Associated Signs and Symptoms of Confirmed Influenza Infections in Ghana

Abstract

Aims: Acute respiratory tract infections are among the most common causes of morbidity and mortality worldwide and rank second among the top causes of health facility attendance in Ghana. A well-positioned case detection system can efficiently identify respiratory illnesses and detect influenza outbreaks early. This study determined clinical signs and symptoms most predictive of confirmed influenza infection among Influenza-Like Illness (ILI) patients. Study Design: A cross-sectional study was conducted. Place and Duration of Study: A data repository of case-based records from the regional sentinel sites available at the Disease Surveillance Department of the Ghana Health Service for the period of January to December 2011 was used as secondary data. The dataset was first accessed in October 2016. Methods: A cross-sectional study was conducted with 2,089 anonymized case-based records. The associations between the combinations of fever and cough together with other signs and symptoms of an influenza positive infection were explored. Frequencies, sensitivity, specificity, positive predictive value, negative predictive value, and odds ratio were computed using Epi Info 7.0 (CDC, Atlanta, GA). Results: The median age was 22 years (IQR: 6.5 to 37 years). Twenty-one percent were children under 5 years. Females formed 1,190 (57%) of the total patients studied. Fever and cough together with a sore throat had the highest significant association (ORa: 1.52; 95%CI: 1.15 - 2.02) with a positive influenza laboratory results in the adjusted analysis. Conclusion: The case definition for Influenza-Like Illness (ILI) was loosely applied considering the variation between the frequencies of patients with fever or a cough. Since an ILI case definition encourages the inclusion of other respiratory signs, observations from this study suggest the inclusion of a sore throat, together with fever and cough as the best predictor of an influenza infection.