Study	Reason for exclusion
Abuelghar 2013	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results. Reasons for losses to follow‐up not given. Not ITT analysis
Ashrafinia 2009	This study compared metformin with LOD. The interventions were not blinded and the only reproductive outcome was menstrual frequency.
Aubuchon 2009	A cross‐over study including 8 participants who were analysed to metformin vs placebo. Participants were asked to use barrier contraception during the entire study period. The study was mechanistic and was not for ITT.
Ayaz 2013b	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results. The study was very similar to one we currently assigned to 'awaiting classification" (Ayaz 2013a), therefore we have contacted the authors to ask for confirmation as to whether HCG was used.
Aygen 2007	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Billa 2005	Article not found
Bonakdaran 2012	Quasi randomisation based on day referred to clinic
Chaudhury 2008	Quasi‐randomised ‐ alternation used for randomisation
Chou 2003	Participants were asked to use barrier contraception during the entire study period and the only reproductive outcome was menstrual frequency.
Eisenhardt 2006	The study determined effect of metformin vs placebo on insulin resistance. The only reproductive outcome was menstrual frequency and participants that conceived dropped out of the study.
Elgafor 2013	This study compared metformin and letrozole with LOD. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Fayed 2009	This study compared metformin and CC vs rosiglitazone and CC. There were no relevant comparisons or placebo.
Gada 2000	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Hashim 2010	This study compared metformin and CC vs letrozole. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Hashim 2011	This study compared metformin and CC vs LOD. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Hwu 2005	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Katica 2014	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Kazerooni 2009	This study evaluated the effect of short‐course pretreatment with metformin on hyperandrogenism, insulin resistance, cervical scores and pregnancy rates in women with CC‐resistant PCOS HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Kocak 2002	Quasi‐randomised trial comparing combined CC and metformin with CC on ovulation in CC‐resistant women with PCOS. Inadequate randomisation and sequence generation (sequential by order of admission). Admission determined by day of menses. Allocation performed by nurse blinded to the study. Odd numbers allocated metformin, even numbers allocated placebo. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Kore 2007	The diagnosis of PCOS was based on US features alone. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Leanza 2014	Participants underwent intrauterine insemination and assisted reproduction is an exclusion criteria for this review. Aspects of the methodology are missing from the article.
Maciel 2004	This study compared metformin with placebo however, there are no reproductive outcomes reported.
Maged 2015	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Mayhew 2011	This study is a review.
Melli 2010	This study compared metformin and CC with metformin and CC and fluoxetine. There are no relevant comparisons or placebo.
Moghetti 2000	This study had 2 protocols. Firstly, metformin was compared with placebo however the only reproductive outcome was menstrual frequency. Secondly, long‐term effects of metformin on ovulation were assessed however, there was no placebo/control.
Neveu 2007	This is not a RCT as women could choose which treatment: metformin and CC, CC alone or metformin alone
Palomba 2005b	This is a follow‐on study from Palomba 2005a where all participants who did not ovulate following 6 months' treatment of metformin or LOD/placebo, were given CC.
Palomba 2005c	This is a commentary of the previous paper Palomba 2005a.
Palomba 2007	This is a non‐RCT comparing metformin vs CC.
Pinnow 2008	This study is a review.
Ramzy 2003	An open‐labelled, randomised trial comparing metformin 500 mg 3/d with placebo 6 weeks prior to CC treatment. In addition, randomisation was performed using alternate numbers. These factors introduced significant bias. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Rezk 2018	This study compared metformin and CC with letrozole. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Ronsini 2006	Participants in this study underwent intrauterine insemination and assisted reproduction is an exclusion criteria for this review. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Sahin 2004	HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Santonocito 2009	The objective of this study was to compare CC with metformin on ovulation rates. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Savic 2003	Participants in this study underwent assisted reproduction, which is an exclusion criteria for this review.
Sohrabvand 2006	This study compared metformin and CC with metformin and letrozole. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Trolle 2007	The participants were asked to use barrier contraception during the entire study period and the only reproductive outcome was menstrual frequency.
Weerakiet 2011	This study compared different doses of metformin (100 mg/d and 1700 mg/d) and added CC if no evidence of ovulation. There was no placebo/control. HCG hormone was used as an ovulation trigger, which may have added additional heterogeneity to the results.
Wisniewski 2009	This study is a review.
Xiaolin 2014	Human menopausal gonadotrophin hormone was used to stimulate the ovaries. Participants undergoing assisted reproduction is an exclusion criteria for this review.

CC: clomiphene citrate; FSH: follicle‐stimulating hormone; hCG: human chorionic gonadotrophin; IVF: in vitro fertilisation; ITT: intention‐to‐treat; PCOS: polycystic ovary syndrome; RCT: randomised controlled trial; US(S): ultrasound (scan)