Sero-epidemiological evaluation of changes in <it>Plasmodium falciparum </it>and <it>Plasmodium vivax </it>transmission patterns over the rainy season in Cambodia

• Main Authors: Cook Jackie, Speybroeck Nico, Sochanta Tho, Somony Heng, Sokny Mao, Claes Filip, Lemmens Kristel, Theisen Michael, Soares Irene S, D'Alessandro Umberto, Coosemans Marc, Erhart Annette
• Format: Article
• Language: English
• Published: BMC 2012-03-01
• Series: Malaria Journal
• Subjects: Malaria; Serology; Classification and regression tree; Elimination; Cambodia
• Online Access: http://www.malariajournal.com/content/11/1/86

<p>Abstract</p> <p>Background</p> <p>In Cambodia, malaria transmission is low and most cases occur in forested areas. Sero-epidemiological techniques can be used to identify both areas of ongoing transmission and high-risk groups to be targeted by control interventions. This study utilizes repeated cross-sectional data to assess the risk of being malaria sero-positive at two consecutive time points during the rainy season and investigates who is most likely to sero-convert over the transmission season.</p> <p>Methods</p> <p>In 2005, two cross-sectional surveys, one in the middle and the other at the end of the malaria transmission season, were carried out in two ecologically distinct regions in Cambodia. Parasitological and serological data were collected in four districts. Antibodies to <it>Plasmodium falciparum </it>Glutamate Rich Protein (GLURP) and <it>Plasmodium vivax </it>Merozoite Surface Protein-1₁₉(MSP-1₁₉) were detected using Enzyme Linked Immunosorbent Assay (ELISA). The force of infection was estimated using a simple catalytic model fitted using maximum likelihood methods. Risks for sero-converting during the rainy season were analysed using the Classification and Regression Tree (CART) method.</p> <p>Results</p> <p>A total of 804 individuals participating in both surveys were analysed. The overall parasite prevalence was low (4.6% and 2.0% for <it>P. falciparum </it>and 7.9% and 6.0% for <it>P. vivax </it>in August and November respectively). <it>P. falciparum </it>force of infection was higher in the eastern region and increased between August and November, whilst <it>P. vivax </it>force of infection was higher in the western region and remained similar in both surveys. In the western region, malaria transmission changed very little across the season (for both species). CART analysis for <it>P. falciparum </it>in the east highlighted age, ethnicity, village of residence and forest work as important predictors for malaria exposure during the rainy season. Adults were more likely to increase their antibody responses to <it>P. falciparum </it>during the transmission season than children, whilst members of the Charay ethnic group demonstrated the largest increases.</p> <p>Discussion</p> <p>In areas of low transmission intensity, such as in Cambodia, the analysis of longitudinal serological data enables a sensitive evaluation of transmission dynamics. Consecutive serological surveys allow an insight into spatio-temporal patterns of malaria transmission. The use of CART enabled multiple interactions to be accounted for simultaneously and permitted risk factors for exposure to be clearly identified.</p>