We investigated the relationship between mefloquine antimalarial treatment and the outcome of pregnancy in Karen women living in an area along the western border of Thailand where multidrug-resistant Plasmodium falciparum infections are common. Of 3,587 pregnancies investigated, 208 (5.8%) were exposed to mefloquine, 656 (18.3%) to quinine only, and 909 (25.3%) to other antimalarials, and 2,470 (68.9%) had no documented malaria. There were 61 stillbirths and 313 abortions. Women who received mefloquine treatment during but not before pregnancy had a significantly greater risk of stillbirth than did women treated with quinine alone (odds ratio [OR], 4.72; 95% confidence interval [CI], 1.7-12.7), women exposed to other treatments (OR, 5.10; 95% CI, 2-13.1), and women who had no malaria (OR, 3.50; 95% CI, 1.6-7.6) (P < .01). This association remained after adjustment for all identified confounding factors. Mefloquine was not associated with abortion, low birth weight, neurological retardation, or congenital malformations. Mefloquine treatment during pregnancy was associated with an increased risk of stillbirth.