Abstract
OBJECTIVES: Although several studies have investigated the effects of temperature on the incidence of tuberculosis (TB) in a single city or region, few studies have investigated the variations in this association using nationwide data. This study aimed to quantify the association between temporal variations in TB incidence and temperature across Japan. METHODS: The data on the weekly number of newly confirmed TB cases and meteorological variables in 47 Japanese prefectures from 2007 to 2019 were collected. The exposure-response relationships between TB incidence and temperature were quantified using a distributed lag nonlinear model for each prefecture, and estimates from all prefectures were then pooled using a meta-regression model to derive nationwide average associations. RESULTS: This study included 335,060 patients with TB. Compared to those with minimum risk temperature on TB incidence (10(th) percentile at 4.45°C), people who were exposed to the highest temperature concentrations had a 52.0% (relative risk 1.52, 95% confidence interval 1.04-2.23) higher risk for TB incidence at the 99(th) percentile (30.1°C). Our results also emphasized the heterogeneity of these associations in different prefectures. Conclusions: Strengthening monitoring and public health strategies aimed at controlling temperature-related TB may be more effective when tailored to region-specific meteorological conditions.