Assessed using the validated Physical Activity Questionnaire for Children (PAQ-C).The PAQ-C is a self-administered,7-day recall instrument, which provides a summary physical activity score derived from eight items, each scored on a 5-point scale. Item 1 (spare time activity) from no activity = 1; 7 times or more = 5. Items 2 to 7 (PE, lunch, right after school, evening, weekends) from lowest activity response = 1 or highest activity response = 5. Item 8 (mean of all days of the week) from none = 1; very often = 5. Item 9 (identifies students who are unusual active during the previous week). By adding up all means of the first eight items in PAQ-C, a summative score of physical activity is obtained. A score of 1 indicates low physical activity level, whereas a score of 5 indicates high physical activity level.

Waist circumference is measured with a measuring tape to the nearest 0.1 cm, following the World Health Organisation's assessment protocol.

Soft drink consumption during the past week was derived from the parental questionnaire ranging from never (0) to every day (7).

Water consumption during school hours was derived from the child questionnaire ranging from never (0) to every day (3).

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter. To shed more light on the nutritional value of the children's lunches, the different food types consumed were summed, and a dichotomous variable was created indicating whether children consumed at least two of the food types during lunch.

As the present study investigated the effectiveness of health-promoting activities under real-world conditions and all participating schools were free to implement activities of their choice, there were no intervention or control groups. Instead, schools were categorised on the basis of the implemented activities' intensity. This resulted in the formation of two groups: medium implementers (six schools) and low implementers (five schools). The project's effectiveness on various health outcomes was compared between these two groups.

Total of all reporting groups

The number analysed in this row differs from the overall number of participants, as we could not collect baseline data in all participants for this variable.

The number analysed in this row differs from the overall number of participants, as we could not collect baseline data in all participants for this variable.

Children's Body Mass Index (BMI) was assessed by height and weight; age- and sex-specific BMI cut-off points were used to define overweight and obesity. BMI z-scores were calculated using Dutch reference values. A z-score of 0 represents the population mean. A z-score > 0 means a BMI higher than the population mean, a z-score of < 0 means a BMI lower than the population mean, both can pose specific health-related risks.

Children's Body mass index (BMI) was assessed by height and weight; age- and sex-specific BMI cut-off points were used to define overweight and obesity.

The child questionnaire was used to gain insight into children's PA behaviour. The questionnaire contained ten items derived from the International Physical Activity Questionnaire for Children (IPAQ-C), which has acceptable validity. Activity scores between 1 and 5 were obtained for each item, after which the mean of these scores was calculated to obtain the total activity summary score (ranging from 1 (low PA) to 5 (high PA)).

A composite score for healthy dietary behaviours was computed from four separate questions. This score was calculated by averaging the weekly consumption (ranging from never (0) to every day (7)) of breakfast consumption and intake of fruit, warm and raw vegetables, and water throughout the day, as reported in the parental questionnaire.

Soft drink consumption during the past week was derived from the parental questionnaire ranging from never (0) to every day (7).

Water consumption during school hours was derived from the child questionnaire ranging from never (0) to every day (3).

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter. To shed more light on the nutritional value of the children's lunches, the different food types consumed were summed, and a dichotomous variable was created indicating whether children consumed at least two of the food types during lunch.

A composite score for healthy dietary behaviours was computed from four separate questions. This score was calculated by averaging the weekly consumption (ranging from never (0) to every day (7)) of breakfast consumption and intake of fruit, warm and raw vegetables, and water throughout the day, as reported in the parental questionnaire.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

Assessed using the validated Physical Activity Questionnaire for Children (PAQ-C).The PAQ-C is a self-administered,7-day recall instrument, which provides a summary physical activity score derived from eight items, each scored on a 5-point scale. Item 1 (spare time activity) from no activity = 1; 7 times or more = 5. Items 2 to 7 (PE, lunch, right after school, evening, weekends) from lowest activity response = 1 or highest activity response = 5. Item 8 (mean of all days of the week) from none = 1; very often = 5. Item 9 (identifies students who are unusual active during the previous week). By adding up all means of the first eight items in PAQ-C, a summative score of physical activity is obtained. A score of 1 indicates low physical activity level, whereas a score of 5 indicates high physical activity level.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

Waist circumference is measured with a measuring tape to the nearest 0.1 cm, following the World Health Organisation's assessment protocol.

Number analysed at T1 and T2 corresponds to number of participants that were not LOF at T1 and T2.

Soft drink consumption during the past week was derived from the parental questionnaire ranging from never (0) to every day (7).

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

Water consumption during school hours was derived from the child questionnaire ranging from never (0) to every day (3).

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

The child lunch questionnaire assessed children's consumption of certain food types during lunch. The items were summarised into six dichotomous (yes/no) food types: fruits, vegetables, grains (bread and cereals), dairy (milk/yoghurt and cheese), water, and butter. To shed more light on the nutritional value of the children's lunches, the different food types consumed were summed, and a dichotomous variable was created indicating whether children consumed at least two of the food types during lunch.

The number analysed at T1 and T2 corresponds to the number of participants that were not LOF at T1 and T2.

Linear mixed model analyses were used to assess longitudinal intervention effects on children's dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. This analysis reports T2-T0.

Linear mixed model analyses were used to assess longitudinal intervention effects on children's PA score. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. This analysis reports T2-T0.

Linear mixed model analyses were used to assess longitudinal intervention effects on children's waist circumference. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. This analysis reports T2-T0.

Linear mixed model analyses were used to assess longitudinal intervention effects on children's dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. This analysis reports T2-T0.

Linear mixed model analyses were used to assess longitudinal intervention effects on children's dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. This analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T1-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T2-T0.

Logistic mixed model analyses were used to assess longitudinal intervention effects on children's binary dietary behaviours. A two-level model with measurements as first level and participants as second level was used. The model's fixed part consisted of group (medium/low implementers), time (T0/T1/T2), interaction term of group with time, and covariates sex (boy/girl), study year at baseline (four/five/six), SES (low/middle/high), and children's BMI z-score at baseline. Analysis reports T1-T0.