Objective To assess the attributable burden of non-optimal temperatures to causes of death from circulatory system diseases in Xi'an, China, 2015—2019, and explore the sensitivity of disease subtypes and vulnerable populations to non-optimal temperatures.

Methods We collected the meteorological data, air pollutant data, and mortality data of permanent residents in Xi'an from 2015 to 2019. A distributed lag non-linear model was used to quantify the exposure-response relationships between daily mean temperature and death from various subtypes of circulatory diseases. Stratified analyses were conducted by age and sex. Attributable risks were calculated using the forward perspective method based on the distributed lag non-linear model.

Results A total of 133 362 circulatory disease deaths were recorded in Xi'an from 2015 to 2019, with 30 627 excess deaths associated with non-optimal temperatures. The overall attributable fraction (AF) was 22.96%(95%CI: 15.30%-28.63%), with 20.08%(95%CI: 12.47%-26.49%) for low temperatures and 2.88%(95%CI: 2.18%-3.49%) for high temperatures. Among major disease categories, the AFs for non-optimal temperatures were 27.70%(95%CI: 18.72%-34.61%) for cerebrovascular diseases and 19.07%(95%CI: 9.12%-27.83%) for ischemic heart disease. Specifically, hemorrhagic stroke had the highest AF (37.72%), particularly among females (43.12%) and individuals aged < 65 years (44.96%). The AF for ischemic stroke was 23.41%. Acute myocardial infarction had an AF of 19.61%, with higher values in females (23.66%) and individuals aged ≥65 years (21.65%). Chronic ischemic heart disease had an AF of 16.05%, with a higher value in males (24.23%).

Conclusion Non-optimal temperatures were identified as significant environmental risk factors for circulatory disease mortality in Xi'an, with the attributable burden predominantly associated with low temperature exposure. Vulnerable populations varied across disease subtypes.