Improving healthcare access management by predicting patient no-show behaviour

David Barrera Ferro,Sally Brailsford,Cristián Bravo,Honora Smith

DOI: https://doi.org/10.1016/j.dss.2020.113398

2020-12-10

Abstract:Low attendance levels in medical appointments have been associated with poor health outcomes and efficiency problems for service providers. To address this problem, healthcare managers could aim at improving attendance levels or minimizing the operational impact of no-shows by adapting resource allocation policies. However, given the uncertainty of patient behaviour, generating relevant information regarding no-show probabilities could support the decision-making process for both approaches. In this context many researchers have used multiple regression models to identify patient and appointment characteristics than can be used as good predictors for no-show probabilities. This work develops a Decision Support System (DSS) to support the implementation of strategies to encourage attendance, for a preventive care program targeted at underserved communities in Bogotá, Colombia. Our contribution to literature is threefold. Firstly, we assess the effectiveness of different machine learning approaches to improve the accuracy of regression models. In particular, Random Forest and Neural Networks are used to model the problem accounting for non-linearity and variable interactions. Secondly, we propose a novel use of Layer-wise Relevance Propagation in order to improve the explainability of neural network predictions and obtain insights from the modelling step. Thirdly, we identify variables explaining no-show probabilities in a developing context and study its policy implications and potential for improving healthcare access. In addition to quantifying relationships reported in previous studies, we find that income and neighbourhood crime statistics affect no-show probabilities. Our results will support patient prioritization in a pilot behavioural intervention and will inform appointment planning decisions.

Machine Learning,Applications

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to improve the accessibility management of healthcare by predicting patients' no - show behavior (i.e., patients fail to keep their appointments on time). Specifically, the paper aims to: 1. **Increase the attendance rate of medical appointments**: A low attendance rate is associated with adverse health outcomes, such as delays in diagnosis and treatment, an increase in premature mortality, and an increase in the use of emergency services. Therefore, increasing the attendance rate helps to improve patients' health outcomes. 2. **Reduce the efficiency problems of medical service providers**: When patients fail to keep their appointments on time, it will lead to vacant appointments, which will affect the treatment opportunities of other patients, increase medical costs, and lead to idle time for doctors and consulting rooms. By accurately predicting the probability of patients' no - show, resource allocation and scheduling can be optimized, thereby improving the efficiency of medical services. 3. **Support the decision - making process**: Given the uncertainty of patients' behavior, generating relevant information about the probability of no - show can help medical managers make better decisions, such as adjusting resource allocation policies or implementing intervention measures to encourage patients to keep their appointments on time. To achieve these goals, the paper develops a decision - support system (DSS) that uses machine - learning methods to classify patients' no - show behavior and makes the following three main contributions: - **Evaluate the effects of different machine - learning methods**: In particular, the capabilities of Random Forest and Neural Networks in modeling nonlinearity and variable interactions. - **Introduce Layer - wise Relevance Propagation (LRP)**: To improve the interpretability of neural network predictions, thereby obtaining insights in the model. - **Identify the factors that affect the probability of no - show**: Especially in the context of developing countries, study the effects of factors such as income and community crime statistics on the probability of no - show, and explore their policy implications and the potential for improving medical accessibility. Through these methods, the paper aims to provide useful information for medical managers to optimize appointment scheduling and reduce the no - show rate, thereby improving the quality and efficiency of healthcare services.

Improving healthcare access management by predicting patient no-show behaviour

Predicting no-show appointments in a pediatric hospital in Chile using machine learning

Enhancing Health Equity by Predicting Missed Appointments in Health Care: Machine Learning Study

Decision analysis framework for predicting no-shows to appointments using machine learning algorithms

Predicting Patient No-Shows: Situated Machine Learning with Imperfect Data

Public health nurse perspectives on predicting nonattendance for cervical cancer screening through classification, ensemble, and deep learning models

Photoacoustic imaging and characterization of the microvasculature.

Predicting Patient No-Shows in Community Health Clinics: A Case Study in Designing a Data Analytic Product

Predicting Patient Wait Times by Using Highly Deidentified Data in Mental Health Care: Enhanced Machine Learning Approach

Prediction of Pediatric No-Show Rates Before and During the COVID-19 Pandemic Utilizing Machine Learning.

A Decision-Making tool based on historical data for service time prediction in outpatient scheduling

Reducing Waiting Times to Improve Patient Satisfaction: A Hybrid Strategy for Decision Support Management

Application of a Machine Learning Algorithm to Develop and Validate a Prediction Model for Ambulatory Non-Arrivals

Optimizing Doctor Availability and Appointment Allocation in Hospitals through Digital Technology and AI Integration

Enhancing Uncertain Demand Prediction in Hospitals Using Simple and Advanced Machine Learning

No-Show Passenger Prediction for Flights

Predicting Admissions From a Paediatric Emergency Department - Protocol for Developing and Validating a Low-Dimensional Machine Learning Prediction Model

A Framework for Predicting Impactability of Healthcare Interventions Using Machine Learning Methods, Administrative Claims, Sociodemographic and App Generated Data

A Predictive Decision Analytics Approach for Primary Care Operations Management: A Case Study of Double-Booking Strategy Design and Evaluation

A Systematic Review of Features Forecasting Patient Arrival Numbers

Identification of high-risk patients for referral through machine learning assisting the decision making to manage minor ailments in community pharmacies