006 - Integrating economic considerations into predictive analytics: A new application for clinical decision support systems in facilitating value-based healthcare

Objective
Clinical prediction models providing binary categorisations for clinical decision support require the selection of a probability threshold, or 'cutpoint', to classify individuals. Existing cutpoint selection approaches typically optimise test-specific metrics, including sensitivity and specificity, but overlook the consequences of correct or incorrect classification. We introduce a new cutpoint selection approach considering downstream consequences using net monetary benefit (NMB) and through simulations compared to alternative approaches in preventing inpatient falls.
Materials and Methods
Parameter estimates for costs and effectiveness from prior studies were included in Monte Carlo simulations. For each use case we simulated the expected NMB resulting from the model-guided decision support using a range of cutpoint selection approaches, including our new value-optimising approach. Sensitivity analyses applied alternative event rates and model discrimination performance.
Results
The proposed approach that considered expected downstream consequences was frequently NMB-maximising compared to other methods. Sensitivity analysis demonstrated that it was or closely tracked the optimal strategy under a range of scenarios. Under scenarios of relatively low event rates and discrimination that may be considered realistic for falls (prevalence=0.036, AUC=0.70), our proposed cutpoint method was either the best or similar to the best of the compared methods regarding NMB.
Discussion
Our results highlight the value of conditioning cutpoints on the implementation setting, particularly for rare and costly events, which are often those subject to prediction model development research.
Conclusions
This study proposes a cutpoint selection method that may optimise clinical decision support systems towards value-based care.

net monetary benefit

clinical prediction model

cutpoint

simulation study