Prediction of outpatient waiting time: using machine learning in a tertiary children's hospital

Area of Science:

• Healthcare Management
• Artificial Intelligence in Medicine
• Pediatric Healthcare

Background:

• Accurate prediction of patient waiting times is essential for efficient hospital operations.
• Informing patients of wait times allows for better visit planning and reduced anxiety.
• This study focuses on predicting outpatient waiting times in a Chinese pediatric hospital.

Purpose of the Study:

• To evaluate the efficacy of machine learning algorithms in predicting outpatient waiting times.
• To compare the performance of different machine learning models for wait time prediction.
• To enhance patient satisfaction and hospital management through accurate wait time forecasting.

Main Methods:

• A novel classification method based on medical knowledge and statistical analysis was developed.
• Four machine learning algorithms were employed: linear regression (LR), random forest (RF), gradient boosting decision tree (GBDT), and K-nearest neighbor (KNN).
• Prediction models were constructed for patient waiting times across four department categories.

Main Results:

• Gradient boosting decision tree (GBDT) and random forest (RF) models significantly outperformed linear regression (LR).
• The RF model achieved the lowest mean absolute error (MAE) for the Internal Medicine Department I (5.03 minutes), a 47.60% improvement over LR.
• The GBDT model was optimal for the other three categories, with MAE reductions of 28.26%, 35.86%, and 33.10% compared to LR.

Conclusions:

• Machine learning models demonstrate high accuracy in predicting pediatric outpatient waiting times.
• Accurate wait time predictions can alleviate patient anxiety and improve the overall healthcare experience.
• This research highlights the potential of AI to enhance efficiency and patient-centered care in pediatric hospitals.