Predicting all-cause risk of 30-day hospital readmission using artificial neural networks

Abstract

Avoidable hospital readmissions not only contribute to the high costs of healthcare in the US, but also have an impact on the quality of care for patients. Large scale adoption of Electronic Health Records (EHR) has created the opportunity to proactively identify patients with high risk of hospital readmission, and apply effective interventions to mitigate that risk. To that end, in the past, numerous machine-learning models have been employed to predict the risk of 30-day hospital readmission. However, the need for an accurate and real-time predictive model, suitable for hospital setting applications still exists. Here, using data from more than 300,000 hospital stays in California from Sutter Health's EHR system, we built and tested an artificial neural network (NN) model based on Google's TensorFlow library. Through comparison with other traditional and non-traditional models, we demonstrated that neural networks are great candidates to capture the complexity and interdependency of various data fields in EHRs. LACE, the current industry standard, showed a precision (PPV) of 0.20 in identifying high-risk patients in our database. In contrast, our NN model yielded a PPV of 0.24, which is a 20% improvement over LACE. Additionally, we discussed the predictive power of Social Determinants of Health (SDoH) data, and presented a simple cost analysis to assist hospitalists in implementing helpful and cost-effective post-discharge interventions.

Fig 1. Total number of records for each hospital under study, and their respective readmission rates.

Fig 2. Data breakdown by hospital admission year.

Fig 3. Neural Network model architecture (Note: Layer sizes are assuming all features are used).

Fig 4. Comparison of NN model performance (with retrospective validation) vs number of features.

Fig 5. Comparison of artificial neural network model with LACE in 4 different age brackets.

Fig 6. Comparison of the model performance among top five Sutter Health hospitals by the number of inpatient records.

Fig 7. Comparison of the neural network model’s performance among subgroups with varying medical conditions.

Fig 8. The projected saving values as a function of the intervention rate, with the example parameters given for the cost-savings analysis in the results section.