Miranda 2005.

Methods	Design: split‐mouth RCT Number of participant: 80 Setting: dental clinic Country: Brazil Unit of randomisation: tooth Unit of analysis: tooth pairs Follow‐up: 6 and 12 months Dropout: 3.75% after 6 months and 12.5% after 12 months
Participants	Number randomised (participants): 80 children/160 teeth (80 ART group and 80 CT group) Number analysed: 70 children/140 teeth Age mean and SD (range): 5.71 years (3‐9 years) Gender: female 33 (41.25%), male 47 (58.75%) Average DMFT score: not reported Dentition: primary Type of caries lesion: single and multiple surface caries lesion Inclusion criteria Child between 3‐9 years ≥ 2 primary molars with similar carious lesions (equal number of surfaces involved, extent and similar depths) Carious lesions in dentin with access in enamel > 1 mm and that was accessible to hand instruments Teeth without pulp exposure Exclusion criteria Children without ability to co‐operate in treatment
Interventions	Two treatment arms: Group 1: ART approach + H‐GIC Group 2: CT + amalgam Teeth in the ART group were treated with hand instruments only. The restorative material was glass ionomer (Ketak‐Molar 3‐M ESPE). In CT group, cavities were filled with silver amalgam (SDI), after removing carious tooth tissues and preparation of cavities with high and low‐speed drill. Both treatments were started without use of anaesthesia. The interventions were conducted by 1 dentist
Outcomes	Success rate was assessed by ART criteria after 6 and 12 months (0 = present, in good condition, 1 = present, local marginal defect (0.5 mm), no repair needed, 2 = present, unique defect > 0.5 and < 1 mm, repair needed, 3 = present, gross marginal defect, repair needed, 4 = not present, restoration partly or completely missing, 5 = not present, restoration replaced by another restoration, 6 = tooth missing, 7= present, wear < 0.5 mm, no repair needed, 8 = present, wear > 0.5 mm, repair needed, 9 = restoration not assessed, participant not present. Codes 0, 1 and 7 were considered success and 2, 3, 4 and 8 as failure. Restorations with codes 5, 6 and 9 were excluded from the analysis. Pain during the treatment was classified as absence of pain, little pain or much pain Recurrent caries assessed as caries on the margin of the restorative material
Notes	Funding not stated Trial register number no reported Sample size calculated Intraexaminer reproducibility not assessed
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "We used a simple randomised to two treatment cited by Pocock (1993) and a table of random numbers, randomised formed by digits from 0 to 9 in a sequence from right to left and from top to bottom"
Allocation concealment (selection bias)	Low risk	Quote: "The concealment was performed through sealed envelopes numbered 1‐100, containing inside cards with corresponding number and an indication of the first treatment, obtained by the method mentioned, being sequentially archived. The listing and envelopes were made by a professional different to the researcher."
Blinding of participants and personnel (performance bias) ‐ participant	High risk	Comments: participant aware of different treatments
Blinding of participants and personnel (performance bias) ‐ operator All outcomes	High risk	Comments: blinding not possible ‐ operator knew the intervention
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The restorations were evaluated by paediatric dentist who did not perform any treatment"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comments: low dropout rate (12.5%), reasons for missing outcome data unlikely to be related to true outcome
Selective reporting (reporting bias)	Low risk	Comments: all prespecified (primary and secondary) outcomes reported
Other bias	Low risk	Comments: split‐mouth design with the same baseline diagnosis of the teeth within a tooth pair