Appropriate use of empirical antibiotics in acute diarrhoea: a cross-sectional survey in southern Thailand.

BACKGROUND: Although treatment with empirical antibiotics is recommended for dysenteric or invasive bacterial diarrhoea, overuse is common in clinical practice worldwide. Recent information on the use of antibiotics in Thai children with acute diarrhoea is lacking.
OBJECTIVE: To survey the appropriate use of empirical antibiotics in children with acute diarrhoea in hospitals in southern Thailand and evaluate the association between their use and clinical features.
METHODS: Outpatient medical records of children aged 2 months to 5 years with acute diarrhoea seen at ten hospitals in southern Thailand (five community and five general hospitals) during January to December 2004 were reviewed. Children with diarrhoeal symptoms lasting >7 days, who were immunocompromised, who had an underlying disease or whose medical records were inadequate were excluded. Appropriate use of empirical antibiotics was defined as: use of an antibiotic in a child with a history of bloody-mucous diarrhoea, who had a faecal WBC count of >10 cells/HPF, or no prescription of antibiotics in watery, non-bloody diarrhoea.
RESULTS: There were 2882 patient encounters, 70% of which were children under 2 years of age. Invasive bacterial diarrhoea comprised 6.9% of the total number of cases. Of the 2882 patient encounters, 50.3%, 36.4% and 13.3% were seen by general practitioners, paediatricians or other medical personnel, respectively. Antibiotics had been used appropriately in 44.1% of cases. Of the 55.9% of cases identified as inappropriate use, 55.2% involved unnecessary antibiotic prescriptions in non-indicated cases. The prevalence of appropriate use was significantly higher in general hospitals than in community hospitals (51% vs 37.1%, p<0.05). Medical trainees were more likely to prescribe antibiotics more appropriately than staff physicians (76.9% vs 44.2%, p<0.05). Multivariate analysis showed that a history of fever (OR 1.25, 95% CI 1.04-1.50), watery-mucous stool (OR 2.54, 95% CI 1.94-3.32), mushy-mucous stool (OR 2.62, 95% CI 1.28-5.35), bloody-mucous stool (OR 6.97, 95% CI 4.17-11.64), stool frequency of 5-10 times/day (OR 1.41, 95% CI 1.16-1.70), body temperature 38.6-39.5 degrees C (OR 1.86, 95% CI 1.17-2.98) and a faecal WBC count of even 1-10 cells/HPF (OR 3.24, 95% CI 1.55-6.77) were associated with antibiotic use.
CONCLUSION: Overuse of antibiotics in the treatment of acute diarrhoea in children is common. The prevalence differs between different levels of hospitals and different types of medical personnel. Physicians might mistake some patients' clinical features of diarrhoea as features of dysentery, leading to unnecessary prescription of antibiotics.