The cost implications of changing from a policy of presumptive diagnosis to one of microscopy-based diagnosis in the management of uncomplicated malaria in an urban hospital adult outpatient clinic in Malawi were studied. Costs were measured in three separate weeks during the rainy season. In weeks I and II all uncomplicated malaria cases were treated on the basis of a presumptive diagnosis. In week II, blood films were taken but the results were not made available and did not affect drug dispensing. In week III, antimalarial drugs were restricted to parasitaemic patients. In week I, a total of 7216 prescriptions were written and dispensed, of which 2883 (39.9%) were for antimalarial drugs. The proportion of antimalarial prescriptions fell to 1171/5556 (21.1%) in week II and 357/5377 (6.6%) in week III. We estimate annual savings from microscopy-directed treatment in this setting to be 52,000 Malawi kwacha (US$ 14,000). This represents 3% of the annual drugs budget for the hospital, and is large enough to justify a change in policy.