Background: A few studies have explored the association between meteorological factors and hand, foot, and mouth disease (HFMD) with inconsistent results. Besides, studies about the effects of air pollutants on HFMD are very limited.
Methods: Daily HFMD cases among children aged 0-14 years in Shenzhen were collected from 2009 to 2017. A distributed lag nonlinear model (DLNM) model was fitted to simultaneously assess the nonlinear and lagged effects of meteorological factors and air pollutants on HFMD incidence, and to further examine the differences of the effect across different subgroups stratified by gender, age and childcare patterns.
Results: The cumulative relative risk (cRR) (median as reference) of HFMD rose with the increase of daily temperature and leveled off at about 30 °C (cRR: 1.40, 95%CI: 1.29, 1.51). There was a facilitating effect on HFMD when relative humidity was 46.0% to 88.8% (cRR at 95th percentile: 1.18, 95%CI: 1.11, 1.27). Short daily sunshine duration (5th vs. 50th) promoted HFMD (cRR: 1.07, 95%CI: 1.02, 1.11). The positive correlation between rainfall and HFMD reversed when it exceeded 78.3 mm (cRR: 1.41, 95% CI: 1.22, 1.63). Ozone suppressed HFMD when it exceeded 104 µg /m3 (cRR at 99th percentile: 0.85, 95%CI: 0.76, 0.94). NO2 promoted HFMD among infants and the cRR peaked at lag 9 day (cRR: 1.47, 95%CI: 1.02, 2.13) (99th vs. 50th). Besides, children aged below one year, males and scattered children were more vulnerable to high temperature, high relative humidity, and short sunshine duration.
Conclusions: Temperature, relative humidity, sunshine duration, rainfall, ozone and NO2 were significantly associated with HFMD, and such effects varied with gender age and childcare patterns. These findings highlight the need for more prevention effort to the vulnerable populations and may be helpful for developing an early environment-based warning system for HFMD.
Keywords: air pollution; and mouth disease; climatic effect; distributed lag nonlinear model; foot; hand; subgroup analysis.