Lee 2012.

Methods	Study design: prospective cohort study. Analyses methods for cohort: multivariate linear regression modelling for 2 years BMI change of 1st graders and 4th graders. Predictor variables were environmental factors, parental and lifestyle habits. Dependent variables were BMI change between 4 and 6 years' follow‐up. Model adjusted for age, sex, sexual maturation at 6 years' follow‐up (Tanner stage I, II, III, IV, V), baseline BMI, and exercise frequency, screen time, sleep duration, household income, parental BMI, parental education, maternal job, family structure, energy intake, meal skipping and snacking. They only adjusted for the BMI in the 4th survey at 6 years' follow‐up. How were missing data handled? Analytic sample taken of total number of children participating in study. Analytic sample was of children who participated at 4 and 6 years' follow‐up; total of 1504 participants. Original sample was of 893 but new participants were recruited over years (2776 participants at 5 years' follow‐up and 2770 at 6 years' follow‐up). Number of study contacts: 3 (baseline, 1 and 2 years). Period of follow‐up (total period of observation): both 1st graders and 4th graders were followed up for 2 years. Period of recruitment: baseline: 2005. New recruitment in 2008. Sample size justification adequately described? No. Sampling method: in 2005, all 1st graders of 4 elementary schools in Gwacheon city, Seoul were included. In 2008, 1st and 4th graders from 2 elementary schools in Jung‐gu, Seoul and 5 elementary schools in southwestern Gyeonggi province were added to the cohort. Study objective: to assess risk factors associated with children's BMI and their changes over a 2‐year period based on the analysis of the Obesity and Metabolic Disorders Cohort in Childhood registry. Study population: children in elementary school, grades 1 and 4.
Participants	Baseline characteristics (reported for 1 overall group) 1st graders (n = 474); 4th graders (n = 1030) Age (mean in years): 1st graders: 7.3 (SD 0.3); 4th graders 10.0 (SD 0.4). Sex: 1st graders: 52.3% girls; 4th graders: 50.7% girls. Ethnicity: NR. Education: maternal education for 1st graders: ≤ 12 years 32.1%; 13‐16 years 63.3%; ≥ 17 years 4.6%. Paternal education for 1st graders: ≤ 12 years 20.3%; 13‐16 years 68.1%; ≥ 17 years 11.6%. Maternal education for 4th graders: ≤ 12 years 35.2%; 13‐16 years 58.4%; ≥ 17 years 6.3%. Paternal education for 4th graders: ≤ 12 years 25.7%; 13‐16 years 58.7%; ≥ 17 years 15.5%. Income: 1st graders: 78.5% of households earned > 3 million KRW per year. 4th graders: 62.2% of households earned > 3 million KRW per year. Pubertal stage: NR. Parental BMI (kg/m2): maternal BMI for 1st graders: 81.0% < 23; 11.6% 23‐24.9; 7.4% ≥ 25; mean (SD) 21.3 (2.4). Paternal BMI for 1st graders: 36.5% < 23; 36.1% 23‐24.9; 27.4% ≥ 25; mean (SD) 23.7 (2.7). Maternal BMI for 4th graders: 75.8% < 23; 15.0% 23‐24.9; 9.1% ≥ 25; mean (SD) 21.7 (2.5). Paternal BMI for 4th graders: 36.9% < 23; 37.1% 23‐24.9; 26.0% ≥ 25; mean (SD) 23.8 (2.5). Child total energy (kJ): 1st graders: 7531.2 (SD 1255.2); 4th graders: 7112.8 (SD 673.6). Child total fat (%TE): 1st graders: 26.6 (SD 4.9); 4th graders: 25.2 (SD 5.1). Child total protein: NR. Child total CHO: NR. Child physical activity: 1st graders who exercised ≤ 1/week, n (%): 128 (27); 2‐4 times/week: 266 (56.1); ≥ 5 times/week: 80 (16.9). 4th graders who exercised ≤ 1/week, n (%): 321 (31.2); 2‐4 times/week, n (%): 576 (55.9), ≥ 5 times/week, n (%): 133 (12.9). Child physical inactivity or screen time or both: 1st graders screen time: 1.4 (SD 0.8) hours/week. 4th graders screen time: 1.6 (SD 1.0) hours/week. Child CVD risk (excluding fatness): NR. Child body fatness: BMI 1st graders (kg/m2): 16.0 (SD 2.3); BMI 4th graders (kg/m2): 18.1 (SD 3.0); BMI percentile 1st graders (kg/m2): ≤ 10%: 43 (SD 9.1); 10.1‐84.9%: 374 (SD 78.9); 85‐94.9%: 36 (SD 7.6); ≥ 95%: 21 (SD 4.4); BMI percentiles 4th graders: ≤ 10%: 35 (SD 3.5); 10.1‐84.9%: 816 (SD 79.2); 85‐94.9%: 110 (SD 10.6); ≥ 95%: 69 (SD 6.7). Included criteria: NR. Excluded criteria: NR. Brief description of participants: 474 1st graders (31.5%) and 1030 4th graders (68.5%). Mean ages: 1st graders: 7.3 (SD 0.3) years; 4th graders: 10.0 (SD 0.4) years. Mean BMI of 1st graders 16.0 (SD 2.3) kg/m2 with 12.0% being over 85th percentile of BMI curve, whereas mean BMI of 4th graders was 18.1 (SD 3.0) kg/m2 with 17.3% being over 85th percentile of BMI curve. Total numbers completed in cohort study: analytic sample taken from entire cohort: 1504. Total number enrolled in cohort study: 893 children enrolled in 2005, and another 1847 children enrolled in 2008, thus total 2740. However, in Figure 1 for the 5 years' follow‐up, it showed that there were, at one point, 2776 children enrolled.
Interventions	Description of exposure for cohort Time span: 2 years. Dietary assessment method: 24‐hour dietary recall. Frequency of dietary assessment: multiple 24‐hour recall at baseline and 1 per year (at 1 and 2 years). Dietary intake recorded for 2 week days and 1 weekend day during each assessment. 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 (kg/m2, 2‐year change).
Identification	Sponsorship source: NR. Country: Korea. Setting: Elementary schools, Gwacheon city, Seoul. Comments: study name: Obesity and Metabolic Disorders Cohort in Childhood. Author's name: Hyun Hye Lee. Institution: Department of Family Medicine, Inje University College of Medicine, Seoul, Korea. Email: drparkhyunah@gmail.com. Declaration of Interests: Yes. "No potential conflict of interest relevant to this article was reported." Study ID: Lee 2012. Type of record: journal article.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Were adequate outcome data for cohorts available? All outcomes	High risk	Authors used an analytical sample and did not analyse entire cohort, which consisted of 2776 children. Reasons for this not provided. Loss to follow‐up not discussed.
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	Adjusted for age, sex, sexual maturation at 6 years' follow‐up, baseline BMI, exercise, screen time, sleep duration, household income, parental BMI and education, maternal job, family structure, energy intake, meal skipping and snacking.
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	Trained researchers measured height and weight; used sex‐specific 2007 growth charts for Korean children.
Can we be confident in the assessment of exposure? All outcomes	Low risk	Authors reported: "Dietary intake was recorded for two weekdays and one day on the weekend by a 24‐hour recall method." Large sample size with multiple assessments to provide usual intake estimation.
Can we be confident in the assessment of presence or absence of prognostic factors? All outcomes	Low risk	Over the 2‐year follow‐up period physical activity and screen time was assessed at least twice, with detailed definitions for moderate and vigorous activity to guide parents and children with this.
Was selection of less‐exposed and more‐exposed groups from the same population? All outcomes	Low risk	NA as study did not divide participants into exposed and unexposed groups. All participants were sampled from similar locations.