Most patients with suspected malaria do not receive diagnostic confirmation before beginning antimalarial treatment. We investigated the extent to which uncertainty about malaria diagnosis contributes to patient nonadherence to artemether–lumefantrine (AL) treatment through a randomized controlled trial in central Uganda. Among 1,525 patients purchasing a course of AL at private drug shops, we randomly offered 37.6% a free malaria rapid diagnostic test (RDT) and then assessed adherence through home visits 3 days later. Of these subjects, 68.4% tested positive for malaria and 65.8% adhered overall. Patients who tested positive did not have significantly higher odds of adherence than those who were not offered the test (adjusted odds ratio [OR]: 1.07, 95% confidence interval [CI]: 0.734–1.57, P = 0.719). Patients who received a positive malaria test had 0.488 fewer pills remaining than those not offered the test (95% CI: −1.02 to 0.043, P = 0.072). We found that patients who felt relatively healthy by the second day of treatment had lower odds of completing treatment (adjusted OR: 0.532, 95% CI: 0.394–0.719, P < 0.001). Our results suggest that diagnostic testing may not improve artemisinin-based combination therapy adherence unless efforts are made to persuade patients to continue taking the full course of drugs even if symptoms have resolved.