J Pediatr Infect Dis 2010; 05(02): 155-159
DOI: 10.3233/JPI-2010-0244
Georg Thieme Verlag KG Stuttgart – New York

The accuracy of clinical diagnosis of influenza in Thai children

Thanong Prasarnphanich
a   Department of Pediatrics, Prapokklao Hospital, Chantaburi province, 22000, Thailand. Tel.: +66 8039301156; E-mail: Dhanong@hotmail.com
› Author Affiliations

Subject Editor:
Further Information

Publication History

06 January 2009

10 June 2009

Publication Date:
28 July 2015 (online)

Abstract

There is a lack of data assessing the accuracy of clinical diagnosis of influenza in children. This study aimed to identify the clinical predictors of influenza infections and to validate the use of clinical criteria of influenza-like illness definition for the diagnosis of influenza in children. Virological influenza surveillance was conducted by collecting throat swab specimens for viral culture and clinical data from febrile children with acute respiratory tract infection at Prapokklao Hospital from July 2005 to September 2007. The criteria of influenza-like illness definition was modified for children and defined by the presence of fever > 38°C plus two of the following three symptoms: headache, cough and sore throat. Two hundred and four of 838 specimens (24%) yielded positive results for influenza virus isolates. Influenza A virus infections were more common than influenza B virus infections. In the multivariate analysis positive predictors of influenza infections were cough [odds ratio (OR) = 2.07 95% confidence interval (CI) = 1.26–3.39] and age > 5 years old (OR = 2.30, 95% CI = 1.66–3.19), whereas diarrhea was inversely associated with influenza (OR = 0.29, 95% CI = 0.12–0.69). Both the World Health Organization criteria (OR = 1.82, 95% CI = 1.04–3.18) and the modified criteria for children (OR = 1.86, 95% CI = 1.38–2.56) showed a positive association with influenza infections. World Health Organization criteria showed high sensitivity and negative predictive value but low specificity and positive predictive value. The modified criteria for children showed fair sensitivity and specificity, high negative predictive value but low positive predictive value. According to the average proportion of positive isolates of 24%, influenza should be recognized as an important cause of acute respiratory tract infections in children especially in the endemic season, but the diagnosis of pediatric influenza cannot be established on clinical criteria alone.