Ismail 2011

Methods	Study design: randomised controlled trial Conducted in: Detroit, Michigan, USA Unit of randomisation: household (child + caregiver) Unit of analysis: children Setting: low‐income African American population in Detroit, Michigan Funded by: "This study was supported with funding from the National Institute of Dental and Craniofacial Research, the Delta Dental Fund of Michigan, and the University of Michigan’s Office of Vice Presidential Research" Duration of the study: 6 years
Participants	Inclusion criteria: housing units with families making < 250th percentile of the poverty line and having ≥ 1 African American child < 5 years old Exclusion criteria: only 1 child from birth to 5 years of age per family was selected for inclusion Age at baseline: intervention group = 4.63 years, control group = 4.51 years Total N at (baseline): 1021 N (controls baseline): 515 N (controls follow‐up): 300 N (interventions baseline): 506 N (interventions follow‐up): 299 Recruitment: Participating families were recruited in a longitudinal study of determinants of dental caries in 1021 randomly selected children (0 to 5 years) and their caregivers Gender Intervention group: 55.5% = female, 44.5% = male Control group: 53% = female, 47% = male
Interventions	Intervention: 15‐minute educational video and motivational interview session. Follow‐up phone call within 6 months of receipt of the intervention by the caregiver. Personalised oral health brochure outlining child’s oral health goals Control: 15‐minute educational video Duration of intervention: 15‐minute DVD + average 40‐minute intervention sessions
Outcomes	New non‐cavitated, new cavitated, new untreated lesions
Implementation related factors	Theoretical basis: social‐cognitive theory, socio‐ecological model, motivational interviewing Resources for implementation: 2‐day training session covering basic principles of MI8, DVD, educational material such as glossy brochure and magnet, motivational interviewer Who delivered the intervention: trained motivational interviewer PROGRESS categories assessed at baseline: place, race, gender, SES, education PROGRESS categories analysed at outcome: not reported Outcomes related to harms/unintended effects: not reported Intervention included strategies to address diversity or disadvantage: African American population, low socioeconomic group Economic evaluation: approximate cost per Swedish crown provided (refer to page 90, Nylander 2001)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A random number was generated for each child using the RAND function in MS Excel. Random numbers were classified into odd and even numbers, and each child was assigned to 1 of the 2 groups
Allocation concealment (selection bias)	Low risk	Assignment of children was masked to participants, project staff (with the exception of co‐ordination desk and interviewing staff), examining dentists and analysis
Incomplete outcome data (attrition bias) All outcomes	Low risk	A small number of missing values (< 4% for any single item) was imputed using IVEware (51), a SAScallable software application. IVEware imputes missing and non‐substantive (‘don’t know’ or ‘refused’) responses using a multiple imputation method in which a sequence of regression models are fit, and values are drawn from predictive distributions. Missing values for dental outcomes were not imputed
Selective reporting (reporting bias)	Unclear risk	Unclear
Other bias	Unclear risk	Unclear
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Assignment of children was masked to participants, project staff (with the exception of co‐ordination desk and interviewing staff), examining dentists and analysis
Blinding of outcome assessment (detection bias) All outcomes	High risk	Examining dentists, desk co‐ordinating officer and analysts were not blinded