Physicians' Academy for Cardiovascular Education

Community-based intervention reduces BP in South Asian adults with hypertension

A Community-Based Intervention for Managing Hypertension in Rural South Asia

Literature - Jafar TH, Gandhi M, de Silva HA et al., - N Engl J Med. 2020. doi: 10.1056/NEJMoa1911965.

Introduction and methods

CV risk can be reduced with treatment of hypertension, still less than one third of people with hypertension have controlled BP [1-3]. Case fatality rates for CV disease are highest in rural areas in low and middle-income countries, where uncontrolled BP is particularly prevalent [4, 5]. An intervention of training of physicians and home health education by community health workers was cost-effective and lowered BP [6, 7]. However, this intervention would not be sustainable or scalable, because it was not integrated into the existing community infrastructure.

The present study evaluates the effectiveness of a scalable, multicomponent intervention, designed specifically for hypertension management in rural areas [8], compared to usual care. A cluster-randomized, controlled trial (Control of BP and Risk Attenuation–Bangladesh, Pakistan, and Sri Lanka [COBRA-BPS]) was performed in 30 rural communities, with a cluster for randomization consisting of 250 to 300 households, in Bangladesh, Pakistan and Sri Lanka over a period of 2 years. Included were participants with age ≥40 and hypertension, defined as current use of antihypertensive medications or persistently elevated BP (systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg) based on each set of the last two of three measurements from 2 separate days. Clusters of 250-300 households were randomized to the multicomponent intervention (n=1330) or usual care by the community center (n=1315).

The multicomponent intervention consisted of five components: (1) BP monitoring and the use of checklists to guide monitoring and referral to physicians, (2) home health education by government community health workers, (3) training of physicians in BP monitoring, management of hypertension, and use of the checklist, (4) a designated hypertension triage reception desk and hypertension care coordinator at the government clinics, and (5) compensation for additional health services and targeted subsidies. The prespecified primary outcome was the mean change in systolic BP from baseline to 24 months. The mean of the second and third BP readings was used for all analyses, and the first was discarded.

Main results


A community-integrated multicomponent intervention including training of physicians and home health education by community health workers led to a greater reduction in BP than usual care in adults with hypertension in rural communities in South Asian countries.

Editorial comment

Neil Poulter lists several advantages of the study by Jafar et al. He emphasizes the generalizability of the findings (at least to South Asia) due to the observed BP reduction across different countries. Poulter also emphasizes the exemplary 90% retention rate among recruits and the unique and inexpensive pragmatic incorporation of hypertension management into the existing community infrastructure. The observed relative reduction of 5.2 mm Hg in systolic BP could underestimate the benefits, because usual care received in the study was superior to care provided in many parts of low- and middle-income countries [2, 9].

He continues by mentioning some limitations. There is no mention in the study of the percentage of people, among the total of eligible patients, who agreed to participate, which could hide a biased sample. Poulter further notes that almost two thirds of the participants were female and that 42% of the participants had chronic kidney disease, which may affect generalizability of the findings.

The change in mean number of antihypertensive medication (0.11 higher in the intervention group) is expected to generate very little BP lowering. Poulter notes that, instead of reporting the 6.3 mg daily dose increase, it would have been more helpful to report changes in the proportions of mean standard doses used.

Jafar et al. do not discuss potential benefits of nonpharmacologic measures or relative benefits of specific drugs that were used in both groups, which may have been critical. It is also not clear what the “compensation for additional health services and targeted subsidiaries” consisted of, which might have been critical to the success of the intervention. Finally, according to Poulter, concealing randomization status from the community health workers during follow-up, potentially linked to the compensation costs, must have been extremely difficult, particularly if they measured BP and provided health education.


Show references

Find this article online at N Engl J Med. Find the editorial article online at N Engl J Med.

Share this page with your colleagues and friends: