Brixval 2009.

Methods	Study design: prospective cohort study. Analyses methods for cohorts: regression analysis in boys and girls related fat intake to a change in BMI‐for‐age z‐score after 3 and 6 years' follow‐up. Adjusted model after 3 years' follow‐up was adjusted for baseline z‐score, physical activity level, pubertal stage at baseline, energy intake and dietary volume. Adjusted model at 6 years' follow‐up also included parent's income level, inactivity and number of overweight parents. How were missing data handled? At 3 years' follow‐up: participants with missing information on any measurement at baseline (n = 41) and incomplete follow‐up (attrition 25.5%; 150/589) excluded from analyses. Dropout analysis revealed baseline characteristics of anthropometrics and dietary information did not differ between participants (n = 308) that did and participants who did not complete follow‐up (all P > 0.05; data not shown). At 6 years' follow‐up: 384 children were re‐examined (attrition 34.8%; 205/589). Possible dropout effects examined indirectly by comparing baseline age, BMI and fat intake of those children participating only at baseline with children participating at both baseline and follow‐up, which showed no difference between groups (no data or statistical tests reported by authors). According to ethical considerations, it was not permitted to contact children who decided not to participate at follow‐up. Sample size justification adequately described? No. Sampling method: state schools in Odense (Denmark) stratified according to school type, location and SES profile. From each stratum, a proportional, 2‐stage sample of children was randomly selected. From the selected schools, 1356 pupils were invited, and 1020 (75.2%) (589 3rd graders and 421 ninth graders) agreed to participate. Periods of recruitment: 1997‐1998. Period of follow‐up (total period of observation): 6 years. Number of study contacts: 3 (baseline, 3 and 6 years). Study objective: objective 1: to examine associations between DED or fibre intake and 3‐year change in BMI‐for‐age z‐score among 8‐ to 10‐year old boys and girls. Objective 2: to investigate the association between fat intake and weight development among a cohort of children aged 9‐10 years at baseline and 15‐16 years at follow‐up, and whether parents' obesity was modifying the association. Study population: children aged 9‐10 years attending schools in Odense, Denmark.
Participants	Baseline characteristics (reported for 1 overall group) Age (mean in years): overall (n = 308) 9.64 (SD 0.4); boys (n = 138) 9.7 (SD 0.4); girls (n = 170) 9.6 (SD 0.4). Sex: 54.5% girls. Ethnicity: NR. Education: % with < 10 years' school attendance (n = 308): father 16.9% (girls), 19.4% (boys); mother 14.1% (girls), 12.4% (boys). Income: % with lowest income category (n = 308): father 5.1% (girls), 6.1% (boys); mother: 7.8% (girls), 9.1% (boys). Pubertal stage: Tanner stage: overall (n = 308) 2.17 (SD 0.4); boys (n = 138) 2 (SD 0.0); girls (n = 170) 2.3 (SD 0.5). Parental BMI (kg/m2): father (n = 308) 25.4 (SD 3.4); mother (n = 308) 23.5 (SD 3.8). Child total energy (kJ): overall (n = 308) 9113.64 (SD 419.29); boys (n = 138) 9500 (SD 260); girls (n = 170) 8800 (SD 210). Child total fat: in g: overall (n = 308) 79.95 (SD 27.97); boys (n = 138) 82.6 (SD 30); girls (n = 170) 77.8 (SD 26); in %TE: overall: 32.76 (SD 6.68); boys 33.3 (SD 6.7); girls 32.1 (SD 6.6). Child total protein: NR. Child total CHO: NR. Child physical activity: % who exercised regularly: boys (n = 138) 61.9%; girls (n = 170) 47.6%. Child physical inactivity or screen time or both: hours/day: overall (n = 308) 5.72 (SD 1.51); boys (n = 138) 6 (SD 1.6); girls (n = 170) 5.5 (SD 1.4). Child CVD risk (excluding fatness): NR. Child body fatness: BMI (kg/m2): overall (n = 308) 17.16 (SD 2.23); boys (n = 138) 17.1 (SD 2); girls (n = 170) 17.2 (SD 2.4); BMI‐for‐age z‐score: overall 0.36 (SD 1.06); boys 0.3 (SD 1.0); girls 0.4 (SD 1.1). Included criteria: 9‐ to 10‐year‐old boys and girls attending 3rd grade at selected schools in Odense, Denmark. Excluded criteria: NR. Total number enrolled in cohort study: 589. Total number completed in cohort study: 398 (after 3 years); 384 (after 6 years). Brief description of participants: 9‐ to 10‐year‐old children attending 3rd grade at schools in Odense, Denmark, who participated in the European Youth Heart Study.
Interventions	Description of exposure for cohorts Time span: 6 years. Dietary assessment method used: 24 hour‐recall. Frequency: single 24‐hour recall by children at baseline validated by an estimated food record (completed by parents for the same 24‐hour period). See Table 9; Table 10; Table 11; Table 12; Table 13; Table 14; Table 15; Table 16; Table 17; Table 18 for details of total fat intake exposure per outcome.
Outcomes	BMI BMI‐for‐age z‐score.
Identification	Sponsorship source: NR. Country: Denmark. Setting: schools in Odense. Comments: Danish component of the European Youth Heart Study. Author's name: Carina S Brixval. Institution: Research Unit for Dietary Studies, Institute of Preventive Medicine, Copenhagen, Denmark. Email: blh@ipm.regionh.dk; SI@ipm.regionh.dk. Declaration of Interests: yes. "The authors declared no conflict of interest." Study ID: Brixval 2009. Type of record: journal article.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Were adequate outcome data for cohorts available? All outcomes	Low risk	Participants with missing information on any measurement at baseline (n = 41) and incomplete follow‐up (attrition 25.5% (150/589) over 3 years) excluded from analyses. Dropout analysis revealed that baseline characteristics of anthropometrics and dietary information did not differ between participants (n = 308) who did and who did not complete the follow‐up (all P > 0.05). At 6 years' follow‐up, 384 children were re‐examined (attrition 34.8% (205/589)). Possible dropout effects examined indirectly by comparing baseline age, BMI and fat intake of those children participating only at baseline with children participating at both baseline and follow‐up, which showed no difference between groups (no data or statistical tests reported by authors). According to ethical considerations, it was not permitted to contact children who decided not to participate at follow‐up.
Was there matching of less‐exposed and more‐exposed participants for prognostic factors associated with outcome or were relevant statistical adjustments done? All outcomes	Low risk	Regression model adjusted for most important prognostic variables.
Did the exposures between groups differ in components other than only total fat? All outcomes	Low risk
Can we be confident in the assessment of outcomes? All outcomes	Low risk	Height (cm) measured to the nearest 0.1 cm with stadiometer. bodyweight (kg) measured to nearest 0.1 kg with calibrated beam‐scale weight. Participants wore underwear or light garments only.
Can we be confident in the assessment of exposure? All outcomes	High risk	A single 24‐hour dietary recall was performed at baseline. Although it was validated by an estimated food record (completed by parents for the same 24‐hour period) it was not repeated during follow‐up and therefore not likely to reflect the habitual fat intake of children during the study period.
Can we be confident in the assessment of presence or absence of prognostic factors? All outcomes	High risk	Parental BMIs calculated from self‐reported weights and heights. Presence or absence of regular physical exercise assessed at baseline by self‐report. Children's activity level at baseline measured using accelerometers; however, this variable contained significant missing data (33%). Unclear whether pubertal stage of children was based on an assessment or on self‐report.
Was selection of less‐exposed and more‐exposed groups from the same population? All outcomes	Low risk	All participants of the European Youth Heart Study in Denmark.