Objective To analyze the effects of cold and heat waves on the risk of non-accidental hospital admissions in Nanjing, China, from 2019 to 2021.

Methods The daily number of non-accidental hospital admissions in four tertiary hospitals(2 grade IIIA hospitals, and 2 grade III comprehensive hospitals) in Nanjing and the corresponding meteorological and air pollution data were collected from January 1, 2019 to December 31, 2021. The distributed lag non-linear model(DLNM) was performed to analyze the effects of cold and heat waves on the risk of non-accidental hospital admissions(including gender, age, and disease groups) of residents.

Results A total of 137 369 non-accidental admissions were collected in the cold season and 141 313 in the warm season. It was found that an increased risk of non-accidental hospital admissions(in gender, age, cardiovascular disease, respiratory disease, and pneumonia groups) was significantly associated with cold waves, however, the differences in gender, age, and disease groups were not statistically significant. The association between heat waves and increased risk of non-accidental hospital admissions was significant only in female (relative risk (RR)=1.207, 95%CI: 1.043, 1.396) and hospital admissions due to transient ischemic attacks and related syndromes(RR=1.473, 95%CI: 1.023, 2.120). The lag risk of non-accidental hospital admissions due to cold waves was the highest on day 0, and the cumulative lag risk showed an upward trend with statistical significance on days 0 to 21. The lag risk and cumulative lag effects caused by heat waves showed no obvious trend, and most of them were not significant.

Conclusion Cold waves in Nanjing from 2019 to 2021 have a more significant effect on the risk of non-accidental hospital admissions than heat waves, which are accompanied by some lag risks and cumulative lag effects. It is suggested to pay more attention to population health during and after cold waves.