Malaria incidence in Limpopo Province, South Africa, 1998–2007

Abstract

BackgroundMalaria is endemic in the low-altitude areas of the northern and eastern parts of South Africa with seasonal transmission. The aim of this descriptive study is to give an overview of the malaria incidence and mortality in Limpopo Province for the seasons 1998–1999 to 2006–2007 and to detect trends over time and place.MethodsRoutinely collected data on diagnosed malaria cases and deaths were available through the provincial malaria information system. In order to calculate incidence rates, population estimates (by sex, age and district) were obtained from Statistics South Africa. The Chi squared test for trend was used to detect temporal trends in malaria incidence over the seasons, and a trend in case fatality rate (CFR) by age group. The Chi squared test was used to calculate differences in incidence rate and CFR between both sexes and in incidence by age group.ResultsIn total, 58,768 cases of malaria were reported, including 628 deaths. The mean incidence rate was 124.5 per 100,000 person-years and the mean CFR 1.1% per season. There was a decreasing trend in the incidence rate over time (p < 0.001), from 173.0 in 1998–1999 to 50.9 in 2006–2007. The CFR was fairly stable over the whole period. The mean incidence rate in males was higher than in females (145.8 versus 105.6; p < 0.001); the CFR (1.1%) was similar for both sexes. The incidence rate was lowest in 0–4 year olds (78.3), it peaked at the ages of 35–39 years (172.8), and decreased with age from 40 years (to 84.4 for those ≥ 60 years). The CFR increased with increasing age (to 3.8% for those ≥ 60 years). The incidence rate varied widely between districts; it was highest in Vhembe (328.2) and lowest in Sekhukhune (5.5).ConclusionInformation from this study may serve as baseline data to determine the course and distribution of malaria in Limpopo province over time. In the study period there was a decreasing trend in the incidence rate. Furthermore, the study addresses the need for better data over a range of epidemic-prone settings.