Related Experiment Video
Updated: Aug 13, 2025
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Predicting Attrition Patterns from Pediatric Weight Management Programs
Hamed Fayyaz1, Thao-Ly T Phan2, H Timothy Bunnell2
1University of Delaware, Newark, DE, USA.
Insights
Predicting dropout in pediatric weight management is crucial. Machine learning accurately forecasts attrition and BMI changes, enabling timely interventions for better child obesity treatment outcomes.
Area of Science:
- Pediatric Endocrinology
- Public Health
- Data Science in Medicine
Background:
- Childhood obesity is a significant public health issue requiring effective interventions.
- Pediatric weight management programs are essential but face high attrition rates, hindering treatment success.
- Existing methods for predicting attrition have shown limited success due to small datasets and static predictor focus.
Purpose of the Study:
- To develop and validate a machine learning pipeline for predicting attrition in pediatric weight management.
- To forecast changes in Body Mass Index (BMI) percentile among children in these programs.
- To enable earlier, personalized interventions by identifying children at risk of dropout.
Main Methods:
- Collected a comprehensive five-year dataset of 4,550 children from diverse backgrounds across four US pediatric weight management programs.
- Developed a customized machine learning pipeline to process longitudinal and interrelated prediction tasks.
- Utilized advanced techniques to predict both attrition likelihood and BMI percentile changes over time.
Main Results:
- The machine learning pipeline demonstrated strong predictive performance.
- Achieved an average Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.77 for predicting attrition.
- Attained an average AUROC of 0.78 for predicting weight outcomes (BMI percentile changes).
Conclusions:
- Machine learning offers a powerful approach to predict attrition and weight outcomes in pediatric weight management.
- Accurate prediction facilitates proactive and personalized interventions, potentially reducing high attrition rates.
- This data-driven strategy can improve the effectiveness of childhood obesity treatment programs.
Abstract:
Obesity is a major public health concern. Multidisciplinary pediatric weight management programs are considered standard treatment for children with obesity who are not able to be successfully managed in the primary care setting. Despite their great potential, high dropout rates (referred to as attrition) are a major hurdle in delivering successful interventions. Predicting attrition patterns can help providers reduce the alarmingly high rates of attrition (up to 80%) by engaging in earlier and more personalized interventions. Previous work has mainly focused on finding static predictors of attrition on smaller datasets and has achieved limited success in effective prediction. In this study, we have collected a five-year comprehensive dataset of 4,550 children from diverse backgrounds receiving treatment at four pediatric weight management programs in the US. We then developed a machine learning pipeline to predict (a) the likelihood of attrition, and (b) the change in body-mass index (BMI) percentile of children, at different time points after joining the weight management program. Our pipeline is greatly customized for this problem using advanced machine learning techniques to process longitudinal data, smaller-size data, and interrelated prediction tasks. The proposed method showed strong prediction performance as measured by AUROC scores (average AUROC of 0.77 for predicting attrition, and 0.78 for predicting weight outcomes).
Related Concept Videos
01:52Regression Toward the Mean
02:20Longitudinal Research
01:20Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
01:12Analysis of Population Pharmacokinetic Data
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
01:24Kaplan-Meier Approach