Effects of Implementing a Standardized Risk Stratification and Triage Workflow for an Established Pulmonary Embolism...

Rationale: The pulmonary embolism response team (PERT) model was developed to facilitate multispecialty decision-making and expedite therapeutic interventions for patients with pulmonary embolism (PE). PERT implementation has previously been associated with survival benefit in some studies, although specific workflow components that confer survival benefit have not been identified. Objectives: To measure the effects of PERT workflow revisions based upon risk stratification on clinical outcomes in an existing PERT. Methods: As a quality improvement initiative, we implemented three specific workflow interventions to an existing PERT program at an academic medical center: 1) designating triage responsibility to a specific group of providers; 2) assigning guideline-based risk stratification to all calls at triage; and 3) establishing intensive care unit admission guidelines on the basis of risk stratification. We used electronic medical records to review clinical outcomes for all PERT calls for 2 years after implementing the revised workflow and compared these with outcomes for the preceding 2-year period. We used logistic regression to compare the odds of in-hospital mortality before and after the workflow revision, with multiple models adjusted for clinically relevant variables. Results: During the study period (2019-2023), there were 420 unique patient PERT activations with confirmed PE; 253 patients were managed using the revised workflow, and 167 patients were managed using the historical workflow. The proportion of patients meeting the primary endpoint of in-hospital death at 30 days after the workflow revisions was significantly lower than during the historical period (6.3% vs. 18.0%; P < 0.001). Logistic regression analysis demonstrated the revised-PERT workflow to have a protective effect against in-hospital mortality (odds ratio = 0.31; 95% confidence interval = 0.16-0.59; P < 0.001). This mortality benefit remained significant after adjustment for demographics, clinical factors, hemodynamic instability, Pulmonary Embolism Severity Index class, and stage in the Bova scoring system. The workflow revisions were also associated with increased use of advanced therapies but did not change the proportion of patients with major bleeding or length of stay in the intensive care unit or hospital. Conclusions: In an existing PERT program, implementation of three specific workflow revisions centered around risk stratification and level-of-care triage improved survival for patients with PE. These findings suggest that incorporation of a standardized approach and risk stratification are valuable components of the PERT response.