Sociodemographic and spatial influences on the malaria incidence of infants in the Ashanti Region/Ghana

Abstract number: 1733_270

Kreuels B., Kobbe R., Adjei S., Adjei on behalf of the Agona IPTi Trial Team O.

Introduction: Malaria incidence rates have been shown to vary greatly over short distances and periods of time in areas of low to medium transmission. In areas of high, perennial transmission, due to early acquired immunity malaria incidence variation is considered to be far less various. However in small children who have not yet developed immunity this is not necessarily true.

Methods: We recruited a group of 535 children from 9 villages in an area of high transmission in Ghana at the age of 3 months and followed them monthly over a period of 21 months. Malaria was defined as a new fever episode together with a positive thick smear of at least 500 parasites/ml. Socio-economic characteristics were evaluated with the help of a questionnaire on recruitment. Geographical characteristics of the villages and distances of households to the forest fringe were analysed on satellite images with a Geographic Information System (GIS). Ecological analyses and multivariate Poisson regressions were performed to determine the most important factors responsible for the spatial variation of malaria incidence.

Results: Malaria incidence in the study area was highly heterogeneous between the 9 villages with a variation of 0.83–2.20 episodes per person year at risk. Incidence rates were strongly correlated to village area (r2 = 0.74) and village population (r2 = 0.68) while altitude variation in the study area was low and did not correlate to incidence rates. We found significant lower malaria incidences in children born in May (p < 0.001), September (p < 0.01) or October (p < 0.01), in children with literate mothers (p < 0.02), in those with use of bednets (p < 0.001) and window-screens (p < 0.001), and those born into families with a higher financial status (p < 0.01). On the other hand malaria incidence was higher in children of Northern Ghanaian ethnicity (p < 0.001) and children whose mothers worked as farmers (p < 0.01). An independent and strongly significant influence could be seen for the distance of the household from the forest fringe with a linear rate reduction of 0.2 episodes every 50 meters further away from the forest (p < 0.001).

Conclusion: Our results indicate that the variation of malaria incidence rates in children from rural areas of high transmission may have been underestimated in the past. Taking this variation into account when designing new and implementing current intervention measures could possibly increase their efficiency and cost-effectiveness.

Session Details

Date: 31/03/2007
Time: 00:00-00:00
Session name: European Society of Clinical Microbiology and Infectious Diseases
Location: ICC, Munich, Germany
Presentation type:
Back to top