The objective ofthis study was to assess, in a developing country setting, the effect of dexamethasone therapy on bacterial meningitis outcomes. A prospective double blind placebo controlled trial was conducted in 89 children aged from 2 months to 12 years suffering from bacterial meningitis. Neurological, developmental, and hearing assessments were conducted at one, four, and 12 months after discharge. Forty eight patients received dexamethasone and 41 placebo. Initial antimicrobial drugs used were ampicillin and chloramphenicol. For all patients at the time of admission the mean duration of illness was 5.7 days; 47% had had seizures and 56% had impaired consciousness. Seventeen of 89 (19%) patients died. The mortality for the dexamethasone group was 25% as compared with 12% in the group receiving placebo. Presentation to the hospital after four days of symptoms and with impaired conscious state were independent predictors of death. Of the dexamethasone group survivors, 26.5% had neurological sequelae and 42.3% had hearing impairment, whereas in the placebo group it was 24% and 30% respectively. Altered state of consciousness was a predictor of neurological sequelae. The presence of neurological sequelae and high cerebrospinal fluid protein independently predicted hearing loss. No beneficial effect of dexamethasone was observed on morbidity or mortality of this group of patients with bacterial meningitis. Dexamethasone is therefore not useful in developing countries as adjunctive treatment in patients seriously ill with bacterial meningitis, who present late for treatment and have been partially treated. (Arch Dis Child 1996;75:482-488)