The Establishment And Application Of School Absenteeism Surveillance System In Rural Areas Of Hubei Province

Objective To build a web-based school absenteeism surveillance system in rural areas,and analyze its function in prevention and control of infectious diseases in schools. Toexplore the optimal model and its related parameters for school absenteeismsurveillance so as to provide technical support for developing an Auto Warning System(AWS).Methods A web-based school absenteeism surveillance system has been designed anddeveloped in32primary schools from Qianjiang City and Shayang County of HubeiProvince. Information of absentees and absentees due to illness with syndrome-specificinformation were collected and reported daily to the system. Multiple quality controlmeasures were taken to ensure the implementation of the system. Descriptive analysiswas conducted to explore the distribution characteristics of absenteeism. Suspectedoutbreaks were checked and verified daily and the aggregations in time and space wereobserved. Exponentially Weighted Moving Average (EWMA) and Cumulative Sums(CUSUM, C1, C2, C3) models were applied on data of absentees due to illnessretrospectively to detect suspected outbreaks. Youden’s index was calculated tooptimize parameters before an optimal model was finally chosen. Custom indicators fordata quality assessment were used to assess the completeness, timeliness and accuracyof the reported data.Results (1) The school absenteeism surveillance system was established successfully, with a total of2892absentees reported by all schools under surveillance from April1,2012to January31,2013. The daily absenteeism rate was0.87‰.2184absentees weredue to illness, accounting for75.52%of total absentees. Most of the absentees due toillness were with respiratory symptoms or diseases (66.35%), followed by rash (22.21%)and fever (14.01%).(2) A total of27cases of suspected outbreaks were detected, and ofwhich92.59%were suspected as respiratory infectious diseases outbreaks. They mainlyoccurred in April to May and November to December, shown as bimodal distribution.Only one varicella outbreak was confirmed after investigation and verification.(3) Theefficacy of models with optimal parameters: the Youden’s index of EWMA (λ=0.25,c=2.0) was0.82, of C1(k=0.5, c=2.0) was0.95, of C2(k=0.5, c=3.0) was0.98, of C3(k=0.5, c=5.0) was0.97, which suggested that C2(k=0.5, c=3.0) had the highest efficacy.(4) During the surveillance period, the total missing report rate, total zero report rate,total absence report rate, total timely report rate and total consistency rate were5.39%,69.83%,24.78%,68.89%, and90.24%, respectively. Data quality control measures aswell as school holidays had a certain impact on the quality of reported data.Conclusion The school absenteeism surveillance system could timely detect theaggregations of absentees and absentees due to illness in time and space, so as toprovide effective early warning and prevention and control of infectious diseasesoutbreaks in schools. The successful establishment and implementation of the system inthe rural areas of Hubei Province could provide effective experience for otherresource-poor countries or regions.