Discrete event simulation model of an acute stroke treatment process at a comprehensive stroke center: Determining the ideal improvement strategies for reducing treatment times

Abstract

Background

Fast treatment is crucial for ischemic stroke patients; the probability of good patient outcomes increases with faster treatment. Treatment times can be improved by making changes to the treatment process. However, it is challenging to identify the benefits of changes prior to implementation. Simulation modelling, which mimics the treatment process, can be used to evaluate changes without patient involvement. This study models the acute stroke treatment process using discrete event simulation (DES) and identifies improvement strategies to reduce treatment times.

Method

The model was developed for a comprehensive stroke center in Nova Scotia, using Python. All treatment pathways and sub-tasks were identified via an observational time and motion study conducted in the center. Nine process change scenarios were tested individually and in combinations. The primary outcome measures were door-to-CT time (DTCT), door-to-needle time (DNT), and door-to-groin puncture time (DGPT). The model simulated 500 patients 30 times.

Results

Collecting patient history on the way to the radiology department (rather than in ED) showed the highest reduction among individual scenarios for DTCT (14.2 vs 12.4 min, p < 0.001). Combining all scenarios in the door-to-CT process resulted in a reduction of the DTCT by approximately 28 %. Thrombolysing patients in the imaging department's waiting area resulted in the lowest DNT (39.4 vs 34.8 min, p < 0.001) among all individual scenarios. The highest reduction in DGPT, among all individual scenarios, was achieved by implementing Rapid angiosuite preparation (67.7 vs 51.4 min, p < 0.001). The combinations of all scenarios resulted in the lowest DTCT (14.2 vs 10.1 min, p < 0.001), DNT (39.4 vs 23.0 min, p < 0.001), and DGPT (67.9 vs 38.5 min, p < 0.001).

Conclusions

The study identified various improvement strategies in the acute stroke treatment process through a discrete-event simulation model. Combining all scenarios resulted in significant reductions for all outcome measures.