A systematic review of models for forecasting the number of emergency department visits

Abstract

The ability to predict patient visits to emergency departments (ED) is crucial for designing strategies aimed at avoiding overcrowding. A good working knowledge of the mathematical models used to predict patient volume and of their results is therefore essential. Articles retrieved by a Medline search were reviewed for studies designed to predict patient attendance at ED or walk-in clinics. Nine studies were identified. Most of the models used to predict patient volume were either linear regression models including calendar variables or time series models. These models explained 31-75% of patient-volume variability. Although the day of the week had the strongest effect, this variable explained only part of the variability. Other causes of this variability are to be defined. However, the performance of the models was good, with errors ranging from 4.2% to 14.4%. Adding meteorological data failed to improve model performance. The mathematical methods developed to predict ED visits have a low rate of error, but the prediction of daily patient visits should be used carefully and therefore does not allow day-to-day adjustments of staff. ED directors or managers should be aware of the model limitations. These models should certainly be used on a larger scale to assess future needs.

Forecasting Hourly Patient Visits in the Emergency Department to Counteract Crowding

“…Rather, forecasts of hourly arrivals are less accurate than forecasts of daily arrivals. It may be noted that the MAPE of 11% for daily arrivals in ED2 is within the range (4.2-14.4%) reported by Wargon et al [24] in a review of models for forecasting the number of daily ED visits, thereby suggesting that a worsening of the MAPE from 11% (daily) to 49% (hourly) may not be unusual. Part of the reason for the worse MAPE for hourly arrivals likely is that the averaging approach of the models means that they underestimate the level of short-term variability and the size of occasional surges [15].…”

“…It also speaks to the advantage of our models that they do not overfit the data. In addition, models that extend calendar variables with weather variables, such as temperature readings and rain-/snowfall data, do not achieve better accuracy [24]. Adding variables about the demand for inpatient resources has also been found not to improve the accuracy of models of ED visits [22].…”

From Model to Forecasting: A Multicenter Study in Emergency Departments

“…7 Numerous authors have already become interested in the prediction of patient visits in EDs, [8][9][10][11][12][13][14][15] as well as in walk-in clinics, 6,[16][17][18] mainly using calendar variables in multiple linear regression models. 19 These methods examine correlations between patient visits and a number of independent determinants, mostly calendar variables. Other techniques have been employed, especially time series analysis predicting future values from past values.…”

Modelling Vulnerability to Severe Weather

“…Wargon et al 2009 have done a review on studies concerned with the modelling and forecasting of emergency department visits. It is found that the number of patient visits at emergency departments or walk-in clinics can be modelled with rather good performance.…”

“…Mostly based on predictors such as the day of the week or season these models explain between 31% and 75% of patient-volume variability. However including meteorological data apparently failed to improve model performance (Wargon et al 2009). Findings of more recent studies however do find that weather factors such 5 as temperature and humidity play a role in the demand for ambulance services and demonstrate that including weather forecast data can in fact improve forecasts of daily ambulance demand (Wong and Lai (2014)).…”