Salehi 2013.

Study characteristics
Methods	2‐arm parallel RCT
Participants	Number recruited: 142 Sex (M:F): 59:83 Mean age (years): group 1: 18.1 (SD 5.23), group 2: 18.2 (SD 4.81) Inclusion criteria: people treated with standard edgewise fixed appliances, good oral hygiene, healthy periodontium, no previous bonded retainer Exclusion criteria: deep overbite, traumatic parafunctional habits such as bruxism and clenching Setting: clinic of 1 operator in Iran; participants paid for treatment
Interventions	Two types of upper and lower fixed retainers Group 1: fibre‐reinforced composite retainer ('Ribbond') was cut to the correct length, pre‐treated with adhesive bis‐GMA sealant and bonded with 'Heliosite' orthodontic composite resin Group 2: multistrand wire retainer was ⌀ 0.0175 inch passive multistrand wire Both types were bonded to all teeth, canine to canine Both retainers bonded under rubber dam with Heliosite composite resin using Fluoro Bond adhesive bis‐GMA sealant
Outcomes	Survival: retainer assessed as failed if it debonded from tooth or fractured Time points: at 18 months
Notes	No assessment of stability, adverse effects on health or patient satisfaction
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random table number table used
Allocation concealment (selection bias)	Unclear risk	No concealment described. Study authors contacted to clarify this but no reply received
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Study authors noted that the statistician analysing the data was blinded
Incomplete outcome data (attrition bias) All outcomes	Low risk	Number of dropouts clearly reported and described
Selective reporting (reporting bias)	Low risk	No evidence or suggestion of selective reporting
Other bias	Low risk	No other suggestions of bias noted