Cruz 2016.

Methods	Design: cluster, parallel RCT (a child is a cluster) Number of participants: 75 Setting: nursing home Country: Colombia Unit of randomisation: participant Unit of analysis: tooth Follow‐up: 6 months Dropout: 14.9 % after 6 months
Participants	Number randomised: 75 participants; 174 teeth (73 ART group and 101 CT group) Number analysed: 64 participants/148 teeth Age mean and SD (range): 74.9 years (60‐101) Sex: female 36 (48%), male 39 (52%) Average DMFT score: not reported Dentition: permanent Type of caries lesion: root caries Inclusion criteria: root caries defined as the softening of the root dentin to a depth of ≥ 0.5 mm Exclusion criteria: teeth with extraction indication, lesion close to the dental pulp or pain symptomatology
Interventions	Two treatment arms: Gp 1: ART approach + RM‐GIC Gp 2: CT + RM‐GIC ART was performed using only manual instrumentation to remove decayed tissue. Cotton rolls and a retraction cord were used to obtain relative isolation of the operative field. 2% chlorhexidine (Clorhexol 0.2 g/100 mL; Farpag®, Bogota, Colombia) was applied for 1 min and the cavity was dried and sealed with aglass ionomer cement modified with light‐curing composite resin (Vitremer™®, 3M ESPE, Seefeld, Germany). Interproximal metal and paper strips were used. Conventional technique was performed using a high‐speed handpiece with irrigation and round diamond burs of different diameters. Cavities were restored with RM‐GIC. Use of anaesthesia was not reported in any group. The interventions were conducted by 2 dentists.
Outcomes	Success rate and survival rate according to following criteria: 'successful' if the restoration was present and without marginal defects or secondary caries; 'survival' if the restoration was present with a marginal defect of 0.5 mm or less and without secondary caries; and 'failure' if the restoration was absent, if there was a marginal defect greater than 0.5 mm, or if there were secondary caries Secondary caries defined as softened root dentin with the contact of the periodontal probe on the margin of the restorative material
Notes	Funding: COLCIENCIAS for the Young Researcher Scholarship‐Internship Program Trial register number not reported Sample size calculated Intraexaminer and interexaminer reproducibility not assessed
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "A series of random numbers was used to fabricate sealed envelopes that were only opened for the random allocation of the participants to each working group (ART or conventional technique with rotary instruments)"
Allocation concealment (selection bias)	Low risk	Quote: "A series of random numbers was used to fabricate sealed envelopes that were only opened for the random allocation of the participants to each working group (ART or conventional technique with rotary instruments)"
Blinding of participants and personnel (performance bias) ‐ participant	High risk	Comment: no information provided, but the participants could tell whether manual or rotary instruments were used
Blinding of participants and personnel (performance bias) ‐ operator All outcomes	High risk	Comment: blinding not possible ‐ operator knew the intervention
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "After six months, the condition of the restorations was assessed by two different prosthodontists, without awareness of the technique that was performed in each participant"
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "After six months, 64 participants were evaluated (32 men and 32 women) and 26 restorations (14.9%) were lost. Seven participants changed geriatric institutions and were lost to follow‐up, two died, and the two remaining participants were unreachable at the institution during the time of revision”
Selective reporting (reporting bias)	Low risk	Comment: all outcomes listed in the methods sections were included.
Other bias	High risk	Comment: no information provided about baseline characteristics of included participants. The analysis did not consider the pair data.