Mifepristone was recently approved in the United States. Regimens with shorter intervals may be more acceptable. The objective of this study was to determine whether the oral route of misoprostol was as effective as the vaginal route of misoprostol 1 day after mifepristone. A prospective, open-labeled, randomized trial of healthy adult women up to 63 days pregnant and wanting a medical abortion were randomized to use either two doses of oral misoprostol 400 microg taken 2 h apart or misoprostol 800 microg vaginally. Women self-administered misoprostol 1 day after taking one-third of the standard dose of mifepristone (200 mg) orally. Women then returned to the clinic up to 5 days later for a repeat sonogram evaluation. A dose of vaginal misoprostol was administered to women with a continuing pregnancy who then returned 1 day later to Day 15. The primary outcome measures were a complete medical abortion by the first or by the second follow-up visits. Surgical intervention was indicated for continuing pregnancy at the second follow-up visit, excessive bleeding, or persistent products of conception 5 weeks later. One thousand one hundred sixty-eight women were enrolled. Of the 1144 (98%) women who complied with their random assignment, two oral doses of misoprostol (800 microg total) were 90% effective at inducing an abortion by the first follow-up visit, compared with one dose of misoprostol by vagina of 97% (chi(2) = 23.95, p = 0.001). By the second follow-up visit, the complete abortion rate was 95% for oral misoprostol and 99% for vaginal misoprostol (chi(2) = 21.76, p = 0.001). There were minimal differences in side effects. Women preferred the oral route. The trial demonstrated that although two doses of oral misoprostol were effective, the vaginal misoprostol was more effective at inducing an early medical abortion at 1 day after low-dose mifepristone, and the regimen could be extended to 63 days gestation.