Evaluation of models for multi-step forecasting of hand, foot and mouth disease using multi-input multi-output: A case study of Chengdu, China.

BACKGROUND: Hand, foot and mouth disease (HFMD) is a public health concern that threatens the health of children. Accurately forecasting of HFMD cases multiple days ahead and early detection of peaks in the number of cases followed by timely response are essential for HFMD prevention and control. However, many studies mainly predict future one-day incidence, which reduces the flexibility of prevention and control. METHODS: We collected the daily number of HFMD cases among children aged 0-14 years in Chengdu from 2011 to 2017, as well as meteorological and air pollutant data for the same period. The LSTM, Seq2Seq, Seq2Seq-Luong and Seq2Seq-Shih models were used to perform multi-step prediction of HFMD through multi-input multi-output. We evaluated the models in terms of overall prediction performance, the time delay and intensity of detection peaks. RESULTS: From 2011 to 2017, HFMD in Chengdu showed seasonal trends that were consistent with temperature, air pressure, rainfall, relative humidity, and PM10. The Seq2Seq-Shih model achieved the best performance, with RMSE, sMAPE and PCC values of 13.943~22.192, 17.880~27.937, and 0.887~0.705 for the 2-day to 15-day predictions, respectively. Meanwhile, the Seq2Seq-Shih model is able to detect peaks in the next 15 days with a smaller time delay. CONCLUSIONS: The deep learning Seq2Seq-Shih model achieves the best performance in overall and peak prediction, and is applicable to HFMD multi-step prediction based on environmental factors.

City-level meteorological conditions modify the relationships between exposure to multiple air pollutants and the risk of pediatric hand, foot, and mouth disease in the Sichuan Basin, China.

Background: Several studies have examined the effects of city-level meteorological conditions on the associations between meteorological factors and hand, foot, and mouth disease (HFMD) risk. However, evidence that city-level meteorological conditions modify air pollutant-HFMD associations is lacking. Methods: For each of the 17 cities in the Sichuan Basin, we obtained estimates of the relationship between exposures to multiple air pollutants and childhood HFMD risk by using a unified distributed lag nonlinear model (DLNM). Multivariate meta-regression models were used to identify the effects of city-level meteorological conditions as effect modifiers. Finally, we conducted subgroup analyses of age and sex to explore whether the modification effects varied in different subgroups. Results: The associations between PM2.5/CO/O3 and HFMD risk showed moderate or substantial heterogeneity among cities (I2 statistics: 48.5%, 53.1%, and 61.1%). Temperature conditions significantly modified the PM2.5-HFMD association, while relative humidity and rainfall modified the O3-HFMD association. Low temperatures enhanced the protective effect of PM2.5 exposure against HFMD risk [PM2.5 <32.7 µg/m3 or PM2.5 >100 µg/m3, at the 99th percentile: relative risk (RR) = 0.14, 95% CI: 0.03-0.60]. Low relative humidity increased the adverse effect of O3 exposure on HFMD risk (O3 >128.7 µg/m3, at the 99th percentile: RR = 2.58, 95% CI: 1.48-4.50). However, high rainfall decreased the risk of HFMD due to O3 exposure (O3: 14.1-41.4 µg/m3). In addition, the modification effects of temperature and relative humidity differed in the female and 3-5 years-old subgroups. Conclusion: Our findings revealed moderate or substantial heterogeneity in multiple air pollutant-HFMD relationships. Temperature, relative humidity, and rainfall modified the relationships between PM2.5 or O3 exposure and HFMD risk.