Field performance of clinical case definitions for influenza screening during the 2009 pandemic

The aim of this study was to assess the performance of 3 different influenza-like illness (ILI) case definitions, adopted by the European (European-CDC), USA (USA-CDC), and Taiwan Centers for Disease Prevention and Control (Taiwan-CDC), as screening tools for influenza during the 2009 H1N1 pandemic.

From August 15 to 30, 2009, all emergency department patients with clinical symptoms or at epidemiologic risk for influenza were enrolled in an observational cohort study. Influenza diagnosis was established by positive rapid influenza diagnostic test or virus isolation. Sensitivity, specificity, positive predictive value, and negative predictive value of the European-, USA, and Taiwan-CDC ILI case definitions for screening were determined.

A total of 870 patients were screened during the study period. Rapid influenza diagnostic test was positive in 315 patients, 273 (85.6%) of whom had fever duration less than 72 hours. Virus isolation identified 4 more patients with influenza A initially negative by rapid influenza diagnostic test. The mean (SD) age of these 319 patients was 24.3 (18.1) years. Of the 870 screened patients, 670 (77.0%), 476 (54.7%), and 325 (37.4%) met the European-, USA-, and Taiwan-CDC ILI case definition, respectively. Screening sensitivity was 95%, 77.7%, and 57.7% and specificity was 33.4%, 58.6%, and 74.4%, respectively. Differences in sensitivity and specificity between any 2 of the 3 groups were statistically significant (P < .05).

First-line physicians should recognize the advantage and limitation of different ILI case definitions in influenza screening, especially confronted by pandemic or highly pathogenic avian influenza in the future.