Increased safety of oral contraceptives (OC) has resulted from a reduction in the estrogen and progestin content per tablet. A reduction in the number of hormonally active pills and their placement at critical points within the cycle may provide a novel regimen for further reducing the hormonal content of OC per cycle and their attendant side effects without compromising efficacy. The objective of this study was to determine the effectiveness of two OC regimens that incorporate a delayed start and limited midcycle use of the combination of ethinyl estradiol and norethindrone, and limited use of norethindrone only during the second half of the cycle. Main outcome measures were defined as ovulation, serum concentrations of estradiol (E2), luteinizing hormone (LH), follicle stimulating hormone (FSH), progesterone (P), follicular diameters, and endometrial thickness. Volunteers were issued blister packs containing 28 pills and randomized to one of two groups. Group 1 used a combination of 50 micrograms ethinyl estradiol and 1 mg norethindrone per tablet day 6-10, and 0.70 mg norethindrone only day 11-19. Placebo tablets were used on days 1-5 and day 20-28. Group 2 used a combination of 50 micrograms ethinyl estradiol and 1 mg norethindrone per tablet on day 8-12, and 0.70 mg norethindrone only on day 13-21. Placebo tablets were used on day 1-7 and day 22-28. A total of 20 cycles were studied using 10 volunteers. To assess any possible carryover effect, two successive cycles were studied for each subject. Serum sampling for E2, FSH, LH, and P, and transvaginal ultrasound imaging to assess endometrial thickness and follicle diameter were carried out at 4 day intervals throughout the cycle. One ovulation occurred in 10 cycles in group 1. Five ovulations occurred in 10 cycles in group 2. All ovulations, regardless of group, occurred in the second cycle. Peak E2 concentrations were not significantly different between groups (152.04 +/- 107.1 pg/mL vs 162.1 +/- 56.1 pg/mL [mean +/- SD] for groups 1 and 2, respectively] but occurred earlier in the cycle in group 1. No differences were noted between the groups in serum concentrations of FSH or LH for any given cycle day. Maximum follicle diameters were not different between groups 1 and 2, regardless of ovulatory status (20.5 +/- 8.1 mm2 vs 20.6 +/- 14.2 mm2, respectively). Ultrasound imaging assessment of midcycle follicle growth revealed diameters ranging from 18.5 mm2 to 34.0 mm2 with gradual resolution through the second half of the cycle in anovulatory cycles, and 16.0 mm2 to 23.5 mm2 with abrupt disappearance in ovulatory cycles. Endometrial thickness did not exceed 10 mm for any anovulatory cycle regardless of group, but ranged from 6 to 9 and 6 to 11 during the luteal phase of ovulatory cycles of groups 1 and 2, respectively. Peak serum P concentrations at midluteal phase in ovulatory cycles ranged from 9.2 ng/ml to 18.2 ng/ml. Data from this preliminary study suggest that ovulation may be prevented with a combination of ethinyl estradiol and norethindrone started as late as cycle day 6 and limited to 5 days' duration using norethindrone only for 9 days during the second half of the cycle. Such a restricted regimen may offer both an effective method of contraception and a means of further reducing both estrogen and progestin content per cycle and the possible short and long term adverse side effects of these hormones.