BackgroundMalaria represents a huge burden for the health care services across Africa. Describing malaria attending health services contributes to quantify the burden and describe the epidemiology and clinical presentation.MethodsRetrospective analysis of data collected through the Manhiça morbidity surveillance system (Mozambique) on all paediatric visits (<15 years) to the outpatient clinic from June 2003 to May 2005. Age-specific minimum community-based incidence rates (MCBIRs) of malaria were calculated using demographic surveillance system data. Malaria was defined as fever or history of fever in the preceding 24 hours with asexual Plasmodium falciparum parasitaemia of any density in the blood smear.ResultsA total of 94,941 outpatient visits were seen during the study period, of which 30.5% had malaria. Children younger than three years accounted for almost half of the total malaria cases and children aged ≥ 5 years represented 36.4% of the cases. Among children who presented with malaria, 56.7% had fever and among children who presented with fever or a history of fever only 37.2% had malaria. The geometric mean parasitaemia in malaria cases was 8582.2 parasites/μL, peaking in children aged two to three years. 13% of malaria cases had a PCV<25% and the mean PCV in malaria cases increased gradually with age, ranging from 27.8% in children aged 2–12 months to 34.4% in ≥ 5 years. The percentage of cases admitted or transferred showed a clear decreasing trend with age. MCBIRs of outpatient malaria per 1,000 child years at risk for the whole study period were of 394 in infants, 630 in children aged 1 to <5 years and 237 in children aged five years or more. A clustering of the cases was observed, whereby most children never had malaria, some had a few episodes and very few had many episodes.ConclusionPreventive measures should be targeted at children younger than three years, as they carry the highest burden of malaria. Children aged 5–15 years represent around a third of the malaria cases and should also be included in control programmes. Concern should be raised about presumptive treatment of fever cases with artemisinin-combination therapies, as many children will, according to IMCI guidelines, receive treatment unnecessarily.