industries, Otsuka, United Biosource, EHE
No relationship to disclose
No relationship to disclose
No relationship to disclose
| 1. | A Berglund, A Wigertz, J Adolfsson, etal: Impact of comorbidity on management and mortality in women diagnosed with breast cancer Breast Cancer Res Treat 135:281–289,2012 Crossref, Medline, Google Scholar |
| 2. | H Cho, AB Mariotto, BS Mann, etal: Assessing non-cancer-related health status of US cancer patients: Other-cause survival and comorbidity prevalence Am J Epidemiol 178:339–349,2013 Crossref, Medline, Google Scholar |
| 3. | BK Edwards, AM Noone, AB Mariotto, etal: Annual report to the nation on the status of cancer, 1975-2010, featuring prevalence of comorbidity and impact on survival among persons with lung, colorectal, breast, or prostate cancer Cancer 120:1290–1314,2014 Crossref, Medline, Google Scholar |
| 4. | WL Read, RM Tierney, NC Page, etal: Differential prognostic impact of comorbidity J Clin Oncol 22:3099–3103,2004 Link, Google Scholar |
| 5. | American Cancer Society: Breast Cancer Facts & Figures 2013-2014. http://www.cancer.org/acs/groups/content/@research/documents/document/acspc-042725.pdf Google Scholar |
| 6. | LH Land, SO Dalton, MB Jensen, etal: Influence of comorbidity on the effect of adjuvant treatment and age in patients with early-stage breast cancer Br J Cancer 107:1901–1907,2012 Crossref, Medline, Google Scholar |
| 7. | LH Land, SO Dalton, MB Jensen, etal: Impact of comorbidity on mortality: A cohort study of 62,591 Danish women diagnosed with early breast cancer, 1990-2008 Breast Cancer Res Treat 131:1013–1020,2012 Crossref, Medline, Google Scholar |
| 8. | LH Land, SO Dalton, TL Jørgensen, etal: Comorbidity and survival after early breast cancer: A review Crit Rev Oncol Hematol 81:196–205,2012 Crossref, Medline, Google Scholar |
| 9. | LK Mell, JH Jeong, MA Nichols, etal: Predictors of competing mortality in early breast cancer Cancer 116:5365–5373,2010 Crossref, Medline, Google Scholar |
| 10. | S Yasmeen, RT Chlebowski, G Xing, etal: Severity of comorbid conditions and early-stage breast cancer therapy: Linked SEER-Medicare data from 1993 to 2005 Cancer Med 2:526–536,2013 Crossref, Medline, Google Scholar |
| 11. | R Haque, M Prout, AM Geiger, etal: Comorbidities and cardiovascular disease risk in older breast cancer survivors Am J Manag Care 20:86–92,2014 Medline, Google Scholar |
| 12. | K Gibson, CL O’Bryant: Screening and management of osteoporosis in breast cancer patients on aromatase inhibitors J Oncol Pharm Pract 14:139–145,2008 Crossref, Medline, Google Scholar |
| 13. | JF Carney, J Davis: Emerging bone health issues in women with breast cancer in Hawai’i Hawaii Med J 66:164–166,2007 Medline, Google Scholar |
| 14. | R Roberts, M Miller, M O’Callaghan, etal: Bone health management of Australian breast cancer survivors receiving hormonal therapy Intern Med J 45:1182–1185,2015 Crossref, Medline, Google Scholar |
| 15. | C Giani, P Fierabracci, R Bonacci, etal: Relationship between breast cancer and thyroid disease: Relevance of autoimmune thyroid disorders in breast malignancy J Clin Endocrinol Metab 81:990–994,1996 Medline, Google Scholar |
| 16. | HB El-Serag, S Sweet, CC Winchester, etal: Update on the epidemiology of gastro-oesophageal reflux disease: A systematic review Gut 63:871–880,2014 Crossref, Medline, Google Scholar |
| 17. | M Kalder, P Hadji: Breast cancer and osteoporosis: Management of cancer treatment-induced bone loss in postmenopausal women with breast cancer Breast Care (Basel) 9:312–317,2014 Crossref, Medline, Google Scholar |
| 18. | Garner CN, Ganetzky R, Brainard J, et al.: Increased prevalence of breast cancer among patients with thyroid and parathyroid disease. Surgery 142:806-813, 2007 Google Scholar |
| 19. | M Søgaard, RW Thomsen, KS Bossen, etal: The impact of comorbidity on cancer survival: A review Clin Epidemiol 5:3–29,2013 suppl 1 Crossref, Medline, Google Scholar |
| 20. | GS Calip, RA Hubbard, A Stergachis, etal: Adherence to oral diabetes medications and glycemic control during and following breast cancer treatment Pharmacoepidemiol Drug Saf 24:75–85,2015 Crossref, Medline, Google Scholar |
| 21. | GS Calip, DM Boudreau, ET Loggers: Changes in adherence to statins and subsequent lipid profiles during and following breast cancer treatment Breast Cancer Res Treat 138:225–233,2013 Crossref, Medline, Google Scholar |
| 22. | MM Zanders, HR Haak, MP van Herk-Sukel, etal: Impact of cancer on adherence to glucose-lowering drug treatment in individuals with diabetes Diabetologia 58:951–960,2015 Crossref, Medline, Google Scholar |
| 23. | DW Shin, JH Park, JH Park, etal: Antihypertensive medication adherence in cancer survivors and its affecting factors: Results of a Korean population-based study Support Care Cancer 19:211–220,2010 Crossref, Medline, Google Scholar |
| 24. | SH Giordano, YL Lin, YF Kuo, etal: Decline in the use of anthracyclines for breast cancer J Clin Oncol 30:2232–2239,2012 Link, Google Scholar |
| 25. | JL Freeman, D Zhang, DH Freeman, etal: An approach to identifying incident breast cancer cases using Medicare claims data J Clin Epidemiol 53:605–614,2000 Crossref, Medline, Google Scholar |
| 26. | AB Nattinger, PW Laud, R Bajorunaite, etal: An algorithm for the use of Medicare claims data to identify women with incident breast cancer Health Serv Res 39:1733–1749,2004 Crossref, Medline, Google Scholar |
| 27. | HT Gold, HT Do: Evaluation of three algorithms to identify incident breast cancer in Medicare claims data Health Serv Res 42:2056–2069,2007 Crossref, Medline, Google Scholar |
| 28. | I Baldi, P Vicari, D Di Cuonzo, etal: A high positive predictive value algorithm using hospital administrative data identified incident cancer cases J Clin Epidemiol 61:373–379,2008 Crossref, Medline, Google Scholar |
| 29. | Hall AE, Paul C, Bryant J, et al: To adhere or not to adhere: Rates and reasons of medication adherence in hematological cancer patients. Crit Rev Oncol Hematol 97:247-262, 2016 Google Scholar |
| 30. | M daCosta DiBonaventura, R Copher, E Basurto, etal: Patient preferences and treatment adherence among women diagnosed with metastatic breast cancer Am Health Drug Benefits 7:386–396,2014 Medline, Google Scholar |
| 31. | L Noens, MA van Lierde, R De Bock, etal: Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: The ADAGIO study Blood 113:5401–5411,2009 Crossref, Medline, Google Scholar |
| 32. | R Balkrishnan: Predictors of medication adherence in the elderly Clin Ther 20:764–771,1998 Crossref, Medline, Google Scholar |
| 33. | J Tam-McDevitt: Polypharmacy, aging, and cancer Oncology (Williston Park) 22:1052–1055,2008 Medline, Google Scholar |
| 34. | TW LeBlanc, MJ McNeil, AH Kamal, etal: Polypharmacy in patients with advanced cancer and the role of medication discontinuation Lancet Oncol 16:e333–e341,2015 Crossref, Medline, Google Scholar |
