Development, validation and recalibration of a prediction model for prediabetes: an EHR and NHANES-based study
2024

Prediction Model for Prediabetes Using EHR and NHANES Data

Sample size: 22635 publication Evidence: moderate

Author Information

Author(s): Casacchia Nicholas J., Lenoir Kristin M., Rigdon Joseph, Wells Brian J.

Primary Institution: Cleveland Clinic

Hypothesis

Can a new penalized regression model using EHR data effectively predict prediabetes compared to a previously developed model?

Conclusion

The approximation method showed comparable performance to LASSO, making it a viable option for healthcare organizations with limited resources.

Supporting Evidence

  • The LASSO model had an optimism-corrected C-statistic of 0.760, while the approximation model had 0.763.
  • The approximation model discriminated well in the NHANES dataset, achieving a C-statistic of 0.787.
  • 30.1% of NHANES participants met the inclusion criteria for prediabetes.

Takeaway

Researchers created a tool to help doctors find patients at risk for prediabetes using health records, which can help catch the problem early.

Methodology

Logistic regression models were developed using EHR data and validated with NHANES data.

Limitations

External validation of the LASSO model was not possible due to missing predictor variables in NHANES.

Participant Demographics

The EHR cohort included 22,635 patients, with 26% identified as having prediabetes; NHANES cohort included 2,348 participants, with 30.1% having prediabetes.

Digital Object Identifier (DOI)

10.1186/s12911-024-02803-w

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication