Fitzgibbon 2005.

Study characteristics
Methods	Study design: cluster‐RCT Intervention period: 14 weeks Follow‐up period (post‐intervention): 2 years Differences in baseline characteristics: reported Reliable outcomes: reported Protection against contamination: NR Unit of allocation: preschool Unit of analysis: individual To assess possible bias in results because of children leaving school or missing anthropometric data at a specific follow‐up, 2 additional analyses were conducted in which study authors imputed BMI 1 and 2 years post‐intervention from prior (baseline, post‐intervention, or Year 1) or subsequent (Year 2) values of BMI.
Participants	N (controls baseline) = 212 N (controls follow‐up) = post‐intervention (N = 183); 1‐year follow‐up (N = 146); 2‐year follow‐up (N = 154) N (interventions baseline) = 197 N (interventions follow‐up) = post‐intervention (N = 179); 1‐year follow‐up (N = 143); 2‐year follow‐up (N = 146) Setting (and number by study group): preschools (intervention N = 6; control N = 6) Recruitment: 12 Head Start sites administered through the Archdiocese of Chicago and that served primarily African‐American children were recruited to participate. All children at these sites were eligible to participate. Geographic region: Chicago, USA Percentage of eligible population enrolled: NR Mean age: intervention, 48.6 ± 7.6 months; control, 50.8 ± 6.4 months Sex: intervention, 49.7% female; control, 50.5% female
Interventions	Child intervention: 14 weeks (3 times/week) of a diet/PA intervention delivered by trained early childhood educators Each session included: 20 min nutrition activity reflecting the food pyramid 20 min aerobic activity based on overall moderate/vigorous movement Parent intervention: Received weekly newsletters that mirrored the children's curriculum Accompanying homework assignments (N = 12) designed to be an interactive activity between children and parents. Parents received a small monetary incentive for completing and returning homework. Control intervention: 14‐week (once a week) curriculum that taught general health concepts such as seat belt safety, immunisation and dental health. Parents received weekly newsletters that mirrored the curriculum, but no homework assignments Combined effects of dietary interventions and PA interventions vs control
Outcomes	Primary: change in BMI from baseline to Year 1 post‐intervention and Year 2 post‐intervention Secondary: dietary intake PA TV viewing Process evaluation: reported
Implementation‐related factors	Theoretical basis: reported (SCT as the primary framework, and concepts from Self‐determination theory) Resources for intervention implementation (e.g. funding needed or staff hours required): reported Who delivered the intervention: reported PROGRESS categories assessed at baseline: reported (gender, race, education) PROGRESS categories analysed at outcome: NR Outcomes relating to harms/unintended effects: reported Intervention included strategies to address diversity or disadvantage: reported Economic evaluation: NR
Notes	Funding: supported by a grant from the National Heart, Lung, and Blood Institute (Grant HL58871). Intervention design reported in secondary reference for Fitzgibbon 2005 (Fitzgibbon et al Preventive Medicine 2002;34:289‐97). This study is linked with results reported for another 12 preschools servicing Latino communities in Fitzgibbon 2006.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "one (school) of each pair was randomly assigned to the weight control intervention (WCI) or to the general health intervention (GHI)"
Allocation concealment (selection bias)	Unclear risk	NR Quote: "The schools were paired based only on class size, and one member of each pair was randomly assigned to the weight control intervention (WCI) or to the general health intervention (GHI)"
Blinding (performance bias and detection bias) All outcomes	Low risk	Quote: "1) assessments were conducted by trained data collectors who were unaware of group assignment at follow‐up, though not at baseline. 2) Dietary intake data were obtained from the parent of the child for a 24‐hour period by a trained and certified registered dietitian, blinded to treatment group. 3) Because of the nature of the intervention, neither the interventionists nor the participants could be blinded to the content of the intervention."
Incomplete outcome data (attrition bias) All outcomes	Low risk	17%‐20% loss to follow‐up. Performed adjusted analysis using 2 different approaches for imputation of missing data and reported both results
Selective reporting (reporting bias)	Low risk	Protocol found. All outcomes listed in the protocol were reported in results.
Other bias	Low risk	No additional threats to validity
Other bias‐ timing of recruitment of clusters	Low risk	Children were enrolled before randomisation