Using evidence as part of the decision making pathway: A resource to assist practitioners in road safety

Author(s): Richmond, Buchan, Pitt, Pike, Fuselli, Fridman, Macpherson

Slidedeck Presentation:

1A Richmond



Injuries and deaths from involvement in a motor vehicle collision is a significant public health issue. As public health researchers and practitioners, we need to support the work of municipalities by advocating for implementation of effective interventions to reduce this burden. This requires an evidence-based approach; however, there are many interventions embedded in existing road safety policies in Canada that do not have evidence to support their effectiveness. In addition, there is significant variability in effectiveness as well as the quality of reporting of these interventions. This poses significant challenges in decision making for practitioners and policy makers.


The objective of this project was to create a tool that summarizes the evidence on the effectiveness of built environment (BE) interventions, commonly found in road safety policies in large, urban Canadian centres.

Target Group:

The information in this tool can be used to support efforts in making evidence-based decisions for road safety. This tool can be particularly useful when municipalities are considering making changes to the built environment to reduce the number and severity of motor vehicle collisions (MVCs), particularly in vulnerable road users such as pedestrians and bicyclists.


Evidence for this tool was initially retrieved using an environmental scan of road safety policies implemented across five, large urban municipalities in Canada (i.e., Vancouver, Toronto, Peel Region, Calgary and Montreal). Articles were retrieved from citations within the policies themselves, a literature search, and reference to a recently completed systematic review (unpublished). Two independent reviewers completed a literature search that included peer-reviewed articles indexed in PubMed and MEDLINE as well as a grey search in Google Scholar.
Inclusion criteria were those articles that examined a BE intervention, included a collision outcome (e.g., MVCs, pedestrian MVCs, cyclist MVCs) or collision pathway outcome (e.g., vehicle speed, vehicle volume), and included all study designs that evaluated the effectiveness of the intervention (e.g., quasi-experimental, pre-post, time series). Exclusion criteria included studies that were not published in English, non-BE interventions (e.g., enforcement and education interventions), studies with outcomes other than MVCs or pathway outcomes (e.g., modal choice, physical activity), study designs that did not examine the effectiveness of the intervention (e.g., implementation/planning guides, risk factor studies), studies that did not complete rigorous statistical analysis, or studies published more than 20 years ago. For bicycle-related interventions, all study designs were included.
A critical appraisal of all studies included in this tool was completed. A review of all interventions, specific to its study design was completed and each intervention study was scored and divided into three groups: Strong, Moderate, or Weak.
Finally, we completed 18 key informant (KI) interviews with stakeholders (i.e., transportation, police services, school boards, developers, urban planners, public health, policy makers, community groups) to inform the development of the tool. Questions for the KIs included what information would be useful to present as well as how the tool should look.


A web-based tool that summarizes the information on the effectiveness of interventions found in the road safety policies was created. The BE interventions were grouped into three major intervention types: sidewalks and streets, intersections and crossings, and bicycle-related. Each intervention includes a description of what the intervention is, the types of study designs that were found from the environmental scan and literature review, the number of studies that were included in the summary and results of the critical appraisal. For each intervention type, there is a summary of the outcomes presented in each study. We extracted data from each study by intervention, and summarized the results accordingly. Locations in which the included studies took place are also provided for each built environment intervention. Key to the development of this tool was the information collected from the KI interviews. This increased the relevance and usefulness of the tool to our target audience.