Prediction Models for Early Childhood Obesity: Applicability and Existing Issues

Abstract

Statistical models have been developed for the prediction or diagnosis of a wide range of outcomes. However, to our knowledge, only 7 published studies have reported models to specifically predict overweight and/or obesity in early childhood. These models were developed using known risk factors and vary greatly in terms of their discrimination and predictive capacities. There are currently no established guidelines on what constitutes an acceptable level of risk (i.e., risk threshold) for childhood obesity prediction models, but these should be set following consideration of the consequences of false-positive and false-negative predictions, as well as any relevant clinical guidelines. To date, no studies have examined the impact of using early childhood obesity prediction models as intervention tools. While these are potentially valuable to inform targeted interventions, the heterogeneity of the existing models and the lack of consensus on adequate thresholds limit their usefulness in practice.

Evaluation of the Predictive Ability, Environmental Regulation and Pharmacogenetics Utility of a BMI-Predisposing Genetic Risk Score during Childhood and Puberty

“…The joint combination of all models individually surpassing an AUC of 0.6 yielded a considerable improvement in the predictive ability (AUC = 0.81 CI 95% = [0.7-0.93]), which could be sufficient for clinical discrimination. All these results are in concordance with previous insights from Butler et al, (2019) [44], who demonstrated that early clinical factors, including maternal age, prepregnancy maternal (and paternal) BMI, birthweight, gestational age, weight gain during early infancy and other easily and measurable factors, do fairly well in predicting childhood obesity. Moreover, these results suggest that the combination of a high-risk genetic profile along with an unhealthy familial environment (represented in terms of parents BMI and obesity family history) could boost the predisposition to the disease.…”

Polygenic Risk Scores Expand to Obesity

“…In fact, the link between childhood and adulthood obesity is a well-established target for obesity prevention, inspiring the development of other models to predict childhood obesity at birth. Early clinical factors, including maternal age, pre-pregnancy maternal (and paternal) BMI, birth weight, gestational age, weight gain during early infancy, and other easily and measurable factors, do fairly well in predicting childhood obesity (Butler et al, 2019). It is not known how much these risk factors overlap with genetic factors, though the experience with other PRSs would suggest they are partially overlapping and complementary.…”

Application of Receiver Operating Characteristics (ROC) on the Prediction of Obesity

“…A survey has been done for predicting obesity using machine learning classifiers, and the performance has been calculated using three parameters -sensitivity, specificity, and ROC. However, the study did not conclude which classifier is best [25]. The Naive Bayesian and decision tree have been used to predictive the obesity, the increase of 20% is noted via Naïve Bayesian (NB) Tree [26].…”