Adopting a Safe Systems Approach to Road Safety: Using the Consolidated Framework for Implementa- tion Research to Examine Injury Prevention and Transportation Professionals’ Perceptions of Vision Zero in Five Canadian Municipalities

2.1. Safe Systems Approach (SSA) and Vision Zero

As noted above, a SSA offers a new way of thinking about road safety [22, 39, 40]. Developed by the International Transport Forum (ITF) of the Organization for Economic Co-operation and Development (OECD), the SSA for road safety is characterized by four underlying principles: 1) People make mistakes that can lead to crashes. The transport system needs to accommodate human error and unpredictability; 2) The human body has a known, limited physical ability to tolerate crash forces before harm occurs. The impact forces resulting from a collision must, therefore be limited to prevent fatal or serious injury; 3) Individuals have a responsibility to act with care and within traffic laws. A shared responsibility exists with those who design, build, manage and use roads and vehicles to prevent crashes resulting in serious injury or death and to provide effective post-crash care; and 4) All parts of the system must be strengthened in combination to multiply their effects, and to ensure that road users are still protected if one part of the system fails [41]. Vision Zero is an example of such an approach, “which can be used to guide the selection of strategies and then the setting of goals and targets” [22].

The four SSA principles, which apply to every part of the road system, are operationalized in concert with the five pillars of road safety: 1) management; 2) safer roads and mobility; 3) safer vehicles; 4) safer road users; and 5) post-crash response [42]. Road safety strategies, such as the WHO global plan for the Decade of Action for Road Safety 2011-2020 outline a set of national-level activities for implementing road safety strategies according to these five pillars [5]. However, it is important to note that the more recent global plan for the Second Decade of Action for Road Safety 2021-2030 includes speed management as a sixth pillar:

Managing speed is critical to the effective implementation of the Safe System approach. It is a cross-cutting risky factor and is addressed through actions relating to different elements of the Safe System, such as multimodal transport and land use planning, infrastructure, vehicle design and road user behaviour [3]. This illustrates how Safe System implementation requires “reducing reliance on a single pillar of action” and “encouraging a broad range of interventions” [41].

2.2. Critics of Vision Zero

Despite the appeal and success of Vision Zero, it has received criticism in spheres of academic literature and public debate. For example, focus on individual road users as a cause of collisions (SSA principle 1) and, thus, partially responsible for making changes to their driver behaviour (SSA principle 3) has been challenged; specifically, McIlroy and colleagues argue that “the fallibility of drivers should not be a central theme to a road safety philosophy, and that the blame should not be apportioned to only the road users and system designers” [4]. Alternatively, a multitude of organizations and actors contribute to system outcomes and the interconnections (or lack thereof) are of equal importance when crafting road safety strategies and policies [43, 44]. In addition, Abebe and colleagues outline three types of arguments against Vision Zero: 1) moral arguments, 2) arguments concerning the (goal setting) rationality of Vision Zero, and 3) arguments aimed at the practical implication of the goals [23]. The authors provide robust descriptions and literature illustrating the depth and breadth of these arguments, content which falls outside the scope of this current paper. Thus, below, we provide a brief overview of the three argument types with accompanying examples.

Moral criticisms include claims that it is morally acceptable to die on the road, given that driving is a risky activity in which people voluntarily engage [45]; however, Abebe and colleagues assert, “most of those who are killed and seriously injured in road traffic did not wish to take any risks. They had no other choice than to travel in the risk traffic system that we have” [23]. It has also been argued that some Vision Zero measures restrict the freedom, autonomy, and privacy of road users, thus making it unjust [46, 47]. Critics questioning the rationality of Vision Zero have argued that the goal, “no one shall be killed or seriously injured as a consequence of accidents in road traffic” [32], is unrealistic and, thus, an irrational pursuit resulting in counterproductive investment [48]. Vision Zero has also been criticized as irrational because it does not provide precise or serviceable goals for public policy [49, 50]. Lastly, practical criticisms include the ways in which safety is measured [51, 52], that there is too little responsibility assigned to system designers [24], and that there is too much responsibility assigned to system designers, which “will reduce drivers’ sense of responsibility and make them drive more dangerously” [23].

Despite the range of criticisms, Vision Zero has achieved international recognition [10] and been identified as “an effective policy to prevent road traffic injury by the WHO in 2004” [33]. Thus, increasing our understanding of how Vision Zero is understood and implemented across a variety of contexts is crucial [37, 53-55].

2.3. Vision Zero in the Canadian context

Several cities and provinces across Canada of various sizes, both urban and rural, have adopted Vision Zero [56]; however, road safety efforts began prior to Vision Zero implementation. For example, the Canadian Council of Motor Transport Administrators (CCMTA) introduced Canada’s Road Safety Strategy 2025 designed to continue, “national effort in addressing important road safety issues in Canada by providing a framework for governments and other road safety stakeholders to establish their own road safety plans, objectives, and interventions” [57]. As asserted in the national strategy, “each province and territory will take ownership of its own individual strategy, while cooperating and collaborating with all levels of government, as well as with other stakeholders” [34]. Further, Parachute, Canada’s national charity dedicated to injury prevention, has taken a national lead to “share current research and best practices in road safety, support data-driven models, create and disseminate evidence-based resources, and bridge key multisector players together to increase the overall awareness and effectiveness of the Vision Zero approach” [56]. To date, British Columbia, Manitoba, and Quebec are the only provinces that have adopted Vision Zero, along with 25 cities and three regions [58].

Municipalities included in the scope of this study have adopted Vision Zero as a road safety strategy: Vancouver, Calgary, Peel Region, Toronto, and Montréal (Table 1). Consistent across municipal policies is the understanding that “death and serious injuries are not inevitable. They are preventable” [58]. Further, municipal road safety strategies all contain an iteration of an action plan with emphasis or focus areas explaining how Vision Zero goals will be actioned and achieved (Table 2), where emphasis areas are identified by “types of collisions identified as primary safety concerns through public consultation and review of Regional and local municipal collision data” [59]. While there is some variation across municipal road safety strategies VRUs, such as pedestrians and cyclists, are consistently prioritized, as well as the need for collaboration and partnerships to effectively implement Vision Zero goals [6].

3.1. Data Collection

We utilized interview and focus group data for our study. The Human Participants Research Protocol was approved by York University’s Office of Research Ethics (e2019-174).

The procedures performed were in accordance with the ethical standards of York University (Canada), the Canadian Institutes for Health Research, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants.

Purposeful sampling techniques [68, 69] were employed by a member of the research team in each municipality to recruit participants for KI interviews and VFGs. Research team members responsible for conducing the KI interviews were master’s level graduate students supervised by the second, third, fifth, sixth, and seven authors. There was a mix of male and female graduate students involved. Professional road safety colleagues of research team members were contacted during the first round of recruiting, followed by a snowball sampling technique [70] that asked confirmed participants to circulate the opportunity to contribute to this research to other members of their professional networks. E-mail, Twitter, and LinkedIn were used to contact potential participants. The average response rate was 87%. Those who chose not to participate did not provide a reason. Lastly, saturation did not guide our data collection, as it is not required to yield useful and meaningful results [71].

We conducted KI interviews with 42 participants and five VFGs with a total of 59 participants, all of which were injury prevention and transportation professionals working across several sectors involved in road safety: policy/decision-making, transportation, public health, police services, non-profit, university research, community associations, and private (Table 3). Participants were made aware of the purpose and scope of the research. Only the interviewer and interviewee were present for the interviews. There were no repeat interviews; however, several participants (n=21) participated in a KI interview and VFG to ensure that the VFGs were populated. The KI interviews were conducted in person prior to the VFGs. The focus groups were initially planned in person with different participants, but due to pandemic challenges with recruitment, we reached out to KI interview participants and offered them the opportunity to patriciate in the VFG. The questions guiding the research, from which these data are derived, were informed by the first iteration of the CFIR [72, 73] and asked, what are the barriers and facilitators to BE change at a local level? The design prioritized the perspectives and experiences of injury prevention and transportation professionals because, in the field of health and well-being scholarship, “qualitative research offers rich and compelling insights into the real worlds, experiences, and perspectives of patients and health care professionals” [74].

Semi-structured KI interviews were conducted from January to December 2019 by members of the research team in each municipality (Table 4). A semi-structured interview guide was used by research team members to guide the one-on-one conversations conducted in-person or virtually (i.e., Skype or Zoom). In Calgary, one interview was conducted with two participants and in Peel Region, three interviews were conducted each with two participants to accommodate participants’ time. Interviewers attended virtual group meetings to ensure consistency and discuss the interview guide, which asked participants to describe barriers and facilitators experienced in their road safety work. Interview times ranged from 30-90 minutes. Interviews were audio recorded and then transcribed verbatim by the interviewer. French transcripts from Montréal participants were translated into English and reviewed by the Montréal research team leader to ensure accuracy.

Due to restrictions related to the COVID-19 pandemic, focus groups were conducted virtually from July to November 2020. VFGs were facilitated using a third-party platform called Upwords [75], which supported an online discussion board for participants to respond to interview questions and other participants’ contributions within the span of one week (Monday-Friday). The number of participants per VFG varied by municipality (Table 4). Similar to the KI interview guide, participants were asked to reflect on barriers and facilitators they experienced in their road safety work within their organization and sector. In addition, the VFG interview guide asked specifically about the impact of Vision Zero and how it influenced participants’ work. The change in the interview guide was due to the rise in Vision Zero discussion and uptake among road safety professionals in Canada [9, 57] and was prompted by researchers added to the team after KI interviews were complete. Responses from participants were monitored and follow-up questions were asked by the first author and an Upwords facilitator. During this process, the first author identified themes and made hand-written notes. The VFG platform created a transcript of participants’ conversations. VFG and KI interview participants were given the opportunity to review their transcripts and make changes. No participants asked for changes to be made to their transcript.

3.2. Data Analysis

KI interview and VFG data were analyzed inductively using thematic analysis (TA), which is a flexible and well-suited approach that adapts to the “varying needs of a wide variety of research projects” [74]. The flexibility of TA facilitated the identification of patterns across the data [76]. Using this inductive process, we extracted data that explicitly mentioned Vision Zero, specifically, data pertaining to Vision Zero and how it was perceived by injury prevention and transportation professionals working across sectors in road safety. Further, TA is useful when utilizing the CFIR [18] to analyze qualitative data because it enables clear and meaningful organization of themes and codes across the CFIR domains and constructs.

The first author familiarized themselves with the data by reading and re-reading interview and VFG transcripts. Handwritten notes were made during this stage. Using NVivo (QSR International Pty Ltd, version 12.6.1, 2019), data pertaining to Vision Zero were identified before being coded according to the CFIR. A combination of semantic and latent coding was conducted whereby the first author reviewed each data extract in relation to the CFIR domains and constructs and organized the data accordingly. The first author reviewed and circulated the CFIR-informed codes to the second and last authors before solidifying the domains and constructs under which each data extract was housed. This process is akin to the organizing and solidifying of themes in the TA process [76]. Results were written and organized according to the CFIR domains and constructs, which involved “weaving together the analytic narrative and data extracts” [77].

3.3. Analytical Framework

We utilized the Consolidated Framework for Implementation Research (CFIR) [18] to situate and analyze our findings within the broader landscape of implementation science and road safety. To recall, Vision Zero is a contract between road users and system designers [22, 28, 29] and the implementation of this contract rests on the safe design of the entire road system by designers, road users following the rules for using the road transport system set by designers, and “if the road users fail to obey these rules due to lack of knowledge or acceptance or ability, or if injuries occur, the system designers are required to take necessary further steps to counteract people being killed or seriously injured” [29]. To date, the CFIR has not been used to examine the implementation of Vision Zero as an innovation, nor has it been used to contextualize practitioners’ perceptions of the efficacy of Vision Zero in the Canadian context.

Broadly, implementation science is an emerging field of diverse researchers who seek to understand how to better translate research into practice [20, 53, 78, 79]. Beginning in the health field, implementation science “is the scientific study of methods to promote the systematic uptake of research findings and other evidence-based practices into routine practice and, hence, to improve the quality and effectiveness of health services” [20]. To do so, implementation science moves beyond individual care to also account for the provider, the organizations, systems, and policy levels [20, 79]. Implementation science utilizes a range of models and frameworks [80], such as the CFIR, which offer efficient ways of generalizing findings across diverse settings and contexts [19, 81].

The CFIR is “a practical theory-based guide for systematically assessing potential barriers and facilitators to guide tailoring of implementation strategies and adaptations for the innovation being implemented” [72], and Vision Zero, as a philosophy, contract, and approach, is an “innovative road safety policy” [32]. Thus, we chose to utilize the CFIR to examine the adoption and influence of Vision Zero in the Canadian urban context from the perspectives of injury prevention and transportation professionals. The CFIR provides an entry point into exploring how Vision Zero was adopted and implemented, as well as the impact Vision Zero uptake had on municipalities’ road safety strategies. In other words, the CFIR is useful for contextualizing what our participants’ perceptions reveal about how Vision Zero influences their road safety work and what barriers and facilitators they experience.

Since its initial development [73], the CFIR has become one of the most frequently cited implementation science frameworks [82], consisting of 5 domains – intervention characteristics, individual characteristics, inner setting, outer setting, and process – and 37 operationally defined constructs [72]. As an approach to implementation science, the CFIR “embraces the reality that contextual factors are active,” and there are “dynamic forces working for and against implementation efforts in the real world” [18]. However, more recently, Damschroder and colleagues released an updated CFIR that expands the number of constructs within the five domains to more adequately address equity-related determinants in implementation: 48 constructs and 19 subconstructs across five domains (with one domain including two subdomains) [18]. Changes to the original CFIR framework were informed by a literature review examining uses of the CFIR, as well as a survey of the authors included in the literature review, to “elicit in-depth feedback about their experience using the CFIR” [18].

