Retrospective comparison of management of gastro-enteritis in hospitalised children.

In Hong Kong, bacterial pathogens, the majority of them Salmonellae, cause approximately one-third of paediatric admissions for diarrhoea. This study retrospectively reviewed inpatient gastro-enteritis management, with particular focus on antibiotic use. Antibiotics are generally recommended for Salmonella gastro-enteritis in infants under 3 months of age but not for older infants and children unless they are so toxic that bacteraemia is suspected. Three groups of children admitted with acute gastro-enteritis were randomly identified from a computerised discharge database. Based on pathological reports held in the case records department, the final groups for analysis were Salmonella (n = 86), rotavirus (n = 55) and non-specified (n = 126). Epi Info version 6 (CDC, Atlanta) was used for data entry and analysis. Compared with a combined rotavirus/non-specified group, the Salmonella group were significantly more likely to have blood (OR 6.1, 95% CI 3.2-11.7, p < 0.0001) and mucus (OR 4.8, 95% CI 2.6-8.9, p < 0.0001) in the stool, fever during admission (OR 3.6, 95% CI 1.6-8.4, p = 0.001), more stools per day (median 6.2 vs 4.2, p < 0.0001), a longer stay in hospital (median 3.4 vs 2 days, p < 0.0001) and to be younger (median 7.1 vs 14.6 mths, p < 0.0001). The Salmonella group were more likely to have been given antibiotics (38% vs 15%, OR 3.6, 95% CI 1.9-6.9, p < 0.0001) but age did not influence the likelihood that antibiotics would be given.