Abstract
BACKGROUND: Sleeping under a net, particularly a long-lasting insecticidal net (LLIN), is associated with reduced malaria morbidity and mortality, but requires high coverage and adherence. In this study, parasitologically confirmed Plasmodium falciparum infection and a clinical indicator (i.e. fever) were measured among children in three villages of central Côte d'Ivoire (Bozi, N'Dakonankro and Yoho) and associations with net coverage explored. In Bozi and Yoho, LLINs were provided by the national malaria control programme, prior to the study and an additional catch-up coverage was carried out in Bozi. In N'Dakonankro, no net intervention was conducted. METHODS: Three cross-sectional surveys were carried out; two in the dry season (February 2010 and November 2011) and one in the rainy season (May 2012). Among 897 children aged <15 years, P. falciparum infection was determined by microscopy and a rapid diagnostic test (RDT). Fever was defined as an axillary temperature ≥37.5°C. A questionnaire was administered to obtain demographic data and net usage. RESULTS: The proportion of children infected with P. falciparum according to microscopy in the third survey was 74%, 81% and 82% in Yoho, N'Dakonankro and Bozi, respectively. Meanwhile, 46% of the children in N'Dakonankro, 44% in Bozi and 33% in Yoho slept under a net. The risk of P. falciparum infection did not differ between net-sleepers and non-net-sleepers. Fewer children had parasitaemia ≥1,000 parasites/μl of blood in Bozi in the third compared to the first survey. Fever was poorly correlated with P. falciparum infection. The risk of P. falciparum infection did not depend on the village of residence, presence of fever or sleeping under LLIN the night before the survey. Conversely, it was higher in the rainy season and among older children. CONCLUSIONS: In an area where P. falciparum is highly prevalent, the use of nets was associated with significantly lower levels of parasitaemia. The apparent lack of effect on P. falciparum infection and fever might be explained by the relatively low net coverage in Bozi and Yoho and the relatively short period (<2 years) during which the impact of nets was measured.