| Methods |
15 sites in US.
Six‐month wash‐out period for injectable or implant contraceptive users. Six treatment cycles.
Unblinded. |
| Participants |
Sexually active, normal weight women aged 18 to 50 years with regular menses.
Excluded contraindications to oral contraceptive use; smokers over 35 years of age; heavy alcohol use; and lactation. |
| Interventions |
EE 20 µg and levonorgestrel 100 µg (N=154) versus EE 20 µg and desogestrel 150 µg (N=154) versus EE 35 µg and norgestimate 180‐215‐250 µg (N=155). |
| Outcomes |
Cycle control.
'Unscheduled' bleeding was defined as bleeding not continuous with withdrawal bleeding. 'Withdrawal bleeding' was not defined. |
| Notes |
Allocated with masked randomization lists at each site generated with block size of six.
86% (398/463) completed study. |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Allocation concealment (selection bias) |
Unclear risk |
Not reported |
