Objectives: Acute gastroenteritis (AGE) is a common cause of hospitalization among children, particularly in winter, even though effective primary and secondary prevention measures are available. The objectives of this study were to estimate the morbidity of infectious AGE in hospitalized children under 5 years of age in France and to describe its characteristics and costs. Methods: The Medical Information Systems Program (PMSI) covers short-stay public and private hospitalizations. All hospital stays (throughout France) in 1997 with an AGE code 1) as the primary diagnosis or 2) as an associated diagnosis with a symptom or complication of AGE as the primary diagnosis were included. Deaths associated with GEA identified by the PMSI were compared with those identified in 1997 through death certificates recorded by INSERM, CépiDC. Results: 51,125 hospital stays were identified, representing an annual rate of 1,385 cases per 100,000 children under 5 years of age, and 11.4% of hospitalizations in this age group. Based on the coding, these AEs were mainly due to viruses (36%), of which 43% were rotaviruses, or bacteria (8%), of which 60% were salmonella, but most were "presumed infectious" (56%). The seasonal peak occurred in winter for rotavirus, other viral or “presumed infectious” AGE, and in summer for salmonellosis. AGI, particularly rotavirus-related cases, were more frequent the younger the child (from 3,606 cases per 100,000 children under 1 year of age to 257 cases per 100,000 children aged 4 years). A complication, primarily dehydration, was noted in 21% of viral cases and in 17% of bacterial cases. Fourteen deaths were recorded in both the PMSI and the death certificates. The average length of stay (3.2 days) was longer among children under 1 year of age in cases of viral origin, complications, or associated bronchiolitis. Total hospitalization costs were estimated at 408,344 KF (approximately 62 million euros), based on the 1997 national ISA point value. Discussion: For the study of AGE, the PMSI provides a description consistent with data available in France and abroad. The non-systematic performance of stool examinations explains why a majority of cases have no specified etiology. These “presumed infectious” AGE cases have characteristics similar to viral AGE. Despite coding variations, the large sample size revealed trends. The PMSI allows for estimating the significant cost of hospitalizations for AGE and should enable monitoring the impact of potential prevention campaigns.