A modelling tool for capacity planning in acute and community stroke services

Area of Science:

• Healthcare Operations Research
• Health Services Management
• Biostatistics

Background:

• Capacity planning for stroke care often relies on simplistic average-based estimates, underestimating actual needs.
• Complex patient pathways, including variations in complexity, admission rates, and discharge delays, challenge traditional planning methods.
• This study analyzes capacity requirements for acute and community stroke services, identifying bottlenecks and future needs.

Purpose of the Study:

• To identify current capacity bottlenecks impacting patient flow in stroke pathways.
• To forecast future capacity requirements considering increased admissions.
• To evaluate the impact of co-location, bed pooling, and patient subgroups on capacity needs.

Main Methods:

• A discrete-event simulation model was developed using anonymized administrative data.
• The model simulated patient flow from acute admission through community rehabilitation and early supported discharge.
• Predictive analysis focused on identifying and quantifying the probability of admission delays.

Main Results:

• Increasing acute beds from 10 to 14 reduced acute stroke unit delays from 1 in 7 to 1 in 50 patients.
• Co-locating and pooling beds (8 out of 26) decreased acute admission delays to 1 in 29 and rehabilitation delays to 1 in 20.
• Planning by average occupancy resulted in delays for 1 in 5 patients in the acute stroke unit.

Conclusions:

• Average occupancy-based planning is insufficient for managing stroke pathway variability and achieving desired service levels.
• Simulation modeling provides valuable decision support for optimizing bed numbers and service organization in stroke care.
• Underutilized mathematical techniques, when implemented via simulation, offer a robust approach to capacity planning.