Qualitative experience implementing an emergency department-based outpatient low-risk pulmonary embolism management pathway

Acute pulmonary embolism (PE) is often diagnosed in an emergency department (ED) setting. Up to 30–50% of patients with low-risk PE can be safely anticoagulated on an outpatient basis, a practice with professional society guideline support.^{1 –4} Such patients are commonly prescribed oral anticoagulants without hospitalization in parts of Europe and Canada, but unnecessary hospitalization for patients with low-risk PE remains common in the United States.^{5 –7} In this study, we aimed to validate previously described barriers and facilitators of ED-based outpatient low-risk PE management. ⁸ In addition, we used implementation science frameworks in two clinician interview phases to guide the development of and refine a multicomponent intervention (‘the outpatient PE pathway’) to support practice changes and increase ED-based outpatient management of low-risk PE at our institution.

This study was conducted within a single academic ED setting at the University of Michigan. This study was not regulated as human subject research by the institutional review board, and informed consent was not required due to the study’s quality improvement nature. Semistructured interviews were conducted with ED attending physicians, resident physicians, and advanced practice providers. Interviews were conducted from March 15-May 5, 2021 (Phase 1) and August 22, 2022-April 18, 2023 (Phase 2). We planned to interview a minimum of 10 subjects in each interview phase, with expansion if needed to achieve thematic saturation, an approach consistent with prior research showing that all key themes from qualitative interviews can usually be identified after nine individual interviews within a population. ⁹ Nonclinicians (VG, AC) conducted the interviews using an interview guide (supplemental material) to reduce any reporting bias or influence from a clinician interviewer. Phase 1 interviews were developed using the Tailored Implementation of Chronic Diseases’ (TICD) framework, a project designed to tailor implementation efforts to local conditions.^10,11 Following phase 1 interviews, the multicomponent outpatient PE pathway was designed as described herein. After releasing the outpatient PE pathway for clinical use, phase 2 interviews were conducted with a new cohort of clinicians who utilized the pathway during patient care. Phase 2 interviews used the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework, a well-described implementation science evaluation framework with strength in design, dissemination, and external validity. ¹² All interviews were conducted virtually over approximately 30 minutes and subsequently transcribed. Transcripts were reviewed by team members (VG, AC) for accuracy. To facilitate qualitative analysis of transcripts, rapid analysis matrices were constructed using the rigorous and accelerated data reduction (RADaR) method, and key themes were evaluated. ¹³

During phase 1 interviews, 10 out of the 16 ED clinicians who were approached completed interviews. Reported barriers to outpatient management included: (1) low clinician confidence in patient selection/risk stratification, (2) unfamiliarity with clinical evidence, (3) burden of outpatient care coordination, (4) anticoagulant medication access/cost, (5) fear/stigma of PE diagnosis, (6) patient health literacy, (7) patient social support, and (8) concurrent patient illnesses. Facilitators of outpatient management included: (1) institution support/guideline policy, (2) outpatient follow-up systems, (3) social work financial assistance for prescriptions, and (4) department educational initiatives. A detailed summary of barriers and facilitators is provided in Supplemental Tables S1 and S2.

Based on phase 1 interview findings, the outpatient PE pathway was designed with the goal of facilitating increased outpatient management of low-risk PE. Key pathway components targeted early patient identification, clinician notification, streamlined order-sets, coordination of outpatient anticoagulation, rapid follow up, and department education. First, upon ordering a computed tomography (CT) scan for PE, an automated pulmonary embolism severity index (PESI) calculation was performed. In the case of low-risk PESI (≤ 85), a nonintrusive electronic medical record (EMR) best practice advisory (BPA) notified clinicians to consider outpatient management if CT diagnosed a PE. If desired, clinicians could utilize the low-risk outpatient PE pathway order-set to streamline workload, including the initial loading dose of anticoagulant, oral anticoagulant prescription, social work notification for medication vouchers if needed, rapid access clinic scheduling, documentation dot phrases, and patient instructions. Concurrent departmental education initiatives incorporated relevant topics at routine didactic sessions, a pathway-summarizing 3 × 5 note card at workstations, and an accessible website summarizing supportive literature and key clinician tasks for outpatient management of low-risk PE (Figure 1). The outpatient PE pathway was implemented in the ED setting with EMR integration and support from ancillary social work and pharmacy services.

