Physical activity intervention for blood pressure reduction in low and middle-income countries: a systematic review with meta-analysis

Background Hypertension is a leading cause of cardiovascular disease and death, and most people with hypertension live in low-and middle-income countries (LMICs). Trials of physical activity (PA) for blood pressure (BP) are primarily conducted in high-income settings. We therefore summarised the evidence from LMICs separately. Methods The review followed PRISMA 2020 (PROSPERO CRD420261325274). We searched PubMed, MEDLINE, CINAHL, the Cochrane Library and Academic Search Complete from inception to February 2026. Adults with elevated BP or hypertension who received a PA intervention in a country classed as low-or middle-income by the World Bank were eligible. To avoid combining different estimands, we analysed four groups of studies separately: chronic controlled trials with office BP (primary analysis), chronic controlled trials with 24-h ambulatory BP, uncontrolled pre–post studies, and acute crossover trials. We used random-effects models (REML with Hartung–Knapp adjustment), and rated certainty using the GRADE. Results Forty-five studies were included, and 40 (2,304 participants) contributed to a meta-analysis. In chronic controlled trials, PA reduced office systolic BP (SBP) relative to control (17 studies, 1,571 participants; MD −10.23 mmHg, 95% CI −12.56 to −7.89; 95% PI −18.91 to −1.55; moderate certainty) and diastolic BP (DBP) (16 studies; MD −4.68 mmHg, 95% CI −6.11 to −3.25; 95% PI −9.55 to 0.18; low certainty). Heterogeneity was high ( I 2 = 83% and 79%). Effects did not differ clearly by intervention type, study design or region. In six trials with 24-h ambulatory BP, the effects were smaller (SBP −2.50 mmHg, 95% CI −5.76 to 0.76; DBP −1.61 mmHg, −2.77 to −0.45; low certainty). Uncontrolled pre–post studies and acute crossover trials gave supporting evidence of very low certainty. We found no evidence of small-study effects in the primary analysis (Egger p = 0.29 for SBP and 0.94 for DBP), although the test has limited power with this number of studies. Conclusion Structured PA programmes lower office BP in adults with hypertension in LMICs, by about 10 mmHg systolic and 5 mmHg diastolic compared with control. The size of the effect varies across settings, and the evidence for ambulatory BP and for longer-term, community-delivered programmes remains limited.