Frazão 2011

Methods	Study design: randomised controlled trial Conducted in: Sao Vicente, Brazil Unit of randomisation: preschools Unit of analysis: individual Setting: preschool Funded by: "Funding provided by the Foundation for the Support of Research of the State of São Paulo, Brazil and the Department of Health, Sao Vicente City Hall" Duration of the study: 18 months
Participants	Inclusion criteria: All children 5 years of age showing ≥ 1 permanent molar with ≥ 1 surface exposed were considered eligible Exclusion criteria: not reported Age at baseline: mean age in control group: 68.40 months Mean age in test group: 68.56 months N (control baseline): 130 N (intervention baseline): 154 Carious lesions of enamel/dentin were analysed in 284 children who met eligibility criteria. Loss of eligible participants was 3.3% (3/90) and 2.7% (8/288), respectively, at first and third follow‐up Recruitment: not reported Gender Control: F = 66/M = 64 Intervention: F = 94/M = 60
Interventions	Intervention: On test drives, in addition to this conventional activity (oral health education), an oral health auxiliary was able to apply, with permanent molars erupted and both superiors and inferiors, the bucco‐lingual brushing technique using the participant's brush Control: Control units: Conventional programme composed of oral health education and presentation of plaque followed by brushing with fluoride toothpaste (1100 μg/g gF) overseen by a dental health aide was held 4 times a year. Educational component was developed through playful activity in the classroom, lasting 30 to 40 minutes, in which participants were encouraged to identify friends and enemies of the health of teeth. Children were provided with a children's toothbrush, and daily brushing was indirectly supervised by teachers in the covered patio of the school. This was carried out 4 times per year Duration of intervention: 5 times per year
Outcomes	Dental caries (dmft6)
Implementation related factors	Theoretical basis: cost‐effectiveness Resources for implementation: not reported Who delivered the intervention: dentists and teachers PROGRESS categories assessed at baseline: age and gender PROGRESS categories analysed at outcome: age and gender Outcomes related to harms/unintended effects: not reported Intervention included strategies to address diversity or disadvantage: low‐income areas Economic evaluation: For calculation of costs, direct expenses related to human and material resources were considered. Costs were calculated on the basis of the number of work hours and the units of brush and toothpaste consumed. Effectiveness was measured by the number of carious lesions found throughout the study (18 months). Calculation of cost‐effectiveness was expressed by the marginal difference in actual expenditures for injury avoided. To allow comparison between programmes, spending were standardised for 18 months, and values were adjusted for each thousand children, with separate analyses for girls and boys. Values in Brazilian reais were converted on the basis of the commercial dollar exchange rate for the half the study period (29/Apr/2008‐US $ 1 = R $1.70). Modified programme cost about $3.04 per child, resulting in a marginal cost‐effectiveness ratio of $35.50 per avoided injury, with around $10.00 per avoided injury among boys (highest risk group)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation occurred by preschool, and preschool children were distributed randomly to test and control groups. Although the study used preschool as the unit of randomisation to avoid contamination, the individual level was the primary focus of the study outcome and was adopted as the inference level
Allocation concealment (selection bias)	Low risk	Allocation concealment was not possible, given the nature of the intervention. All activities were similar in intervention and control groups, except cross‐brushing on surfaces of first permanent, which was delivered to children in the intervention group. In the control group, oral health education and dental plaque dying followed by toothbrushing with fluoride dentifrice supervised directly by a dental assistant was undertaken 4 times per year. In addition to these activities, children in the test group undertook professional cross‐brushing on surfaces of the first permanent molar delivered by a specially trained dental assistant 5 times per year
Incomplete outcome data (attrition bias) All outcomes	Low risk	The dropout rate in the study was very low. Rates were 3.3% and 2.7%, respectively, because of participant absence from preschool
Selective reporting (reporting bias)	Unclear risk	Unclear
Other bias	Unclear risk	Unclear
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Examiner was kept masked to group assignment, and dental assistant in charge of the control units was kept blinded about differential characteristics of the intervention in test preschools. Participants were kept unaware of whether they belonged to control or test units
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Examiners were blinded to differential characteristics of the intervention in test preschools