BACKGROUND: Despite the uptake of parasitological testing into policy and practice, appropriate prescription of anti-malarials and artemisinin-based combination therapy (ACT) in accordance with test results is variable. This study describes a National Malaria Control Programme-led capacity building intervention which was implemented in 10 States of Nigeria. Using the experience of Niger State, this study assessed the effect on malaria diagnosis and prescription practices among febrile under-fives in rural health facilities. METHODS: The multicomponent capacity building intervention consisted of revised case management manuals; cascade training from national to state level carried out at the local government area (LGA) level; and on the job capacity development through supportive supervision. The evaluation was conducted in 28, principally government-owned, health facilities in two rural LGAs of Niger State, one in which the intervention case management of malaria was implemented and the other acted as a comparison area with no implementation of the intervention. Three outcomes were considered in the context of rapid diagnostic testing (RDT) for malaria which were: the prevalence of RDT testing in febrile children; appropriate treatment of RDT-positive children; and appropriate treatment of RDT-negative children. Outcomes were compared post-intervention between intervention and comparison areas using multivariate logistic regression. RESULTS: The intervention did not improve appropriate management of under-fives in intervention facilities above that seen for under-fives in comparison facilities. Appropriate treatment with artemisinin-based combinations of RDT-positive and RDT-negative under-fives was equally high in both areas. However, appropriate treatment of RDT-negative children, when defined as receipt of no ACT or any other anti-malarials, was better in comparison areas. In both areas, a small number of RDT-positives were not given ACT, but prescribed an alternative anti-malarial, including artesunate monotherapy. Among RDT-negatives, no under-fives were prescribed artesunate as monotherapy. CONCLUSION: In a context of significant stock-outs of both ACT medicines and RDTs, under-fives were not more appropriately managed in intervention than comparison areas. The malaria case management intervention implemented through cascade training reached only approximately half of health workers managing febrile under-fives in this setting. Implementation studies on models of cascade training are needed to define what works in what context.