High Prevalence of Malaria in Zambezia, Mozambique: The Protective Effect of IRS versus Increased Risks Due to Pig-Keeping and House Construction

BACKGROUND African countries are scaling up malaria interventions, especially insecticide treated nets (ITN) and indoor residual spraying (IRS), for which ambitious coverage targets have been set. In spite of these efforts infection prevalence remains high in many parts of the continent. This study investigated risk factors for malaria infection in children using three malaria indicator surveys from Zambezia province, Mozambique. The impact of IRS and ITNs, the effects of keeping farm animals and of the construction material of roofs of houses and other potential risk factors associated with malaria infection in children were assessed. METHODS Cross-sectional community-based surveys were conducted in October of 2006, 2007 and 2008. A total of 8338 children (ages 1-15 years) from 2748 households were included in the study. All children were screened for malaria by rapid diagnostic tests. Caregiver interviews were used to assess household demographic and wealth characteristics and ITN and IRS coverage. Associations between malaria infection, vector control interventions and potential risk factors were assessed. RESULTS Overall, the prevalence of malaria infection was 47.8% (95%CI: 38.7%-57.1%) in children 1-15 years of age, less than a quarter of children (23.1%, 95%CI: 19.1%-27.6%) were sleeping under ITN and almost two thirds were living in IRS treated houses (coverage 65.4%, 95%CI: 51.5%-77.0%). Protective factors that were independently associated with malaria infection were: sleeping in an IRS house without sleeping under ITN (Odds Ratio (OR)= 0.6; 95%CI: 0.4-0.9); additional protection due to sleeping under ITN in an IRS treated house (OR = 0.5; 95%CI: 0.3-0.7) versus sleeping in an unsprayed house without a ITN; and parental education (primary/secondary: OR = 0.6; 95%CI: 0.5-0.7) versus parents with no education. Increased risk of infection was associated with: current fever (OR = 1.2; 95%CI: 1.0-1.5) versus no fever; pig keeping (OR = 3.2; 95%CI: 2.1-4.9) versus not keeping pigs; living in houses with a grass roof (OR = 1.7; 95%CI: 1.3-2.4) versus other roofing materials and bigger household size (8-15 people: OR = 1.6; 95%CI: 1.3-2.1) versus small households (1-4 persons). CONCLUSION Malaria infection among children under 15 years of age in Zambezia remained high but conventional malaria vector control methods, in particular IRS, provided effective means of protection. Household ownership of farm animals, particularly pigs, and living in houses with a grass roof were independently associated with increased risk of infection, even after allowing for household wealth. To reduce the burden of malaria, national control programs need to ensure high coverage of effective IRS and promote the use of ITNs, particularly in households with elevated risks of infection, such as those keeping farm animals, and those with grass roofs.