Background: Malaria rapid diagnostic tests (RDTs) have great potential to improve quality care and rational drug use in malaria-endemic settings although studies have shown common RDT non-compliance. Yet, evidence has largely been derived from limited hospital settings in few countries. This article reviews a PhD thesis that analyzed national surveys from multiple sub-Saharan African countries to generate large-scale evidence of malaria diagnosis practices and its determinants across different contexts. Design: A mixed-methods approach was used across four studies that included quantitative analysis of national household and facility surveys conducted in multiple sub-Saharan African countries at the outset of new guidelines (Demographic and Health Surveys and Service Provision Assessments). Qualitative methods were used to explore reasons for quantitative findings in select settings. Results: There was low (17%) and inequitable test uptake across 13 countries in 2009–2011/12, with greater testing at hospitals than at peripheral clinics (odds ratio [OR]: 0.62, 95% confidence interval [CI]: 0.56–0.69) or community health workers (OR: 0.31, 95% CI: 0.23–0.43) (Study I). Significant variation was found in the effect of diagnosis on antimalarial use at the population level across countries (Uganda OR: 0.84, 95% CI: 0.66–1.06; Mozambique OR: 3.54, 95% CI: 2.33–5.39) (Study II). A Malawi national facility census indicated common compliance to malaria treatment guidelines (85% clients with RDT-confirmed malaria prescribed first-line treatment), although other fever assessments were not often conducted and there was poor antibiotic targeting (59% clients inappropriately prescribed antibiotics). RDT-negative patients had 16.8 (95% CI: 8.6–32.7) times higher odds of antibiotic overtreatment than RDT-positive patients conditioned by cough or difficult breathing complaints (Study III). In Mbarara (Uganda), health workers reportedly prescribed antimalarials to RDT-negative patients if no other fever cause was identified and non-compliance seemed further driven by RDT perceptions, system constraints, and client interactions (Study IV).