Kerstin de Wit

Years since first academic appointment: Less than 5 years

Title: Making Decisions in the Era of the Clinical Decision Rule

Full author list: Kerstin de Wit, Mathew Mercuri, Teresa Chan


CT pulmonary embolism (PE) scanning should be avoided when possible because it exposes patients to radiation, increasing their lifetime risk of cancer, and causes contrast induced nephropathy, There is high quality evidence that CT can be avoided with bedside ‘rule out’ tools such as the Wells score and D-dimer, the PERC rule and age-adjusted D-dimer. Despite this, discrepancies persist between evidence-based guidelines and emergency physician practice.

Most diagnostic PE research was lead by Thrombosis researchers (not emergency researchers). It is unsurprising that the diagnostic tools may not be best suited for use in a high pressured emergency environment and there is inadequate understanding of context-specific barriers in an emergency department.


This study will document how emergency physicians diagnose PE, the knowledge gaps and perceived barriers to using PE clinical decision rules.


Population: Canadian emergency physicians from multiple Provinces in Canada. Purposive sampling will ensure half have been in practice for <10 years and half will be university affiliated.

Intervention: We are using a cognitive task analysis. The interview has three parts (see Figure 2). Physicians will be invited to participate in a one hour confidential interview. The physician will be blind to the specific purpose of the interview other than it will be about ‘clinical decision making’.

Phase 1. Critical incident interviewing. The physician recalls a ‘critical incident’ when they tested for PE. The physician explains how they determined whether the patient had PE or not.

Phase 2. Video prompted mind mapping. The participant is shown two videos of simulated patients with possible PE. They are asked to draw out their diagnostic thinking process on paper (mind mapping) and explain their reasoning.

Phase 3. Knowledge assessment and memory aide usage. The participants take a quiz with the components of the PERC rule, the Wells score and age-adjusted D-dimer rule. We will capture use of reference devices (e.g. phone or desktop computer) and websites.

Outcomes: Qualitative outcomes – we will identify common themes in behaviours, diagnostic sequences and barriers to using clinical decision rules and D-dimer. We will develop a model for emergency physician diagnostic reasoning.

The primary quantitative outcome will be the proportion of physicians who order CT scanning without Wells rule application in Phases 1 and 2 of the interview. The secondary quantitative outcomes will be the median number of components identified correctly for the Wells, PERC and age-adjusted D-dimer rule and the components of the rules most often correctly recalled.

Analyses: Using framework analysis, we will identify common experience, opinion, feelings, knowledge and actions using an inductive approach. We will use the emerging framework to group the data and identify relationships in attitude, diagnostic behaviour and shared ideas.

Figure 2: Overview of physician interview process qualitative analysis

CTA cognitive task analysis









The study has obtained research ethics approval and has already recruited 15 local emergency physicians.


  1. Would you be willing to perform these interviews at your centre? We are looking for an additional 40 interviews from across Canada and have reimbursement for the research department and the participating physicians, per interview.