De Vries 2015.

Study characteristics
Methods	Study name: Groningen expert centre for kids with obesity (GECKO) Study design: cluster‐RCT Intervention period: 11 months Follow‐up period (post‐intervention): 18 months Differences in baseline characteristics: reported Reliable outcomes: reported Protection against contamination: reported Unit of allocation: nurse Unit of analysis: individual accounting for clustering
Participants	N (control baseline) = 65 N (control follow‐up) = 54 N (intervention baseline) = 96 N (intervention follow‐up) = 89 Setting (and number by study group): Well Baby Clinics; intervention: 7 nurses (N = 96 children); control: 6 nurses (N = 65 children) Recruitment: parents were informed about the current study during the 3rd trimester of pregnancy by the general practitioner, midwife or gynaecologist or at their 1st visit to the Well Baby Clinic. Geographic region: Drenthe, one of the northern provinces of the Netherlands Percentage of eligible population enrolled: 70% Mean age: intervention + control: 2 weeks Sex: intervention: 40% female; control: 57% female
Interventions	To evaluate the effect of early stimulation of PA on growth, body composition, motor activity and motor development in toddlers. The intervention group received recommendations from a nurse during a home visit 2 weeks after birth and during regular visits at the Well Baby Clinic at 2, 4, 8 and 11 months of age. After every consultation, parents received a printed copy of the recommendations. 5 visits by participants and parents, the 1st a home visit at 2 weeks old, and the rest to the Well Baby Clinic at 2, 4, 8 and 11 months of age. Follow‐up visit at age 2.5 years took place either at clinic or at home. Before each intervention visit (5 in total), the intervention nurses received special training from child physiotherapists on how to implement the stimulation programme. The focus at 2 weeks was to engage symmetric handling and encourage use of coloured toys and sounds. The focus at 2 months was to encourage variation in the infant’s position and location of play, and the focus at 4 months was to expand on this. At 8 months, the recommendations were to encourage the infant to crawl and thereby enlarge his playing area. Then at 11 months, parents were instructed to encourage their infant to walk without support. Parents in the control group received standard care without activity recommendations PA intervention vs control
Outcomes	Outcome measures Primary outcome: BMI, sum of skinfolds Secondary outcomes: % overweight, weight, height, waist circumference, hip circumference, skinfolds (triceps, biceps, subscapular, supra‐iliacal), % body fat, motor skills (Bayley score), PA Process evaluation: NR
Implementation‐related factors	Theoretical basis: NR Resources for intervention implementation: NR Who delivered the intervention: reported PROGRESS categories assessed at baseline: child: gender; parent: education, SES (income) PROGRESS categories analysed at outcome: gender Outcomes relating to harms/unintended effects: NR Intervention included strategies to address diversity or disadvantage: NR Economic evaluation: NR
Notes	NCT01127412 Funding: this research was funded by an unrestricted grant from Hutchison Whampoa Ltd. and the University of Groningen GECKO also included a birth cohort study; only birthweight was reported at baseline no other anthropometric outcomes were reported at baseline (aged 2 weeks)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Randomisation was carried out manually by a GECKO researcher, who drew pieces of paper from a bag. No further details of allocation. This method is highly susceptible to subversion or alteration.
Allocation concealment (selection bias)	High risk	Randomisation was carried out manually by a GECKO researcher, who drew pieces of paper from a bag. No further details of allocation. This method is highly susceptible to subversion or alteration.
Blinding (performance bias and detection bias) All outcomes	Low risk	Single‐blinded. A trained researcher, who was blinded to the group allocation of the child, performed all follow‐up measurements.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Low attrition rate (17% intervention, 7% control), study reports that dropout did not differ between the intervention (N = 7) and control groups (N = 11, P = 0.06)
Selective reporting (reporting bias)	Low risk	Protocol/trial registration document seen. All outcomes reported
Other bias	Low risk	No further threats to validity
Other bias‐ timing of recruitment of clusters	Unclear risk	NR; it is likely that nurses were randomised first and newborns assigned to them over time later