Abstract
Background:
Sex-race stratification may lead to identification of risk factors for low antihypertensive medication adherence that are not apparent when assessing risk factors in women and men without race stratification. We examined risk factors associated with low pharmacy refill adherence across sex-race subgroups (white women, black women, white men, black men) within the Cohort Study of Medication Adherence among Older Adults (n=2,122).
Methods:
Pharmacy refill adherence was calculated as proportion of days covered (PDC) using all antihypertensive prescriptions filled in the year prior to a baseline risk factor survey. Sex- and sex-race-stratified multivariable Poisson regression models with robust standard errors were used to estimate adjusted prevalence ratios and 95% confidence intervals for associations between participant characteristics and low adherence.
Results:
Prevalence of low adherence was 22.9% versus 40.7% in white versus black women (p<0.001) and 26.3% versus 37.2% in white versus black men (p=0.003). In multivariable models, reducing antihypertensive medication due to cost was associated with low adherence within each sex-race subgroup. Additional factors associated with low adherence included shorter hypertension duration and comorbidities in white women; not being married and depressive symptoms in white men; and ≥6 primary care visits/year and complementary and alternative medicine use in black men. Among men, not being married and reporting depressive symptoms were associated with low adherence for whites, but not blacks.
Conclusions:
Identification of sex-race-specific risk factors for low antihypertensive medication adherence may guide development and implementation of tailored interventions to increase antihypertensive medication adherence and blood pressure control among older patients.
Keywords: medication adherence, hypertension, risk factors, older adults, proportion of days covered (PDC)
Introduction
Hypertension is a leading cause of mortality and disease burden, and a major public health and clinical challenge affecting older adults.1–3 Despite advances in hypertension management and treatment, almost 30% of adults taking antihypertensive medication have uncontrolled blood pressure.1,4 Low adherence to antihypertensive medication is associated with poor blood pressure control in older adults,5 and higher rates of emergency room visits, hospitalizations, cardiovascular events and mortality.6–8 Understanding risk factors for low antihypertensive medication adherence may support development of personalized interventions to promote adherence and improve blood pressure control among older adults.
Sex and race differences in the prevalence of hypertension, blood pressure control, and antihypertensive medication adherence are well documented.1,5,9,10 A higher prevalence of hypertension and uncontrolled blood pressure has been reported among older women compared to older men, and among blacks compared to whites.1,2,11,12 Race differences in antihypertensive medication adherence mirror those for blood pressure control – antihypertensive medication adherence and blood pressure control are lower among blacks compared with whites.13 Studies reporting sex differences in pharmacy refill adherence among men and women are less consistent: some studies report similar prevalence of low pharmacy refill adherence in older men and women,5,14 while others report higher prevalence of low pharmacy refill adherence among older women compared to men.10 Furthermore, sex and race differences research has helped to uncover risk factors for low antihypertensive medication adherence that are sex- or race-specific. For example, Holt and colleagues found that reduced sexual functioning and high BMI were associated with low self-report antihypertensive medication adherence among older, hypertensive men but not women.9 In this case, sex stratification uncovered sex-specific risk factors for low antihypertensive medication adherence that would not have otherwise been identified using a non-stratified approach.
The recent mandate from the NIH15 to include, where appropriate, assessment of sex differences together with the current era of personalized and precision medicine are powering efforts to examine sex and race differences. While growing recognition of the importance of sex and race differences in research and increasing number of sex-stratified and race-stratified assessments are promising, stratification by sex and race (e.g., white women, black women, white men, black men) may be necessary to identify risk factors for antihypertensive medication adherence that would not be identified in a sex- or race-stratified analysis alone. However, few studies have comprehensive risk factor data from a predominantly black and white patient population linked to patient-level pharmacy fill data and self-reported race to evaluate patterns across sex-race subgroups.
The Cohort Study of Medication Adherence among Older Adults (CoSMO) includes data from a comprehensive patient adherence risk factor survey linked to pharmacy fill and medical claims, and medical records in over 2,000 community-dwelling older black and white women and men with treated hypertension.5 We examined socio-demographic, clinical, psychosocial, behavioral, and health care factors5,16 associated with low pharmacy refill adherence in women, men and across four sex-race subgroups of CoSMO participants: white women, black women, white men, and black men. Sex-race stratification may lead to identification of risk factors that are not apparent when assessing risk factors in women and men without race stratification, which, in turn, may guide development and implementation of tailored interventions to increase antihypertensive medication adherence and blood pressure control among older patients.
Methods
Study Population, Study Design, and Data Collection
We conducted a cross-sectional analysis of baseline data from the Cohort Study of Medication Adherence among Older Adults (CoSMO). The CoSMO study design, participant baseline characteristics and response rates have been previously described.5 In brief, CoSMO enrolled 2,194 adults aged 65 years or older with essential hypertension from a large managed care organization in southeastern Louisiana. Recruitment and completion of baseline telephone interviews by participants occurred from August 21, 2006 to September 30, 2007. CoSMO was approved by Ochsner Clinic Foundation Institutional Review Board, and the privacy board of the managed care organization; participants provided verbal consent to participate in this minimal risk study. This analysis included 2,122 individuals with pharmacy fill data for antihypertensive medication at baseline and reporting black or white race. Data from a patient risk factor survey and pharmacy utilization and medical claims databases of the managed care organization were analyzed.5
Outcome: Pharmacy Refill Medication Adherence
Pharmacy refill adherence was calculated as proportion of days covered (PDC) using all antihypertensive prescriptions filled in the year prior to the baseline risk factor survey. PDC was calculated as the number of days with medication available divided by the number of days between the first and last pharmacy fills in the one-year period. A PDC for each antihypertensive medication class was calculated for each participant and an overall PDC was calculated as the average across all antihypertensive medication classes. Low adherence was defined as PDC less than 0.80.7
Risk Factors for Low Antihypertensive Medication Adherence
Using a conceptual model as a guide,8 risk factors were assessed using a telephone-administered comprehensive participant risk factor survey.5
Socio-demographic and Clinical Risk Factors
Socio-demographic and clinical characteristics included sex, race, age, marital status, educational attainment, hypertension duration, weight and height. Body mass index (BMI) was calculated as weight (kg)/height (m2). Hypertension knowledge was assessed using a validated tool;17 participants with scores in the lowest tertile were categorized as having low hypertension knowledge.18 Comorbidities were identified from claims data and used to calculate the Charlson Comorbidity Index score, which was dichotomized as <2 versus ≥2.19,20 The number of classes of antihypertensive medications filled in the year prior to the baseline survey was obtained from a pharmacy utilization database and categorized as <3 classes versus ≥3 classes.5
Healthcare System Risk Factors
Participant satisfaction was assessed using three domains from the Group Health Association of America Consumer Satisfaction Survey: overall satisfaction, access to care and communication.21 Mean scores corresponding to a poor and fair rating were classified as not satisfied. Reducing antihypertensive medication due to cost was ascertained by asking participants if they had taken less high blood pressure medication than was prescribed because of the cost (yes vs no); and participant-reported number of visits to a primary care provider in the year prior to the baseline survey was categorized as <6 visits versus ≥6 visits.5
Psychosocial and Behavioral Risk Factors
Lifestyle behaviors included smoking status and weekly intake of alcoholic beverages.5 Presence of depressive symptoms was defined as a score ≥16 using the 20-item Center for Epidemiologic Studies Depression Scale.22 Low social support was defined as scores in the lowest tertile utilizing the RAND Medical Outcomes Study Social Support Survey,23 and low coping was defined as scores below the median utilizing the shortened version of the John Henry Active Coping scale.24 Perceived high stress was defined as a score in the highest tertile utilizing the Perceived Stress Scale.25
Self-management Risk Factors
Self-management behaviors5 included the use of complementary and alternative medicines at least several times or on a regular basis in the year prior to the baseline interview (CAM; dichotomized as yes or no)26 and the use of lifestyle modifications (i.e., salt reduction, fruit and vegetable consumption, exercise and weight management) to lower blood pressure.27 The total number of lifestyle modifications was summed for each participant and dichotomized as <2 or ≥2.5
Statistical Analysis
Participant characteristics and prevalence of low antihypertensive medication adherence were calculated overall and within each sex-race subgroup. Within each sex, chi-square tests were used to determine the statistical significance of differences in participant characteristics by race. Bivariate associations of participant characteristics with low PDC adherence were assessed within sex and sex-race subgroups. Sex- and sex-race-stratified multivariable Poisson regression models with robust standard errors were used to estimate the adjusted prevalence ratios and 95% confidence intervals for associations between participant characteristics and low adherence. All variables were included simultaneously in the multivariable models, with the exception of stress (due to its collinearity with depressive symptoms) and the healthcare satisfaction domains related to communication and healthcare access (due to their collinearity with overall healthcare satisfaction). We tested for effect modification by race by including multiplicative interaction terms for each main effect (e.g., race * cost) in separate sex-stratified multivariable models. Interaction was tested at the alpha=0.05 level of significance. Analyses were performed using Stata version 14.1 (StataCorp, College Station, TX). The figure was created using R 3.4.3 (R Foundation for Statistical Computing, Vienna, Austria).
Results
Of the 2,122 participants, 48.9% were 75 years old or older, 59.2% were women (37.3% white women, 21.9% black women), and 40.8% were men (31.9% white men, and 8.9% black men). Overall, 29.1% of the study population had low antihypertensive medication adherence (Table 1). The prevalence of low adherence differed between whites and blacks (24.4% and 39.7%, respectively, p<0.001), but not between men and women (28.6% and 29.5%, respectively, p=0.682).
Table 1.
Characteristics of CoSMO Study Participants, Overall and in Sex-Race Sub-groups
| Women | Men | ||||||
|---|---|---|---|---|---|---|---|
| Overall (n=2,122) n (%) |
White (n=792) n (%) |
Black (n=464) n (%) |
p- value |
White (n=678) n (%) |
Black (n=188) n (96) |
p- value |
|
| Socio-Demographic Factors | |||||||
| Age ≥75 years | 1038 (48.9) |
425 (53.7) |
205 (44.2) |
0.001 |
333 (49.1) |
75 (39.9) |
0.025 |
| Not married | 920 (43.4) |
422 (53.3) |
303 (65.3) |
<0.001 | 147 (21.7) |
48 (25.5) |
0.263 |
| Less than high school education | 439 (20.7) |
120 (15.2) |
155 (33.5) |
<0.001 |
78 (11.5) |
86 (45.7) |
<0.001 |
| Low HTN knowledge | 680 (32.1) |
220 (27.8) |
178 (38.4) |
<0.001 |
200 (29.5) |
82 (43.6) |
<0.001 |
| Clinical Factors | |||||||
| HTN duration < 10 years | 779 (36.9) |
332 (42.2) |
148 (32.0) |
<0.001 | 235 (34.7) |
64 (34.0) |
0.864 |
| CCI score ≥2a | 1050 (49.5) |
328 (41.4) |
237 (51.2) |
0.001 | 375 (55.4) |
110 (58.5) |
0.446 |
| BMI ≥25 kg/m2 | 1625 (76.7) |
534 (67.5) |
398 (86.0) |
<0.001 | 535 (78.9) |
158 (84.0) |
0.119 |
| <3 classes of anti-HTN medicationsa | 1180 (55.6) |
469 (59.2) |
238 (51.3) |
0.006 |
389 (57.4) |
84 (44.7) |
0.002 |
| Health Care System Factors | |||||||
| Not satisfied with overall healthcare | 98 (4.6) | 29 (3.7) | 30 (6.5) | 0.023 | 28 (4.1) | 11 (5.9) | 0.316 |
| Not satisfied with healthcare communication | 225 (10.6) |
85 (10.8) |
53 (11.5) |
0.707 | 62 (9.1) |
25 (8.4) |
0.089 |
| Not satisfied with access to healthcare | 96 (4.5) | 31(3.9) | 24 (5.2) | 0.293 | 28 (4.1) | 13 (6.9) | 0.112 |
| Reduced medications due to cost | 77 (3.6) | 15 (1.9) | 32 (6.9) | <0.001 | 19 (2.8) | 11 (5.9) | 0.043 |
| 6+ visits to primary care physiciana | 483 (22.8) |
165 (20.9) |
111 (24.0) |
0.211 | 162 (23.9) |
45 (23.9) |
0.998 |
| Psychosocial/Behavioral Factors | |||||||
| Current or former smoker | 1068 (50.7) |
326 (41.5) |
168 (36.5) |
0.084 | 453 (67.2) |
121 (65.1) |
0.580 |
| ≥2 alcoholic beverages per week | 446 (21.1) |
135 (17.1) |
27 (5.8) |
<0.001 |
243 (36.1) |
41 (22.0) |
<0.001 |
| Depressive symptoms | 276 (13.0) |
113 (14.3) |
72 (15.5) |
0.546 |
61 (9.0) |
30 (16.0) |
0.006 |
| Low social support | 719 (33.9) |
265 (33.5) |
190 (41.0) |
0.008 | 201 (29.7) |
63 (33.5) |
0.308 |
| Low coping | 1007 (47.5) |
383 (48.4) |
227 (48.9) |
0.863 | 302 (44.5) |
95 (50.5) |
0.145 |
| High stress | 718 (33.8) |
273 (34.5) |
195 (42.0) |
0.008 | 188 (27.7) |
62 (33.0) |
0.160 |
| Self-Management Behaviors | |||||||
| Complementary and alternative medicine | 560 (26.4) |
197 (24.9) |
139 (30.0) |
0.050 | 166 (24.5) |
58 (30.9) |
0.078 |
| ≥2 lifestyle modifications | 1736 (81.8) |
630 (79.6) |
417 (89.9) |
<0.001 |
523 (77.1) |
166 (88.3) |
<0.001 |
| Medication Adherence | |||||||
| Low proportion of days covered | 618 (29.1) |
181 (22.9) |
189 (40.7) |
<0.001 |
178 (26.3) |
70 (37.2) |
0.003 |
in the prior year
CCI – Charlson Comorbidity Index; BMI – Body Mass Index; HTN – hypertension; CoSMO-Cohort Study of Medication Adherence among Older Adults
Statistically significant results are indicated by bold text
Black women were more likely to have low adherence compared to white women (Table 1; 40.7% and 22.9%, respectively, p<0.001). Compared to black women, white women were older and were more likely to have had hypertension for less than ten years, to be taking fewer than three classes of antihypertensive medications, and to consume two or more alcoholic beverages per week. Also, white women were less likely to be unmarried, have less than a high school education, have low hypertension knowledge, have two or more comorbidities, have a high BMI, indicate that they were not satisfied with their healthcare, reduce antihypertensive medications due to cost, have low social support, have high stress and report two or more lifestyle modifications for blood pressure management than black women.
The prevalence of low antihypertensive medication adherence associated with participant characteristics, among women overall and for white women and black women, separately, is presented in Table 2. In the multivariable-adjusted model for all women (Table 3), hypertension duration less than ten years, having a comorbidity score ≥ 2, taking fewer than 3 classes of antihypertensive medications, reducing medications due to cost, depressive symptoms, and being black were associated with low adherence. Among white women, reporting hypertension duration less than ten years (PR = 1.40; 95% CI: 1.06, 1.83; p=0.016), having a comorbidity score ≥ 2 (prevalence ratio (PR) = 1.49; 95% CI: 1.13, 1.97; p=0.004), and reducing medications due to cost (PR = 2.31; 95% CI: 1.58, 3.39; p<0.001) were associated with low adherence (Table 3). For black women, only reducing antihypertensive medications due to cost (PR = 2.10; 95% CI: 1.62, 2.73; p<0.001) was associated with low adherence (Table 3). Among women, there was no effect modification by race for any participant characteristic (p-values for all interactions >0.05).
Table 2.
Prevalence of Low Pharmacy Refill Antihypertensive Medication Adherence Associated with Participant Characteristics among CoSMO Women, Overall and by Race
| All women (n=1256) |
White women (n=792) |
Black women (n=464) |
||||
|---|---|---|---|---|---|---|
| % with low PDC |
p- value |
% with low PDC |
p- value |
% with low PDC |
p- value |
|
| Socio-Demographic Factors | ||||||
| Age ≥75 years | 29.5 | 0.959 | 22.6 | 0.848 | 43.9 | 0.216 |
| Age <75 years | 29.4 | 23.2 | 38.2 | |||
| Not married | 31.2 | 0.120 | 23.2 | 0.792 | 42.2 | 0.363 |
| Married | 27.1 | 22.4 | 37.9 | |||
| <High school graduate | 37.5 | 0.001 | 25.8 | 0.399 | 46.5 | 0.080 |
| High school graduate | 27.2 | 22.3 | 38.0 | |||
| Clinical Factors | ||||||
| Low HTN knowledge | 33.7 | 0.026 | 22.7 | 0.958 | 47.2 | 0.026 |
| Not low HTN knowledge | 27.5 | 22.9 | 36.7 | |||
| HTN duration <10 years | 32.7 | 0.042 | 27.7 | 0.004 | 43.9 | 0.353 |
| HTN duration ≥10 years | 27.3 | 18.9 | 39.4 | |||
| CCI score ≥2a | 34.9 | <0.001 | 28.1 | 0.003 | 44.3 | 0.118 |
| CCI score <2a | 25.1 | 19.2 | 37.2 | |||
| BMI ≥25 kg/m2 | 30.9 | 0.051 | 24.0 | 0.294 | 40.2 | 0.662 |
| BMI <25 kg/m2 | 25.2 | 20.6 | 43.1 | |||
| <3 classes of anti-HTN medicationsa | 31.1 | 0.143 | 25.0 | 0.091 | 43.3 | 0.252 |
| 3+ classes of anti-HTN medicationsa | 27.3 | 19.8 | 38.1 | |||
| Health Care System Factors | ||||||
| Not satisfied with overall healthcare | 44.1 | 0.012 | 41.4 | 0.015 | 46.7 | 0.501 |
| Satisfied with overall healthcare | 28.8 | 22.2 | 40.4 | |||
| Not satisfied with healthcare communication | 34.1 | 0.216 | 28.2 | 0.216 | 43.4 | 0.685 |
| Satisfied with healthcare communication | 29.0 | 22.3 | 40.5 | |||
| Not satisfied with access to healthcare | 30.9 | 0.809 | 32.3 | 0.203 | 29.2 | 0.236 |
| Satisfied with access to healthcare | 29.4 | 22.5 | 41.4 | |||
| Reduced medications due to cost | 76.6 | <0.001 | 66.7 | <0.001 | 81.3 | <0.001 |
| Did not reduce medications due to cost | 27.7 | 22.0 | 37.7 | |||
| 6+ visits to primary care physiciana | 34.4 | 0.039 | 28.5 | 0.052 | 43.2 | 0.516 |
| <6 visits to primary care physiciana | 28.0 | 21.4 | 39.8 | |||
| Psychosocial/Behavioral Factors | ||||||
| Current or former smoker | 28.7 | 0.693 | 21.2 | 0.404 | 43.5 | 0.393 |
| Never smoker | 29.8 | 23.7 | 39.4 | |||
| ≥2 alcoholic beverages per week | 22.2 | 0.031 | 21.5 | 0.698 | 25.9 | 0.105 |
| <2 alcoholic beverages per week | 30.5 | 23.0 | 41.7 | |||
| Depressive symptoms | 38.9 | 0.002 | 30.1 | 0.048 | 52.8 | 0.024 |
| No depressive symptoms | 27.8 | 21.7 | 38.5 | |||
| Low social support | 33.2 | 0.029 | 26.4 | 0.091 | 42.6 | 0.488 |
| Medium/high social support | 27.3 | 21.1 | 39.4 | |||
| Low coping | 28.4 | 0.397 | 20.9 | 0.196 | 41.0 | 0.919 |
| Medium/high coping | 30.5 | 24.8 | 40.5 | |||
| High stress | 31.6 | 0.195 | 24.2 | 0.520 | 42.1 | 0.623 |
| Low/medium stress | 28.2 | 22.2 | 39.8 | |||
| Self-management Behaviors | ||||||
| Complementary and alternative medicine use |
35.4 | 0.005 | 26.9 | 0.118 | 47.5 | 0.053 |
| No complementary and alternative medicine use |
27.3 | 21.5 | 37.9 | |||
| ≥2 lifestyle modifications | 30.1 | 0.275 | 23.3 | 0.526 | 40.3 | 0.561 |
| <2 lifestyle modifications | 26.3 | 21.0 | 44.7 | |||
in the prior year
CCI – Charlson Comorbidity Index; BMI – Body Mass Index; HTN – hypertension; pharmacy fill - Proportion of Days Covered (PDC); CoSMO – Cohort Study of Medication Adherence among Older Adults
Statistically significant results are indicated by bold text
Table 3.
