JMIR Publications

ABSTRACT

Background:

Unplanned patient readmissions within 30 days of discharge pose a significant challenge in Canadian healthcare economics. To address this issue, risk stratification, machine learning and linear regression paradigms have been proposed as potential predictive solutions. Ensemble machine learning methods, such as stacked ensemble models with boosted tree algorithms, have shown promise in early risk identification of specific patient groups.

Objective:

This study aimed to implement an ensemble model with submodels for structured data and compare metrics, evaluate the impact of optimized data manipulation with PCA on shorter readmissions, and quantitatively verify the causal relationship between Expected Length of Stay and Resource Intensity Weight Value for a comprehensive economic perspective.

Methods:

This retrospective study used Python 3.9 and streamlined libraries to analyze data obtained from the Discharge Abstract Database (DAD) covering 2016-2021. The study employed two subdatasets, clinical and geographical, to predict patient readmission and analyze economic implications, respectively. A stacking classifier ensemble model was used after Principle Component Analysis to predict patient readmission. Linear regression was performed to determine a relationship between RIW and ELOS.

Results:

The ensemble model achieved a precision and slightly higher recall (0.49 and 0.68) indicating a higher instance of false positives. The model was able to predict cases relatively better than other models compared to the literature. For the ensemble model, readmitted females and males of age 40-44 and 35-39 respectively were more likely to use resources. The regression tables verify the casualty of the model and confirm the trend that patient readmission is much more costly to both the patient and the healthcare system.

Conclusions:

The study validates the use of hybrid ensemble models for predicting economic cost models in healthcare, with the goal of reducing bureaucratic and utility costs associated with hospital readmissions. The availability of robust and efficient predictive models, as demonstrated in this study, can help hospitals focus more on patient care while maintaining low economic costs. The study predicts the relationship between ELOS and RIW, which can indirectly predict patient outcomes by reducing administrative tasks and physicians' burden, thereby reducing cost burdens placed on patients. It is recommended that changes to the general ensemble model and linear regressions be made to analyze new numerical data for hospital costs. The study highlights the benefits of using hybrid ensemble models in predicting economic cost models in healthcare to help hospitals focus more on patient care while reducing administrative and bureaucratic costs.