In areas co-endemic for Plasmodium vivax and Plasmodium falciparum there is an increased risk of P vivax parasitaemia following P falciparum malaria. Radical cure is currently only recommended for patients presenting with P vivax malaria. Expanding the indication for radical cure to patients presenting with P falciparum malaria could reduce their risk of subsequent P vivax parasitaemia. We did a multicentre, open-label, superiority randomised controlled trial in five health clinics in Bangladesh, Indonesia, and Ethiopia. In Bangladesh and Indonesia, patients were excluded if they were younger than 1 year, whereas in Ethiopia patients were excluded if they were younger than 18 years. Patients with uncomplicated P falciparum monoinfection who had fever or a history of fever in the 48 h preceding clinic visit were eligible for enrolment and were required to have a glucose-6-dehydrogenase (G6PD) activity of 70% or greater. Patients received blood schizontocidal treatment (artemether-lumefantrine in Ethiopia and Bangladesh and dihydroartemisinin-piperaquine in Indonesia) and were randomly assigned (1:1) to receive either high-dose short-course oral primaquine (intervention arm; total dose 7 mg/kg over 7 days) or standard care (standard care arm; single dose oral primaquine of 0·25 mg/kg). Random assignment was done by an independent statistician in blocks of eight by use of sealed envelopes. All randomly assigned and eligible patients were included in the primary and safety analyses. The per-protocol analysis excluded those who did not complete treatment or had substantial protocol violations. The primary endpoint was the incidence risk of P vivax parasitaemia on day 63. This trial is registered at ClinicalTrials.gov, NCT03916003. Between Aug 18, 2019, and March 14, 2022, a total of 500 patients were enrolled and randomly assigned, and 495 eligible patients were included in the intention-to-treat analysis (246 intervention and 249 control). The incidence risk of P vivax parasitaemia at day 63 was 11·0% (95% CI 7·5-15·9) in the standard care arm compared with 2·5% (1·0-5·9) in the intervention arm (hazard ratio 0·20, 95% CI 0·08-0·51; p=0·0009). The effect size differed with blood schizontocidal treatment and site. Routine symptom reporting on day 2 and day 7 were similar between groups. In the first 42 days, there were a total of four primaquine-related adverse events reported in the standard care arm and 26 in the intervention arm; 132 (92%) of all 143 adverse events were mild. There were two serious adverse events in the intervention arm, which were considered unrelated to the study drug. None of the patients developed severe anaemia (defined as haemoglobin <5 g/dL). In patients with a G6PD activity of 70% or greater, high-dose short-course primaquine was safe and relatively well tolerated and reduced the risk of subsequent P vivax parasitaemia within 63 days by five fold. Universal radical cure therefore potentially offers substantial clinical, public health, and operational benefits, but these benefits will vary with endemic setting. Australian Academy of Science Regional Collaborations Program, Bill & Melinda Gates Foundation, and National Health and Medical Research Council.