Abstract
Objectives:
To describe the prevalence of, types of, and characteristics associated with self-reporting multiple (≥2) barriers to medication use in older women using chronic cardiovascular and oral hypoglycemic medications.
Methods:
Design: Cross-sectional. Setting: Women’s Health Initiative (WHI) (2005–2010). Participants: Women who were using any chronic medication from three target classes (i.e., antilipemics, antihypertensives, oral hypoglycemics) for at least a month and who had answered questions about barriers to medication use at year 4 (2009) of the study period (N=59,054). Measurements: Common self-reported barriers to medication use, as well as sociodemographic, health characteristic, medication use, and access to care variables were additionally evaluated. Multivariable logistic regression models were used to examine associations between participant characteristics and barriers to medication use.
Results:
Among the participants, 47,846 (81%) reported no barriers, 7105 (12%) reported 1 barrier, and 4103 (6.9%) reported ≥2 barriers to medication use. The most common barriers reported were having concerns about side effects, not liking to take medications, and medications costing too much. Several characteristics were found to be associated with reporting ≥2 barriers in multivariable modeling, including demographic (e.g., lower age, Black race, Hispanic ethnicity) and health/medication (e.g., lower quality of life, lower physical function, higher number of concurrent medications) characteristics.
Conclusion:
Among older women using chronic cardiovascular and oral hypoglycemic medications, approximately 20% reported at least one barrier to medication use, with 7% of women reporting multiple barriers. Pharmacists should prioritize identifying barriers to medication use in older women using chronic medications to improve patient care.
Keywords: medication adherence and persistence, medication safety, geriatrics
BACKGROUND
Medication use is common in older adults aged ≥65 years, with 36% of community-dwelling older adults reporting use of 5 or more medications on a regular basis, frequently resulting in medication non-adherence.1–3 Medication non-adherence can be viewed as two underlying constructs – the extent of (how common) and reasons for (barriers to) medication non-adherence.4 Some of the most common barriers to medication use reported by older adults are regimen complexity, cost, and side effects.5 Clinically, it is important for pharmacists and other healthcare providers to identify the most common barriers to medication use in older adults in order to provide patient-centered care.
Chronic cardiovascular drugs (e.g., antilipemics, antihypertensives) and oral hypoglycemics are some of the most frequently used drugs among older adults.1 As chronic cardiovascular disease and diabetes increase morbidity and mortality, optimal management with medications is important.6 Prior studies on medication use barriers in older adults have been limited by focusing on single disease states (e.g., CHD) and individual barriers (e.g, medication cost), as well as using small sample sizes.7–9 Older adults reporting multiple barriers to, or more concerns toward, medication use have been shown to be at highest risk for medication non-adherence10.
OBJECTIVE
Our objective was to describe the prevalence of, types of, and characteristics associated with self-reporting multiple (≥2) barriers to medication use by older women taking chronic cardiovascular and oral hypoglycemic medications in the Women’s Health Initiative (WHI), a national, population-based study among postmenopausal women.
METHODS
Study Population
This cross-sectional study included community-dwelling postmenopausal women who were enrolled in an Extension Study from 2005–2010 of the WHI (N=115,407). The WHI observational study recruited women between October 1, 1993, and December 31, 1998, at 40 clinical centers in the USA. Women were eligible if they planned to remain in the area where they lived at the time of recruitment and had an estimated survival of at least 3 years. The study methods have been described in detail elsewhere.11 At the closeout of the main WHI study in March 2005, women were invited to participate in an additional 5 years of follow-up in the Extension 1 Study. The consent rate for the Extension Study was 71.3%. Among these women, 97,448 completed the medication inventory at Year 4 (2009), which served as the index date for this analysis. Women were asked to assemble their prescription and over-the-counter medications taken in the past 2 weeks and to write them down on a mailed form. Women were instructed to record medication use information directly from the medication containers. Women were also asked how long they had used each medication. Women who needed assistance with the form could request a phone call to complete the form by interview. We included women who had answered questions about self-reported barriers to medication and who were using any chronic medication from three target classes (i.e., antilipemics, antihypertensives, oral hypoglycemics) for at least one month (N=59,054). Informed consent was obtained, and all protocols were approved by the respective institutional review board at the participating institutions.
