Application of moving epidemic method in surveillance and early warning of hand foot and mouth disease

ObjectiveTo evaluate the feasibility of application of moving epidemic method in surveillance and early warning of hand foot and mouth disease (HFMD).

MethodsMoving epidemic method was used for the cross validation of the incidence data of HFMD in Jingzhou from 4 to 37 weeks during 2010–2017. The evaluation indicators included sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, Matthew correlation coefficient and Yoden index. The selected historical data were used to establish a surveillance and early warning model for 4–37 weeks in 2018 and evaluate the trend of HFMD in Jingzhou in 2018.

ResultsThe cross validation results showed that the fitting effects in 2011 and 2013 were poor, which were eliminated. Data in 2010, 2012, 2014, 2015, 2016 and 2017 were used to establish a surveillance model of the moving epidemic method. The optimal parameter value of \delta was 2.5. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, Matthew correlation coefficient and Yoden index were 0.94, 0.93, 0.92, 0.95, 13.57, 0.06, 0.87 and 0.87, respectively. The epidemic started to occur in 18th week in 2018, the incidence level was low.

ConclusionThe method produced a robust and specific signal to detect HFMD epidemic, and can be used to establish the spring and summer epidemic threshold of HFMD in Jingzhou.