Objective. To assess efficacy and tolerability of vaginal compared with intramuscular progesterone in reducing the rate of recurrent preterm birth before 34 weeks of gestation. Design. Prospective randomized study. Setting. Obstetrics and Gynecology Department, Armed Forces Hospital Southern Region, Kingdom of Saudi Arabia. Sample. Five-hundred and eighteen women with a prior history of preterm birth. Methods. Women were randomized to receive either 90 mg of vaginal progesterone gel once daily or 250 mg of intramuscular progesterone weekly. Treatment began between 14 and 18 weeks of gestation and continued until 36 complete weeks of gestation, delivery or the occurrence of premature rupture of membranes. Main outcome measures. The primary outcome measure was delivery before 34 weeks of gestation. The secondary outcome measures were PTB between 34 and 37 weeks of gestation and neonatal outcomes including birthweight, neonatal death, and the need for admission to the neonatal intensive care unit. Results. The baseline characteristics of the study participants were similar. Two-hundred and thirty-eight (94.1%) patients in the vaginal group and 226 (90.8%) patients in the intramuscular group were compliant with their medications. Vaginal progesterone was associated with a lower percentage of deliveries before 34 weeks of gestation than the intramuscular preparation (p= 0.02). This association was also observed at 28 and 32 weeks of gestation (p= 0.04). Adverse effects were reported in 14.1% of patients in the intramuscular group, but in only 7.5% of patients in the vaginal group (p= 0.017). Conclusions. Vaginal progesterone was more effective than intramuscular progesterone for the prevention of preterm birth and had fewer adverse effects.