The

epidemiology of rotavirus varies by setting [6]. Seasonality of infection is prominent in temperate climates while a low prevalence is maintained throughout the year in tropical countries [7]. The mode of transmission, though believed to be mainly feco-oral, is also possibly airborne and person-to-person because infection occurs in childhood irrespective of sanitary conditions [8]. Rotaviral gastroenteritis is usually accompanied by vomiting Smad3 signaling and fever and results in severe disease among infants [9]. Rotavirus is excreted in large numbers during diarrhea and the virus can remain infectious on inanimate surfaces, moist surfaces and hands. This report describes rotavirus infection detected by stool Raf inhibitor testing in children followed from birth to three years of age, with sampling during and in the absence of diarrhea. This study was conducted from 2002 through 2006 in three contiguous slums in Vellore, India after approval by the institutional review board of the Christian Medical College, Vellore. The study conduct, recruitment, and sample collection methods have been published previously [10]. Briefly, a birth cohort of 452 children was followed from birth till three years of

age, analysis was restricted to the 373 children who completed three years of follow-up. Surveillance of children for rotavirus infection was done by screening bimonthly stool samples and diarrheal stool samples, and clinical data were collected to record diarrheal severity 17-DMAG (Alvespimycin) HCl using the Vesikari score

with scores <5 considered mild, 6–10 moderate, 11–15 severe and 16–20 very severe [11]. In case of a surveillance sample, positive samples detected by ELISA (Dako Rota IDEIA, Ely, UK) were genotyped by reverse-transcription polymerase chain reaction (RT-PCR) for VP7 and VP4 amplification while for a diarrheal sample, irrespective of the ELISA result, one sample per episode was screened using RT-PCR for VP6 before genotyping [12]. The definitions for symptomatic and asymptomatic rotavirus infections used in this study are given in Table 1. Age-specific incidence and seasonality of symptomatic and asymptomatic infections were studied. The incidence rates were obtained by Poisson regression equations and frailty models adjusted for clustering of disease/infection within a child. For cumulative incidence of rotavirus infection, Kaplan–Meier estimates of median time to infection were calculated and compared between children infected with rotavirus overall and with specific genotypes. Factors influencing rotavirus infection as well as disease rates were studied using Poisson regression. To study the risk factors for rotavirus infection, children who experienced rotavirus infection in the first year were compared to children who did not experience rotavirus infection in the first year using multiple unconditional logistic regression.