Thomas 1995.

Methods	Country where data collected: UK Parallel‐group RCT Unit of randomisation: participant Unit of analysis: burn Duration: NR
Participants	Inclusion criteria: < 5% TBSA, presented up to 24 h post burn Exclusion criteria: burns to face, neck, axilla; chemical and electrical burns Participants: 50 participants with 54 burns Mean age (years): NR; children 10/18 vs 7/16 vs 7/16 Male participants: NR; ratios 2:1 vs 1:1.3 vs 1:1.3 no significant difference between groups Burn type: scalds 95% vs 56% vs 88%; no significant difference between groups Burn degree: NR (minor) Burn size (%TBSA): 0.84 vs 0.94 vs 0.79; no significant difference between groups Burn location: NR
Interventions	Intervention arm 1: chlorhexidine tulle‐gras. N = 18 Intervention arm 2: hydrocolloid (granuflex). N = 16 Intervention arm 3: hydrocolloid + SSD. N = 16 Cointerventions: NR
Outcomes	Primary outcome: wound healing Secondary outcome: pain
Notes	Funding: Convatec/Squibb supplied granuflex
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Patients were randomly allocated to one of three treatment groups after obtaining informed consent" Comment: no information on how the randomisation sequence was generated
Allocation concealment (selection bias)	Unclear risk	Quote: "Patients were randomly allocated to one of three treatment groups after obtaining informed consent". Comment: no information on whether allocation concealment was adequate
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quote: "During dressing changes the healing progress of the wound was noted..." Comment: no information on whether observers were blinded; balance of probabilities would be not
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Comment: no specific quote but unclear whether all randomised participants were included in analysis
Selective reporting (reporting bias)	Low risk	Comment: no specific quote but outcomes mentioned in early part of text are reported in findings
Other bias	High risk	Comment: unit of analysis issues as randomisation was at the participant level whilst analysis was at the level of burn wounds (some participants had multiple burns)