Kumar 2011.

Methods	2‐arm parallel randomised controlled trial
Participants	Numbers recruited: 224 (gender split not reported) Mean age: not reported Inclusion criteria: people who had pre‐adjusted edgewise appliance in each arch and: willing to wear upper and lower retainers Exclusion criteria: none reported Setting: university clinic in Department of Orthodontics & Dentofacial Orthopedics, College of Dental Science, Davangere, Karnataka, India
Interventions	Comparison: upper and lower Begg retainers vs. upper and lower thermoplastic retainers Group 1: upper and lower Begg retainers. Acrylic baseplates with labial bow of 0.9 mm stainless steel wire. Labial bow had U‐loops opposite the premolars and extended past the last erupted molars Group 2: upper and lower thermoplastic (vacuum‐formed) ‐ Essix A+, (Raintree Essix, New Orleans, USA) retainers All participants had bonded wire placed in the lower as well The participants were instructed to wear them 24 hours per day for 6 months (and then 12 hours per day for the following 6 months ‐ although the 1‐year data were not reported). They were instructed to eat with the retainers in place and remove after eating for cleaning
Outcomes	Outcomes relevant to review: stability and patient satisfaction Stability assessed using PAR and Little's Irregularity Index. The results for the lower irregularity were reported, but not the upper Participant satisfaction data included the mean, but no standard deviations, so the data could not be analysed using the Review Manager 5 software (RevMan 2014). We contacted the authors to see if they could provide the raw data to allow analysis, but received no reply. Relapse was assessed at 2 time points after debond: 3 and 6 months
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported ‐ authors contacted about this but no reply received
Allocation concealment (selection bias)	Unclear risk	Not reported ‐ authors contacted about this but no reply received
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported ‐ authors contacted about this but no reply received
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Not reported ‐ authors contacted about this but no reply received
Selective reporting (reporting bias)	High risk	Results of irregularity reported in the lower, but not the upper arch
Other bias	Unclear risk	Insufficient data in the paper to determine this. Authors contacted but no reply received