Multivariable-Adjusted Prevalence Ratios for Low Pharmacy Refill Antihypertensive Medication Adherence among CoSMO Women, Overall and for Black and White Women Separately
| All women (n=1228) PR (95% Cl) |
White women (n=775) PR (95% Cl) |
Black women (n=453) PR (95% Cl) |
|
|---|---|---|---|
| Age ≥75 years | 1.10 (0.92, 1.33) | 1.08 (0.81, 1.43) | 1.14 (0.90, 1.44) |
| Not married | 1.03 (0.86, 1.23) | 1.03 (0.78, 1.35) | 1.04 (0.82, 1.32) |
| Less than high school education | 1.16 (0.96, 1.41) | 1.13 (0.80, 1.60) | 1.16 (0.92, 1.45) |
| Low HTN knowledge | 1.07 (0.89, 1.29) | 0.92 (0.67, 1.26) | 1.19 (0.94, 1.50) |
| HTN duration <10 years | 1.28 (1.08, 1.53)** | 1.40 (1.06, 1.83)* | 1.14 (0.90, 1.44) |
| CCI score ≥2 | 1.30 (1.09, 1.56)** | 1.49 (1.13, 1.97)** | 1.15 (0.90, 1.45) |
| BMI ≥25 kg/m2 | 1.09 (0.88, 1.36) | 1.15 (0.86, 1.55) | 1.01 (0.74, 1.38) |
| <3 classes of anti-HTN medications | 1.26 (1.05, 1.50)* | 1.31 (0.98, 1.74) | 1.19 (0.94, 1.49) |
| Not satisfied with overall healthcare | 1.11 (0.80, 1.54) | 1.52 (0.92, 2.51) | 0.94 (0.62, 1.40) |
| Reduced medications due to cost | 2.18 (1.77, 2.68)*** | 2.31 (1.58, 3.39)*** | 2.10 (1.62, 2.73)*** |
| 6+ visits to primary care physician | 1.06 (0.87, 1.29) | 1.24 (0.93, 1.66) | 0.93 (0.71, 1.20) |
| Current or former smoker | 0.99 (0.83, 1.17) | 0.85 (0.65, 1.11) | 1.13 (0.90, 1.41) |
| ≥2 alcoholic beverages per week | 0.97 (0.72, 1.32) | 1.08 (0.75, 1.54) | 0.77 (0.42, 1.40) |
| Depressive symptoms | 1.25 (1.01, 1.55)* | 1.26 (0.90, 1.76) | 1.25 (0.96, 1.63) |
| Low social support | 1.08 (0.91, 1.28) | 1.16 (0.89, 1.52) | 1.02 (0.82, 1.27) |
| Low coping | 0.85 (0.72, 1.01) | 0.79 (0.61, 1.04) | 0.92 (0.74, 1.15) |
| Complementary and alternative medicine use | 1.19 (0.99, 1.43) | 1.14 (0.85, 1.53) | 1.20 (0.96, 1.52) |
| ≥ 2 lifestyle modifications | 1.09 (0.85, 1.41) | 1.14 (0.81, 1.61) | 1.03 (0.71, 1.48) |
| Black | 1.62 (1.35, 1.95)*** | -- | -- |
CCI – Charlson Comorbidity Index; BMI – Body Mass Index; HTN – hypertension; PR – prevalence ratio; CI – confidence interval; CoSMO – Cohort Study of Medication Adherence among Older Adults
Statistically significant results are indicated by bold text
p<0.05,
p<0.01,
p<0.001
The prevalence of low antihypertensive medication adherence was higher among black men than white men (Table 1; 37.2% versus 26.3%, respectively, p=0.003). Compared to black men, white men were more likely to be 75 years or older, be taking fewer than three classes of antihypertensive medication, and consume two or more alcoholic beverages per week. White men were less likely than black men to have less than a high school education, have low hypertension knowledge, reduce medications due to cost, report depressive symptoms, and engage in two or more lifestyle modifications for blood pressure control.
The prevalence of low antihypertensive medication adherence associated with participant characteristics, among men overall and for white men and black men, separately, is presented in Table 4. In the multivariable-adjusted model for men overall (Table 5), reducing medications due to cost was associated with low adherence. In the sex-race subgroup models, not being married (PR = 1.36; 95% CI: 1.00, 1.85; p=0.047), reducing antihypertensive medication due to cost (PR = 2.23; 95% CI: 1.44, 3.44; p<0.001), and experiencing depressive symptoms (PR = 1.63; 95% CI: 1.14, 2.34; p=0.007) were associated with low adherence among white men. In black men, reducing antihypertensive medications due to cost (PR = 1.68; 95% CI: 1.05, 2.67; p=0.030), reporting six or more primary care visits in the last year (PR = 1.54; 95% CI: 1.04, 2.28; p=0.032), and complementary and alternative medicine use (PR = 1.73; 95% CI: 1.15, 2.60; p=0.009) were associated with low adherence. Among men, not being married and having depressive symptoms had a different effect on low adherence among white men compared to black (p-values for the interactions were 0.038 and 0.004, respectively; Figure 1).
Table 4.
Prevalence of Low Pharmacy Refill Antihypertensive Medication Adherence Associated with Participant Characteristics among CoSMO Men, Overall and by Race
| All men (n=866) | White men (n=678) | Black men (n=188) | ||||
|---|---|---|---|---|---|---|
| % with low PDC |
p- value |
% with low PDC |
p- value |
% with low PDC |
p- value |
|
| Socio-Demographics and Patient Characteristics | ||||||
| Age ≥75 years | 27.9 | 0.669 | 27.0 | 0.653 | 32.0 | 0.227 |
| Age <75 years | 29.3 | 25.5 | 40.7 | |||
| Not married | 33.3 | 0.099 | 33.3 | 0.028 | 33.3 | 0.517 |
| Married | 27.3 | 24.3 | 38.6 | |||
| <High school graduate | 35.4 | 0.034 | 32.1 | 0.216 | 38.4 | 0.767 |
| High school graduate | 27.1 | 25.5 | 36.3 | |||
| Clinical/Treatment Variables | ||||||
| Low HTN knowledge | 29.8 | 0.603 | 28.0 | 0.504 | 34.2 | 0.441 |
| Not low HTN knowledge | 28.1 | 25.5 | 39.6 | |||
| HTN duration <10 years | 27.8 | 0.706 | 24.7 | 0.527 | 39.1 | 0.709 |
| HTN duration ≥10 years | 29.0 | 26.9 | 36.3 | |||
| CCI score ≥2a | 30.7 | 0.132 | 28.8 | 0.099 | 37.3 | 0.990 |
| CCI score <2a | 26.1 | 23.2 | 37.2 | |||
| BMI ≥25 kg/m2 | 28.3 | 0.644 | 25.4 | 0.340 | 38.0 | 0.630 |
| BMI <25 kg/m2 | 30.1 | 29.4 | 33.3 | |||
| <3 classes of anti-HTN medicationsa | 28.5 | 0.945 | 26.7 | 0.741 | 36.9 | 0.933 |
| 3+ classes of anti-HTN medicationsa | 28.8 | 25.6 | 37.5 | |||
| Health Care System Variables | ||||||
| Not satisfied with overall healthcare | 38.5 | 0.166 | 32.1 | 0.473 | 54.6 | 0.221 |
| Satisfied with overall healthcare | 28.2 | 26.0 | 36.2 | |||
| Not satisfied with healthcare communication | 33.3 | 0.298 | 30.7 | 0.410 | 40.0 | 0.730 |
| Satisfied with healthcare communication | 28.0 | 25.8 | 36.4 | |||
| Not satisfied with access to healthcare | 34.2 | 0.424 | 28.6 | 0.776 | 46.2 | 0.490 |
| Satisfied with access to healthcare | 28.4 | 26.2 | 36.6 | |||
| Reduced medications due to cost | 60.0 | <0.001 | 57.9 | 0.001 | 63.6 | 0.062 |
| Did not reduce medications due to cost | 27.5 | 25.3 | 35.6 | |||
| 6+ visits to primary care physiciana | 32.4 | 0.178 | 28.4 | 0.486 | 46.7 | 0.133 |
| <6 visits to primary care physiciana | 27.5 | 25.6 | 34.3 | |||
| Psychosocial/Behavioral Variables | ||||||
| Current or former smoker | 28.6 | 0.939 | 26.3 | 0.895 | 37.2 | 0.971 |
| Never smoker | 28.3 | 25.8 | 36.9 | |||
| ≥2 alcoholic beverages per week | 26.8 | 0.431 | 25.1 | 0.702 | 36.6 | 0.875 |
| <2 alcoholic beverages per week | 29.3 | 26.5 | 37.9 | |||
| Depressive symptoms | 39.6 | 0.015 | 44.3 | 0.001 | 30.0 | 0.371 |
| No depressive symptoms | 27.4 | 24.5 | 38.6 | |||
| Low social support | 30.3 | 0.473 | 27.4 | 0.670 | 39.7 | 0.622 |
| Medium/high social support | 27.9 | 25.8 | 36.0 | |||
| Low coping | 30.7 | 0.210 | 28.5 | 0.238 | 37.9 | 0.850 |
| Medium/high coping | 26.9 | 24.5 | 36.6 | |||
| High stress | 31.6 | 0.219 | 28.7 | 0.365 | 40.3 | 0.539 |
| Low/medium stress | 27.4 | 25.3 | 35.7 | |||
| Hypertension Management Behaviors | ||||||
| Complementary and alternative medicine use | 32.1 | 0.178 | 27.1 | 0.773 | 46.6 | 0.078 |
| No complementary and alternative medicine use | 27.4 | 26.0 | 33.1 | |||
| ≥2 lifestyle modifications | 28.6 | 0.954 | 26.2 | 0.949 | 36.1 | 0.396 |
| <2 lifestyle modifications | 28.8 | 26.5 | 45.5 | |||
in the prior year
CCI – Charlson Comorbidity Index; BMI – Body Mass Index; HTN – hypertension; CoSMO – Cohort Study of Medication Adherence among Older Adults
Statistically significant results are indicated by bold text
Table 5:
Multivariable-Adjusted Prevalence Ratios for Low Antihypertensive Pharmacy Refill Medication Adherence among CoSMO Men, Overall and for Black and White Men Separately
| All men (n=850) PR (95% Cl) |
White men (n=666) PR (95% Cl) |
Black men (n=184) PR (95% Cl) |
|
|---|---|---|---|
| Age ≥75 years | 0.92 (0.73, 1.15) | 1.00 (0.76, 1.31) | 0.78 (0.52, 1.18) |
| Not marrieda | 1.16 (0.90, 1.51) | 1.36 (1.00, 1.85)* | 0.79 (0.48, 1.29) |
| Less than high school education | 1.18 (0.90, 1.54) | 1.24 (0.87, 1.76) | 1.27 (0.84, 1.92) |
| Low HTN knowledge | 0.96 (0.76, 1.22) | 0.98 (0.74, 1.31) | 0.89 (0.60, 1.31) |
| HTN duration <10 years | 0.96 (0.76, 1.21) | 0.94 (0.71, 1.25) | 1.05 (0.70, 1.58) |
| CCI score ≥2 | 1.11 (0.87, 1.40) | 1.19 (0.90, 1.59) | 1.03 (0.67, 1.59) |
| BMI ≥25 kg/m2 | 0.95 (0.72, 1.25) | 0.87 (0.64, 1.18) | 1.37 (0.76, 2.49) |
| <3 classes of anti-HTN medications | 1.12 (0.90, 1.40) | 1.13 (0.87, 1.47) | 1.19 (0.77, 1.86) |
| Not satisfied with overall healthcare | 1.14 (0.76, 1.71) | 0.95 (0.56, 1.64) | 1.46 (0.83, 2.58) |
| Reduced medications due to cost | 1.93 (1.36, 2.74)*** | 2.23 (1.44, 3.44)*** | 1.68 (1.05, 2.67)* |
| 6+ visits to primary care physician | 1.14 (0.90, 1.46) | 1.02 (0.75, 1.38) | 1.54 (1.04, 2.28)* |
| Current or former smoker | 0.98 (0.78, 1.24) | 0.93 (0.70, 1.24) | 1.09 (0.74, 1.62) |
| ≥ 2 alcoholic beverages per week | 0.98 (0.77, 1.24) | 1.02 (0.77, 1.34) | 0.81 (0.50, 1.30) |
| Depressive symptomsb | 1.16 (0.84, 1.61) | 1.63 (1.14, 2.34)** | 0.58 (0.30, 1.10) |
| Low social support | 1.00 (0.78, 1.27) | 0.91 (0.68, 1.23) | 1.19 (0.81, 1.76) |
| Low coping | 1.12 (0.89, 1.39) | 1.11 (0.85, 1.44) | 1.10 (0.75, 1.62) |
| Complementary and alternative medicine use | 1.13 (0.90, 1.43) | 1.01 (0.76, 1.35) | 1.73 (1.15, 2.60)** |
| ≥2 lifestyle modifications | 0.96 (0.72, 1.27) | 1.02 (0.74, 1.42) | 0.71 (0.39, 1.30) |
| Black | 1.29 (1.00, 1.67) | -- | -- |
p-value for not married-by-race interaction in overall model = 0.038
p-value for depressive symptoms-by-race interaction in overall model = 0.004
CCI – Charlson Comorbidity Index; BMI – Body Mass Index; HTN – hypertension; PR – prevalence ratio; CI – confidence interval; CoSMO – Cohort Study of Medication Adherence among Older Adults
Statistically significant results are indicated by bold text
p<0.05,
p<0.01,
p<0.001
Figure 1:
Prevalence Ratios for Low Antihypertensive Medication Adherence among White and Black Men for Depressive Symptoms and Marital Status
Model adjusted for age, education, hypertension knowledge, hypertension duration, comorbidities, body mass index, number of classes by antihypertensive medications, healthcare satisfaction, reduction in medications due to cost, number of visits to primary physician in past year, smoking status, alcohol use, social support, coping, complementary and alternative medicine use, and adoption of healthy lifestyle modifications
Discussion
In this study, we identified socio-demographic, clinical, psychosocial, behavioral, self-management, and health care system risk factors associated with low pharmacy refill medication adherence in adults with essential hypertension by sex and within sex-race subgroups. The prevalence of low antihypertensive medication adherence was higher among blacks versus whites, overall and when stratified by sex. No sex differences in the prevalence of low adherence were identified. While reducing medications due to cost was associated with low antihypertensive medication adherence across sex-stratified and sex-race-stratified analyses, other risk factors were identified only after sex-race stratification. Furthermore, statistically significant interactions by race among older men revealed important distinctions in black and white men regarding the association between low adherence and not being married, and between low adherence and having depressive symptoms. This study extends prior research by evaluating a comprehensive set of risk factors linked to an objective measure of medication adherence in women and men and stratifying by four sex-race groups. These results may provide insight into sex-race-specific risk factors for low antihypertensive medication adherence in older women and men and lead to development of personalized strategies to address low adherence and improve blood pressure control in older patients.
Risk factors associated with Low Pharmacy Refill Adherence across sex-stratified and sex-race subgroup analyses
Consistent with previous studies,28,29 reducing medications due to cost, a modifiable risk factor, was associated with low PDC adherence in both men and women, and across sex-race subgroups. While the overall percentage of participants who reported reducing their medications due to cost was low (only 3.6% of participants), the association was strong, signaling that targeted medication cost reduction interventions could produce substantial benefits for a subset of patients for whom cost is a barrier. Cost-related underuse of medications for older adults with chronic disease is associated with health care coverage factors (increasing out of pocket costs and inadequate prescription coverage) as well as the quality of the physician - patient relationship.29,30 A survey of Medicare Beneficiaries found that 39% of seniors who reported cost-related non-adherence had not talked with their physicians about this barrier.30 While deficiencies in health care coverage for medications are amenable to changes at the policy level, there may also be opportunities for healthcare providers to address cost-related nonadherence by addressing this barrier during clinic visits and prescribing lower cost (e.g., generic) medications as appropriate.
In our study, blacks reported higher prevalence of cost-related nonadherence than whites. Similarly, Gellad and colleagues (2007) found that issues related to medication affordability were more prevalent among blacks than whites31 and in a focus group study, Holt and colleagues found that black participants elaborated on cost issues more than white participants.32 Thus, special attention to medication cost barriers among black patients may be warranted.
Risk Factors associated with Low Pharmacy Refill Adherence for women
Among all women, having shorter duration of hypertension, having more comorbidities, taking fewer classes of antihypertensive medications, and reporting depressive symptoms were associated with low antihypertensive medication adherence. These results are consistent with previous reports.6,33–36 Furthermore, sex-race stratification revealed that the associations between comorbidities and low antihypertensive medication adherence, and shorter hypertension duration and antihypertensive medication adherence found among all women were driven primarily by white women. While neither number of comorbidities nor duration of hypertension is “modifiable” in the context of a healthcare encounter, considering these risk factors may help providers to identify patients at risk for low antihypertensive medication adherence.
The associations between taking three or more classes of antihypertensive medications and low medication adherence, and reporting depressive symptoms and low adherence that were detected among all women were no longer significant when the analysis was stratified by sex-race. Because the magnitude of these effects were similar for white women and black women, sex-race stratification decreased the sample size and thus reduced the statistical power to detect a difference in prevalence of low adherence between those with and without these risk factors. In the case of depressive symptoms, for example, the prevalence ratios for white women and black women were nearly identical to the prevalence ratio for all women, but the decreased sample size produced larger confidence intervals spanning 1. Thus, for women, evaluating both sex- and sex-race stratification revealed important information about risk factors for low antihypertensive medication adherence.
Risk Factors associated with Low Pharmacy Refill Adherence for men
While only reduction in medications due to cost was associated with low medication adherence in the overall model for men, further stratification by race revealed risk factors unique to white or black men. Depressive symptoms were associated with low adherence among white men; however, this risk factor had a non-significant protective effect for black men (p-value for interaction = 0.004). While previous research documented an overall association between depression and nonadherence in non-stratified analyses,37,38 and an association between depression and nonadherence among women in sex-stratified analyses,9 our finding may provide new insights on sex-race differences. It is possible that aggregating white and black men in a single model obscures an association between depressive symptoms and low antihypertensive medication adherence for white men. Further research is needed to explain this race difference, and to demonstrate whether interventions to address depressive symptoms can improve antihypertensive medication adherence and blood pressure control among older men with comorbid depression and hypertension.
Race also modified the effect of marriage on low antihypertensive medication adherence among men. Specifically, not being married was associated with low medication adherence among white men only (p-value for interaction term = 0.038). The finding for white men is consistent with studies that have reported marriage or spousal support as a protective factor for low medication adherence, poor cardiovascular outcomes, and increased mortality among elderly men.39,40 Having a spouse may discourage unhealthy lifestyle behaviors, such as alcohol consumption, smoking, or substance abuse, or may provide practical support for remembering doctor appointments, pharmacy visits, previous medical history, and assisting with maintaining medication adherence.39,41 While marital status is not modifiable in the context of a medical encounter, it may serve to flag hypertensive patients at risk for low adherence. We were unable to assess the impact of marital quality on low adherence in black or white men. Recent studies have identified marital quality as a determinant of cardiovascular health outcomes, antihypertensive medication adherence and blood pressure control,42 with some health benefits of high relationship quality unique to blacks.43 Future research is needed regarding marital quality and its influence on health behaviors and antihypertensive medication adherence in men with hypertension.
The current study also found a statistically significant association between having six or more visits to a primary physician in the prior year and low antihypertensive medication adherence only among black men. It is possible that those with lower adherence have poorly controlled disease and worse health, requiring them to visit a physician more often.44 Further research is needed to understand this race difference.
The association between CAM use and low antihypertensive medication adherence, which was specific to black men, is consistent with previous research related to race differences in the use of CAM. Previous studies have reported high prevalence of CAM use among blacks,45,46 and in our study, black men reported higher prevalence of CAM use compared to white men (the difference was marginally significant). Using data from the current cohort, Krousel-Wood and colleagues demonstrated that black participants were more likely than their white counterparts to have low adherence to antihypertensive medications when reporting CAM use.46 It is possible that the difference in the association between CAM use and low adherence between blacks and whites can be explained by race differences in the types of CAM used and whether it is used as a complement or an alternative to antihypertensive medication. This was beyond the scope of the current analysis and should be the focus of future research. Our analysis extends previous findings by demonstrating that the race difference in the association between CAM use and low antihypertensive medication adherence was present only among men and highlights an opportunity for providers to communicate with men about their CAM use and any trade-offs with respect to antihypertensive medication adherence.
Added value of sex-race subgroup analysis
Our results among women demonstrate that sex-race stratification can clarify the effect of race on associations between risk factors and low antihypertensive medication adherence identified in sex-stratified analyses. Furthermore, sex-race stratification may uncover risk factors that are not identified in analyses stratified by race or sex alone, as we found in men. Both of these functions enhance our ability to identify those at risk for low medication adherence and develop interventions targeting relevant risk factors for low antihypertensive medication adherence. In addition, sex-race stratification can aid reproducibility of results: inconsistent results in prevalence rates and determinants of low adherence seen across previous studies may be due to differences in the sex-race distribution of participants.
Limitations and Strengths
The findings should be considered in light of the study limitations. The cross-sectional and observational nature of the study does not allow for causal inferences. Although the study included a relatively large sample of elderly black and white participants, the number of black male participants (n=188) was low compared to other sex-race subgroups, which could limit the statistical power of the study to detect significant associations for this subgroup; future studies including larger sample sizes for subgroup analyses are needed to confirm the findings. Pharmacy refill is an indirect measure of adherence and we cannot confirm that the medications were actually taken correctly by patients; however, prior research has found that pharmacy refill adherence is associated with blood pressure control and outcomes.47 Further, some of the patients may have filled prescriptions outside of the managed care organization’s database and these claims data may not have been included in the analysis. In addition, the inclusion of English-speaking community-dwelling older adults in a managed care organization from one geographical area may not be generalizable to all patients with hypertension. While the factors identified in these analyses can be used as markers for low antihypertensive medication adherence, further research is needed to determine whether or not changes in modifiable risk factors associated with low adherence lead to improved adherence and better blood pressure control.
This study has several strengths, including a diverse sample of black and white participants with respect to socio-demographic characteristics and the presence of cardiovascular risk factors. Antihypertensive medication adherence was assessed using an objective measure of PDC, which has been correlated to blood pressure control and cardiovascular disease outcomes.47,48 We were able to link comprehensive participant risk factor survey data to pharmacy refill adherence and medical claims. The restriction of the study sample to older adults in a managed care setting minimizes the confounding effects of health insurance, access to care, and employment status in the older adult population. Finally, because hypertension is a prevalent disease in older adults, the results of this study may be useful in the management of a substantial portion of the elderly, insured population with hypertension.
Conclusion
Differences in prevalence of and risk factors associated with low pharmacy refill adherence were identified across four sex-race subgroups of CoSMO participants. This study extends prior research by highlighting sex-race differences not identified when looking at sex or race alone and by providing insights on the effect of race on associations between low adherence and risk factors identified in sex-stratified analysis. It is possible that these differences may underlie inconsistencies in adherence research results among women versus men. The identification of risk factor differences in sex-race groups may inform future clinical studies and personalized precision health care aimed at improving antihypertensive medication adherence and blood pressure control.
Acknowledgments
Statement of Financial Disclosure: This work was supported, in part, by K12 HD043451 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health (Williams-Building Interdisciplinary Research Careers in Women’s Health (BIRCWH) Scholar; Krousel-Wood-Principal Investigator; Erin Peacock-Data Analyst) and by R01 AG022536 from the National Institute on Aging (Krousel-Wood-Principal Investigator). The corresponding author has also received support from U54 GM104940 and UL1 TR001417. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIA or the National Institutes of Health (NIH). The funding sponsors of this grant did not play a role in the design, methods, subject recruitment, data collections, analysis and/or preparation of 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.
Potential Conflict of Interest: Paul Muntner has consulted for Amgen for work not related to this manuscript.
This work was presented in part at the Health Disparities Conference at Xavier University of Louisiana in New Orleans, Louisiana in 2016 and at the NIH K12 BIRCWH annual conference in Bethesda, Maryland in 2016.
References
- 1.Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart Disease and Stroke Statistics-2017 Update: A Report from the American Heart Association. Circulation. 2017;135(10):e146–e603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Nwankwo T, Yoon SS, Burt V, Gu Q. Hypertension among Adults in the United States: National Health and Nutrition Examination Survey, 2011–2012. NCHS Data Brief. 2013;(133)(133):1–8. [PubMed] [Google Scholar]
- 3.Mills KT, Bundy JD, Kelly TN, et al. Global Disparities of Hypertension Prevalence and Control: A Systematic Analysis of Population-Based Studies from 90 Countries. Circulation. 2016;134(6):441–450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yoon SS, Gu Q, Nwankwo T, Wright JD, Hong Y, Burt V. Trends in Blood Pressure among Adults with Hypertension: United States, 2003 to 2012. Hypertension. 2015;65(1):54–61 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Krousel-Wood MA, Muntner P, Islam T, Morisky DE, Webber LS. Barriers to and Determinants of Medication Adherence in Hypertension Management: Perspective of the Cohort Study of Medication Adherence among Older Adults. Med Clin North Am. 2009;93(3):753–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Holmes HM, Luo R, Hanlon JT, Elting LS, Suarez-Almazor M, Goodwin JS. Ethnic Disparities in Adherence to Antihypertensive Medications of Medicare Part D Beneficiaries. J Am Geriatr Soc. 2012;60(7):1298–1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Choudhry NK, Shrank WH, Levin RL, et al. Measuring Concurrent Adherence to Multiple Related Medications. Am J Manag Care. 2009;15(7):457–464. [PMC free article] [PubMed] [Google Scholar]
- 8.Krousel-Wood M, Thomas S, Muntner P, Morisky D. Medication Adherence: A Key Factor in Achieving Blood Pressure Control and Good Clinical Outcomes in Hypertensive Patients. Curr Opin Cardiol. 2004;19(4):357–362. [DOI] [PubMed] [Google Scholar]
- 9.Holt E, Joyce C, Dornelles A, et al. Sex Differences in Barriers to Antihypertensive Medication Adherence: Findings from the Cohort Study of Medication Adherence among Older Adults. J Am Geriatr Soc. 2013;61(4):558–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pittman DG, Tao Z, Chen W, Stettin GD. Antihypertensive Medication Adherence and Subsequent Healthcare Utilization and Costs. Am J Manag Care. 2010;16(8):568–576. [PubMed] [Google Scholar]
- 11.Aronow WS, Fleg JL, Pepine CJ, et al. ACCF/AHA 2011 Expert Consensus Document on Hypertension in the Elderly: A Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents Developed in Collaboration with the American Academy of Neurology, American Geriatrics Society, American Society for Preventive Cardiology, American Society of Hypertension, American Society of Nephrology, Association of Black Cardiologists, and European Society of Hypertension. J Am Soc Hypertens. 2011;5(4):259–352. [DOI] [PubMed] [Google Scholar]
- 12.Whelton PK. The Elusiveness of Population-Wide High Blood Pressure Control. Annu Rev Public Health. 2015;36:109–130. [DOI] [PubMed] [Google Scholar]
- 13.Bagchi AD, Esposito D, Kim M, Verdier J, Bencio D. Utilization of, and Adherence to, Drug Therapy among Medicaid Beneficiaries with Congestive Heart Failure. Clin Ther. 2007;29(8):1771–1783. [DOI] [PubMed] [Google Scholar]
- 14.Balkrishnan R Predictors of Medication Adherence in the Elderly. Clin Ther. 1998;20(4):764–771. [DOI] [PubMed] [Google Scholar]
- 15.National Institutes of Health. Enhancing Reproducibility through Rigor and Transparency. 2015;Notice: NOT-OD-15–103.
- 16.Bosworth HB, Powers B, Grubber JM, et al. Racial Differences in Blood Pressure Control: Potential Explanatory Factors. J Gen Intern Med. 2008;23(5):692–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Williams MV, Baker DW, Parker RM, Nurss JR. Relationship of Functional Health Literacy to Patients’ Knowledge of their Chronic Disease. A Study of Patients with Hypertension and Diabetes. Arch Intern Med. 1998;158(2):166–172. [DOI] [PubMed] [Google Scholar]
- 18.Sanne S, Muntner P, Kawasaki L, Hyre A, DeSalvo KB. Hypertension Knowledge among Patients from an Urban Clinic. Ethn Dis. 2008;18(1):42–47. [PubMed] [Google Scholar]
- 19.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A New Method of Classifying Prognostic Comorbidity in Longitudinal Studies: Development and Validation. J Chronic Dis. 1987;40(5):373–383. [DOI] [PubMed] [Google Scholar]
- 20.Deyo RA, Cherkin DC, Ciol MA. Adapting a Clinical Comorbidity Index for Use with ICD-9-CM Administrative Databases. J Clin Epidemiol. 1992;45(6):613–619. [DOI] [PubMed] [Google Scholar]
- 21.Meng YY, Jatulis DE, McDonald JP, Legorreta AP. Satisfaction with Access to and Quality of Health Care among Medicare Enrollees in a Health Maintenance Organization. West J Med. 1997;166(4):242–247. [PMC free article] [PubMed] [Google Scholar]
- 22.Radloff LS. The Use of the Center for Epidemiologic Studies Depression Scale in Adolescents and Young Adults. J Youth Adolesc. 1991;20(2):149–166. [DOI] [PubMed] [Google Scholar]
- 23.Stewart AL, Hays RD, Ware JE Jr. The MOS Short-Form General Health Survey. Reliability and Validity in a Patient Population. Med Care. 1988;26(7):724–735. [DOI] [PubMed] [Google Scholar]
- 24.Fernander AF, Duran RE, Saab PG, Llabre MM, Schneiderman N. Assessing the Reliability and Validity of the John Henry Active Coping Scale in an Urban Sample of African Americans and White Americans. Ethn Health. 2003;8(2):147–161. [DOI] [PubMed] [Google Scholar]
- 25.Cohen S, Kamarck T, Mermelstein R. A Global Measure of Perceived Stress. J Health Soc Behav. 1983;24(4):385–396. [PubMed] [Google Scholar]
- 26.Lengacher CA, Bennett MP, Kipp KE, Berarducci A, Cox CE. Design and Testing of the Use of a Complementary and Alternative Therapies Survey in Women with Breast Cancer. Oncol Nurs Forum. 2003;30(5):811–821. [DOI] [PubMed] [Google Scholar]
- 27.Centers for Disease Control and Prevention. 2003–04 Blood Pressure Questionnaire-BPQ_C. https://www.cdc.gov/nchs/data/nhanes/nhanes_03_04/sp_bpq_c.pdf. Accessed February 15, 2008.
- 28.Brown CM, Segal R. The Effects of Health and Treatment Perceptions on the Use of Prescribed Medication and Home Remedies among African American and White American Hypertensives. Soc Sci Med. 1996;43(6):903–917. [DOI] [PubMed] [Google Scholar]
- 29.Piette JD, Beard A, Rosland AM, McHorney CA. Beliefs that Influence Cost-Related Medication Non-Adherence among the “Haves” and “Have Nots” with Chronic Diseases. Patient Prefer Adherence. 2011;5:389–396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wilson IB, Rogers WH, Chang H, Safran DG. Cost-Related Skipping of Medications and Other Treatments among Medicare Beneficiaries between 1998 and 2000. Results of a National Study. J Gen Intern Med. 2005;20(8):715–720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.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):1572–1578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Holt EW, Rung AL, Leon KA, Firestein C, Krousel-Wood MA. Medication Adherence in Older Adults: A Qualitative Study. Educ Gerontol. 2014;40(3):198–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.DiMatteo MR, Lepper HS, Croghan TW. Depression is a Risk Factor for Noncompliance with Medical Treatment: Meta-Analysis of the Effects of Anxiety and Depression on Patient Adherence. Arch Intern Med. 2000;160(14):2101–2107. [DOI] [PubMed] [Google Scholar]
- 34.Park YH, Kim H, Jang SN, Koh CK. Predictors of Adherence to Medication in Older Korean Patients with Hypertension. Eur J Cardiovasc Nurs. 2013;12(1):17–24. [DOI] [PubMed] [Google Scholar]
- 35.Armario P, Waeber B. Therapeutic Strategies to Improve Control of Hypertension. J Hypertens. 2013;31 Suppl 1:S9–12. [DOI] [PubMed] [Google Scholar]
- 36.Jankowska-Polanska B, Chudiak A, Uchmanowicz I, Dudek K, Mazur G. Selected Factors Affecting Adherence in the Pharmacological Treatment of Arterial Hypertension. Patient Prefer Adherence. 2017;11:363–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Muntner P, Judd SE, Krousel-Wood M, McClellan WM, Safford MM. Low Medication Adherence and Hypertension Control among Adults with CKD: Data from the REGARDS (Reasons for Geographic and Racial Differences in Stroke) Study. Am J Kidney Dis. 2010;56(3):447–457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Krousel-Wood M, Islam T, Muntner P, et al. Association of Depression with Antihypertensive Medication Adherence in Older Adults: Cross-Sectional and Longitudinal Findings from CoSMO. Ann Behav Med. 2010;40(3):248–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Marriage Simeonova E., Bereavement and Mortality: The Role of Health Care Utilization. J Health Econ. 2013;32(1):33–50. [DOI] [PubMed] [Google Scholar]
- 40.Ng TP, Jin A, Feng L, et al. Mortality of Older Persons Living Alone: Singapore Longitudinal Ageing Studies. BMC Geriatr. 2015;15:126-015-0128-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Mayberry LS, Osborn CY. Empirical Validation of the Information-Motivation-Behavioral Skills Model of Diabetes Medication Adherence: A Framework for Intervention. Diabetes Care. 2014;37(5):1246–1253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Trevino DB, Young EH, Groff J, Jono RT. The Association between Marital Adjustment and Compliance with Antihypertension Regimens. J Am Board Fam Pract. 1990;3(1):17–25. [PubMed] [Google Scholar]
- 43.Grewen KM, Girdler SS, Light KC. Relationship Quality: Effects on Ambulatory Blood Pressure and Negative Affect in a Biracial Sample of Men and Women. Blood Press Monit. 2005;10(3):117–124. [DOI] [PubMed] [Google Scholar]
- 44.Calderon-Larranaga A, Diaz E, Poblador-Plou B, Gimeno-Feliu LA, Abad-Diez JM, Prados-Torres A. Non-Adherence to Antihypertensive Medication: The Role of Mental and Physical Comorbidity. Int J Cardiol. 2016;207:310–316. [DOI] [PubMed] [Google Scholar]
- 45.Sternberg SA, Chandran A, Sikka M. Alternative Therapy Use by Elderly African Americans Attending a Community Clinic. J Am Geriatr Soc. 2003;51(12):1768–1772. [DOI] [PubMed] [Google Scholar]
- 46.Krousel-Wood MA, Muntner P, Joyce CJ, et al. Adverse Effects of Complementary and Alternative Medicine on Antihypertensive Medication Adherence: Findings from the Cohort Study of Medication Adherence among Older Adults. J Am Geriatr Soc. 2010;58(1):54–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Krousel-Wood M, Holt E, Joyce C, et al. Differences in Cardiovascular Disease Risk When Antihypertensive Medication Adherence is Assessed by Pharmacy Fill Versus Self-Report: The Cohort Study of Medication Adherence among Older Adults (CoSMO). J Hypertens. 2015;33(2):412–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Chowdhury R, Khan H, Heydon E, et al. Adherence to Cardiovascular Therapy: A Meta-Analysis of Prevalence and Clinical Consequences. Eur Heart J. 2013;34(38):2940–2948. [DOI] [PubMed] [Google Scholar]