The incidence rate in the under six months age group may have been an underestimation if many hospitalisations for acute gastroenteritis occurred in the first six weeks of life. There was no active follow up, only passive surveillance of hospitalisations of study participants. Participants may have moved from the area or died at home, and thus no longer be contributing to the total follow

up time, yet it was assumed that these participants had contributed the full 5 years of follow up time. This would have led selleck screening library to underestimation of incidence rates as the denominator would be inflated. Although CHBH is the referral hospital for all local clinics in Soweto, there is a chance that selleck inhibitor some participants may have consulted

a private practitioner and had an admission at a private hospital. There is also the possibility that those with very severe acute gastroenteritis may have died in the community before arriving at the hospital. These cases would not have been identified as an episode of acute gastroenteritis and included in the numerator in incidence calculations but would have contributed to total person time, leading to an underestimation of the number of admissions for severe acute gastroenteritis and the incidence rates. There were no stool samples collected on admission and so no stool identification of pathogens was possible. As a result the true proportion of

severe acute gastroenteritis caused by rotavirus could not be determined. Despite these limitations the results provide unique information on disease burden estimates in HIV-infected children Acute gastroenteritis is an important cause of hospitalisation in South Africa, especially in children under 2 years of age and those with concomitant HIV infection. The estimated risk of hospitalization for rotavirus associated acute gastroenteritis is two out fold greater in HIV-infected compared to HIV-uninfected children, despite rotavirus being identified in a lower proportion of acute gastroenteritis cases in HIV-infected children. The introduction of rotavirus vaccine, proven to be safe, immunogenic and efficacious in both HIV-infected and uninfected children, into the national immunisation program is likely to decrease the overall burden of severe acute gastroenteritis regardless of HIV infection status. Ongoing surveillance for rotavirus disease as well as a case control study to determine the effectiveness of the vaccine in routine use are currently underway in South Africa. Conflict of Interest Statement: The Phase 3 trial on which this secondary analysis is based was funded by Wyeth. SM has been a paid temporary-consultant /expert board member for Pfizer, GSK, Merck, and Novartis, and has been paid for speaking engagements by Pfizer and GlaxoSmithKline.