Symptomatic Predictors for 2009 Influenza A Virus (H1N1) Infection with an Emphasis for Patients with a Negative Rapid Diagnostic Test

BACKGROUND: The clinical diagnosis of influenza is difficult because it shares nonspecific symptoms with a variety of diseases. Emergency departments and clinics were overwhelmed by a surge of anxious patients during the 2009 influenza A virus (H1N1) outbreak. Our objective was to identify symptomatic predictors of influenza virus infection for patients with a negative rapid diagnostic test. METHODOLOGY/PRINCIPAL FINDINGS: We conducted a retrospective review of 805 patients who presented at Chang Gung Memorial Hospital, from August 1, 2009, to September 30, 2009. Respiratory specimens from these patients were subjected to rapid influenza tests and reverse-transcription polymerase chain reactions. In total, 36% of 308 children and 23% of 497 adults were positive for 2009 influenza A virus (H1N1) infection by polymerase chain reaction or virus culture. For pediatric patients, sore throat and influenza-like illness significantly increased the odds of having 2009 influenza A virus (H1N1) infection, by more than 3-fold (95% confidence interval (CI): 1.9–7.3) and 7-fold (95% CI: 4.00–14.2), respectively. For adult patients, cough and constitutional symptoms increased the odds of having 2009 influenza A virus (H1N1) by greater than 5-fold (95% CI: 3.1–10.2) and 3-fold (95% CI: 2.1–6.7), respectively. The negative likelihood ratio of the combination of fever and cough was 0.096 (95% CI: 0.01–0.69) for children with negative results of rapid influenza diagnostic tests. CONCLUSION/SIGNIFICANCE: In influenza epidemic settings, clinicians should be aware that rapid influenza diagnostic tests are relatively insensitive for the diagnosis of influenza virus infection. For patients with negative rapid influenza diagnostic tests, those lacking fever and cough have a low probability of influenza virus infection. The management strategy should be made individually and depend on the severity of illness.