Effect of Antipyretic Usage on the Accuracy of the World Health Organisation Case Definition for Influenza-Like Illnesses

Abstract

Introduction Influenza is a major cause of morbidity and mortality globally especially during pandemics. For this reason countries have set up surveillance systems to continually monitor Influenza like illnesses (ILI) for early detection of outbreaks and rapid response. However, in sub-Saharan Africa, specifically Ghana where medication can be purchased over the counter, it is more likely that people might have taken some antipyretics before reporting to the health facility. This may result in patients not having a measured axillary temperature ≥37.5°C by the time they present to the health facility. These cases are more likely not to meet the WHO case definition and will be missed by the surveillance system. This study was conducted to assess the effect of antipyretics on ILI case detection. Method A cross sectional study was conducted in 4 health facilities in the Shai-Osudoku and Ningo-Prampram districts of Ghana from September 2013 to May 2014. Nasopharyngeal and oropharyngeal specimens were collected from 321 patients seeking ambulatory care at the health facilities who met the case definition of “History of fever or measure fever ≥37.5°C (Axillary) with cough” and consented to be part of the study. Swabs were tested for influenza virus using molecular assays. Data collection tool used was a structured questionnaire. Univariate and bivariate analysis were carried out using statistical package for the social sciences (SPSS version 17; IBM Corporation, Armonk, NY). University of Ghana http://ugspace.ug.edu.gh xiii

Results Out of 321 participants, 236 (73.52%) had a history of fever and 85 (26.48%) had measured fever ≥ 37.5oC. A total of 95(29.60%) participants took antipyretics before visiting the facility. Out of these 95 participants, 62(65.26%) had a history of fever. Participants who took antipyretics before reporting to the facility were more likely to have a history of fever (OR: 1.78, CI: 1.05 – 3.00, p-value: 0.03). A total of 39 ( 56.52%) of the influenza cases that had a history of fever were missed by the WHO case definition for ILI. However, patients with measured fever ≥ 37.5oC were 3 times more likely to have had an influenza infection compared to patients with history of fever. Sensitivity of influenza case detection was found to be higher in patients with history of fever and cough (56.52%, CI: 43.38 – 66.23) Conclusion A larger proportion of patients with ILI present to the health facility with a history of fever compared to those with measured fever ≥ 37.5oC. Among those with history of fever, a greater proportion takes in antipyretics prior to seeking care at the health facility which results in a measured body temperature < 37.5oC which results in influenza cases been missed. An assessment of the use of antipyretics and history of fever show a positive effect on the sensitivity of Influenza case detection. Keywords: Influenza, fever, antipyretics