Predicting high‐risk periods for weight regain following initial weight loss
Authors' conclusion
Weight regain can be predicted on a proximal, week-to-week level. Future work should investigate the clinical utility of adaptive interventions for weight-loss maintenance and develop more sophisticated predictive models of weight regain.
Study context
- Population
- Adults after initial weight loss; weekly questionnaire and self-monitoring data (incl. daily e-scale) over 40 weeks
- Participants
- 46 / 22 / 30
- Comparison
- Training (Study 1, ≥5% loss) vs Testing (Study 1, <5% loss) vs Testing (Study 2)
- Training (Study 1, ≥5% loss)n = 46
- Testing (Study 1, <5% loss)n = 22
- Testing (Study 2)n = 30
- Outcome
- Prediction of weight regain the following week (week-to-week)
Methodological detailslongitudinal mixed-effects models, random forest regression, external testing data sets
- Longitudinal mixed-effects models
- Random forest regression
- External testing data sets
Performance
As reported in the abstract
Prediction of weight regain the following week
| Test or model | Sensitivity | Specificity |
|---|---|---|
| Initial training data set | 75.6% | 45.8% |
| Testing data sets | 81%-82% | 30%-33% |
Sensitivity
Initial training data set75.6%
Testing data sets81%-82%
Specificity
Initial training data set45.8%
Testing data sets30%-33%
Further findings
Predictors retained in final algorithm
- Frequency of self-monitoring caloric intake and weight, plus self-rated hunger and importance of weight-management goals vs competing life demands
Original publication
Predicting high‐risk periods for weight regain following initial weight loss. Obesity (Silver Spring). Published 2 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →