Perioperative medicine & outcomes
Prediction and risk evaluation of delirium after surgery in older patients: development and internal validation of an algorithm from the prospective BioCog cohort study
Authors' conclusion
Models combining preoperative with precipitating factors during surgery predict POD with high accuracy.
Study context
- Population
- Patients aged ≥65 yr without preoperative dementia (Mini-Mental Status Examination ≥24) undergoing surgery of expected duration ≥60 min, Germany and The Netherlands
- Participants
- 929 patients
- Outcome
- POD according to DSM 5
- Follow-up
- Until the seventh postoperative day
Methodological detailsprediction model development and internal validation, two tertiary care centres, nested cross-validation
- Prediction model development and internal validation
- Two tertiary care centres
- Nested cross-validation
Main finding
As reported in the abstract
POD prediction
| Group | AUC | Brier score |
|---|---|---|
| Gradient-boosted trees (GBT) algorithm | 0.83 [0.79-0.86] | 0.12 (0.12-0.13) |
Gradient-boosted trees (GBT) algorithm
AUC0.83 [0.79-0.86]
Brier score0.12 (0.12-0.13)
- Incidence of postoperative delirium (POD)
- 184 (20%)
Observational data: associations, not effects.
Original publication
Prediction and risk evaluation of delirium after surgery in older patients: development and internal validation of an algorithm from the prospective BioCog cohort study. Br J Anaesth. Published 17 March 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →