Modeling and forecasting emergency department crowding using SARIMA, Holt Winter method, and Prophet models

Mansoor AL Yarubi; Nur Arina Bazilah Kamisan; Siti Mariam Norrulashikin

doi:10.53730/ijhs.v9n1.15567

Authors

Mansoor AL Yarubi Universiti Teknolcgi Malaysia, Fohor Bahru, Malaysia Ministry of Health, Nizwa, Oman
Nur Arina Bazilah Kamisan Universiti Teknologi Malaysia, Johor Bahru, Malaysia
Siti Mariam Norrulashikin
sitimariam@utm.my
Universiti Teknologi Malaysia, Johor Bahru, Malaysia

Keywords:

Emergency department, Holt winter method, Prophet model, SARIMA model, Univariate time series

Abstract

Emergency department (ED) crowding in health care is linked with longer wait times, high mortality rates, and low healthcare quality. Univariate time series models such as Seasonal Autoregressive Integrated Moving Average (SARIMA), Holt-Winters method (HW), and Prophet model (PM) have been widely employed to predict ED crowding. However, there is no consensus on the best fit time series model for ED crowding forecasting. This study compared the predictive precision of three univariate time series models, SARIMA, HW, and PM, in predicting ED crowding at Nizwa Hospital in Oman. The study used hourly patient visits at ED from January to December 2023. The model selection was based on minimizing Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). ED visits showed irregular trends and seasonal effects due to time and day of the week effects. The 24-hour ED visits depicted two peak phases: noon (local maximum) and around 10 PM to midnight (global maximum). The prophet model had better accuracy than the SARIMA and HW models. Adopting the Prophet model predictions can help avoid unexpected ED crowding, reduce waiting times, and improve quality health care management.

Downloads

Download data is not yet available.

References

Becerra, M., Jerez, A., Aballay, B., Garcés, H. O., & Fuentes, A. (2020). Forecasting emergency admissions due to respiratory diseases in high variability scenarios using time series: A case study in Chile. Science of the Total Environment, 706, 134978. https://doi.org/10.1016/j.scitotenv.2019.134978 DOI: https://doi.org/10.1016/j.scitotenv.2019.134978

Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2015). Time series analysis: Forecasting and control (5th ed.). Wiley.

Calegari, R., Fogliatto, F. S., Lucini, F. R., Neyeloff, J., Kuchenbecker, R. S., & Schaan, B. D. (2016). Forecasting daily volume and acuity of patients in the emergency department. Computational and Mathematical Methods in Medicine, 2016, 3863268. DOI: https://doi.org/10.1155/2016/3863268

Chen, W., Linthicum, B., Argon, N. T., Bohrmann, T., Lopiano, K., Mehrotra, A., Travers, D., & Ziya, S. (2020). The effects of emergency department crowding on triage and hospital admission decisions. The American Journal of Emergency Medicine, 38(4), 774–779. https://doi.org/10.1016/j.ajem.2019.06.039 DOI: https://doi.org/10.1016/j.ajem.2019.06.039

Chiu, I.-M., Lin, Y.-R., Syue, Y.-J., Kung, C.-T., Wu, K.-H., & Li, C.-J. (2018). The influence of crowding on clinical practice in the emergency department. The American Journal of Emergency Medicine, 36(1), 56–60. https://doi.org/10.1016/j.ajem.2017.07.011 DOI: https://doi.org/10.1016/j.ajem.2017.07.011

Duarte, D., & Faerman, J. (2019). Comparison of time series prediction of healthcare emergency department indicators with ARIMA and Prophet. Proceedings of the Computer Science and Information Technology Conference, 123-133. DOI: https://doi.org/10.5121/csit.2019.91810

Etu, E. E., Monplaisir, L., Masoud, S., Arslanturk, S., Emakhu, J., Tenebe, I., ... & Krupp, S. (2022, June). A comparison of univariate and multivariate forecasting models predicting emergency department patient arrivals during the COVID-19 pandemic. In Healthcare (Vol. 10, No. 6, p. 1120). MDPI. DOI: https://doi.org/10.3390/healthcare10061120

