Patterns and etiology of diarrhea in three clinical settings

Koopman, J. S. (School of Public Health, U. of Michigan, Ann Arbor, Ml 48109), V. J. Turkish, A. S. Monto, V. Gouvea, S. Srivastava and R. E. Isaacson. Patterns and etiology of diarrhea in three clinical settings. Am J Epidemiol 1984;119:114-23.Acute diarrhea of young children was studied from September 1978 to April 1981 to determine age and time patterns, clinical characteristics and microbial association in two pedlatric practices and in a hospital population in southern Michigan. The practice population sizes were estimated so that rates of diarrhea could be determined. Care was sought for about 0.85 episodes per child in the first year of life and 0.4 episodes in the second year of life. Bacterial pathogens were rarely identified in the practices and were identified in only 5% of hospitalized patients. Rotavlrus was identified in 16% of the episodes in children under two years of age in the practices. These rotavirus diarrheas occurred mainly in the winter and were clearly more severe than nonrotavirus diarrheas. The rotavirus-infected patients did not, however, have more frequent respiratory symptoms. Respiratory symptom frequency was related to practice setting and income but not etiology. In the fall, before the rotavirus seasonal peak, a peak of nonrotaviral, nonbacterial diarrhea was seen in the practices. The symptoms were mild and a corresponding peak was not seen in hospital patients. The total direct costs of dlarrheal Illness to society were estimated to be due more to hospltalization than to outpatient care.