A prospective cohort study of the SEARCH integrated HIV/hypertension community health worker-led intervention in rural Kenya and Uganda

Introduction

Clinic-based hypertension screening and treatment for people with and without HIV depends on consistent clinic engagement. Retention is challenging in rural areas, especially for people with severe hypertension, which typically requires more frequent visits than clinically stable HIV. We hypothesised that Ministry of Health (MoH) community health workers (CHWs) could improve severe hypertension detection and treatment through an integrated hypertension/HIV intervention.

Methods

In rural Uganda and Kenya, we added HIV testing and a status-neutral hypertension intervention to CHW workflow in an ongoing cluster-randomised population-level study (SEARCH:NCT05768763). Data spans March 2023–August 2024. Trained CHWs screened all adults aged ≥ 40 years in intervention communities for hypertension, referring those with blood pressure (BP) ≥ 140/90 mmHg to MoH HIV/primary care clinics. After initial in-clinic evaluation, adults with BP ≥ 160/100 mmHg were offered choice of clinic-based or telehealth (CHW home visit, clinician telehealth evaluation, medication delivery) follow-up care. Telehealth used a MoH-compatible CHW smartphone app that syncs with electronic clinic records, prompts CHW follow-up visits and facilitates clinician telehealth assessment/medication prescribing. We report hypertension control achieved through the implementation of CHW-supported screening and telehealth and used targeted minimum loss-based estimation to estimate the change in population prevalence of uncontrolled hypertension from baseline to 1 year.

Results

Across eight communities, 198 CHWs measured BP in 14,378/15,879 adults aged ≥ 40 years at baseline (91%) and 13,334/15,879 after 1 year (84%); 55% were female and 19% living with HIV. Estimated population prevalence of BP ≥ 140/90 mmHg decreased from 16.0% at baseline to 6.4% at year 1 (9.6% absolute decrease, 95% CI 8.6%, 10.6%). Among people with HIV aged ≥ 40 years (n = 3036), the prevalence of BP ≥ 140/90 mmHg decreased from 10.5% to 4.7% (5.9% absolute decrease, 95% CI 3.0%, 8.8%). In the subset with BP ≥ 160/100 who enrolled in the intervention (n = 919), 96% received antihypertensive medication, 81% were retained in care at 1 year and 79% achieved BP control; people with HIV (n = 120) had similar retention (80%) and BP control (80%).

Conclusions

Within the context of a pragmatic trial, leveraging existing CHWs in an integrated HIV/hypertension model reduced the population-level prevalence of uncontrolled hypertension by 60% among people with and without HIV, extending health services into the community at scale.