Hampton 2001.

Methods	Randomized controlled trial. Randomization in a 3:3:3:2 ratio in blocks of size 11:9. The method of randomization was not described. NGM/EE regimens were blinded and Loestrin Fe open. Communication with the authors indicated a computer‐generated random allocation sequence. The randomization was balanced using permuted blocks and stratified by study center.
Participants	6022 women at 110 sites in the USA and Canada. One‐third of the women participated in the study for 13 cycles, two‐thirds of the women participated for 6 cycles. Inclusion criteria were women aged 18 to 45 years who had regular menstrual cycles, were sexually active, at risk of pregnancy and agreed to use only the study drug as contraception. Exclusion criteria were positive serum beta‐hCG pregnancy test, seated systolic/diastolic blood pressure more than 140/90 mm Hg, lactation or pregnancy within 42 days of study admission, any disorders that were contraindications to steroid hormonal therapy, uncontrolled thyroid disorder, cervical dysplasia, smoking in women older than 35 years, exposure to etretinate, receipt of an experimental drug, device, hepatic enzyme‐inducing drug, isotretinoin or tretinoin within 30 days of screening, receipt of Depo‐Provera within 6 months of screening and alcohol or substance abuse within 12 months of screening. More than 60 percent of the women used oral contraceptives less than 2 months before admission.
Interventions	Triphasic norgestimate/ethinylestradiol (180‐215‐250 μg NGM and 25 μg ethinylestradiol in a 7/7/7 days regimen and 7 days of placebo tablets, N=1723 for 6 cycles of whom N=487 continued for an additional 6 cycles) [Ortho Tri‐Cyclen Lo] versus monophasic norethindrone acetate/ethinylestradiol and ferrous fumarate (1000 μg NETA and 20 μg EE for 21 days and 7 days of 75 mg ferrous fumarate, N=1171 of whom N=318 continued for an additional 6 cycles) [Loestrin‐Fe] versus 'cyclophasic' norgestimate/ethinylestradiol (250‐180 μg NGM and 25 μg EE in an alternating 2‐day regimen) [Cyclophasic‐25] versus 'cyclophasic' norgestimate/ethinylestradiol (180‐60 μg NGM and 20 μg EE in an alternating 2‐day regimen) [Cyclophasic‐20].
Outcomes	Primary outcomes are: pregnancy; cycle control, side effects; laboratory changes; body weight; vital signs; changes in physical examination. Use of daily diary cards to collect data on pill‐intake, cycle control and side effects. Data on side effects were recorded if reported in response to a general question or observed during physical examination. In the original article the recommendations by Belsey 1986 were used for analyzing bleeding patterns. Breakthrough bleeding was defined as bleeding that requires sanitary protection of more than one pad or tampon per day and occurs during the active tablet‐taking interval but is not contiguous with menses. Breakthrough spotting  was defined as bleeding that requires one or less pad or tampon during the active tablet‐taking interval but is not contiguous with menses. The definition of amenorrhea was two consecutive cycles without any bleeding or spotting. In a secondary article the bleeding data were re‐analyzed with the new recommendations by Mishell 2007. Unscheduled bleeding was defined as any bleeding that occurs while taking active hormones, regardless of the duration of the regimen, with the exception of bleeding that begins during the hormone‐free interval and continues through days 1‐4 of the subsequent active cycle and bleeding reported during days 1‐7 of the first cycle of combined hormonal contraceptive therapy. The definition of amenorrhea was absence of all bleeding or spotting. A third article examined bleeding patterns by age and weight subgroups. Outcomes for 'cyclophasic' norgestimate/ethinylestradiol groups are not described.
Notes	The report does not provide an a priori hypothesis. The paper states that the sample size was determined to meet the US regulatory requirements of at least 10,000 cycles for the evaluation of the safety and efficacy of oral contraceptives with at least 200 participants evaluated for 13 cycles. Study duration: 6 and 13 cycles. In the group of participants enrolled for a trial period of 6 cycles, 258 women taking triphasic pills and 176 women taking monophasic pills discontinued early. In the group of participants enrolled for a trial period of 13 cycles, 204 women using triphasic pills and 126 women using monophasic pills discontinued early. The reasons for discontinuation are partially described. The paper reports that 6.5% of the women in the triphasic group and 5.8% of the women in the monophasic group were lost to follow up but provides no numerator. The number of women withdrawn because of protocol violations is not mentioned. The paper states that the evaluation of contraceptive efficacy was based on an intent‐to‐treat analysis. The evaluation of cycle control was not according to the intention‐to‐treat principle. Cycles in which data on dosing and bleeding was lacking and cycles with incorrect pill‐intake were excluded from the analysis. The trial was sponsored by the manufacturer of the studied triphasic norgestimate/ethinylestradiol pill (Johnson & Johnson).
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	Not described in report. Communication with the authors indicated allocation concealment by a centralized voice‐activated randomization system.