Effects and interaction of meteorological factors on hemorrhagic fever with renal syndrome incidence in Huludao City, northeastern China, 2007-2018

Abstract

Background: Hemorrhagic fever with renal syndrome (HFRS), a rodent-borne disease, is a severe public health threat. Previous studies have discovered the influence of meteorological factors on HFRS incidence, while few studies have concentrated on the stratified analysis of delayed effects and interaction effects of meteorological factors on HFRS.

Objective: Huludao City is a representative area in north China that suffers from HFRS with primary transmission by Rattus norvegicus. This study aimed to evaluate the climate factors of lag, interaction, and stratified effects of meteorological factors on HFRS incidence in Huludao City.

Methods: Our researchers collected meteorological data and epidemiological data of HFRS cases in Huludao City during 2007-2018. First, a distributed lag nonlinear model (DLNM) for a maximum lag of 16 weeks was developed to assess the respective lag effect of temperature, precipitation, and humidity on HFRS incidence. We then constructed a generalized additive model (GAM) to explore the interaction effect between temperature and the other two meteorological factors on HFRS incidence and the stratified effect of meteorological factors.

Results: During the study period, 2751 cases of HFRS were reported in Huludao City. The incidence of HFRS showed a seasonal trend and peak times from February to May. Using the median WAT, median WTP, and median WARH as the reference, the results of DLNM showed that extremely high temperature (97.5th percentile of WAT) had significant associations with HFRS at lag week 15 (RR = 1.68, 95% CI: 1.04-2.74) and lag week 16 (RR = 2.80, 95% CI: 1.31-5.95). Under the extremely low temperature (2.5th percentile of WAT), the RRs of HFRS infection were significant at lag week 5 (RR = 1.28, 95% CI: 1.01-1.67) and lag 6 weeks (RR = 1.24, 95% CI: 1.01-1.57). The RRs of relative humidity were statistically significant at lag week 10 (RR = 1.19, 95% CI: 1.00-1.43) and lag week 11 (RR = 1.24, 95% CI: 1.02-1.50) under extremely high relative humidity (97.5th percentile of WARH); however, no statistically significance was observed under extremely low relative humidity (2.5th percentile of WARH). The RRs were significantly high when WAT was -10 degrees Celsius (RR = 1.34, 95% CI: 1.02-1.76), -9 degrees Celsius (1.37, 95% CI: 1.04-1.79), and -8 degrees Celsius (RR = 1.34, 95% CI: 1.03-1.75) at lag week 5 and more than 23 degrees Celsius after 15 weeks. Interaction and stratified analyses showed that the risk of HFRS infection reached its highest when both temperature and precipitation were at a high level.

Conclusions: Our study indicates that meteorological factors, including temperature and humidity, have delayed effects on the occurrence of HFRS in the study area, and the effect of temperature can be modified by humidity and precipitation. Public health professionals should pay more attention to HFRS control when the weather conditions of high temperature with more substantial precipitation and 15 weeks after the temperature is higher than 23 degrees Celsius.

Fig 1. The geographical location of Huludao City in China.

The map was created by ArcGIS 10.3 (Environmental Systems Research Institute; Redlands, CA, USA). The base map was acquired from the data center for geographic sciences and natural sources research, CAS (http://www.resdc.cn/data.aspx?DATAID=201).

Fig 2. Time series plot of the HFRS cases and meteorological factors.

Fig 3. Monthly distribution of HFRS cases.

Fig 4. The lag effect between WAT, WTP, WARH, and HFRS infection (Abbreviations: RR, relative risk; WAT, weekly average temperature; WTP, weekly total precipitation; WARH, weekly average relative humidity).

Fig 5. The lag-specific effect of climate factors on HFRS (Abbreviations: RR, relative risk; WAT, weekly average temperature; WTP, weekly total precipitation; WARH, weekly average relative humidity).

Fig 6. The effect interactions of the association among temperature, relative humidity, precipitation and HFRS in Huludao, 2007–2018.

(Abbreviations: RR, relative risk; WAT, weekly average temperature; WTP, weekly total precipitation; WARH, weekly average relative humidity. A: the interaction effect between WAT and WTP; B: the interaction effect between WAT and WARH; C: the interaction effect between WARH and WTP).

Fig 7. The associations between temperature and HFRS with different strata of relative humidity and precipitation (Abbreviations: RR, relative risk; CI: confidence interval).