SW-RCT Implementation of Canadian Syncope Risk Score Based Practice Recommendations

Multi-Centre Cluster-Randomized Implementation of Canadian Syncope Risk Score Based Practice Recommendations for Emergency Department Syncope Management

Syncope is a common reason for emergency department (ED) presentation. While often benign, some patients have serious and life-threatening underlying causes, both cardiac and non-cardiac, which may or may not be apparent at the time of the initial ED assessment. Identifying which patients will benefit from further investigation, ongoing monitoring and/or hospital admission is essential to reduce both adverse outcomes and high costs. The research team has spent over a decade developing the evidence base for a risk stratification tool directed at optimizing the accuracy of ED decisions: the Canadian Syncope Risk Score (CSRS). This tool is now ready for the final phase of its introduction into clinical practice, namely a robust, multicentre implementation trial of the CSRS-based practice recommendations to demonstrate its real-world effectiveness. These recommendations, if applied, could lead to reduction in hospitalization with only 6% of high-risk patients requiring hospitalization, shorter ED lengths of stay for the 76% of ED syncope patients who are at low risk for 30-day serious outcomes, and more standardized disposition decisions, specifically discharge of 18% of medium-risk patients after appropriate discussion. Hence, the investigators hypothesize that an important reduction in hospitalization and ED disposition time can be achieved by implementing the CSRS-based recommendations with potential improvements in patient safety. The overall objective of this study is to evaluate the effectiveness of the knowledge translation (KT) of the CSRS-based practice recommendations in multiple Canadian EDs using a stepped wedge cluster randomized trial (SW-CRT) on health care efficiency and patient safety.

The investigators will conduct a SW-CRT involving 20 participating ED clusters across Canada. The total study duration is 18 months. All clusters start the trial in a control period (usual care) for three months with no intervention being delivered at any site, then sequentially cross over from the control period to the intervention period in random sequence, with 5 clusters (EDs) crossing over after third months, until all sites have adopted the intervention (CSRS based practice recommendation). The following principal research questions will be addressed 1) What is the effect of the knowledge translation and implementation of the CSRS-based practice recommendations on health resource utilization? The health resource utilization measures include the proportion hospitalized, the proportion investigated in the ED, and ED disposition time 2) How and why did the implementation achieve the observed effect? The embedded process evaluation measures align with the RE-AIM (Reach Effectiveness Adoption Implementation Maintenance) framework and include: a) adoption - the proportion of physicians who attended the educational sessions and the proportion who adopted the CSRS in practice; b) reach- the proportion of eligible patients for whom the CSRS was utilized during their ED visit; c) intervention fidelity - the proportion of patients for whom the resulting CSRS recommendations were followed d) maintenance- whether observed adoption rates remain stable or increase over time. 3) What is the effect of the knowledge translation strategy on patient safety, as measured by 30-day serious outcome identification, 30-day and 1-year return ED visits, hospitalizations, and mortality? and 4) validate the ultra-low-risk criteria in a new cohort of patients and assess if the CSRS can be improved in its ability to predict 30-day serious outcomes.

The components of the practice recommendations include: 1) evidence-informed systematic clinical evaluation with appropriate history, physical examination and in-ED investigations (e.g., troponin testing, work-up for pulmonary embolism and CT head) for detecting serious underlying conditions and predicting 30-day serious outcomes; 2) application of the CSRS for risk-stratification at the end of ED visit after no serious underlying conditions for the syncope were identified; 3) use of patient information materials to aid in disposition; and 4) the use of 15-day outpatient cardiac monitoring for CSRS medium and high-risk patients upon ED discharge. All the components of the practice recommendations will be applied by the ED physician treating the patient.

The ED disposition time is defined as the time interval between ED physician initial assessment and ED disposition. We chose this time interval rather than ED length of stay which is the time interval between ED arrival/registration and departure from the ED. The ED disposition time more accurately reflects the active phase of ED physician care for collecting clinical information, work-up of patients and resolution of diagnostic uncertainty. The longer interval of ED length of stay includes wait times both before and after, which are often determined by triage acuity as well as extraneous system factors (e.g., ED and hospital crowding, staffing patterns, sudden influx of patients, availability of inpatient beds and overcapacity protocols, and availability of transportation for a subgroup of patients for whom a decision to discharge has been made);

Physicians: Inclusion criteria: ED physicians involved in ED syncope care Non-ED physicians involved in ED syncope care Physician's delegates involved in ED syncope care Exclusion criteria: ED physicians not involved in ED syncope care Non-ED physicians not involved in ED syncope care Physician's delegates not involved in ED syncope care Patients: Inclusion criteria: Patients who are adults (aged > 18 years) Patients who present to the ED within 24 hours of syncope. Exclusion criteria: Patients who do not fulfill the definition of syncope, namely those with a prolonged loss of consciousness (i.e., > 5 minutes), Glasgow Coma Scale < 15 in patients without dementia (or a change in the mental status from baseline in those with dementia); Patients with witnessed obvious seizure, or head trauma preceding the loss of consciousness; and those who are unable to provide proper details (e.g., alcohol intoxication or other substance use). Patients who had a serious underlying for the syncope identified during the index ED evaluation and those who were consulted to an inpatient service or hospitalized for reasons other than syncope workup (e.g., social reasons such as inability to cope at home, pain due to the fall, significant trauma).