Associations of short-term ambient temperature and directional temperature change with cardiovascular admissions among older adults: a time-stratified case-crossover study

Older adults are particularly vulnerable to the cardiovascular effects of short-term ambient temperature exposure. However, evidence remains limited on the lagged, non-linear associations of daily mean temperature and short-term directional temperature change with cardiovascular hospital admissions among large populations of older adults. We conducted a time-stratified case-crossover study among adults aged 65 years or older in Henan Province, China, from 2019 to 2024. Cardiovascular disease (CVD) admissions were identified from inpatient discharge records and linked to elderly health management rosters. Environmental exposures were assigned using the geocoded locations of participants’ affiliated primary healthcare institutions as exposure-location proxies. Distributed lag non-linear models were used to estimate cumulative and lag-specific associations between daily mean temperature and CVD admissions over lag days 0–14. We additionally examined short-term directional temperature change within the lag 0–3-day window after adjustment for concurrent mean temperature, relative humidity, and fine particulate matter (PM₂.₅). A total of 8,876,140 older adults were identified, of whom 8,875,258 were assigned exposure-location proxies. The primary CVD analysis included 2,598,149 admission events and 11,424,206 case-day and control-day records. The cumulative temperature-admission curve was non-linear. Relative to the temperature at which the fitted cumulative exposure-response function was minimized (28.5 °C), the odds ratio (OR) was 1.032 (95% confidence interval (CI), 1.009–1.056) at − 3.6 °C and 1.064 (95% CI, 1.045–1.084) at 0.2 °C; the fitted maximum was approximately 14.2 °C (OR, 1.194; 95% CI, 1.177–1.213), and the OR was 1.015 (95% CI, 1.003–1.027) at 32.0 °C. At 0.2 °C, the association emerged after approximately 2 days and persisted through lag day 14, whereas the association at 32.0 °C was smaller and more immediate. After adjustment for concurrent mean temperature, directional warming was also associated with CVD admissions (OR, 1.030; 95% CI, 1.023–1.038 at + 8.5 °C). Temperature-admission associations varied across the temperature range and by lag, whereas the directional-warming association was small and should be considered hypothesis-generating. Because absolute admission risk and attributable burden were not estimated, the clinical and public health relevance remains uncertain.