A machine learning framework for optimizing obesity care by simulating clinical trajectories and targeted interventions
Authors' conclusion
This study discovered novel insights into the specific dynamics underlying low WLS use rates. This methodology permits in silico testing of interventions designed to optimize obesity care prior to implementation.
Study context
- Population
- Adults seen by a primary care provider (PCP) and eligible for weight loss surgery (WLS), 2016–2017, at a large institution
- Participants
- 5,876 eligible patients
- Outcome
- Progression between states: PCP visit, endocrine weight management referral, endocrine weight management visit, WLS referral, WLS visit and WLS
- Follow-up
- 2 years
Methodological detailsmultistate model, machine learning, simulation
- Multistate model
- Machine learning
- Simulation
Main finding
As reported in the abstract
Progression from initial PCP visit (Kaplan-Meier estimates)
- Endocrine weight management referral
- 35% (n = 2063)
- Endocrine weight management visit
- 15.6% (n = 930)
- WLS referral
- 6.3% (n = 400)
- WLS visit
- 4.7% (n = 298)
- WLS
- 1.0% (n = 69)
Observational data: associations, not effects.
Further findings
Predictors of progression
- Individual providers and clinics differed significantly in referral practices
- Female, younger, higher BMI and trainee-seen patients more likely to progress
Simulated intervention to increase PCP referrals · 3 years
- Additional WLS procedures
- about 49
Original publication
A machine learning framework for optimizing obesity care by simulating clinical trajectories and targeted interventions. Obesity (Silver Spring). Published 16 October 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →