Predicting no-show appointments in a pediatric hospital in Chile using machine learning
J Dunstan1,2, F Villena1, J P Hoyos3
1Center for Mathematical Modeling (CNRS IRL2807), University of Chile, Santiago, Chile.
Health Care Management Science
|January 27, 2023
Summary
Machine learning models effectively predict pediatric patient no-shows in Chile
Area of Science:
- Health Informatics
- Machine Learning in Healthcare
- Public Health Systems Analysis
Background:
- Chile's public health system faces high medical appointment no-show rates, averaging 19%, exceeding the national goal of 15%.
- Doctor Luis Calvo Mackenna Hospital, a pediatric facility, experiences particularly high no-show rates, reaching 29% in some specialties.
- Predicting and reducing no-shows is crucial for optimizing public healthcare resource allocation and patient access.
Purpose of the Study:
- To develop and evaluate machine learning (ML) models for predicting pediatric patient no-shows.
- To identify key demographic, social, and historical patient variables associated with no-shows.
- To propose and assess cost-effective metrics for intervention strategies to reduce appointment no-shows.
Main Methods:
- Analysis of 395,963 appointments (2015-2018) using Random Forest, Logistic Regression, Support Vector Machines, and AdaBoost algorithms.
- Application of class imbalance algorithms (RUS Boost, Balanced Random Forest, Balanced Bagging, Easy Ensemble) due to low no-show prevalence.
- Computation of alternative prediction thresholds based on cost-effectiveness criteria to minimize Type I and II errors.
Main Results:
- Overall no-show rate was 20.4%, with ophthalmology at 29.1%.
- Highest no-show rates were observed in the most socioeconomically deprived groups and during the second infancy period.
- Patient history of non-attendance, reservation delay, and socioeconomic indicators were the most relevant predictors.
- An 8-week intervention using a reminder strategy reduced no-shows by 10.3 percentage points compared to a control group.
Conclusions:
- Machine learning models, particularly those incorporating cost-effective metrics, can accurately predict pediatric patient no-shows.
- Patient historical behavior and socioeconomic factors are critical predictors for no-show risk.
- A reminder strategy targeting high-risk patients significantly decreased no-show rates, demonstrating the effectiveness of data-driven interventions in public healthcare.
More Related Videos
Related Concept Videos
01:03Prediction Intervals
2.3K
The interval estimate of any variable is known as the prediction interval. It helps decide if a point estimate is dependable.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
2.3K
01:18Steps in Outbreak Investigation
164
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
164