Home-Based Intervention Drops Blood Pressure in Rural South Africa

MADRID, Spain—Leaning on the support of community health workers (CHWs) and the ability of nurses to prescribe medications helped bring blood pressure under control among adults with hypertension in rural South Africa, according to the results of the IMPACT-BP trial.

The effort, with or without the automated transfer of data from home BP monitors to primary care clinics, provided a greater reduction in mean systolic readings by 6 months compared with standard care—by 8 to 9 mm Hg, Thomas Gaziano, MD (Mass General Brigham, Boston, MA), reported earlier this week at the European Society of Cardiology Congress 2025.

By 1 year, the reduction had risen to about 10 mm Hg. Rates of hypertension control—to below 140/90 mm Hg—increased dramatically as well.

“In rural South Africa in a very impoverished community, we were able to reduce blood pressure up to 10 mm Hg, persistent through 12 months of the study, and increase control of blood pressure by 20 to 30%,” Gaziano said.

The findings were published simultaneously in the New England Journal of Medicine.

The IMPACT-BP Trial

Globally, Gaziano noted, only 56% of individuals with hypertension are aware they have high BP, 31% are receiving treatment for it, and 18% have it under control, with even worse numbers in certain parts of the world where there are greater challenges accessing healthcare.

In South Africa, for example, the waiting time to see a provider within the public healthcare system is up to 3.5 hours, with individuals having to deal both with travel costs and time lost at work, he said.

“The staff are overworked,” said Gaziano. “Most of the clinics and primary care clinics in South Africa in rural areas are not serviced by physicians, but by nurses only. They’re extremely overworked with the burden of chronic disease as well as infectious disease, and people living with HIV in particular.”

The South African government has been interested in lessening the crowding of clinics and moving care into the communities, he said. That has been done for antiretroviral therapy for HIV, and Gaziano’s team wanted to test whether it was achievable for hypertension management.

IMPACT-BP was conducted in the uMkhanyakude district of the KwaZulu-Natal province of South Africa, with investigators recruiting 774 adults with hypertension (BP > 140/90 mm Hg on two measurements separated by more than 6 months) from primary care clinics. The mean age of the participants was 62 years, and 76% were women. At baseline, mean systolic BP was 147 mm Hg, with 20.2% of patients having a reading of at least 160 mm Hg. Nearly half (46.5%) were HIV-positive and only a minority had running water in their home (14.5%) or were employed (11.2%).

The patients were randomized to three arms:

• Standard of care: Participants would go to the clinic, where a nurse would measure BP and prescribe medications, if necessary. Patients would go to the pharmacy to pick up the antihypertensive drugs.
• CHW-led care: Participants would get a home-based BP monitor and a trained CHW would ensure that it was working and record measurements off the device. Those readings would be entered into a clinical decision support tool. Nurses would examine the data and prescribe antihypertensives. The CHW would then take the medications to patients’ homes.
• CHW-plus care: This was similar to the CHW-led model, but with a home BP monitor equipped with a cuff designed to automatically transfer data to the nurse at the clinic. CHWs would still visit patients to deliver medications and assess adherence.

The primary outcome was the change in systolic BP at 6 months, and that favored both CHW arms of the trial. At that time, average systolic BP was 145.8 mm Hg with standard care, 137.5 mm Hg with CHW-led care, and 136.5 mm Hg with CHW-plus care (P < 0.001 for both CHW arms versus standard of care). Those figures were 144.8, 134.1, and 134.0 mm Hg, respectively, at 1 year.

The rate of BP control to a goal of less than 140/90 mm Hg at 6 months was 57.6% with standard care, 76.9% with CHW-led care, and 82.8% with CHW-plus care. The proportions in the two intervention arms increased—to 82.8% and 85.7%—by 1 year.

There were no adverse events deemed to be related to the study.

‘Robust Proof-of-Concept’

The discussant for the study, Tazeen Jafar, MD (Duke-NUS Medical School, Singapore, and Duke Global Health Institute, Durham, NC), said an important finding is that the added electronic element—automated data transfers from the BP cuff—did not enhance the BP reductions obtained with CHW-led care.

“The overall message from this [is] that interaction of humans with the patients was key to improving blood pressure control” in the trial, she said.

Similar multicomponent, CHW-led interventions in other low- and middle-income countries—like the one studied in COBRA-BPS, which Jafar led, for instance—have provided generally consistent results, with some variation in the magnitude of the BP impact, she added.

The mean reduction in systolic BP was about 5 mm Hg with the intervention tested in COBRA-BPS, and the smaller effect compared with IMPACT-BP could be related to how care was delivered, Jafar suggested. In COBRA-BPS, the intervention was delivered by CHWs who worked for the public health sector, whereas in IMPACT-BP, the CHWs were hired specifically for the trial and were not performing other tasks. In addition, medications were not delivered for free in COBRA-BPS.

“I think when IMPACT-BP is scaled up and integrated into the health sector, one would expect to see attenuation of the benefit that we are seeing in the standalone program,” Jafar said.

Another challenge for wider applicability of the approach studied in IMPACT-BP is that nurses were responsible for prescribing medications, something that is not available in many countries. And it will be important moving forward to move recruitment of patients beyond the clinics, where participants were enrolled for IMPACT-BP, Jafar said, noting that many people with hypertension are not seeking care at healthcare facilities.

These issues aside, IMPACT-BP was an excellent study that “fills a large knowledge gap by providing robust proof-of-concept evidence regarding the success of this community health worker-led intervention in the region,” she said. “Community health workers are a readily available resource in many low- and middle-income countries, and they can be leveraged upon to scale up the intervention.”

She added that this type of intervention could have relevance, too, for underserved populations in higher-income countries.