03:30 PM - 04:00 PM
MRI Capacity Assessment in Ontario: A Wait Times Estimation Model
Saba Vahid, Methodologist, Analytics and Informatics, Cancer Care Ontario
AliVahit Esensoy, Acting Senior Manager, Analytics and Informatics, Cancer Care Ontario
Jonathan Norton, Senior Team Lead, Access To Care Informatics, Cancer Care Ontario
Zain Mujtaba, Implementation Lead, Health System Funding and Quality, Ontario Ministry of Health and Long Term Care
Jason Dang, Senior Data Analyst, Analytics and Informatics, Cancer Care Ontario
Managing wait times for MRI is part of the Ontario government’s strategy to transform healthcare in the province. In this project, an MRI capacity assessment model at the regional level was delivered to the Ministry of Health and Long term Care (MOHLTC). The tool provides MRI demand forecasts based on historical data from the Wait Times Information System (WTIS) in Ontario. Time series forecasting was used to produce regional level demand forecasts.
In addition, a wait times estimation module was developed using multivariate linear regression to support the MOHLTC in assessing incoming funding requests based on their estimated impact on regional wait times. Finally, non-linear programming was used to recommend optimal regional funding levels to reach provincial wait time targets.
This interactive model allows MOHLTC to investigate different funding policies, such as the wait times impact of blitz versus sustained funding.
04:00 PM - 04:30 PM
A Rolling Horizon Approach to Forecast Emergency Department Patient Arrivals.
The mismatch between the demand for ED services and the available resources have direct and indirect negative consequences, such as long wait times, overcrowding, poor patient outcomes, and productivity loss. Moreover, ED physician pay in some jurisdictions reflects pay-for-performance contracts based on operational benchmarks. To assist in capacity planning and meeting these benchmarks, we built a forecasting model to produce short-term forecasts of ED arrivals. The variability in patient presentation rates and relative allocation of resources require the need for separate models for high and low acuity patients. We used regression and time series techniques to model total ED arrivals as well as separate forecasts for high (resuscitation, emergent) and low (urgent, less-urgent, non-urgent) acuity patients. Several accuracy measures have been calculated and compared to validate our forecasting models. We advocate the use of the rolling horizon approach to accurately forecast ED patient arrivals. Our analysis provides ED managers with valuable insight to efficiently allocate ED resources.
04:30 PM - 05:00 PM
A characterization of the last 20 years of Operations Research on Hospital Admission Systems
This paper uses a collaborative platform, produced by the authors that enables one to show an updated review of Hospital Admission Systems (HAS) and outline an overview of the subject, based upon articles published in the last six decades. The main characteristics of Operational Research (OR) as applied to HAS within the 1995–2015 period are described and analysed. The information contained in relevant publications on this theme is catalogued in accordance with specific key elements previously chosen; a database containing this information is built. Moreover, the progresses on OR technics applied to HAS on the period under concern are evaluated, chiefly in terms of the related applications and their practical implementation. The goal is to provide an insight on the developments of OR technics applied to HAS over the last years and point out new pathways for the near future.