Predicting Weight Outcomes From Obesity Medications in a Paediatric Population
Authors' conclusion
The OMs studied had similar outcomes, with Semaglutide demonstrating slightly better outcomes, but not all patients had successful outcomes. Predictive models can inform clinical decision-making about OMs based on individual characteristics. Future studies validating models prospectively are needed.
Study context
- Population
- Patients aged 12–21 years with obesity and without DM dispensed an obesity medication, 2003–2025, paediatric healthcare system; 20% Hispanic, 27% Black, 49% public insurance, 62% severe obesity
- Participants
- 595 patients
- Exposure
- Liraglutide vs phentermine vs phentermine/topiramate vs semaglutide
- Outcome
- ≥ 5% reduction in %BMIp95 (patients dispensed the medication ≥ 6 months)
Methodological detailsreal-world, electronic health record data, machine learning prediction model (Catboost gradient-boosting)
- Real-world
- Electronic health record data
- Machine learning prediction model (Catboost gradient-boosting)
Main finding
As reported in the abstract
Average percentage change in %BMIp95 (dispensed > 12 months)
Liraglutide
8.76
average · %
Phentermine
10.90
average · %
Phentermine/Topiramate
9.34
average · %
Semaglutide
12.87
average · %
- Patients with ≥ 5% final %BMIp95 reduction (across medications)
- 64%-80%
Observational data: associations, not effects.
Further findings
Predictive model performance
- AUROC
- ≥ 0.75
- Associated predictors: OM duration, baseline %BMIp95 and number of specialty visits
Original publication
Predicting Weight Outcomes From Obesity Medications in a Paediatric Population. Pediatr Obes. Published 12 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →