Hospitalizations for varicella in the United States, 1988 to 1999

BACKGROUND: Varicella epidemiology is changing with increasing use of the varicella vaccine. METHODS: To describe the epidemiology of severe varicella disease before and after vaccine introduction, data from the National Hospital Discharge Survey (NHDS) for 1988 to 1999 were analyzed. Incidental cases of varicella in persons hospitalized for a different indication were excluded. RESULTS: In the prevaccination era (1988 to 1995), there were 10 632 varicella hospitalizations annually. The most common complications were viral pneumonitis (20.9%), fluid/electrolyte disturbances (19.3%) and soft tissue infections (17.8%). Most (89.1%) persons had no severe underlying immunocompromising conditions. The mean length of hospitalization was 5.4 days, corresponding to approximately 57 000 days of hospitalization annually. In the first years after vaccine licensure (1996 to 1999), vaccine coverage reached 59%. Although not statistically significant, there was a trend toward decreased hospitalizations and a decline in mean length of hospitalization. CONCLUSIONS: Varicella morbidity was higher in the prevaccination era than previously reported. Although no significant decline is evident, a trend toward decreased hospitalizations is emerging in the first years after vaccine introduction.