Overweight in adolescence can be predicted at age 6 years: a CART analysis in German cohorts

PLoS One. 2014 Mar 27;9(3):e93581. doi: 10.1371/journal.pone.0093581. eCollection 2014.

Abstract

Objective: To examine, whether overweight in adolescents can be predicted from the body mass index (BMI) category, at the age of 6, the mother's education level and mother's obesity and to quantify the proportion of overweight at the age of 14 that can be explained by these predictors.

Method: Pooled data from three German cohorts providing anthropometric and other relevant data to a total of 1 287 children. We used a classification and regression tree (CART) approach to identify the contribution of BMI category at the age of 6 (obese: BMI > 97th percentile (P97); overweight: P90 < BMI ≤ P97; high normal weight: P75<BMI ≤ P90; third quartile: P50<BMI ≤ P75; below the median: ≤ P50), maternal education level and maternal obesity for prediction of overweight/obesity (BMI>P90) at the age of 14.

Results: While 4.8% [95%CI: 3.2;7.0] of 651 boys and 4.1% [95%CI: 2.6;6.2] of 636 girls with a BMI<P75 at age 6 were overweight/obese in adolescence, prevalence increased to 41.3% [95%CI: 31.9;51.1] and 42.5% [95%CI: 33.8;51.6], respectively, in those with BMI ≥ P75. The lowest prevalence was 1.9% [95%CI: 0.8;3.8] in boys with a BMI ≤ P50 and the highest prevalence 91.7% [95%CI: 61.5;99.8] with a BMI>P97 (similar results for girls). BM I ≥ P75 at the age of 6 explained 63.5% [95%CI: 51.1;74.5]) and 72.0% [95%CI: 60.4;81.8] of overweight/obesity at the age of 14 in boys and girls, respectively.

Conclusions: Overweight/obesity in adolescence can be predicted by BMI category at the age of 6 allowing for parent counselling or risk guided interventions in children with BMI ≥ P75, who accounted for >2/3 of overweight/obesity in adolescents.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Age Factors
  • Anthropometry*
  • Body Mass Index
  • Body Weight
  • Child
  • Cohort Studies
  • Educational Status
  • Female
  • Germany / epidemiology
  • Humans
  • Male
  • Models, Statistical*
  • Mothers / psychology
  • Overweight / diagnosis*
  • Overweight / epidemiology
  • Overweight / physiopathology
  • Prevalence
  • Regression Analysis
  • Sex Factors

Grants and funding

The DONALD Study is funded by the Ministry of Science and Research of North Rhine Westphalia, Germany. KOPS was funded by Deutsche Forschungsgemeinschaft (DFG Mü 5.1, 5.2, 5.3 and 5.5). This analysis was supported by the German Federal Ministry of Education and Research (Competence Network Obesity; project: “Life Course Approach To Obesity Research: From Epidemiology to Future Strategies of Prevention – EPI Germany” [FKZ: 01GI1121A]). Neither of the funding agencies had any influence on the study conduct and report. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.