Objective To document the investigation and control of an outbreak of gastroenteritis in City G,South China,and provide a reference for preventing future outbreaks.Methods An ambispective cohort study was designed.Attack rate (AR) and relative risks (RR) were calculated to identify the causes of gastroenteritis.Investigations using questionnaires included personal interviews with patients and doctors,reviews of medical records,laboratory examinations of fecal specimens and continuous hygiene monitoring of water samples from the waterworks.Results Overall,427/71534 (AR=5.97‰) cases were identified between October 31 and November 12 2010.Geographic distribution was highly localized,with 80% of cases occurring in the areas supplied by waterworks‐A.Consumption of water provided solely by waterworks‐A was found to be associated with illness (RR=8.20,95 CI%:6.12-10.99) compared with that from waterworks‐B.Microbiological analyses confirmed the presence of Norovirus in six of eight fecal samples from symptomatic patients,two water samples from waterworks‐A and two sewage samples.After taking effective measures,the hygienic indices of waterworks‐A met health criteria again on November 9 and no cases were reported 3 days later.Conclusion The outbreak reported here was caused by drinking tap water contaminated with sewage at the source.Early identification of possible contamination sources and awareness of changes that might negatively impact water quality are important preventive measures to protect public health. Objective To document the investigation and control of an outbreak of gastroenteritis in City G, South China, and provide a reference for preventing future outbreaks.Methods An ambispective cohort study was designed .Attack rate (AR) and relative risks (RR) were calculated to identify the causes of gastroenteritis. Investigations using using countires personal interviews with patients and doctors, reviews of medical records, laboratory examinations of fecal specimens and continuous hygiene monitoring of water samples from the waterworks. Results Overall, 427/71534 (AR = 5.97 ‰) Cases were identified between October 31 and November 12 2010. Geographic distribution was highly localized, with 80% of cases occurring in the areas provided by waterworks-A.Consumption of water provided solely by waterworks-A was found to be associated with illness (RR = 8.20, 95% CI: 6.12-10.99) compared with that from waterworks-B. Microbiological analyzes confirmed the presence of Norovirus in six of eight fe cal samples from symptomatic patients, two water samples from waterworks-A and two sewage samples. After taking effective measures, the hygienic indices of waterworks-A met health criteria again on November 9 and no cases were reported 3 days later. Confluence The outbreak reported here was caused by drinking tap water contaminated with sewage at the source. Early identification of possible contamination sources and awareness of changes that might negatively impact water quality are important preventive measures to protect public health.