Assessing the acceptability of a malaria prevention intervention in infant by healthcare providers: the contribution of structural equation modeling (SEM)

IntroductionDespite evidence of its efficacy and cost-effectiveness, and being recommended by WHO since 2010, the implementation of preventive malaria prophylaxis for infants (strategy now called perennial malaria chemoprevention) has remained limited in sub-Saharan Africa. The MULTIPLY project aimed to pilot, in rural Togo, this strategy extended in the 2nd year of life and integrated within facility-based and outreach vaccination services. Our aim was to conduct an in-depth study of the acceptability of perennial malaria chemoprevention prior to its implementation, in order to inform future programmatic deployment of this strategy. MethodsA cross-sectional Knowledge, Attitudes and Practices questionnaire was self-administered pre-implementation of perennial malaria chemoprevention in Haho district, among 231 health care providers. Three acceptability components, derived from the Theoretical Framework of Acceptability, were investigated: ease of administration, perceived effectiveness and necessity of the strategy. Structural equation modeling was used to identify the complex relationships between acceptability components, feasibility factors, knowledge, attitudes and practices. ResultsThe overall pre-implementation acceptability of perennial malaria chemoprevention was high. Administration of perennial malaria chemoprevention was perceived as being all the easier if it was integrated into a setting where feasibility criteria, such as the presence of human resources ({beta}= 0.18; p<0.01), were met. Perceived effectiveness of the strategy was associated with the knowledge of the strategy ({beta}= 0.20; p<0.01) and perception of drug effectiveness ({beta}= 0.48; p<0.01). Perceived effectiveness of malaria control programs ({beta}= 0.23; p<0.01) and low perception of the effectiveness of insecticide-treated bednets ({beta}= -0.10; p=0.04) was associated with the perceived need for perennial malaria chemoprevention. Health care providers seniority was associated, directly or indirectly, with all three dimensions of the acceptability of perennial malaria chemoprevention. ConclusionThe use of structural equation modeling allowed a comprehensive and nuanced quantitative assessment of pre-intervention acceptability of perennial malaria chemoprevention, highlighting the importance of considering feasibility factors such as access to and availability of human and material resources and attitudes towards malaria prevention when promoting the adoption of this childhood preventive strategy among health care providers in a rural district in Togo.