de Droog 2017

Methods	Study design: Randomised controlled trial Funding: “This work was supported by a grant from the Dutch Ministry of Health, Welfare and Sport (grant number: 201400117.014.013). The Ministry's sole role was funding, and, thus, was not involved in the design, data collection, data analyses, data interpretation, and writing of the report. None of the authors had a potential conflict of interest.”
Participants	Description: Children aged 2‐3 years in nursery schools in Rotterdam, the Netherlands N (Randomised): 163 children Age: Mean = 2.63 years % Female: 48% SES and ethnicity: “The sample consisted of toddlers from mostly low‐SES households with various cultural backgrounds.” Inclusion/exclusion criteria: “Only schools without formal fruit and vegetables programs were selected” Recruitment: Not specified Recruitment rate: 99% (197/199) Region: The Netherlands
Interventions	Number of experimental conditions: 4 Number of participants (analysed): Passive with puppet: 36 Passive without puppet: 40 Interactive with puppet: 41 Interactive without puppet: 37 Description of intervention: Children were read a picture book “Rabbit’s brave rescue”. The embedded message in the book was that “eating carrots makes you strong”. Reading sessions were conducted in a quiet room within the nursery school during one workweek. The reading sessions were being held in small groups of 3‐5 toddlers, and took about 10 minutes. Reading was performed either with or without a hand puppet (hand puppets were developed that resembled the physical appearance of the main character in the picture book, ‘Rabbit’). Children allocated to the passive groups (with or without a puppet) were not asked questions during reading time and children allocated to the interactive groups (with or without a puppet) were asked questions during reading time. Duration: 4 days Number of contacts: 4 reading sessions (1 per day) Setting: Preschool Modality: Face‐to‐face Interventionist: Women with pedagogical education Integrity: The reading sessions were monitored. Date of study: Recruited in February and March 2015 Description of control: N/A
Outcomes	Outcome relating to children's fruit and vegetable consumption: Consumption of carrots (proportion): “The proportion of consumed carrots was calculated by dividing the pieces of carrots the child had eaten by the total number of pieces of foods the child had eaten.” “Proportional scores were used, rather than absolute scores, because the proportional scores take into account the total amount of foods eaten.” Outcome relating to absolute costs/cost‐effectiveness of interventions: Not reported Outcome relating to reported adverse events: Not reported Length of follow‐up from baseline: 4 days Length of follow‐up post‐intervention: Immediately Subgroup analyses: None Loss to follow‐up: “Children who were absent on the last reading day (n = 34), were excluded from the analyses.” “The total drop‐out was evenly spread across conditions.” Overall: 17% (not specified by group) Analysis: Unknown if sample size calculation was performed.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	“On the first day, the storytellers picked up the children from class in order of the name list provided by the school, and randomly assigned them to one of the four reading conditions, ensuring balance in gender.” No mention of how the randomization sequence was generated.
Allocation concealment (selection bias)	High risk	The allocation was done by the person delivering the intervention.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	“For the reading sessions, four women with a pedagogical education were recruited and trained to perform all the different reading styles and puppetry conditions. These storytellers were teamed up with four female experimenters who observed the toddlers during the readings. With each team being allocated to a specific day of the week, all the toddlers in the study were exposed to all the storytellers and observers.” Those delivering the intervention were aware of group allocation, however this is unlikely to have impacted the outcomes.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	“The experimenter conducting the eating task was blinded to group assignment, because the reading sessions and eating tasks took place in different rooms.”
Incomplete outcome data (attrition bias) All outcomes	High risk	Dropouts were 23% at short‐term follow‐up (in text). However in Consort flowchart, it appears that people were excluded prior to randomization. In the text it says that most were excluded due to not attending on the final measurement day. This sounds like the dropouts should be removed at the analysis/data collection stage.
Selective reporting (reporting bias)	Low risk	All outcomes are reported as pre‐specified in the trial registration.
Other bias	Low risk	No other sources of bias identified