Use of healthcare worker sickness absenteeism surveillance as a potential early warning system for influenza epidemics in acute care hospitals.

Dear Editor, The global spread of emerging infectious diseases can strain resources and result in healthcare staff absenteeism, as was the case during the severe acute respiratory syndrome (SARS) outbreak in Singapore in 2003.1,2 The recent emergence of a novel infl uenza A (H1N1-2009) pandemic3 has again reminded us of the potential impact of such infections on hospital operating capacity.4 Syndromic surveillance systems have been proposed for the early detection of community and institutional outbreaks of emerging infections5,6 but there is little work on the usefulness of such systems for hospital personnel. Following the SARS experience, a syndromic surveillance system for monitoring healthcare personnel sickness absenteeism was established in Tan Tock Seng Hospital (TTSH), a 1100-bed general hospital in Singapore. As part of preparations for pandemic H1N1-2009 infl uenza, we did a quick assessment of whether our system could serve as a potential early warning system for pandemic infl uenza. Following the nosocomial outbreaks of SARS, TTSH implemented an on-line staff sickness absenteeism surveillance system. The web-based user interface includes pre-populated demographic and employment details of hospital personnel, including name and work area. In each work area, there are at least 2 designated staff in charge of daily submission of data on medical certifi cates (MCs) for sickness absenteeism. Data captured include start and end dates of MC, area of work, and the reason for the staff being on MC, either as a diagnosis [e.g., pneumonia, URTI (upper respiratory tract infection)] or a set of self-reported symptoms (fever, cough, breathlessness, and diarrhoea). On a daily basis, a team of epidemiologists monitors healthcare worker reports of URTI, gastroenteritis and conjunctivitis, with clusters being identifi ed and actively investigated. Hospital personnel may be advised on enhanced infection control measures and staying away from work if necessary. For our rapid assessment, we included all staff MCs from 31 December 2006 to 29 December 2007 which had URTI/ infl uenza as a diagnosis and/or the following self-reported symptoms: fever, and/or respiratory symptoms (including nasal discharge, sore throat, laryngitis, cough, sputum, and breathlessness). We refer to the sum of these healthcare sickness absenteeism submissions as reports of acute respiratory illness (ARI). Syndromic reports of healthcare staff ARI were compared to national level polyclinic data on URTI, and national laboratory surveillance data on the proportion of specimens positive for infl uenza A for the year 2007, presented by epidemiological week.7 We also looked at potential clustering of staff ARI reports in a given Use of Healthcare Worker Sickness Absenteeism Surveillance as a Potential Early Warning System for Infl uenza Epidemics in Acute Care Hospitals