Using machine learning to reduce unnecessary rehospitalization of cardiovascular patients in Saudi Arabia

Using machine learning to reduce unnecessary rehospitalization of cardiovascular patients in Saudi Arabia

Patients with cardiovascular disease are frequently readmitted to the hospital within 30 days. Aside from being expensive, this is also sometimes avoidable. However, there's a shortage of research on this issue in Saudi Arabia, which limits the ability to address the problem, in line with the goals of Vision 2030.

What did we do?

To investigate cardiovascular patients' readmission risks, we utilized data from 48 Ministry of Health hospitals in Saudi Arabia. We excluded patients with incomplete information or stays exceeding one year from our analysis. Using this data, we developed a machine learning model to stratify cardiovascular disease patients based on their risk of readmission. Additionally, we calculated healthcare utilization costs for cardiovascular-related readmissions based on hospitalization days and services provided, using the Ministry of Health's price list.

What did we find?

• The model correctly identified 71% of readmissions (recall) and predicted 57% (precision), achieving an overall effectiveness score of 62%.
• Enhancing the model with more robust clinical data and advanced techniques, such as deep learning and transfer learning, could improve its performance significantly.

What does this mean?

The model has moderately succeeded in identifying cardiovascular-related readmission encounters. After enhancing the model, results could be used to flag high-risk patients of readmission and trigger specific action plans to minimize the risk. Ultimately, this approach can lead to more efficient utilization of healthcare resources.