A joint analysis of influenza-associated hospitalizations and mortality in Hong Kong, 1998-2013

Abstract

Influenza viruses may cause severe human infections leading to hospitalization or death. Linear regression models were fitted to population-based data on hospitalizations and deaths. Surveillance data on influenza virus activity permitted inference on influenza-associated hospitalizations and deaths. The ratios of these estimates were used as a potential indicator of severity. Influenza was associated with 431 (95% CrI: 358-503) respiratory deaths and 12,700 (95% CrI: 11,700-13,700) respiratory hospitalizations per year. Majority of the excess deaths occurred in persons ≥65 y of age. The ratios of deaths to hospitalizations in adults ≥65 y were significantly higher for influenza A(H1N1) and A(H1N1)pdm09 compared to A(H3N2) and B. Substantial disease burden associated with influenza viruses were estimated in Hong Kong particularly among children and elderly in 1998-2013. Infections with influenza A(H1N1) was suggested to be more serious than A(H3N2) in older adults.

Figure 1

Weekly influenza virus activity in Hong Kong from 1998 through 2013, measured by ILI+ proxies defined for each influenza type/subtype as the product of the weekly proportion of outpatient consultations associated with influenza-like-illness in sentinel clinics and the weekly proportions of laboratory specimens testing positive for influenza A(H1N1), A(H3N2), A(H1N1)pdm09 and B viruses respectively.