The Promise for Reducing Healthcare Cost with Predictive Model: An Analysis with Quantized Evaluation Metric on...

Area of Science:

• Healthcare economics
• Health informatics
• Machine learning applications

Background:

• Rising healthcare costs are linked to patient readmission rates.
• The Hospital Readmission Reduction Program penalizes hospitals for excessive 30-day rehospitalizations.
• Predictive analytics are increasingly used to mitigate readmission risks.

Purpose of the Study:

• To propose a quantized evaluation metric for assessing the cost-effectiveness of predictive readmission models.
• To evaluate the impact of machine learning on transitional care and readmission rates using the proposed metric.

Main Methods:

• Development of a novel quantized evaluation metric for cost-effectiveness.
• Application of machine learning models to predict patient readmissions.
• Evaluation of model performance and associated healthcare cost savings.

Main Results:

• The proposed metric provides a methodological approach to evaluate cost-effectiveness.
• Machine learning models demonstrated potential for significant healthcare savings.
• The final model estimated net healthcare savings exceeding $1 million with a 50% readmission prevention rate.

Conclusions:

• Predictive analytics, particularly machine learning, can be a cost-effective strategy for reducing hospital readmissions.
• The developed metric aids in assessing the economic viability of healthcare solutions.
• Implementing AI-driven transitional care can lead to substantial financial benefits for healthcare systems.