Self-Reported Barriers to Medication Use
Self-reported barriers to medication use were collected once during Year 4 of Extension 1 of WHI. Participants were asked, “Have any of the following barriers prevented you from obtaining or taking any medications that have been prescribed for you?” The barriers represented a list of 10 of the most common self-reported barriers to medication use, developed by WHI investigators.5 These items are aligned with one component of Necessity-Concerns Framework – i.e., concerns about taking medication.12 Participants also had the option to respond that they have not experienced any barriers to taking prescription medications.
Covariates
Sociodemographic information included self-reported race and ethnicity, education, and family income. Age was expressed as age at index date. For health characteristics, information closest to the index date was used. Women were asked to rate their quality of life on a scale from 0 (worst) and 10 (best). This self-reported score was then multiplied by 10. Quality of life was defined as low (0–50), moderate (50–69), and high (≥70).13 Activities of Daily Living (ADL) were measured using four activities: feeding, dressing, transferring, and bathing. Dependence in ADLs was operationalized as reporting needing “some help” in any of the four activities. Physical function was assessed using the 10-item Scale from the Rand 36-item Health Survey (SF-36) and operationalized as low, moderate, or high based on tertile.14
The total number of prescription and non-prescription medications that women reported taking was calculated. A dichotomous measure (user/non-user) was created for use of any of the three target classes – antihypertensive, antilipemic, and/or oral hypoglycemic medications – and a count variable (range=1–3) was generated.
Medicare coverage was determined using a Centers for Medicare and Medicaid Services (CMS) enrollment file that WHI receives from CMS. Women were defined having Medicare coverage if they were enrolled in Medicare Part A or B, with or without health maintenance organization (HMO) enrollment. Self-reported living assistance was measured as receiving any special service (such as help with transportation, meals, medicines, or bathing). Women who engage in preventive healthcare may be more likely to have better access to care and, in turn, fewer barriers to medication use. To assess preventive health behavior, we examined self-reported receipt of a mammogram, flexible sigmoidoscopy, or bone density scan in the four years prior to the index date.
Statistical Analysis
Descriptive characteristics were reported by number of barriers (no barriers, 1 barrier, and ≥2 barriers). The mean and standard deviation is presented for continuous variables, and the frequencies and percentages for categorical variables. For continuous variables, F-tests (ANOVA) were used to compare the three groups. For categorical variables, chi-squared tests were used to compare groups. We described the prevalence of individual barriers among those reporting only 1 barrier (for descriptive purposes) and among women reporting ≥2 barriers (for modeling). Because older adults reporting multiple barriers to, or concerns toward, medication use have been shown to be at highest risk for medication non-adherence, multivariable logistic regression models were used to assess the factors associated with reporting multiple (≥2) barriers compared to those reporting no barriers. A single model was used including all sociodemographic, health characteristic, medication use, and access to care variables.2 The covariates were selected from the literature as those potentially associated with having barriers to medication use.2 Adjusted odds ratios and 95% confidence intervals were produced for each characteristic. P-values were considered statistically significant at values <0.05. All statistical analyses were conducted using SAS software version 9.4 (SAS Institute Inc., Cary, NC). The financial sponsors played no role in the design, execution, analysis and interpretation of data, or writing of the current study.
RESULTS
The mean (SD) age of the study participants was 76.6 (6.5) years. Approximately 7% and 3% of women were Black or Hispanic, respectively. The majority of women had high self-reported quality of life and independence in activities of daily life (Table 1). The mean number of prescription and non-prescription medications used was 5.1 and 3.0, respectively. Of the three target medication classes in this analysis, antihypertensive use was most common (85%), followed by antilipemic (59%) and oral hypoglycemic use (13%).
Table 1.
Characteristics of Participants Overall and by Self-Reported Barriers to Medication Use
| Characteristicsa, N (%) | Total N=59,054 | No Barriers N=47,846 | 1 Barrier N=7,105 | ≥2 Barriers N=4,103 | P-Value |
|---|---|---|---|---|---|
| Ageb | 76.6 (6.5) | 76.7 (6.5) | 76.54 (6.6) | 75.88 (6.5) | <0.0001 |
| Race/ethnicity | -- | -- | -- | -- | <0.0001 |
| White | 51054 (86.5) | 41772 (87.3) | 6018 (84.7) | 3264 (79.6) | |
| Black | 4339 (7.4) | 3235 (6.8) | 630 (8.9) | 474 (11.6) | |
| Hispanic | 1560 (2.6) | 1140 (2.4) | 213 (3.0) | 207 (5.1) | |
| American Indian | 193 (0.3) | 135 (0.3) | 34 (0.5) | 24 (0.6) | |
| Asian/Pacific Islander | 1225 (2.1) | 1036 (2.2) | 120 (1.7) | 69 (1.7) | |
| Education | -- | - | - | - | <0.0001 |
| 0–8 Years | 387 (0.7) | 277 (0.6) | 55 (0.8) | 55 (1.3) | |
| Some High School | 1503 (2.6) | 1106 (2.3) | 242 (3.4) | 155 (3.8) | |
| High School Diploma/GED | 9791 (16.6) | 8000 (16.7) | 1163 (16.4) | 628 (15.3) | |
| School after High School | 22063 (37.4) | 17591 (36.8) | 2789 (39.3) | 1683 (41.0) | |
| College degree or higher | 24930 (42.2) | 20557 (43.0) | 2821 (39.7) | 1552 (37.8) | |
| Family income, $ | -- | -- | -- | -- | <0.0001 |
| <10,000 | 1434 (2.4) | 1001 (2.1) | 242 (3.4) | 191 (4.7) | |
| 10,000–19,999 | 5293 (9.0) | 3838 (8.0) | 879 (12.4) | 576 (14.0) | |
| 20,000–34,999 | 12914 (21.9) | 10216 (21.4) | 1687 (23.7) | 1011 (24.6) | |
| 35,000–49,999 | 12131 (20.5) | 9894 (20.7) | 1433 (20.2) | 804 (19.6) | |
| 50,000–74,999 | 12396 (21.0) | 10354 (21.6) | 1316 (18.5) | 726 (17.7) | |
| ≥75,000 | 11822 (20.0) | 10114 (21.1) | 1145 (16.1) | 563 (13.7) | |
| US region | -- | -- | -- | -- | <0.0001 |
| Northeast | 14255 (24.1) | 11641 (24.3) | 1673 (23.6) | 941 (22.9) | |
| Southeast | 14542 (24.6) | 11419 (23.9) | 1898 (26.7) | 1225 (29.9) | |
| Midwest | 13763 (23.3) | 11198 (23.4) | 1629 (22.9) | 936 (22.8) | |
| West | 16494 (27.9) | 13588 (28.4) | 1905 (26.8) | 1001 (24.4) | |
| Quality of life | -- | -- | -- | -- | <0.0001 |
| Low | 2093 (3.5) | 1348 (2.8) | 399 (5.6) | 346 (8.4) | |
| Moderate | 8323 (14.1) | 5980 (12.5) | 1419 (20.0) | 924 (22.5) | |
| High | 48086 (81.4) | 40112 (83.8) | 5196 (73.1) | 2778 (67.7) | |
| Activities of daily livingc | -- | -- | -- | -- | <0.0001 |
| No Help | 55773 (94.4) | 45411 (94.9) | 6620 (93.2) | 3742 (91.2) | |
| Some Help | 1981 (3.4) | 1403 (2.9) | 331 (4.7) | 247 (6.0) | |
| Physical function | -- | -- | -- | -- | <0.0001 |
| Low | 21388 (36.2) | 16181 (33.8) | 3162 (44.5) | 2045 (49.8) | |
| Moderate | 21295 (36.1) | 17738 (37.1) | 2319 (32.6) | 1238 (30.2) | |
| High | 12087 (20.5) | 10594 (22.1) | 1017 (14.3) | 476 (11.6) | |
| No. of prescription drugsb | 5.1 (2.6) | 4.96 (2.6) | 5.57 (2.68) | 6.17 (2.8) | <0.0001 |
| No. of non-prescription drugsb | 3.0 (1.5) | 3.02 (1.5) | 3.14 (1.6) | 3.26 (1.7) | <0.0001 |
| Antihypertensive use | 50137 (84.9) | 40457 (84.6) | 6126 (86.2) | 3554 (86.6) | <0.0001 |
| Antilipemic use | 34712 (58.8) | 28564 (59.7) | 3885 (54.7) | 2263 (55.2) | <0.0001 |
| Oral hypoglycemic use | 7614 (12.9) | 5715 (11.9) | 1100 (15.5) | 799 (19.5) | <0.0001 |
| No. of different classes of target medicationsd | <0.0001 | ||||
| 1 | 30543 (51.7) | 24753 (51.7) | 3762 (53.0) | 2028 (49.4) | |
| 2 | 23613 (40.0) | 19296 (40.3) | 2680 (37.7) | 1637 (39.9) | |
| 3 | 4898 (8.3) | 3797 (8.0) | 663 (9.3) | 438 (10.7) | |
| Had Medicare coverage | 53029 (89.8) | 43072 (90.0) | 6340 (89.2) | 3617 (88.2) | 0.0002 |
| Living assistance | 2731 (4.9) | 2072 (4.6) | 394 (6.0) | 265 (7.0) | <0.0001 |
| Mammogram | 55197 (93.5) | 44847 (93.7) | 6564 (92.4) | 3786 (92.3) | <0.0001 |
| Flexible sigmoidoscopy | 28883 (48.9) | 23183 (48.5) | 3591 (50.6) | 2109 (51.4) | <0.0001 |
| Bone density scan | 38282 (64.8) | 30844 (64.5) | 4696 (66.1) | 2742 (66.9) | 0.0005 |
Abbreviations: GED, General Equivalency Diploma, SD, standard deviation; US, United States
Missing/unknown variables: race/ethnicity, n=683 (1.2%); education, n=380 (0.6%); family income, n=3064 (5.2%); quality of life, n=552 (0.9%); activities of daily living, n=1300 (2.2%); physical function, n=4284 (7.3%)
Mean (standard deviation)
Activities of daily living included feeding, dressing, transferring, and bathing.
Target medication classes include antihypertensives, antilipemics, and/or oral hypoglycemics
Among the 59,054 women meeting eligibility criteria for this analysis, 47,846 (81%) reported no barriers, with a total of 11,208 (19.0%) women reporting one or more barriers to medication use. Among those reporting one or more barriers, 7105 (12%) reported 1 barrier and 4103 (6.9%) reported ≥2 barriers. The most common barriers reported were having concerns about side effects, not liking to take medications, and medications costing too much (Table 2). Among women reporting ≥2 barriers, the most common combinations of barriers were: health insurance would not cover/medication costs too much (n=742); concerned about side effects/do not like taking medications (n=671); and taking too many medications/do not like taking medications (n=330).
Table 2.
Self-Reported Barriers to Medication Use among Postmenopausal Women Using Chronic Cardiovascular and Oral Hypoglycemic Medications
| Barriers to Medication Use | Any Barrier N = 11,208a | 1 Barrier N = 7,105 (%) | ≥2 Barriers N = 4,103 (%)a |
|---|---|---|---|
| Concerned about side effects | 4,780 (42.7) | 2455 (34.6) | 2325 (56.7) |
| Do not like taking medications | 3,560 (31.8) | 1494 (19.5) | 2066 (50.4) |
| Medication costs too much | 3,390 (30.3) | 1389 (19.5) | 2001 (48.8) |
| Health insurance would not cover | 2,859 (25.5) | 1247 (17.6) | 1612 (39.3) |
| Taking too many medications | 1,583 (14.1) | 302 (4.3) | 1281 (31.2) |
| Problem getting to medical facility | 311 (2.8) | 100 (1.4) | 211 (5.1) |
| Family discouraged me | 228 (2.0) | 27 (0.4) | 201 (4.9) |
| Taking medication is inconvenient | 207 (1.9) | 55 (0.8) | 152 (3.7) |
| Friends discouraged me | 146 (1.3) | 16 (0.2) | 130 (3.2) |
| Concerned about missing work | 88 (0.8) | 20 (0.3) | 68 (1.7) |
Column adds up to >100% because participants could report multiple barriers
Several participant characteristics were found to be associated with reporting ≥2 barriers in multivariable analyses, after adjusting for all sociodemographic, health characteristic, medication use, and access to care variables (Table 3). Lower age (OR 0.95, 95% CI 0.95–0.96), Black race (reference: White; OR 1.59, 95% CI 1.40–1.81), Hispanic ethnicity (reference: White; OR 1.93, 95% CI 1.60–2.33), lower family income (reference: <$10,000; ≥$75,000 OR 0.42, 95% CI 0.34–0.52), lower quality of life (reference: high; low OR 2.46, 95% CI 2.10–2.88), lower physical function (reference: high; low OR 1.88, 95% CI 1.66–2.13), and higher number of concurrent medications (prescription: OR 1.14, 95% CI 1.12–1.16; non-prescription: OR 1.13, 95% CI 1.11–1.16) were significantly associated with increased odds of reporting ≥2 barriers. Women receiving a mammogram and those taking two or three of the target medication classes (compared to those taking one) had lower odds of reporting ≥2 barriers (p<0.05).
Table 3.
Multivariable Association between Participant Characteristics and Self-Reported Multiple (≥2) Barriers to Medication Use Compared to those Reporting No Barriersa
| Characteristics | ≥2 Barriers to Medication Use |
|---|---|
| -- | Adjusted ORb (95% Cl) |
| Age | 0.95 (0.95, 0.96) |
| Race/ethnicity | -- |
| White | Reference |
| Black | 1.59 (1.40, 1.81) |
| Hispanic | 1.93 (1.60,2.33) |
| American Indian | 1.64 (0.95,2.82) |
| Asian/Pacific Islander | 1.16 (0.87, 1.54) |
| Education | -- |
| High School Diploma/GED | Reference |
| 0–8 Years | 1.20 (0.78–1.85) |
| Some High School | 1.35 (1.07–1.70) |
| School after High School but no college degree | 1.34 (1.20–1.50) |
| College degree or higher | 1.28 (1.14–1.45) |
| Family income, $ | -- |
| <10,000 | Reference |
| 10,000–19,999 | 1.05 (0.85, 1.30) |
| 20,000–34,999 | 0.72 (0.59, 0.88) |
| 35,000–49,999 | 0.57 (0.46, 0.70) |
| 50,000–74,999 | 0.49 (0.40,0.61) |
| >75,000 | 0.42 (0.34, 0.52) |
| US region | -- |
| Northeast | Reference |
| Southeast | 1.07 (0.96, 1.19) |
| Midwest | 0.95 (0.85, 1.06) |
| West | 0.85 (0.76, 0.95) |
| Quality of life | -- |
| High | Reference |
| Moderate | 1.60 (1.45, 1.76) |
| Low | 2.46 (2.10,2.88) |
| Physical function | -- |
| High | Reference |
| Moderate | 1.43 (1.27, 1.61) |
| Low | 1.88 (1.66,2.13) |
| No. of prescription drugs | 1.14 (1.12, 1.16) |
| No. of non-prescription drugs | 1.13 (1.11, 1.16) |
| No. of different classes of target medicationsc | -- |
| 1 | Reference |
| 2 | 0.79 (0.73, 0.86) |
| 3 | 0.73 (0.64, 0.83) |
| Mammogram | 0.78 (0.67, 0.90) |
| Bone density scan | 1.14 (1.05, 1.23) |
Abbreviations: GED, General Equivalency Diploma; OR, odds ratio
Only those variables reaching statistical significance (p<0.05) are included in the table for presentation. Variables for which results are not shown because they did not reach statistical significance are: Medicare coverage, living assistance, receipt of flexible sigmoidoscopy, and activities of daily living. Those reporting only 1 barrier not included in model.
Adjusted for all covariates in Table 1
Target medication classes include antihypertensives, antilipemics, and/or oral hypoglycemics
DISCUSSION
In a large, diverse, US cohort of older women, we found that approximately one in five participants reported at least one barrier to chronic disease medication use. Among the 7% of women reporting multiple barriers to medication use, the most common combination of barriers related to high medication cost and lack of prescription medication coverage (even in the post-Part D implementation period). Black race, Hispanic ethnicity, lower family income, lower quality of life, lower physical function, and higher number of medications were associated with higher odds of multiple barriers to medication use.
Many studies have reported on individual barriers to medication use in older adults, but few have described patients reporting multiple barriers.2,5,12,15 Consistent with prior research, we found that medication cost and concerns about side effects were two of the most common barriers reported.5 In addition, we found that women reported a global dislike for taking medications as an important and common barrier. Beliefs about medications have been shown to be significantly associated with medication adherence, and thus should be addressed by pharmacists managing medications for older adults.9,12,15 When examining common combinations among women reporting multiple barriers, we also found that women frequently reported a general sense of taking too many medications. On average, participants were taking 5 prescription and 3 non-prescription medications at the time of the Medication Inventory, indicating a significant medication burden. To translate this finding to clinical practice, a simple screening item could be employed by asking older adults taking multiple medications, “Do you think the number of medications that you are currently taking is: too many or just right?” Based on the response, conversations around discontinuation could follow for those expressing that they are taking too many medications.16 Future research should explore the validity of simple screening items to predict important process and outcome measures related to medication use, such as subsequent medication non-adherence and healthcare utilization.
Several factors were found to be associated with reporting multiple barriers to medication use. Consistent with prior research, we observed differences in medication use by race/ethnicity.17–19 Compared to White women, Black and Hispanic women were significantly more likely to report multiple barriers, after adjusting for potential access difficulties that may disproportionately affect minorities (e.g., income, education). This supports existing evidence suggesting that increased efforts are needed to improve access to culturally appropriate healthcare and adequate drug benefits for older women.20
Not surprisingly, we found that women taking more medications (prescription and non-prescription) overall were more likely to report multiple barriers. Specifically, for every additional prescription medication, there was a 14% increased odds of reporting multiple barriers. Polypharmacy has been consistently shown to be an important medication barrier in older adults.21 Conversely, among the three target medication classes used for inclusion criteria in this analysis (antihypertensives, antilipemics, and oral hypoglycemics), women taking more of these medication classes were less likely than those only taking a single class to report multiple barriers. We were unable to measure the duration of cardiovascular disease and diabetes, but it is possible that women taking multiple cardiovascular and/or diabetes medications had surmounted early barriers in their treatment plan.
We found that women with better self-reported quality of life and physical function were less likely to report multiple barriers. This aligns with previous findings that older adults with poor quality of life and physical function often suffer from greater comorbidity burden and face several logistical barriers (e.g., transportation) to successful medication management.22 Finally, we found that women reporting receipt of a mammogram in the previous four years were less likely to report multiple barriers to medication use. This may be evidence of greater healthcare access among those women receiving a mammogram and, in turn, fewer barriers to medication use.
Strengths of our study include the large and diverse cohort of postmenopausal women and the measurement of a wide array of potential barriers to medication use. We view the focus on women as a strength because women are disproportionately represented among older adults, and are a population heavily burdened by chronic conditions including cardiovascular disease and diabetes. There are important limitations worth mentioning. First, we measured medication barriers in general rather than related to individual classes of medications. This approach is similar to other studies determining overall medication use barriers.7,23,24 However, barriers to specific medications may vary from general barriers to medications so it is important to clearly report which construct is being measured. Second, this was a cross-sectional analysis and, thus, does not reflect patterns of barriers to medication use over time. Future studies should explore longitudinal patterns of medication barriers. Third, the ten barriers to medication use included in this analysis are not derived from a validated instrument. However, these barriers have face validity and potential utility in clinical practice. Future research is needed to validate simple, freely available, and clinically useful instruments for assessing medication barriers.
CONCLUSION
In conclusion, we found that approximately one in five older women reported at least one barrier to medication use, with 7% of women reporting multiple barriers to medication use. These findings highlight the need to screen for and initiate conversations about medication beliefs. This may prove fruitful for pharmacists caring for older women, allowing them to more easily identify opportunities for discussions about perceived and structural barriers and possible medication regimen simplification.
ACKNOWLEDGEMENT
Availability of data and materials: The datasets supporting the conclusions of this article are available at: https://www.whi.org/SitePages/WHI%20Home.aspx.
Funding/Support: The WHI program is funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C, and HHSN268201600004C. Dr. Marcum was supported by an Agency for Healthcare Research & Quality grant (K12HS022982). Dr. Tom was supported by a National Institute on Aging grant (K01 AG050723).
ROLE OF THE FUNDING SOURCE
The funding source had no role in the study design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interest: None of the authors has a conflict to report.
REFERENCES
- 1.Qato DM, Wilder J, Schumm LP, Gillet V, Alexander GC. Changes in Prescription and Over-the-Counter Medication and Dietary Supplement Use Among Older Adults in the United States, 2005 vs 2011. JAMA Intern Med 2016;176(4):473–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Gellad WF, Grenard JL, Marcum ZA. A systematic review of barriers to medication adherence in the elderly: looking beyond cost and regimen complexity. Am J Geriatr Pharmacother 2011;9(1):11–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Steinman MA, Hanlon JT. Managing medications in clinically complex elders: “There’s got to be a happy medium”. JAMA 2010;304(14):1592–1601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Voils CI, Maciejewski ML, Hoyle RH, et al. Initial validation of a self-report measure of the extent of and reasons for medication nonadherence. Med Care 2012;50(12):1013–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Marcum ZA, Gellad WF. Medication adherence to multidrug regimens. Clin Geriatr Med 2012;28(2):287–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Munger MA, Van Tassell BW, LaFleur J. Medication nonadherence: an unrecognized cardiovascular risk factor. MedGenMed 2007;9(3):58. [PMC free article] [PubMed] [Google Scholar]
- 7.Gehi AK, Ali S, Na B, Whooley MA. Self-reported medication adherence and cardiovascular events in patients with stable coronary heart disease: the heart and soul study. Arch Intern Med 2007;167(16):1798–1803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Williams J, Steers WN, Ettner SL, Mangione CM, Duru OK. Cost-related nonadherence by medication type among Medicare Part D beneficiaries with diabetes. Med Care 2013;51(2):193–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Park HY, Seo SA, Yoo H, Lee K. Medication adherence and beliefs about medication in elderly patients living alone with chronic diseases. Patient Prefer Adherence 2018;12:175–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Horne R, Chapman SC, 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):e80633. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Langer RD, White E, Lewis CE, Kotchen JM, Hendrix SL, Trevisan M. The Women’s Health Initiative Observational Study: baseline characteristics of participants and reliability of baseline measures. Ann Epidemiol 2003;13(9 Suppl):S107–21. [DOI] [PubMed] [Google Scholar]
- 12.Clifford S, Barber N, Horne R. Understanding different beliefs held by adherers, unintentional nonadherers, and intentional nonadherers: application of the Necessity-Concerns Framework. J Psychosom Res 2008;64(1):41–6. [DOI] [PubMed] [Google Scholar]
- 13.Naughton MJ, Brunner RL, Hogan PE, et al. Global Quality of Life Among WHI Women Aged 80 Years and Older. J Gerontol A Biol Sci Med Sci 2016;71 Suppl 1:S72–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ware JE Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30(6):473–83. [PubMed] [Google Scholar]
- 15.Wei L, Champman S, Li X, et al. Beliefs about medicines and non-adherence in patients with stroke, diabetes mellitus and rheumatoid arthritis: a cross-sectional study in China. BMJ Open 2017;7(10):e017293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Turner JP, Tannenbaum C. Older Adults’ Awareness of Deprescribing: A Population-Based Survey. J Am Geriatr Soc 2017;65(12):2691–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gellad WF, Haas JS, Safran DG. Race/ethnicity and nonadherence to prescription medications among seniors: results of a national study. J Gen Intern Med 2007;22(11):1572–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gerber BS, Cho YI, Arozullah AM, Lee SY. Racial differences in medication adherence: A cross-sectional study of Medicare enrollees. Am J Geriatr Pharmacother 2010;8(2):136–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Trinacty CM, Adams AS, Soumerai SB, et al. Racial differences in long-term adherence to oral antidiabetic drug therapy: a longitudinal cohort study. BMC Health Serv Res 2009;9:24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.American Geriatrics Society Ethnogeriatrics Committee. Achieving High-Quality Multicultural Geriatric Care. J Am Geriatr Soc 2016;64(2):255–60. [DOI] [PubMed] [Google Scholar]
- 21.Patton DE, Hughes CM, Cadogan CA, Ryan CA. Theory-Based Interventions to Improve Medication Adherence in Older Adults Prescribed Polypharmacy: A Systematic Review. Drugs Aging 2017;34(2):97–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Boyd CM, Darer J, Boult C, Fried LP, Boult L, Wu AW. Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. JAMA 2005;294(6):716–24. [DOI] [PubMed] [Google Scholar]
- 23.Ho PM, Bryson CL, Rumsfeld JS. Medication adherence: its importance in cardiovascular outcomes. Circulation 2009;119(23):3028–35. [DOI] [PubMed] [Google Scholar]
- 24.Ovbiagele B, Campbell S, Faiz A, Chambless LE, Investigators VS. Relationship between non-specific prescription pill adherence and ischemic stroke outcomes. Cerebrovasc Dis 2010;29(2):146–53. [DOI] [PMC free article] [PubMed] [Google Scholar]