Abstract
Objectives: Cancer Care Ontario (CCO) is the provincial governmental organization responsible for planning hematopoietic cell transplantation (HCT) services in Ontario, Canada. The objective of this project is to develop a capacity planning model to investigate the effects on wait times of adding extra bed capacity for allogeneic transplant (ALLOHCT) in HCT centers.
Approach: A high-level process flow diagram was generated to understand patient flow at a 6-bed HCT unit within a hospital in Ontario and validated through consultation. This flow diagram was used to construct a system dynamics model to simulate patient flow. The model was parameterized with data from CCO, Discharge Abstract Database, and with hospital and clinical expert input. The effects at six months were projected for five scenarios: 1) current state; 2) increase bed capacity by 1 bed; or 3) increase bed capacity by 2 beds; 4) increasing patient demand by 20 patients per year; 5) combination of scenarios 3 and 4. Provincial clinical consensus established a benchmark wait time of 42 days for ALLOHCT from ready to transplant to the transplant date. In addition, the estimated number of beds required to reduce the wait times to the provincial benchmark within 1 year was calculated.
Results:The addition of 1 ALLOHCT bed resulted in a reduction of 22% and 11% to the ALLOHCT wait times and wait lists, respectively. The addition of 2 beds resulted in a reduction of 38% and 22% to the wait times and wait lists, respectively. If the demand increases by 20 patients per year, the addition of 2 beds resulted in a reduction of 16% in the wait times and while the wait list may experience a brief reduction, after 6 months, the wait list size will have increased by 9% as a result of the increased demand. In order to reduce the wait times to the provincial benchmark within 1 year, an additional 8 beds are needed.
Considerations: Concurrent planning for additional health human resources (physicians, nurses, etc…) needs to be done to ensure the additional beds are adequately staffed. This model also only considers the effects of adding beds within 1 year. There may be instances where bed space cannot be immediately opened and new capital is required. Additionally, the demand for ALLOHCT continues to increase, which in turn drives up the number of arrivals to the queue. A multi-year model will be built to account for timing of bed openings and increasing demand for ALLOHCT.
Conclusion:Using a system dynamics model, we are able to quantify the relationship between ALLOHCT bed capacity and wait times at an HCT center. This model can be used to estimate the ALLOHCT bed requirements for sites in other jurisdictions where ALLOHCT demand and wait time benchmarks are known.
Disclosures
Kouroukis: Janssen: Research Funding; Karyopharm: Research Funding.
A System Dynamics Simulation Applied to Healthcare: A Systematic Review