Despite estimates suggesting Leptospira spp. being endemic in Southeast Asia, evidence remains limited. Diagnostic accuracy evaluations based on Leptospira ELISA mainly rely on hospitalized and severe patients; therefore, studies measuring the pathogen burden may be inaccurate in the community. We evaluated the Panbio Leptospira ELISA IgM among 656 febrile outpatients attending primary care in Chiangrai, Thailand, and Hlaing Tha Yar, Yangon, Myanmar. ELISA demonstrated limited diagnostic accuracy for the detection of acute leptospiral infection using the manufacturer recommended cutoff, with a sensitivity of 71.4% and specificity of 36.4%, and an area under the receiver operator characteristic curve value of 0.65 (95% CI: 0.41-0.89), compared with our reference test, the PCR assay. ELISA also performed poorly as a screening tool for detecting recent exposure to Leptospira spp. compared with the "gold-standard" microscopic agglutination test, with a specificity of 42.7%. We conclude that the utility of the Leptospira IgM ELISA for both serodiagnosis and seroprevalence is limited in our setting.