Revisiting Decision Curve Analysis
Author Information
Author(s): Rousson Valentin, Zumbrunn Thomas
Primary Institution: University of Lausanne, Institute for Social and Preventive Medicine
Hypothesis
How can decision curve analysis improve the evaluation of prediction models in clinical settings?
Conclusion
The study presents extensions to decision curve analysis that enhance its interpretation and broaden its application.
Supporting Evidence
- The overall net benefit combines the net benefits for treated and untreated individuals.
- Decision curve analysis can be applied to case-control studies with some modifications.
- The complex model showed a higher net benefit than the simple model across various threshold probabilities.
Takeaway
This study helps doctors decide when to treat patients by showing how useful different prediction models are.
Methodology
The study discusses decision curve analysis and its application to case-control studies, emphasizing the overall net benefit.
Limitations
The choice of threshold probability can be subjective and may vary among clinicians.
Participant Demographics
The study analyzed data from a sample of 4658 individuals from the Framingham Heart Study.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website