Abstract P1146: A Multifaceted Blood Pressure Intervention Improves Medication Adherence in Low Income Adults with Hypertension

Abstract

Methods: A secondary analysis from the I mplementation of M ultifaceted Pa tient- C entered T reatment S trategies for Intensive BP Control (IMPACTS) cluster randomized trial in federally qualified health centers in the Gulf South was done. Participants were randomized to intervention (standardized BP measurement education, stepped-care treatment algorithm, team-based care, health coaching, home BP monitoring) versus EUC. Adherence was measured with the Krousel-Wood Medication Adherence Scale-4 item (K-Wood-MAS-4; range 0-4); score<1=high adherence. SBP&diastolic BP (DBP) were assessed with a standard protocol using automated BP devices. A multivariable mixed effects logistic regression model with group-by-time interaction estimated predicted probabilities of high adherence for each group at each time point; between-group differences were assessed. A multivariable mixed effects linear regression model estimated difference in SBP for high versus low adherence across all time points. Models were adjusted for baseline socio-demographics, SBP, hypertension duration, number antihypertensive drugs, and comorbidities. Results: Data from 1,272 hypertensive adults from 36 clusters (mean cluster size 35.3, SD 10.6, range 21-60) were analyzed: mean age was 58.3 years (SD 8.9), 56.7% female, 66.5% Black; 62.9% household income <$20K. At baseline, mean SBP&DBP were 147.6 mm Hg (SD 12.5) and 84.6 mm Hg (SD 10.9), respectively; mean K-Wood MAS-4 score was 0.82 (SD 0.89); only 44.3% reported high adherence. Adjusted percent of high adherence at baseline, 6-, 12-&18 months: Intervention group: 43% (95% CI 39%, 47%), 53% (95% CI 48%, 57%), 53% (95% CI 49%, 58%)&54% (95% CI 50%, 59%), respectively; EUC group : 46% (95% CI 42%, 50%), 44% (95% CI 40%, 48%), 48% (95% CI 44%, 53%)&55% (95% CI 50%, 59%), respectively. Between-group differences in probability of high adherence were higher in intervention versus EUC groups at 6 (p=0.001)&12 months (p=0.018) but not 18 months (p=0.463). Across all visits, high versus low adherence was associated with mean 4.35 mm Hg lower SBP (p<0.001). Conclusion: The multifaceted intervention versus EUC increased adherence at 6M and 12M post randomization; higher adherence was associated with lower SBP.