Using the near real-time effective reproduction number Rt as an early-warning tool for seasonal bronchiolitis and influenza-like illness epidemics

Abstract

Yearly bronchiolitis and influenza-like illness epidemics in France often involve high morbidity and mortality, which severely impact health care. Epidemics are declared by the French National Institute of Public Health based on syndromic surveillance of primary care and emergency departments (EDs), using statistics-based alarms. Although the effective reproduction number (Rt) is used to monitor the dynamics of epidemics, it has never been used as an early-warning tool for bronchiolitis or influenza-like illness epidemics in France. We assessed whether Rt is useful for detecting seasonal epidemics by comparing it to the tool currently used (MASS) by epidemiologists to declare epidemic phases. We used anonymized ED syndromic data from the Île-de-France region in France from 2010 to 2022. We estimated Rt and compared the indication of accelerated transmission (Rt > 1) to the MASS epidemic alarm time points. We computed the difference between those 2 time points, time to epidemic peak, and the daily cases documented at first indication and peak. Rt provided alarms for influenza-like illness and bronchiolitis epidemics that were, respectively, a median of 6 days (IQR, 4, 8) and 64 days (IQR, 52, 80) earlier than the alarms provided by MASS. Rt detected earlier signals of bronchiolitis and influenza-like illness epidemics. Using this early-warning indicator in combination with others to declare an annual epidemic could provide opportunities to improve health care system readiness.

Keywords: bronchiolitis; early warning; epidemic; influenza; reproduction number.