Abstract
The combined impact of nonoptimal temperatures and air pollution on hospitalizations remains understudied. This study investigated how major criterion air pollutants (PM(2.5), NO(2), and O(3)) modified the short-term temperature-hospitalization associations and quantified the excess attributable burden. Daily time-series data of emergency hospitalizations and environmental exposures (2000-19) in Hong Kong were analyzed using two comparative models with product terms (1) between temperature and lower and higher strata for pollutants, and (2) between temperature and the count of elevated pollutants. Over 10 million noncancer nonexternal (NCNE) admissions occurred, including ∼14% circulatory and ∼20% respiratory admissions. PM(2.5) consistently amplified temperature effects, leading to 1.5-2.6% significantly higher NCNE risks with low and high temperatures and 2.8-3.7% higher circulatory risks with low temperatures. NO(2) also intensified circulatory risks. Temperature effects were 1.4-2.6% higher on polluted days (≥1 elevated pollutants) than on unpolluted days (0 elevated pollutants), with positive linear trends as pollutant counts increased (p-trend <0.05). The observed temperature-attributable numbers (fraction, %) without considering pollution modification were 68,032 (0.7%), 44,290 (3.2%), and 39,995 (2.0%) for NCNE, circulatory, and respiratory admissions, respectively. These could be reduced by 0.5-0.8% in counterfactual low-pollution scenarios. Combined air pollution exposure exacerbates temperature-related hospitalization risks in Hong Kong, particularly for NCNE and circulatory admissions.