The burden of hand, foot, and mouth disease attributable to climate variability: A nationwide time-series modeling study in Japan.

Substantial evidence suggests that meteorological variables are associated with hand, foot, and mouth disease (HFMD) incidence; however, geographic variation in these associations and in the corresponding attributable burden remains insufficiently characterized. We conducted a nationwide time-series analysis to quantify the burden of HFMD attributable to non-optimal ambient temperature and relative humidity in Japan, while examining between-prefecture variation in the estimated associations. For each of the 47 prefectures (2000-2019), exposure-lag-response functions for temperature and humidity were modelled using distributed lag non-linear models and then pooled via multivariate meta-regression incorporating prefecture-level characteristics as potential effect modifiers. From these pooled models, we derived attributable fractions (AFs) for non-optimal temperature and humidity. Both exposures exhibited non-linear, J-shaped relationships with HFMD incidence. Using the minimum-incidence temperature (13.5 °C; 42nd percentile) and humidity (45.7%; 1st percentile) as references, cumulative relative risks at the 95th percentiles were 2.39 (95% CI: 1.95-2.94) for heat (28.4 °C) and 1.61 (95% CI: 1.45-1.80) for wet conditions (82.7%). Between-prefecture variability was partly accounted for by several factors: latitude, longitude, and temperature modified temperature-HFMD associations, while the economic power index, urbanization rate, and numbers of primary school students modified humidity-HFMD associations. Nationwide AFs indicated that 27.57% (95% empirical CI [eCI]: 2.76-30.94) of HFMD cases were attributable to non-optimal temperature and 15.32% (95% eCI: 2.92-23.03) to non-optimal humidity. These findings enhance our understanding of the complex non-linear relationships between HFMD incidence and meteorological variables, contributing to projections of future transmission risk under climate variability.