Trends in hospitalizations for diarrhea in United States children from 1979 through 1992: estimates of the morbidity associated with rotavirus.

Objectives: To examine trends in the hospitalizations of children for diarrheal disease in the U.S. and to provide estimates for the burden of disease associated with rotavirus diarrhea.

Methods: Data for diarrheal hospitalizations among U.S. children ages 1 month through 4 years were compiled from the National Hospital Discharge Survey for the years 1979 through 1992. Between 1979 and 1992, 12% of all hospitalizations of U.S. children 1 month through 4 years of age had an International Classification of Diseases code for diarrhea listed in one of the top three positions on the discharge diagnosis.

Results: The annual rate of diarrheal hospitalizations, 97 per 10 000 persons (average, 185 742 per year), did not change substantially during the 14-year study period and accounted annually for 724 394 inpatient days (3.9 days per hospitalization). For most diarrheal hospitalizations (75.9%) no causative agent was specified in the National Hospital Discharge Survey records; of the remaining 24.8%, viruses were most commonly reported (19.3%), followed by bacteria (5.1%) and parasites (0.7%). The proportion of hospitalizations associated with viral diarrheas rose from 13% to 27% during the 14-year study period, whereas the proportion of hospitalizations for noninfectious diarrhea declined from 79% to 60%. Every year the number of hospitalizations peaked from November through April, the "winter" months, among children ages 4 through 35 months; this peak began in the West during November and December and reached the Northeast by March.

Conclusions: Diarrhea continues to be a common cause of hospitalization among children in the United States and the winter seasonality estimated to be caused in large part by rotavirus would be expected to decrease if rotavirus vaccines currently being developed were introduced. Our analysis of temporal trends in diarrheal hospitalizations provides a unique surrogate with which to estimate the disease burden associated with rotavirus diarrhea.