Feng, T., Zheng, Z., Xu, J., Liu, M., Li, M., Jia, H., & Yu, X. (2022). The comparative analysis of SARIMA, Facebook Prophet, and LSTM for road traffic injury prediction in Northeast China. Frontiers in public health, 10, 946563. DOI: https://doi.org/10.3389/fpubh.2022.946563

Hoot, N. R., & Aronsky, D. (2008). Systematic review of emergency department crowding: causes, effects, and solutions. Annals of emergency medicine, 52(2), 126-136. https://doi.org/10.1016/j.annemergmed.2008.03.014 DOI: https://doi.org/10.1016/j.annemergmed.2008.03.014

Hoot, N. R., LeBlanc, L. J., Jones, I., Levin, S. R., Zhou, C., Gadd, C. S., & Aronsky, D. (2009). Forecasting emergency department crowding: A prospective, real-time evaluation. Journal of the American Medical Informatics Association, 16(3), 338–345. DOI: https://doi.org/10.1197/jamia.M2772

Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts.

Jeyaraman, M. M., Copstein, L., Al-Yousif, N., Alder, R. N., Kirkland, S. W., Al-Yousif, Y., Suss, R., Zarychanski, R., Doupe, M. B., Berthelot, S., Mireault, J., Tardif, P., Askin, N., Buchel, T., Rabbani, R., Beaudry, T., Hartwell, M., Shimmin, C., Edwards, J., … Abou-Setta, A. M. (2021). Interventions and strategies involving primary healthcare professionals to manage emergency department overcrowding: A scoping review. BMJ Open, 11(5), e048613. DOI: https://doi.org/10.1136/bmjopen-2021-048613

Pines, J. M., Hollander, J. E., Localio, A. R., & Metlay, J. P. (2006). The association between emergency department crowding and hospital performance on antibiotic timing for pneumonia and percutaneous intervention for myocardial infarction. Academic Emergency Medicine, 13(8), 873-878. DOI: https://doi.org/10.1197/j.aem.2006.03.568

Pines, J. M., Iyer, S., Disbot, M., Hollander, J. E., Shofer, F. S., & Datner, E. M. (2008). The effect of emergency department crowding on patient satisfaction for admitted patients. Academic Emergency Medicine, 15(9), 825-831. DOI: https://doi.org/10.1111/j.1553-2712.2008.00200.x

Rémi, J., Pollmächer, T., Spiegelhalder, K., Trenkwalder, C., & Young, P. (2019). Sleep-related disorders in neurology and psychiatry. Deutsches Ärzteblatt International, 116(41), 681. DOI: https://doi.org/10.3238/arztebl.2019.0681

Rosychuk, R. J., Youngson, E., & Rowe, B. H. (2016). Presentations to emergency departments for COPD: A time series analysis. Canadian respiratory journal, 2016(1), 1382434. DOI: https://doi.org/10.1155/2016/1382434

Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37-45. DOI: https://doi.org/10.1080/00031305.2017.1380080

Tuominen, J., Roine, A., Saviauk, T., Seppo, A., Pihlaja, M., Ovaska, J., ... & Oksala, N. (2021). Forecasting Daily Arrivals and Peak Occupancy in a Combined Emergency Department. DOI: https://doi.org/10.21203/rs.3.rs-138768/v1

Vieira, A., Sousa, I., & Dória-Nóbrega, S. (2023). Forecasting daily admissions to an emergency department considering single and multiple seasonal patterns. Healthcare Analytics, 3, 100146. https://doi.org/10.1016/j.health.2023.100146 DOI: https://doi.org/10.1016/j.health.2023.100146

Villani, M., Earnest, A., Nanayakkara, N., Smith, K., De Courten, B., & Zoungas, S. (2017). Time series modelling to forecast prehospital EMS demand for diabetic emergencies. BMC health services research, 17, 1-9. DOI: https://doi.org/10.1186/s12913-017-2280-6