The "timeless" use of influenza-like illness criteria for influenza detection in the tropics

Abstract

Objective: We assessed the performance of influenza-like illness (ILI) case definitions by the Centers for Disease Control and Prevention (CDC), European Centers for Disease Control and Prevention and World Health Organization (WHO) in the tropics where seasonal patterns of respiratory viruses in acute upper respiratory tract infections (AURTIs) are ill-defined.

Methods: Clinical data and samples for respiratory multiplex polymerase chain reaction test were collected from 717 consecutive patients attending an emergency department in Singapore for uncomplicated AURTI in 2016-2018.

Results: Influenza (20.6%), rhinoviruses (14.4%), and coronaviruses (3.6%) were the most common viral pathogens identified. Biannual peaks with year-round activity were identified for influenza. Although higher rhinovirus activity was observed in inter-influenza seasonal periods, rhinoviruses and coronaviruses circulated year-round without distinct seasonal patterns. During high influenza activity months, the CDC and WHO ILI case definitions had moderate-to-high positive likelihood ratio (LR+) of 3.8-6.8 and 4.5-10.7, respectively, for ruling in influenza. They had moderately-high LR + of 3.3-3.8 and 3.9-4.6 for diagnosing influenza during other months. The ILI case definitions had high specificity (77.2%-85.4%) for rhinoviruses and coronaviruses.

Conclusion: The CDC and WHO ILI case definitions can be applied to clinically diagnose influenza in the tropics, regardless of the time of the year.

Keywords: Acute upper respiratory tract infection; Influenza-like illness; Respiratory viruses; Seasonality; Tropics.