Improving healthcare access management by predicting patient no-show behaviour

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.

Improving intervention design to promote cervical cancer screening among hard-to-reach women: assessing beliefs and predicting individual attendance probabilities in Bogotá, Colombia

“…These variables have been found to have good predictive value for medical appointment attendance [ 30 ]. Five of these variables (age, lead time, month, and day) were previously used to model no-show behaviour for preventive care appointments in Bogotá [ 31 ]. We also retrieved data from historical administrative records (dataset 2) relating to appointments scheduled for 23,384 women between 2017 and 2019 as part of the ACS program.…”

“…The methodology is described in detail in [ 31 ] and is summarised briefly here. For age and lead time we used decision trees to build categorical variables aiming at increasing model stability [ 32 ].…”

Optimizing Pension Participation in Kenya through Predictive Modeling: A Comparative Analysis of Tree-Based Machine Learning Algorithms and Logistic Regression Classifier

“…Figures A4 and A5 display the connection between different variables and pension uptake using the dataset that majorly comprised categorical variables and few continuous variables. To examine the associations between these variables, a correlogram was generated using the Cramer's V correlation coefficient, which measures the association between categorical variables (Barrera Ferro et al 2020). A Cramer's V correlation coefficient function was utilized to generate a correlation matrix for the categorical columns, and Seaborn's heatmap function was employed for correlogram visualization.…”

Studying the appointment scheduling window considering patient no-show behavior in one public and one private outpatient clinics

“…The high rate of patient absence is one of the most important challenges that out-patient health-care centers have to cope with in the appointment scheduling process (Ferro et al , 2020). Overbooking, open access appointment system, simulation and queueing theory are the most widely used methods to reduce the negative effects of high patient no-show rate.…”

“…These systems provide an opportunity for health centers to respond to the maximum number of patients and reduce some of their costs, such as costs of patient waiting time, patient rejection and physician idle time (Saremi et al , 2013). Most of the health providers have to cope with efficiency problems and poor health outcomes, which are caused by low attendance levels in medical appointments (Ferro et al , 2020). High no-show rate of patients in the appointed time leads to waste of resources along with an increase in total costs that consequently decreases the efficiency of the health-care center.…”