Two-part Statistical Model for Identifying Baseline Predictors of Chronic Postsurgical Pain
Authors' conclusion
A two-part model may offer a superior statistical approach to linear and logistic regression, better identifying patient- and clinical care-related risk factors of chronic postsurgical pain, primarily by distinguishing factors that drive the occurrence of chronic pain from those that are associated with pain severity.
Study context
- Population
- Surgical patients in a single-center chronic postsurgical pain data settraining 3,000, testing 925
- Participants
- 3,925 patients
- Comparison
- Two-part model vs linear and logistic regression
- Outcome
- Predictive performance for postsurgical pain (R 2, root mean square error, mean absolute error) on the testing set
- Follow-up
- 3 months
Methodological detailssecondary analysis of a prospective single-centre cohort, single-center, prospectively collected data set, random training/testing split repeated 400 times
- Secondary analysis of a prospective single-centre cohort
- Single-center
- Prospectively collected data set
- Random training/testing split repeated 400 times
Performance
As reported in the abstract
Predictive performance
Mean R 2
0.075 vs . 0.050
P < 0.0001
Two-part model
Root mean square error
1.466 vs . 1.485
P < 0.0001
Two-part model
Mean absolute error
1.020 vs . 1.030
P < 0.0001
Two-part model
Further findings
Baseline predictors of postsurgical pain (15 baseline factors)
- Two-part model identified 7 baseline covariates missed by logistic or linear models or both: self-reported race, education level, ASA classification, overall body pain, widespread pain, symptom severity, anxiety level
- Significant in all three models: patient sex, surgical type, surgical site pain
Original publication
Two-part Statistical Model for Identifying Baseline Predictors of Chronic Postsurgical Pain. Anesthesiology. Published 3 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →