Developing and validating a prediction model for frequent attenders at a Swedish emergency department using an electronic medical record system, a retrospective observational study

Abstract

Background: Frequent attenders (FA) account for a significant number of emergency department (ED) visits but to date there is no prediction model to identify patients at risk of becoming a FA. The aim of this research was to identify and describe FA using readily available data provided by electronic medical records and create a prediction model to identify future FA Method: Adults ≥18 years that visited the ED during 2015 were included. Patients with ≥4 visits were defined as FA, and patients with ≤3 visits were placed in the control group. Numerous variables were analyzed and differences between the groups compared. Logistic regression analysis was used to determine the predictor variables and the model validated using Receiver Operating Characteristic (ROC) on an independent sample. Results: 6635 patients were included in developing the model: 15.3 (n=1012) were classified as FA and 15.4 (n=1011) as the control group. Variables associated with at risk of becoming a FA were the following: age above 60 years OR 1.52 [CI 1.27 – 1.82], ED arrival by ambulance or helicopter OR 1.31 [CI 1.08 – 1.58], sheltered living OR 3.82 [CI 2.37 – 6.17], previous contact with psychiatric department OR 1.52 [CI 1.23 – 1.89], 10 outpatient care visits or more OR 4.81 [CI 3.81 – 6.08] and 10 outpatient care physician visits or more OR 3.94 [CI 3.25 – 4.78]. The ROC in the validation set had an area under the curve of 0.85 [CI 0.84 – 0.86]. Conclusion: Data from electronic medical record software can be used to create and validate the risk of becoming a FA in the ED. We found that age over 60 years, ED arrival by ambulance or helicopter, sheltered living, previous contact with psychiatric departments, and frequent visits at outpatient care together predict the risk of becoming a FA.