Effectiveness of interventions to reduce mortality, morbidity, and emergency service demand during extreme heat: a systematic review with meta-analyses

Background As climate change increases the frequency and intensity of extreme heat, effective interventions are needed to reduce heat-related mortality, morbidity, and demand for emergency services. We systematically reviewed the effectiveness of interventions to reduce adverse health outcomes during periods of high temperature. Methods We searched PubMed, Embase, Web of Science, and Google Scholar from inception to 09.02.2026, and grey literature. Randomised and nonrandomised intervention studies assessing the effect of heat-protection interventions on mortality, morbidity and emergency demand in the general population or in heat-vulnerable groups were included. Risk of bias was assessed with the JBI checklist. Following a pre-registered protocol which was co-produced with an international expert panel (CRD420251020849), we conducted a systematic review and random-effects meta-analysis where studies were sufficiently comparable; otherwise, findings were synthesised narratively. Findings Of 32,149 records screened, 49 studies were included. Most studies evaluated complex interventions, particularly heat-health action plans (HHAP; n=18) and heat warning systems (HWS=12). Fewer studies assessed urban and building planning measures, education and awareness campaigns, community interventions, occupational interventions, and emergency preparedness measures. Mortality was the most frequently reported outcome. In meta-analysis of pre–post studies, HHAP implementation was associated with reduced mortality risk (RR=0·93 [95% CI 0·89; 0·99], p < 0 ·05). Evidence for HWS alone was inconsistent. Three studies suggested that air conditioning in public institutions was associated with lower mortality risk than no air conditioning. Evidence for other interventions was limited or inconsistent. Interpretation Comprehensive heat–health action plans appear to reduce heat-related mortality, but evidence for warning systems alone and other intervention types remains limited or inconsistent. More rigorous evaluations are needed, particularly for morbidity, emergency service demand, and implementation in low-resource settings.