External Validation of a Multivariable Prediction Model for Placenta Accreta Spectrum
Authors' conclusion
The performance of the Weiniger model is variable based on the case-mix of the population with regard to PAS clinical risk factors and ultrasound features, highlighting the importance of spectrum bias when applying this PAS prediction model to distinct populations. The model showed benefit for predicting PAS in populations with substantial case-mix heterogeneity at threshold probability of >25%.
Study context
- Population
- Patients at risk of placenta accreta spectrum delivering at 2 US tertiary centres2 validation cohorts
- Participants
- 253 vs 99
- Comparison
- Weiniger placenta accreta spectrum (PAS) prediction model applied in two external cohorts
- Cohort An = 253
- Cohort Bn = 99
- Outcome
- Model discrimination (c-statistic) for surgical and/or pathological diagnosis of PAS
Methodological detailstwo-centre, decision curve analysis
- Two-centre
- Decision curve analysis
Performance
As reported in the abstract
Discrimination (c-statistic)
A
0.728
c-statistic · 95% CI 0.662–0.794
B
0.866
c-statistic · 95% CI 0.754–0.977
Further findings
| Outcome | Cohort A | Cohort B |
|---|---|---|
| Calibration intercept and slope | ||
| Calibration intercept (95% CI 0.154-0.980) | 0.537 | — not stated in the abstract |
| Calibration slope (95% CI 0.170-0.532) | 0.342 | — not stated in the abstract |
| Calibration intercept (95% CI 1.899- 4.335) | — not stated in the abstract | 3.001 |
| Calibration slope (95% CI −0.166 to 1.221) | — not stated in the abstract | 0.604 |
— not stated in the abstract
Calibration intercept and slope
- Model underestimated true PAS risks on average; evidence of overfitting in both cohorts
Clinical usefulness (decision curve analysis) vs treat-all strategy
- Greater net benefit of the model in Cohort A at threshold probability
- >0.25
- No net benefit over treat-all in Cohort B at any threshold probability
Original publication
External Validation of a Multivariable Prediction Model for Placenta Accreta Spectrum. Anesth Analg. Published 7 October 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →