Kim 2011.

Methods	Parallel group study. Number of women randomized: 120. Number of withdrawals: 0. Number of women analyzed: 120. Duration of study: 1 cycle.
Participants	Country of authors: South Korea. 120 poor responders (repeated day 3 levels of FSH > 8.5 mIU/mL or antral follicle count ≤ 5, or both). 40 in each group. Inclusion criteria: not clearly stated. Exclusion criteria: PCOS (Rotterdam criteria). Mean age ± SD: group 1: 36.7 ± 3.1 years; group 2: 35.9 ± 2.8 years; group 3: 36.4 ± 3.3 years. Setting: university‐based infertility clinic, Seoul. Korea.
Interventions	Pretreatment was ethinyl oestradiol 0.03 mg and levonorgestrel 0,15 mg for 21 days in the cycle preceding COS. Group 1: GnRH antagonist multiple dose protocol after OCP pretreatment. Ovarian stimulation started 5 days after OCP discontinued using rFSH (225 IU/day, dose adjusted every 3‐4 days). Cerotide (0.25 mg) started when lead follicle was 14 mm diameter and continued until day of hCG injection. Group 2 GnRH antagonist multiple dose protocol without OCP pretreatment. Ovarian stimulation started on cycle day 3 using rFSH (225 IU/day, dose adjusted every 3‐4 days). Cerotide (0.25 mg) started when lead follicle was 14 mm diameter and continued until day of hCG injection. Group 3: GnRH agonist luteal low‐dose long protocol without OCP pretreatment. Daily injection of decapeptyl (0.1 mg) started from mid‐luteal phase and continued until menses followed by a dose reduction to 0.05 mg daily and continued until day of hCG injection.
Outcomes	Primary: Number of mature oocytes retrieved. Secondary: Total amount and days of rFSH. Number of fertilised oocytes and grade I and II embryos. Implantation rate ongoing pregnancy rate. Clinical pregnancy rate per cycle and live birth rate per cycle. Miscarriage rate.
Notes	Power calculation: yes. ITT analysis: yes. Earlier publications were Kim 2005 and Kim 2009.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Computer generated lists."
Allocation concealment (selection bias)	Unclear risk	Quote: "The sequence of allocation to the three groups was provided to the investigating physicians."
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Not reported.
Incomplete outcome data (attrition bias) All outcomes	Low risk	In groups 1 and 3 there were no losses, withdrawals or cancellations. In group 2, 1 cycle was cancelled before embryo transfer; all women randomized were included in data analysis.
Selective reporting (reporting bias)	Low risk	All outcomes listed were reported although multiple pregnancy was reported it is not listed a priori.
Other bias	Low risk	Groups similar at baseline with respect to demographic characteristics.