Decline and change in seasonality of US rotavirus activity after the introduction of rotavirus vaccine.

BACKGROUND: In 2006, routine immunization of US infants against rotavirus was initiated. We assessed national, regional, and local trends in rotavirus testing and detection before and after vaccine introduction.

METHODS: We examined data for July 2000 through June 2008 from a national network of approximately 70 US laboratories to compare geographical and temporal aspects of rotavirus season timing and peak activity. To assess trends in rotavirus testing and detection, we restricted the analyses to 33 laboratories that reported for >or=26 weeks per season from 2000 to 2008.

RESULTS: Nationally, the onset and peak of the 2007-2008 rotavirus season were delayed 15 and 8 weeks, respectively, compared with prevaccine seasons from 2000-2006. Delays were observed in each region. The 2007-2008 rotavirus season lasted 14 weeks compared with a median of 26 weeks during the prevaccine era. Of 33 laboratories, 32 reported fewer positive results and a lower proportion of positive test results in 2007-2008 compared with the median in 2000-2006, with a 67% decline in the number and a 69% decline in the proportion of rotavirus-positive test results. The proportion of positive test results in 2007-2008 compared with the median in 2000-2006 declined >50% in 79% of the laboratories and >75% in 39% of the laboratories.

CONCLUSIONS: The 2007-2008 US rotavirus season seems substantially delayed, shorter, and diminished in magnitude compared with seasons before vaccine implementation. The extent of change seems greater than expected on the basis of estimated vaccine coverage, suggesting indirect benefits to unvaccinated individuals. Monitoring in future seasons is needed to confirm these trends.