Muckelbauer 2009.

Methods	Design: NRCT Timing: Prospective Allocation to group: Schools in 1 city (Dortmund, Germany) served as intervention sites, and schools in another city (Essen, Germany) served as control. Quote: "Schools in Dortmund represented the intervention group and schools in Essen the control group. For each city, 20 schools were selected randomly" Number of clusters or sites: 33 schools Number of individuals: 2950 Length of intervention: 10 months
Participants	General description of participants: Elementary school children in deprived neighbourhoods in Dortmund and Essen, Germany Age: Childen (mean age 8.3 years) Inclusion criteria: School‐level (quote): "Schools were eligible for participation if they were located in deprived areas, as defined with the following criteria: unemployment rate of >15%, proportion of social welfare recipients of >5%, and proportion of non‐German residents of >5%, as indicated by the local public authorities." Student‐level: Parental consent Exclusion criteria: N/R Recruitment: N/R Weight status at baseline: The prevalence of overweight was 23% at baseline in the IG and 26% in the CG. Mean baseline BMI SDS was 0.23 in the IG and 0.30 in the CG SSB consumption at baseline: At baseline, SSB consumption was 260 ml/day in the IG and CG Equity considerations: Quote: "The study population comprised children attending the second and third grades of elementary schools in deprived neighborhoods of 2 neighboring cities, namely, Dortmund and Essen, Germany. Both cities have a population of ∼600 000 and are located in the Ruhr Area, a conglomerate of formerly industrial cities. (...) From a public health perspective, it is of importance that this intervention was effective in a deprived population, in which the prevalence of obesity was up to 3 times greater than that among children of a higher socioeconomic background.” A secondary analysis showed that the intervention was more effective in children without migration background (i.e. children whose parents and grandparents were born in Germany). Quote: "After intervention, the risk of being overweight was reduced in the IG compared to the CG among non‐MIG [i.e. children without migration background] (odds ratio = 0.51, 95% confidence interval (CI): 0.31–0.83), but not among MIG children (odds ratio = 1.02, 95% CI: 0.63–1.65)"
Interventions	Intervention: Improved access to drinking water at school. (Quote: "In each IG school, 1 water fountain (Sodamaster‐Aquatower 200; IONOX‐Wassertechnologie, Obertraubling, Germany), or 2 for schools with 150 participants, was installed. The fountains provided cooled, filtered, plain or optionally carbonated water. In addition, each child received a plastic water bottle (500 mL). (...) In month 5 after the baseline assessment, each participant received a new water bottle with an improved handling design") Behavioural co‐intervention: Organised water bottle fill‐ups and nutrition education focused on water. (Quote: "[T]eachers were encouraged to organize filling of the water bottles each morning for all children in the corresponding classes. The educational intervention consisted of four 45‐minute classroom lessons dealing with the water needs of the body and the water circuit in nature. At the beginning of the study, teachers received a booklet with the prepared curriculum and necessary materials to implement the lessons in the formal school curriculum. The lessons were developed by using the results of empirical teaching research and were intended to improve the constructs of intention, attitudes, and perceived behavioral control, on the basis of the theory of planned behavior. Three months after the beginning of the study, teachers introduced a motivation unit (ie, booster sessions) that used a goal‐setting strategy to reach a sustained increase in water consumption by giving quantitative targets and feedback") Control: No intervention (Quote: "CG schools did not receive any intervention")
Outcomes	Measures of SSB intake: SSB intake, assessed with a 24‐hour recall questionnaire that was self‐completed under teachers’ supervision at baseline and 10 months Measures of intake of alternatives to SSB: Water and juice intake, assessed with a 24‐hour recall questionnaire that was self‐completed under teachers’ supervision at baseline and 10 months Anthropometric measures: Prevalence of overweight and BMI SD‐Scores, assessed by trained healthcare professionals at baseline and 10 months Adverse outcomes: The study does not report how data on adverse outcomes were collected, but states that none were observed Other outcomes: Target group and stakeholder perceptions and uptake, assessed with interviews and questionnaires and by reading the integrated flow meters at baseline and at 6 control visits during the intervention period
Context and implementation	Setting: Schools Sector: Education Country: Germany Year(s) when implemented: 2006 ‐ 2007 Mode of implementation: Pilot trial by researchers Level of implementation: Setting‐based intervention
Declarations	COI: "The authors have indicated they have no financial relationships relevant to this article to disclose" Funding: "This trial was carried out by the Research Institute of Child Nutrition Dortmund (Dortmund, Germany), and was supported by grant 05HS026 from the German Federal Ministry of Food, Agriculture, and Consumer Protection. Intervention materials (water fountains, bottles, and lesson booklets) were provided by the Association of the German Gas and Water Industries. Ms Muckelbauer and Mr Libuda received research funding from grant 05HS026 from the German Federal Ministry of Food, Agriculture, and Consumer Protection" Trial registration: ClinicalTrial.gov number NCT00554294 Protocol availability: N/R
Notes	None
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	This was a NRCT, and allocation to the IG and CG was not random. Quote: "Schools in Dortmund represented the intervention group (IG) and schools in Essen the control group (CG). For each city, 20 schools were selected randomly"
Allocation concealment (selection bias)	High risk	This was a NRCT, and allocation was not random and not concealed (see quote above)
Similarity of baseline outcome measurements (selection bias)	Low risk	Quote: "The IG and CG did not differ in baseline characteristics regarding prevalence of overweight, BMI SDS, gender, age, and migrational background. Water and soft drink consumption levels at baseline were similar in the IG and the CG, but the level of juice consumption was slightly higher in the IG than in the CG"
Similarity of other baseline characteristics (selection and performance bias)	Low risk	Baseline characteristics were assessed and similar (see table 1 of the study's primary report)
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	A per‐protocol analysis was performed, and the attrition rate was higher in the CG than in the IG (1 control school withdrew, stating time requirements for data collection), but baseline characteristics of dropouts and non‐dropouts were similar, and overall attrition was in an acceptable range. For the questionnaire data a substantial number of participants (33%) were excluded due to implausible values. Quote: "Of 3190 children screened at baseline, a total of 2950 children (92%) were also measured at the follow‐up assessment and were considered for analysis. Dropouts (n = 240) were similar to analyzed participants with respect to the prevalence of overweight (24.6% vs 24.5%; P = .741), mean BMI SDS (0.26 vs 0.26; P = .807), mean age (8.27 vs 8.30 years; P = .574), proportion of boys (50.4% vs 50.2%; P = .772), and proportion of children with migrational background (42.1% vs 44.3%; P = .568). (...) Overall, 1987 (67%) of 2950 analyzed children (IG: 65%; CG: 70%) had plausible questionnaires on beverage consumption at both baseline and follow‐up assessments"
Blinding (performance and detection bias) Subjective outcomes	High risk	Participants were not blinded, and social desirability bias, which might have been reinforced by the behavioural components of the intervention, may have biased the results for beverage intake data, which was self‐reported
Blinding (performance and detection bias) Objective outcomes	Low risk	Body weight outcomes were objective, and assessed by trained health professionals following a standardised protocol
Contamination (performance bias)	Low risk	Intervention and control schools were located in different cities, and it is unlikely that contamination substantially affected results
Selective reporting (reporting bias)	Low risk	The trial was registered after the start but before the end of the study, and all important outcomes mentioned in the Methods section are reported
Other bias	Low risk	No other concerns