Acute respiratory disease hospitalizations as a measure of impact of epidemic influenza.

Current measures of the health impact of epidemic influenza focus on analyses of death certificate data which may underestimate the true health effect. Previous investigations of influenza-related morbidity either have lacked virologic confirmation of influenza virus activity in the community or were not population-based. Community virologic surveillance in Houston, Texas has demonstrated that influenza viruses have produced epidemics each year since 1974. This study examined the relationship of hospitalizations with acute respiratory disease to the occurrence of influenza epidemics. Discharge records of 13,297 acute respiratory disease hospitalizations that occurred between July 1978 and June 1981 were obtained from 11 hospitals with 48.4% of hospital beds available in Harris County (metropolitan Houston). The correlation of adult acute respiratory disease hospitalizations with established indices of community acute respiratory disease morbidity was strong (r = 0.74) and indicated that the peak of adult acute respiratory disease hospitalizations followed the peak of influenza virus isolations by one week. Only 23.2% of persons hospitalized were 65 years of age or older, compared to 60-70% of persons who die during influenza epidemics. Although the highest rates of acute respiratory disease hospitalizations occurred among infants and persons aged 65 years or older, the rates for adults 45-64 years and preschool children aged 1-4 years were greater than 1 per 1,000 persons. Surveillance of acute respiratory disease hospitalizations can improve the measurement of serious morbidity associated with epidemic influenza, and can better define the characteristics of persons at risk for development of illness requiring hospitalization.