Abstract
BACKGROUND: Identifying and addressing barriers to antihypertensive medication adherence is critical for hypertension control. Relatively little is known about how social determinants of health (SDoH) influence adherence, particularly when measurable from linked electronic health record and pharmacy dispensing data, respectively. METHODS: We conducted a secondary analysis of a pragmatic trial of patients with uncontrolled hypertension (≥140/90 mm Hg) across 12 primary care clinics in a large healthcare system in Illinois, leveraging linked data on prescription dispenses and SDoHs recorded in electronic health record data by clinic staff. We examined 3 SDoHs thought to be related to hypertension: transportation needs, financial resource strain, and food insecurity, and measured adherence using the proportion of days covered over a 1-year follow-up period. We evaluated relationships between each SDoH and being adherent (proportion of days covered ≥80%) using logistic regression after 1:3 propensity score matching for baseline covariates. In secondary analyses, we fit linear regressions to estimate the association between each SDoH exposure and the absolute proportion of days covered. RESULTS: In total, 5386 patients were included (mean [SD] age, 65.9 [12.3] years, 63.5% female, 49.2% Black race); 2.8%, 3.4%, and 6.8% reported transportation needs, financial resource strain, and food insecurity, respectively. Only 2955 (54.9%) of patients were adherent. After matching, financial resource strain was associated with 45% lower odds (odds ratio, 0.55 [95% CI, 0.39-0.79]) of being adherent to antihypertensive medications versus not experiencing financial resource strain. Transportation needs and food insecurity were not significantly associated with being adherent (odds ratio, 0.94 [95% CI, 0.61-1.45]; odds ratio, 0.89 [95% CI, 0.69-1.16], respectively). For the secondary outcome, after matching, patients reporting financial resource strain (-5.8% [95% CI, -10.4% to -1.2%]) also had a significantly lower proportion of days covered than those who did not. CONCLUSIONS: SDoH may have important associations with antihypertensive medication adherence. Identifying and addressing social needs in primary care practice may help improve hypertension control and health outcomes.