Related Experiment Video
Updated: Sep 24, 2025
05:37
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
2.2K
External validation of prognostic models predicting outcome after chronic subdural hematoma
Dana C Holl1,2,3, Ana Mikolic4, Jurre Blaauw5
1Department of Neurosurgery, Erasmus Medical Centre, Erasmus MC Stroke Centre, Dr Molewaterplein 40, 3015 GD, Rotterdam, The Netherlands. d.holl@erasmusmc.nl.
Acta Neurochirurgica
|May 2, 2022
Summary
Existing prediction models for chronic subdural hematoma (CSH) outcomes show poor performance. Further research is needed to develop reliable prognostic tools for CSH patients.
Area of Science:
- Neurosurgery
- Clinical Epidemiology
- Medical Statistics
Background:
- Several prognostic models for chronic subdural hematoma (CSH) treatment outcomes exist.
- These models lack sufficient external validation for routine clinical use.
- Accurate outcome prediction in CSH patients remains a clinical challenge.
Purpose of the Study:
- To assess the performance of existing prognostic models for predicting outcomes in patients with CSH.
- To evaluate the calibration and discrimination of validated models using a large retrospective database.
Main Methods:
- Systematic literature search up to February 2021 for CSH prognostic models.
- External validation of identified models using a retrospective database of 2384 patients.
- Assessment of model calibration and discrimination (concordance index C).
Main Results:
- Three models were externally validated: one for 30-day mortality and two for hematoma recurrence.
- Models overestimated outcomes (e.g., 11% overestimation for mortality).
- Discriminative ability was poor to modest (C-indices ranging from 0.46 to 0.70).
Conclusions:
- Existing prognostic models demonstrate inadequate predictive performance for CSH outcomes.
- The heterogeneity of CSH patients contributes to prediction difficulties.
- Development of high-quality prognostic models with unified predictors and outcomes is warranted.
