Glazener 2017.

Methods	RCT: 2 parallel comparing A:B (mesh trial) A:C (graft trial) 35 centres in UK 65 surgeons Remote Web‐based randomisation 2‐Year follow‐up Modified ITT analysis
Participants	1352 randomised, 320 randomised within posterior repair subgroup n = 1348 Mesh trial (overall not just posterior) A: 111 (430) B: 111 (435) Graft trial A: 93 (367) C: 98 (368) 35 centres Primary anterior or posterior repairs
Interventions	A: native tissue B: synthetic mesh C: biological graft
Outcomes	Repeat surgery for prolapse Recurrent posterior vaginal wall prolapse (objective failure) Prolapse outcomes Mean postop Bp, Ba, C QOL (POP‐SS) ‐ this is the primary outcome in this trial Adverse events ‐ postoperative complications
Notes	HTA‐funded study in UK
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated online programme
Allocation concealment (selection bias)	Low risk	Sequentially numbered, opaque, sealed envelopes
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Patients were blinded until 12 months unless asked
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Reviewers attempted to remain blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	20% at 2 years. Specific attrition rates for posterior repair groups not given
Selective reporting (reporting bias)	Low risk	Reported on all primary outcomes of this review, protocol available
Other bias	Low risk	Funding declared and no COI Concomitant procedures included anterior and apical repair procedures ‐ same numbers in comparison groups had concomitant procedures