Background
In Gabon, the impact of intermittent preventive treatment of malaria in infants (IPTi) was not statistically significant on malaria reduction, but the impact on moderate anaemia was, with some differences between the intention to treat (ITT) and the according to protocol (ATP) trial analyses. Specifically, ATP was statistically significant, while ITT analysis was borderline. The main reason for the difference between ITT and ATP populations was migration.

Methods
This study estimates the cost-effectiveness of IPTi on the reduction of anaemia in Gabon, comparing results of the ITT and the ATP clinical trial analyses. Threshold analysis was conducted to identify when the intervention costs and protective efficacy of IPTi for the ATP cohort equalled the ITT cost-effectiveness ratio.

Results
Based on IPTi intervention costs, the cost per episode of moderate anaemia averted was US$12.88 (CI 95% 4.19, 30.48) using the ITT analysis and US$11.30 (CI 95% 4.56, 26.66) using the ATP analysis. In order for the ATP results to equal the cost-effectiveness of ITT, total ATP intervention costs should rise from 118.38 to 134 US$ ATP or the protective efficacy should fall from 27% to 18.1%. The uncertainty surrounding the cost-effectiveness ratio using ITT trial results was higher than using ATP results.

Conclusions
Migration implies great challenges in the organization of health interventions that require repeat visits in Gabon. This was apparent in the study as the cost-effectiveness of IPTp-SP worsened when drop out from the prevention was taken into account. Despite such challenges, IPTi was both inexpensive and efficacious in averting cases of moderate anaemia in infants.