Figure 1.

Process of implementing ED-based low-risk outpatient PE pathway. (A) Preimplementation interviews and intervention design phase. (B) Postimplementation interviews and iterative modification phase.

During phase 2 interviews, 10 out of the 14 ED clinicians approached completed interviews. The median time between outpatient PE pathway use in patient care and completion of phase 2 interviews was 18 days. During phase 2 interviews, clinicians reported familiarity with low-risk PE management, facilitated by departmental education lectures, workspace-posted informative 3 × 5 note cards, and embedded BPA notifications in the EMR. All patients managed via the outpatient PE pathway were scheduled for rapid-access clinic follow-up appointments before being discharged from the ED. Clinician feedback during phase 2 interviews guided iterative modifications to the outpatient PE pathway. Key modifications included corrections to PESI automated calculations, changing anticoagulant prescriptions from a ‘starter pack’ (unavailable at many pharmacies) to traditional single pill strength prescriptions (with free text instructions), improved coordination of rapid access clinic appointments, and corrections to the EMR order-set.

In this qualitative study, phase 1 interviews with ED clinicians validated barriers and facilitators to outpatient management of low-risk PE previously described by Westafer et al. ⁸ Key actionable barriers included: (1) clinician identification of low-risk patients, (2) discharge care coordination burdens, (3) medication access, and (4) rapid outpatient follow up. Interview feedback guided implementation strategies as part of a single-center, multilevel implementation intervention. Phase 2 interviews guided process refinements that improved the workload for clinicians and care coordination for patients.

This study provides a framework for implementation science and quality improvement within a range of clinical contexts, particularly within multidisciplinary care coordination and clinical practice culture change. The generalizability of this study is improved by the interview approach to sample individual experiences from distinct training backgrounds of attendings, residents, and advanced practice providers. The use of phase 1 interviews allowed the implementation team to design the outpatient PE pathway to best fit the preexisting workflow of ED clinicians, improve clinician comfort with discharging low-risk PE, and decrease the workload to accomplish such discharges. The use of phase 2 interviews allowed the implementation team to overcome residual functional barriers for clinicians in a ‘real-time’ manner.

As with all qualitative studies, a key limitation of this study is that the limited number of interview subjects may incompletely capture clinician perspectives on the outpatient management of acute PE. Nevertheless, we achieved thematic saturation within our sample, and expanding to a larger interview population within our institution would have been unlikely to yield new themes. Given medical practice variability between institutions and across geography, our use of single-center interview subjects may limit the external validity of our findings to other clinical settings, and necessitates further study in additional populations.

In conclusion, we have performed a single-center qualitative study, guided by implementation science frameworks, to investigate and intervene on ED clinician low-risk PE management practices. Based on two rounds of qualitative interviews, we were able to tailor the design of and iteratively optimize an outpatient PE pathway. The implementation strategies described here may be considered for custom application and extrapolation to other healthcare settings.

Supplemental Material

sj-pdf-1-vmj-10.1177_1358863X251326585 – Supplemental material for Qualitative experience implementing an emergency department-based outpatient low-risk pulmonary embolism management pathway

Supplemental material, sj-pdf-1-vmj-10.1177_1358863X251326585 for Qualitative experience implementing an emergency department-based outpatient low-risk pulmonary embolism management pathway by S Nabeel Hyder, Valerie Gavrila, Anthony Cuttitta, Colin Greineder and Geoffrey D Barnes in Vascular Medicine

Supplemental Material

sj-pdf-2-vmj-10.1177_1358863X251326585 – Supplemental material for Qualitative experience implementing an emergency department-based outpatient low-risk pulmonary embolism management pathway

Supplemental material, sj-pdf-2-vmj-10.1177_1358863X251326585 for Qualitative experience implementing an emergency department-based outpatient low-risk pulmonary embolism management pathway by S Nabeel Hyder, Valerie Gavrila, Anthony Cuttitta, Colin Greineder and Geoffrey D Barnes in Vascular Medicine