DEVELOPING A SIMULATION-BASED DECISION SUPPORT TOOL TO IMPROVE CANCER CARE RESOURCE USE AND PATIENT ACCESS

The healthcare system delivers processes that involve complex interactions among different types of staff, equipment, and the patients that receive diverse medical services. These interactions often result in long waiting times for patients and poor utilization of staff and resources, aspects that need to be improved. This dissertation presents a study to address issues in these areas by considering complex dynamic behaviors, namely uncertainty, ambiguity, and incompleteness, present in such a system. The overall objective of this dissertation is to develop a simulation system for aiding decision-makers in managing operations, such as scheduling and other administrative policies making. A specific healthcare facility, the Saskatoon Cancer Clinic (SCC), is taken as the study vehicle, both for illustration and validation purposes. This dissertation involves experiments, based on the simulation of five different scenarios in the SCC operations environment, which represent the activities of the staff and patients, patients‟ pathways in the SCC, and physical resources (Reception, Medical Review Office, Phlebotomy, Examination Rooms). The system makes the following assumptions: (1) the patients are divided into two classes (New Patients and Returning Patients) and (2), cancers are classified into eight types. The patients are further characterized with other attributes: (a) the punctuality of coming to the clinic, (b) no-show, (c) cancelation, (d) the need of isolation (being infectious), (e) time spent with various staff (14 different types in total), (f) the number of times the patient received services by each staff during their visit, (g) the frequency of the appointments, (h) the time span between appointments, and (i) patient treatment plans. Each staff is characterized by (1) schedules, (2) work shifts, (3) vacation, (4) type of disease they have the expertise to serve, (5) types of patients they serve, (6) the time assigned for service, and (7) actual time of service. The modeling tool used for building the simulation system is Discrete Event Simulation (DES), specifically AnyLogic software, because it is best suited for the conceptualization and granularity level made for this system. The simulation system consists of (1) a BAseline Simulation model (BAS) (i.e. current operations), (2) five different What-If Scenarios (WISes) (variations of different structural changes in the system). The BAS is created first, followed by its extensive validation. The five WIS models are created to generate results for fourteen different Key Performance Indicators (KPIs) of the system. The KPIs characterize the patients, staff, and resource in the system, e.g. number of patients waiting for their first appointment with an oncologist, length of stay of patients in the clinic, utilization of examination rooms, etc. The KPIs are analyzed individually as well as aggregated with an equal weight; the domain of the real number values of the KPIs is a value range of (1, 6), where „6‟ denotes the worst and „1‟ denotes the best KPI. The result of the simulation for the five WISes is as follows: (1) Scenario 1 has a value of 1.93, which is ranked the first among all the five scenarios (having two oncologists sharing three examination rooms); (2) Scenario 3 has a value of 2.57, which is ranked the second among all the five scenarios – this scenario has a flexible lunch hour, and it has three new patients to be consulted per each four hour shift block; (3) Scenario 5 (3.36) is ranked the third, followed by Scenario 2 (3.71), and Scenario 4 (4.5), and the BAS (4.79) ranks last. This dissertation draws the following conclusions. First, there are ways to increase the efficiency and effectiveness of oncology clinics, as well as other ambulatory clinics. Second, patients benefit from the re-design of certain clinical administrative policies. Third, increasing the efficiency of the clinic, Canadian health dollars can be saved. Previous studies focused only on specific aspects, such as (1) patient wait times, (2) patient's schedules, (3) utilization of chemotherapy chairs, (4) utilization of nurses, (5) staff overtime, (6) staff schedule, in a non-integral manner, while this study takes an integral approach with the finest system granularity. The main scientific contributions of this dissertation in the field of operation management of complex healthcare systems are: (1) validating the effectiveness of the approach of human collaborative decision making based on the simulation of individual to individual interactions in medical treatment centers; (2) developing the procedure of constructing a discrete event dynamics simulation model for medical treatment centers with highly uncertain dynamics, including model construction and validation; (3) demonstrating the benefit of the approach along with the simulation system in reducing the waiting time of patients to receive treatments. This dissertation has provided evidence of improving the efficiency of complex healthcare systems, generating a positive impact to the quality of Canadian healthcare.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.