Key result
VHD prediction models show a ~27% decline in discrimination during external validation.
Why the study?
While many clinical prediction models exist to guide valvular heart disease treatment decisions, their relative performance is largely unknown.
How do clinical prediction models for valvular heart disease perform in external validations?
Systematic Review (n=49)
How do clinical prediction models for valvular heart disease perform in external validations?
Many clinical prediction models for valvular heart disease lack external validation, and those that are validated often show substantial declines in discriminatory performance, highlighting the need for rigorous external validation before clinical use.
Clinicians should require rigorous external validation before using valvular prediction models; extends evidence of frequent performance decay upon validation.
No abstract yet. The journal hasn't published one we can show.
No takes yet. Share an insight, caveat, or question.
Wessler et al. (2019) conducted a systematic review in Valvular heart disease (n=49). Clinical prediction models (CPMs) was evaluated on Performance in external validations (discrimination) (median percentage change -27.1%). A systematic review of 49 clinical prediction models for valvular heart disease revealed substantial heterogeneity in external validation performance, with a median 27.1% decline in discrimination.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: