Best Strategies for Implementation of Clinical Pathways in Emergency Department Settings

To meet the first objective, we will use Michie's Theoretical Domains Framework (TDF) to identify and organize CP-relevant behaviours. We will conduct qualitative interviews, guided by the TDF and involving knowledge users from participating sites, to identify barriers and facilitators for change relevant to the proposed CP implementation. Experience from other implementation initiatives will also be used to guide the intervention. An intervention mapping technique will be used to select and organize appropriate strategies to affect behaviour changes. Further input will be sought from knowledge user site partners. The intervention will begin with a site visit to assess readiness for adoption, infrastructure requirements, and ability to implement the pathway.

To meet objective two, we will use a cluster randomized control trial (RCT) design, through measurement of relevant clinical and process outcomes in 16 community EDs in Ontario. Using a balanced incomplete block design, we will study this implementation intervention with two clinical pathways: i) Asthma and ii) Vomiting & Diarrhea (V&D). We will recruit and randomize 16 community EDs to receive the intervention for one of the CPs and serve as control for the alternate pathway. Primary clinical outcomes for each CP include the following: i) Asthma: the proportion of asthmatic patients with moderate to severe exacerbation who received steroids within 60 minutes from ED presentation; and ii) V&D: the proportion of children with mild to moderate dehydration treated with oral rehydration therapy. The primary process outcome measure will be the presence of an appropriately completed CP, as per pre-defined criteria, on the chart for relevant patients. Secondary clinical outcomes include disease-specific treatments, ED length of stay, revisits and hospitalizations. Secondary process outcomes include CP use based on shift-level data relating to the busyness of the ED. A mixed methods evaluation will be used to assess clinical and process outcomes. Data sources include chart audits, administrative databases, environmental scans and knowledge user interviews. Sample size calculations for the primary process outcome yielded the highest number of required charts. A total sample size of 6400 patient charts is required to detect an absolute difference of 12% between control and intervention hospitals using a two-sided test at the 5% level of significance.

To meet objective three, a process log will be used to track progress with site deadline targets, implementation issues, degree of uptake, local adaptations and adherence.

And finally, to meet objective four, comparisons of CP implementation and health care costs will be done for all patients with Asthma or V&D in the pre and post implementation phases at all sites. Additional costs and outcomes will be synthesized in an incremental cost-effectiveness ratio (ICER) comparing intervention and control sites. Cost-effectiveness analyses (CEAs) will be done to relate ICERs to the primary clinical outcome for each CP.