Introduction: Cambodia aims to eliminate malaria by 2025. However, heterogeneity in malaria risk is considered a challenge for malaria elimination. This study aimed to describe and explain micro-epidemiological variation in Plasmodium infection prevalence at household and village level in a malaria-endemic region of Cambodia.
Methods: The study was set in three villages in Ratanakiri Province, Cambodia. A cross-sectional survey to investigate individual and household-level risk factors for Plasmodium infection was conducted. The primary outcome was Plasmodium infection. A two-level logistic regression model with a random intercept fitted for each household was used to model the odds of infection, with sequential adjustment for individual-level then household-level risk factors.
Results: There were 713 adolescent and adult respondents and 435 children living in 258 households who were included in the survey. Cumulative malaria prevalence varied between the three villages, from 6.5% to 16.6% in adolescents and adults, and 4.1% to 11.9% in children. Individual-level risk factors for Plasmodium infection included hammock net use and frequency of evening outdoor farm gatherings in adults, and older age in children. Household-level risk factors included house wall material, crop types, and satellite dish and farm machine ownership. Individual-level risk factors did not explain differences in odds of Plasmodium infection between households or between villages. In contrast, once household-level risk factors were taken into account, there was no significant difference in odds of Plasmodium infection between households and between villages.
Conclusion: This study showed the importance of household-level risk factors in explaining differences in Plasmodium infection prevalence between households and between villages, and points to the importance of ongoing indoor and peridomestic transmission in a region where forest workers and mobile populations have previously been the focus of attention. Bed net coverage should be improved, and interventions targeting malaria risk at household level should be further explored.