Electronic Monitoring of Truancy and Influenza Surveillance, Hong Kong – Volume 18, Number 5 — May 2012 – Journal of Emerging Infectious Diseases

For the publisher: Potentially useful public health interventions, such as school closures, must be introduced in a timely manner during the course of an ongoing epidemic to significantly affect community transmission (1,2). In most traditional surveillance systems that include data on healthcare utilization, however, considerable delays occur between data collection and feedback, leading to sub-optimal and inappropriate information to guide. evidence-based public health decisions. New syndromic surveillance approaches have been tried to improve the timeliness by targeting events earlier in the care-seeking journey and by promoting the real-time collection and processing of surveillance data using information technology. modern (3,4). Building on an existing platform of an electronic school management system, we developed an automated truancy monitoring system for influenza-like illness (ILI) in Hong Kong and evaluated its performance using data collected from March 2008 to June 2011. The Institutional Review Board of the University of Hong Kong / Hospital Authority, Hong Kong West Cluster, approved the study.

We have partnered with a commercial supplier who develops and delivers e-learning platforms and management systems for educational institutions, including 337 primary and secondary schools in Hong Kong attended by children aged 6-18. Invitations to participate in the new absenteeism system were sent to all schools subscribing to the electronic school management system, and 62 schools across Hong Kong were recruited in phases during the study period. We started with 18 schools (17,255 students) from February to June 2008, then expanded to 45 schools (37,087 students) in 2008-2009, 50 schools (41,765 students) in 2009-2010 and 62 schools (49,425 students) in 2010-11.

The absenteeism system worked as follows. A pupil identification chip card was issued to each pupil at all participating schools, and pupils were required to swipe their card on a sensor at the entrance of the school as an electronic attendance record, which replaced a traditional paper roll call. Reasons for absence, including SG, were asked during phone calls to a subset (37%) of schools, and responses were manually entered into the system by teachers. Daily aggregate data, including the total number of students and the number of absentees in each year (stratified by reason for absence), was compiled each afternoon. Individual children were not identifiable. Data cleaning, aggregation, analysis and reporting has been automated with version R 2.12.1 (R Foundation for Statistical Computing, Vienna, Austria). The aggregate weekly absenteeism rates were calculated as the total number of absentees divided by the total number of students in all schools. Regular weekly and ad hoc reports on absenteeism patterns, with an interpretation of overall influenza disease activity in the community, were distributed to all participating schools through the same school management system and disseminated to the general public. via an existing influenza monitoring dashboard (5).

Figure

Figure. . . Influenza surveillance data, Hong Kong, February 23, 2008 – June 18, 2011. A) Weekly overall truancy rate. B) School absenteeism rate specific to weekly influenza-like illness (ILI). C) Weekly SG (defined as …

School absenteeism rates and baseline data from 2 traditional surveillance systems existing in Hong Kong between March 2008 and June 2011 covered a total of 7 influenza seasons (Figure). Data for both systems came from all age groups across the country. The virus isolation rate in the laboratory as a gold standard highlighted the typical seasonality of influenza in Hong Kong, which peaked each year around February in winter and around August in summer (figure, panel D) (6). Clear, sharp peaks were detectable from aggregate and ILI-specific data during most of these influenza seasons (Figure, panels A, B), which typically occurred 1 to 3 weeks before peaks in laboratory data (range 1 to 5 and -1 to 4 weeks; median of 3 and 0.5 weeks for aggregate and ILI data, respectively). The data generally showed much sharper peaks than ambulatory sentinel data, possibly related to better coverage of disease activity in the community, unlike sentinel data, which only captured episodes leading to visits. in outpatient clinics (7). The limitations of our truancy data included a relatively lower perceived sensitivity of the ILI-specific truancy rate (as data is only provided by 37% of participating schools) and the presence of data gaps during school days. school holidays or school closures due to public health measures to mitigate seasonal flu in March 2008 (1) and pandemic influenza in June 2009 (2).

This study demonstrated the feasibility and potential benefits of using automatically captured truancy data as a complementary data stream for influenza surveillance. Real-time monitoring of truancy, an early event in the healthcare research journey, can improve situational awareness and help inform appropriate public health decisions and interventions in a more timely and informed manner. on evidence. Electronic data capture from pre-existing smart card systems is an attractive and cost-effective option that does not require additional resources, systems or manpower, unlike other approaches (8ten). The growing popularity of smart card technology in various situations could also offer potential opportunities for innovative surveillance systems.

Mountain peak

We thank the Public Health Laboratory Service, the Hospital Authority and the Center for Health Protection, Department of Health of the Hong Kong Special Administrative Region Government, for posting influenza surveillance data online. We thank the Department of Microbiology, Queen Mary Hospital, Hong Kong, for providing reference laboratory data. In addition, we thank BroadLearning Education (Asia) Ltd and the participating schools for providing data on absenteeism.

This work received financial support from the Research Fund for the Control of Infectious Diseases (grant n ° 11101092) and the Area of ​​Excellence Scheme of the University of Hong Kong Grants Committee (grant n ° AoE / M-12/06). DKMI received research funding from Hoffmann-La Roche Inc., and BJC received research funding from MedImmune Inc.

Author affiliations: University of Hong Kong, Hong Kong Special Administrative Region, People’s Republic of China

The conclusions, findings, and opinions expressed by the authors contributing to this review do not necessarily reflect the official position of the United States Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or institutions affiliated with the. authors. . Use of trade names is for identification purposes only and does not imply endorsement by any of the groups mentioned above.


Source link

Leave a Reply

Your email address will not be published. Required fields are marked *