629Influenza-like illness surveillance in Australia: comparison of general practice data with laboratory-confirmed influenza notifications

Abstract

Abstract Background Influenza is a respiratory infection responsible for 645,000 annual deaths worldwide. Surveillance systems provide valuable data for monitoring influenza in order to detect outbreaks and guide public health responses. This study aimed to investigate the epidemiology of influenza-like illness (ILI) using two Australian general practice databases (MedicineInsight and the Australian Sentinel Practice Research Network (ASPREN)) and compare them with laboratory-confirmed influenza from the National Notifiable Diseases Surveillance System (NNDSS). Methods All patients who had a consultation in MedicineInsight general practices or ASPREN and all laboratory-confirmed influenza reported by the NNDSS between 2015-2017 were included. Weekly ILI rates per 1,000 consultations (MedicineInsight/ASPREN) were compared with influenza notifications (NNDSS). Results Data was consistent among sources, with higher cases in 2017, among women and patients aged 20-49 years. The peak rate in MedicineInsight almost doubled in 2017 compared to 2015, while in ASPREN it was less pronounced. MedicineInsight ILI curves more closely resembled NNDSS patterns (shape, the start of the season, peaks) than ASPREN, although both were highly correlated with NNDSS (r = 0.90 to 0.97 and r = 0.88 to 0.98, respectively). Conclusions MedicineInsight and ASPREN provided consistent ILI results, both resembling confirmed influenza epidemic curves, suggesting the potential use of routinely collected electronic medical records (MedicineInsight) in influenza surveillance. MedicineInsight provides comprehensive medical data, such as underlying conditions, medications prescribed and vaccination status, which could be used to improve accuracy on influenza detection. Key messages Electronic medical records could be used to monitor ILI in combination with ASPREN for effective early detection of outbreaks.