Hospital-based influenza surveillance in Korea: hospital-based influenza morbidity and mortality study group.

Influenza epidemics occur annually with variations in size and severity. Hospital-based Influenza Morbidity & Mortality was established to monitor influenza epidemics and their severity, which is composed of two surveillance systems: emergency room-based and inpatient-based surveillance. Regarding emergency room-based surveillance, influenza-like illness index (influenza-like illness cases per 1,000 emergency room-visiting subjects), number of laboratory-confirmed cases and the distribution of influenza types were estimated weekly. Inpatient-based surveillance included monitoring for hospitalization, complications, and mortality. The emergency room influenza-like illness index correlated well with the number of laboratory-confirmed influenza cases, and showed a bimodal peak at Week 4 (179.2/1,000 emergency room visits) and Weeks 13-14 (169.6/1,000 emergency room visits) of 2012. Influenza A was the predominant strain during the first epidemic peak, while influenza B was isolated exclusively during the second peak. In 2011-2012 season, the mean admission rate of emergency room-visiting patients with influenza-like illness was 16.3% without any increase over the epidemic period. Among the hospitalized patients with influenza, 33.6% (41 out of 122 patients) were accompanied by complications, and pneumonia (28.7%, 35 out of 122 patients) was the most common. Most fatal cases were caused by influenza A (96.2%) after the first epidemic peak. In conclusion, Hospital-based Influenza Morbidity & Mortality was effective for monitoring the trends in circulating influenza activity concurrently with its severity. In the 2011-2012 season, the influenza epidemic persisted for a ≥ 5-month period, with a bimodal peak of influenza A and B in sequence. Overall, influenza A was more severe than influenza B.