Evaluation of the diagnostic utility of case definitions to detect influenza virus infection in Vojvodina, Serbia

Abstract

Introduction/Objective A case definition recommended by the World Health Organization is commonly used for influenza surveillance worldwide. The aim of this study was to evaluate prognostic values of proposed case definitions of Influenza Like Illness (ILI), Severe Acute Respiratory Illness (SARI) and Acute Respiratory Distress Syndrome (ARDS) for laboratory confirmed-influenza and to compare the age distribution of influenza patients across virus types and subtypes in Vojvodina. Methods We conducted a descriptive epidemiological study using surveillance reports and laboratory data from October 1, 2010 to May 20, 2017 (seven surveillance seasons). Results We included 2,937 participants, 48.6% of whom were laboratory-confirmed influenza cases, and most of the confirmed cases (30.1%) were detected in February. In the 15–29 years age group, the type A influenza (H3N2) was more frequent among patients with ILI (54.9% vs. 34.2%, p = 0.040), and less frequent in patients with SARI (39.4% vs. 65.8%, p = 0.009) compared with influenza type B. In patients aged 30–64 years with ARDS, influenza type B was more common than influenza type A (H3N2) (13.4% vs. 6.2%, p = 0.032), but less common in compared to influenza type A (H1N1) pdm09 (13.4% vs. 25.7%, p = 0.017). The SARI case definition of influenza was associated with an increased likelihood of laboratory-confirmed influenza for all age groups (p < 0.05). During the epidemic period, it was observed that the ILI case definition had the highest diagnostic value for influenza in the age group 5–14 (AUC = 0.733; 95% CI: 0.704–0.764), while the SARI and ARDS case definitions were the best predictors of influenza for patients 15–29 years of age (AUC = 0.565; 95% CI: 0.504–0.615 and AUC = 0.708; 95% CI: 0.489–0.708, respectively). The case definition of ARDS had the maximum sensitivity (100%) among patients 15–29 years of age. Conclusion The proposed case definitions of influenza appeared to be good predictors of influenza and therefore can be useful for influenza surveillance, especially in the countries with limited laboratory capacities.