Multicomponent Blood Pressure Control Program in Medicare Beneficiaries with Hypertension Primarily of Hispanic Origin in a Large Health System in South Florida

Background Health care systems play a key role in hypertension control in large populations, which remains challenging. This study assessed: (a) blood pressure (BP) control rate changes pre-vs. post-implementation of a hypertension control program in the understudied Hispanic population in South Florida; and (b) compared annual BP control rate agreement rates calculated using time-average BP (AVG-BP) vs. BP from most recent visit (LAST-BP). Methods Leon Medical Centers (LMC), a major integrated healthcare services provider to Medicare and dual eligible patients in Miami-Dade, implemented a multicomponent hypertension control program in 2011. Electronic health records (EHRs) data from patients aged 65-89 years with hypertension were analyzed. Longitudinal analysis involved calculation of age- and diabetes-standardized annual BP control rates, and random effects modeling. Systolic<140 mmHg and diastolic<90 mmHg defined controlled BP. Kappa statistic was calculated. Results From 2008 (n=4,710) to 2018 (n=21,540), mean age ± SD increased (73.4±5.3 to 77.1±6.1). Proportion of Hispanics remained constant (>98%). BP increased substantially, similarly in women and men post-program implementation; e.g., in women, standardized control rates (AVG-BP) increased ranged from 68.9-73.3% in 2008-2010 to 88.8-92.3% in 2013-2018. Improvement observed using LAST-BP was analogous, though latter absolute rates were systematically lower. Kappa between AVG-BP- and LAST-BP-derived rates in 2018 was 0.296. Conclusions Program implementation was associated with a meaningful, sustained increase in BP control up to high levels. Agreement between AVG-BP and LAST-BP control rates was only fair. Whether longitudinal EHRs longitudinal BP data-derived metrics may offer added-value in health services-related assessments remains to be determined.