Background: Primaquine is used to eliminate Plasmodium vivax hypnozoites, but its optimal dosing regimen remains unclear. We undertook a systematic review and individual patient data meta-analysis to investigate the efficacy of different primaquine dosing regimens to prevent P. vivax recurrence.Methods: P. vivax efficacy studies published between January 1, 2000 and February 16, 2021 were identified by a systematic review of Medline, Web of Science, Embase and Cochrane Central. Shared individual patient data were pooled and standardised. The effects of primaquine regimen duration and total dose of primaquine on the rate of first P. vivax recurrence between day 7 and 180 were assessed by multivariable Cox regression with random effects for study site. The effect of primaquine daily dose on the composite incidence of gastrointestinal symptoms on days 5-7 was assessed by multivariable logistic regression. PROSPERO registration CRD42019154470.Results: Following identification of 210 studies, 6,879 patients from 23 studies and 16 countries were included. In 1,470 patients treated without primaquine, the risk of recurrence was 51×0% (95%CI 48×2-53×9) by day 180, compared with 19×3% (16×9-21×9) in 2,567 patients treated with low dose primaquine (~3×5 mg/kg total) and 8×1% (7×0-9×4) in 2,813 patients treated with high dose primaquine (~7 mg/kg total), regardless of duration. Compared to patients treated without primaquine, the rate of vivax recurrence was lower following treatment with both low dose (adjusted hazard ratio (AHR)=0×21, 0×17-0×27; p<0×0001) and high dose primaquine (AHR=0×10, 0×08-0×12; p<0×0001). Findings: were similar in high and low relapse periodicity regions. Gastrointestinal symptoms on days 5-7 were reported by 4×0% (0×6-7×4) of patients treated without primaquine compared to 9×5% (3×3-15×6) treated with high daily dose primaquine (~1 mg/kg/day); p=0×0020. ConclusionsRaising the total dose of primaquine to 7 mg/kg provides greater efficacy in preventing P. vivax recurrences in many endemic regions with a small dose-related increase in gastrointestinal adverse events.Funding: RJC, BEB, MJG, JAS and RNP are supported by Australian National Health and Medical Research Council (NHMRC) Investigator Grants (1194702, 2016792, 2017436, 1196068, 2008501). CSC works at Shoklo Malaria Research Unit which is supported by the Wellcome Trust (220211). NJW is a Wellcome Trust Principal Fellow (093956/Z/10/C). JAW is a Sir Henry Dale Fellow funded by the Wellcome Trust (223253/Z/21/Z). JH receives salary support from the U.S. President’s Malaria Initiative. This research was supported by grants from Bill & Melinda Gates Foundation and Medicines for Malaria Venture.Declaration of Interest: All authors declare no competing interests.Ethical Approval: Shared data were obtained according to ethical approvals from the country of origin and original study. The data were anonymised and unable to be linked to individuals. As such, our analysis did not require additional ethical approval according to the guidelines of the Oxford Central University Research Ethics Committee.