Validity of Scottish predictors of child obesity (age 12) for risk screening in mid-childhood: a secondary analysis of prospective cohort study data—with sensitivity analyses for settings without various routinely collected predictor variables

Objective: To analyse the Growing Up in Scotland (GUS) cohort to identify predictors, available by school entry (age 5–6), of obesity at age 12. Methods: Literature-based, routinely collectible risk factors and proxies for Adverse/Protective Childhood Experiences (ACEs/PCEs) were considered. Missing predictor data were multiply imputed using Multiple Chained Equations (30 datasets). Predictor reduction used multivariable logistic regression with backward/forward stepwise selection (p=0.06), followed by internal validation via bootstrapping and shrinkage. Performance was evaluated with AUROC and sensitivity/specificity; optimal cut-offs were selected using Youden’s J. Two models were developed: an Optimum Data model (including variables not routinely available everywhere) and a Scottish Data model (restricted to routinely available Scottish data). Results: Among 2787 children with outcome data (obesity prevalence 18.3% at age 12), the Optimum Data model retained six predictors: maternal BMI, indoor smoking, equivalized income quintile, child’s sex, child’s BMI at age 5–6, and ACEs. After internal validation, AUROC was 0.855 (95% CI 0.852–0.859). At the Youden cut-off, sensitivity was 76.3% and specificity 77.6%, with 37.0% of children screening positive. The Scottish Data model replaced equivalized income with SIMD and ACEs with age at introduction of solid foods; AUROC was 0.849 (95% CI 0.846–0.852), sensitivity 76.2%, specificity 79.2%, and referral burden 30.8%. Conclusion: Universally collected, machine-readable, and linkable data at age 5–6 can predict obesity at age 12 with good discrimination. However, the implied referral burden is substantial in the current Scottish context; broader screening criteria and system capacity considerations are required.

Gabriela Carrillo-Balam, Lawrence Doi, Louise Marryat, Andrew James Williams, Paul Bradshaw, John Frank