Background and aim.Varicella is normally a self-limited disease of childhood that does not require hospitalization. In the prevaccine era varicella caused >9000 hospitalizations per year. To determine whether the varicella vaccine, licensed in 1995, has decreased hospitalizations because of varicella, we examined national rates of varicella-related hospital discharges (VRHD) covering a 12-year period that included pre- and postvaccine data.Methods.Data from the 1988 to 1999 National Hospital Discharge Survey and population estimates from the National Center for Health Statistics were used to calculate biennial rates of VRHD. To control for coding consistency, rates of invasive disease caused byHaemophilus influenzaewere calculated for the same time period.Results.The rate of VRHD for 1998 to 1999 (4.42 hospitalizations per 100 000 person-years) was the lowest of any of the periods measured, but this difference was not statistically significant. The same was true of VRHD limited to cases with varicella coded as the primary diagnosis. A trend toward a decrease in VRHD was observed in all age groups examined, although none was statistically significant. Calculated rates from this national data set were in agreement with prior studies using active surveillance, and the previously documented fall in hospitalizations caused by invasiveH. influenzaedisease was demonstrated using these methods.Conclusions.Although it is uncommon for children with varicella to require hospitalization, these cases are an important contributor to cost and morbidity of varicella. In contrast to predictions of prelicensure mathematical models, there has not been a significant decrease in total or first diagnosis VRHD since the vaccine became available. Current coverage levels are below those used in prelicensure models. Increased acceptance of the varicella vaccine by parents and practitioners may aid in the further decrease of varicella-related hospitalizations.