Abstract
BACKGROUND: Cambodia has made significant progress towards achieving malaria elimination by 2025. Cases continue to decrease and are primarily concentrated in forested areas. Forest-goers are most at risk of malaria due to their proximity to the forest, poor sleeping conditions, frequent mobility, and distance from health services. Consistent use of long-lasting insecticidal nets or hammock nets (LLINs/LLIHNs), early diagnosis and treatment of cases are central to reducing disease burden. The aim of this study was to understand forest-goers' knowledge, attitudes, and practices related to malaria prevention and care-seeking, and to identify key behavioural determinants of LLIN/LLIHN use and prompt care-seeking within 24 h of developing a fever. METHODS: A mixed-methods study design consisting of a cross-sectional survey and qualitative in-depth interviews was implemented in two Cambodian provinces. Survey participants (N = 654) were recruited using respondent driven sampling. Interview participants (N = 28) were selected using purposive sampling. Findings from the survey were analysed using univariate and bivariate analysis and multivariate weighted logistic regression. Interviews were coded and analysed using thematic content analysis. RESULTS: All study participants had heard of malaria and 98% knew that malaria was transmitted by mosquitoes. LLIN/LLIHN ownership was high (94%). Although 99% of participants perceived LLIN/LLIHN use as an important malaria prevention measure, only 76% reported using one during their last visit to the forest. Only 39% of survey participants who reported seeking care did so within the recommended 24 h from fever onset during their last febrile illness. Among all study participants, 43% did not seek any healthcare during their last febrile episode. In controlled regression models, perceived community social norms were significantly associated with LLIN/LLIHN use (OR: 2.7, 96% CI 1.99-2.64) and care-seeking within 24 h of fever onset (OR: 1.7, 95% CI 1.00-2.88). Social support from other forest-goers was also significantly associated with LLIN/LLIHN use (OR: 4.9, 95% CI 1.32-18.12). CONCLUSIONS: Study findings are consistent with other studies on LLIN/LLIHN use and care-seeking behaviours. While rates of LLIN/LLIHN ownership were high among the study population, rates of use were not as high. More concerning were the delayed care-seeking behaviours. Social behaviour change activities should incorporate social norms and social support as mechanisms for behaviour change given the identified positive correlations with LLIN/LLIHN use and prompt care-seeking.