| 35. | AI Neugut, M Subar, ET Wilde, etal: Association between prescription co-payment amount and compliance with adjuvant hormonal therapy in women with early-stage breast cancer J Clin Oncol 29:2534–2542,2011 Link, Google Scholar |
| 36. | G Kimmick, R Anderson, F Camacho, etal: Adjuvant hormonal therapy use among insured, low-income women with breast cancer J Clin Oncol 27:3445–3451,2009 Link, Google Scholar |
| 37. | L Atkins, L Fallowfield: Intentional and non-intentional non-adherence to medication amongst breast cancer patients Eur J Cancer 42:2271–2276,2006 Crossref, Medline, Google Scholar |
| 38. | DL Hershman, J Tsui, J Meyer, etal: The change from brand-name to generic aromatase inhibitors and hormone therapy adherence for early-stage breast cancer J Natl Cancer Inst 106:dju319,2014 Crossref, Medline, Google Scholar |
| 39. | DP Goldman, GF Joyce, Y Zheng: Prescription drug cost sharing: Associations with medication and medical utilization and spending and health JAMA 298:61–69,2007 Crossref, Medline, Google Scholar |
| 40. | RE Johnson, MJ Goodman, MC Hornbrook, etal: The effect of increased prescription drug cost-sharing on medical care utilization and expenses of elderly health maintenance organization members Med Care 35:1119–1131,1997 Crossref, Medline, Google Scholar |
| 41. | A Babazono, T Tsuda, E Yamamoto, etal: Effects of an increase in patient copayments on medical service demands of the insured in Japan Int J Technol Assess Health Care 19:465–475,2003 Crossref, Medline, Google Scholar |
| 42. | SM Curkendall, N Thomas, KF Bell, etal: Predictors of medication adherence in patients with type 2 diabetes mellitus Curr Med Res Opin 29:1275–1286,2013 Crossref, Medline, Google Scholar |
| 43. | R Kazerooni, M Bounthavong, JH Watanabe: Association of copayment and statin adherence stratified by socioeconomic status Ann Pharmacother 47:1463–1470,2013 Crossref, Medline, Google Scholar |
| 44. | R Nieuwlaat, N Wilczynski, T Navarro, etal: Interventions for enhancing medication adherence Cochrane Database Syst Rev 11:CD000011,2014 Medline, Google Scholar |
|
| Condition | Medication |
|---|---|
| Hypertension | Diuretics, calcium channel blockers, beta blockers, ACE inhibitors, angiotensin II receptor blockers, alpha blockers, alpha-2 receptor agonist, peripheral adrenergic inhibitors |
| Hyperlipidemia | Antihyperlipidemic combinations, bile acid sequestrants, cholesterol absorption inhibitors, fibric acid derivatives, miscellaneous antihyperlipidemic agents, PC SK9 inhibitors, statins |
| Thyroid diseases | Hicon, iosat, methimazole, northyx, pima syrup, potassium iodide, propylthiouracil, PTU, sodium iodide I-131, SSKI, tapazole |
| Diabetes | Insulin, alpha-glucosidase inhibitors, amylin analogs, antidiabetic combinations, incretin mimetics, meglitinides, non-sulfonylureas, SGL T-2 inhibitors, sulfonylureas, thiazolidinediones |
| GERD | Antacids, H2 blockers, proton pump inhibitors |
| Osteoporosis | Bisphosphonates, estrogens, SERMs, calcitonin, monoclonal antibodies |
Abbreviations: ACE, angiotensin-converting enzyme; GERD, gastroesophageal reflux disease; PC SK9, proprotein convertase subtilisin/kexin type 9; PTU, propylthioruacil; SERM, selective estrogen receptor modulator; SGLT-2, sodium-glucose co-transporter 2; SSKI, saturated solution of potassium iodide.