There are some significant modifications made to the language used in the updated CFIR that increase its utility, making it more flexible and applicable to a variety of innovations. For example, Damschroder and colleagues shift the language from patients in the first iteration to recipients, referring to “individuals intended to benefit from the innovation” [18]. Further, the language of stakeholders has been removed and, instead, refers to “people who have influence and/or power over the outcome of implementation efforts” [18]. This enhanced focus aligns with the municipal road safety plans represented in this study, which prioritizes the safety of VRUs (e.g., pedestrians, cyclists, children, and older adults) in how road systems and safety interventions are designed and implemented [23, 39].

4.1. Innovation

Data collected from participants about Vision Zero mapped onto two constructs within the Innovation domain: 1) Innovation Evidence Base and 2) Innovation Complexity.

4.2. Outer Setting

Participants discussed the contexts in which Vision Zero is implemented and how this influences the implementation process. Data significant to the Outer Setting domain mapped on to three constructs: 1) local attitudes, 2) local conditions, and 3) partnerships and connections.

4.3. Implementation Process

Several participants’ responses mapped onto the Assessing Context construct within the Implementation Process domain, which refers to the phase of implementation that collects information specifically to identify barriers and facilitators. For example, a Peel Region (policy/decision-maker) participant explained, “not all residents understand the impacts of not doing Vision Zero,” while a Toronto (non-profit) participant stated, “sometimes policies/regulations are not in line with transportation equity principles of Vision Zero, so this can limit the types of projects that get approved.” In addition, a Toronto (transportation) participant discussed challenges with recruiting expertise to implement Vision Zero programming: “with Vision Zero and active transportation still being relatively new focus areas, it is difficult to find consultants with the required expertise and experience to roll out these programs.”

Alternatively, some participants discussed facilitators for implementing Vision Zero programming and projects. For example, a Calgary (policy/decision-maker) explained the importance of funding: “with full control over the City’s Transportation department budget, we would be able to fund an implementation plan that more closely aligned with Vision Zero because we could allocate the needed funds to support traffic calming.” This was affirmed by a Vancouver (public health) participant: “we really wanted to be strategic in what we were funding. We didn’t want to just fund pedestrian or cyclist education initiatives, unless, there was a Vision Zero component in that.” In addition, a Toronto (policy/decision-maker) emphasized the importance of political support for Vision Zero:

We have a City Councillor who has been active in supporting the expansion of cycling infrastructure, advancing Vision Zero initiatives, etc. He’s the champion for stuff like strengthening our Vision Zero Program or doing other things to make it safer for pedestrians and cyclists across the city.

This passage illustrates the importance of political will and advocacy in order to successfully implement Vision Zero interventions. Overall, these results illuminate some contextual challenges for implementing Vision Zero.

5.1. Vision Zero and the CFIR

Given that KI interview and VFG data were collected for a larger pan-Canadian project examining barriers and facilitators to BE change, our results highlight meaningful connections between Vision Zero and modifying the BE to improve VRU safety in Canada. Results from the larger study utilized the first iteration of the CFIR [73] and identified cross-sectoral collaboration and data and evidence as significant facilitators for BE changes for VRUs [17]. Those findings align with the results of this current work that map onto the Innovation Evidence Base within the Innovation Domain and Partnerships and Connections within the Outer Setting domain. Utilizing the CFIR as the lens through which to examine these data provides insight helpful for municipalities by highlighting significant contextual factors that influence the adoption and implementation of Vision Zero, such as the importance of making evidence-informed changes (Innovation Evidence Base) and ensuring that the public is educated about the aims of Vision Zero and accompanying BE changes (Local Attitudes).

The updated CFIR [18] uses the language of innovation instead of intervention, aligning more so with Vision Zero as a broad SSA for road safety. It is important to note that, as an innovation and SSA, components of Vision Zero are relevant to all domains and constructs across the CFIR; however, these results highlight the particular perceptions of Vision Zero from the perspectives of professionals working across sectors involved in road safety and, thus, illustrate the most significant implementation factors for Vision Zero from their unique positions. As noted in Section 4.0, most data mapped onto the Outer Setting Domain, illustrating that, for these participants, the specific contexts in which Vision Zero is adopted and attempted are significantly influenced by contextual particularities, thus requiring more municipality-specific knowledge about the nature of resistance and opportunities for support for Vision Zero implementation.

It is also important to note that the updated CFIR includes Human Equality-Centredness as a subconstruct under the Inner Setting domain; specifically, it is housed under the Culture construct, which defines Human Equality-Centredness as “shared values, beliefs, and norms about the inherent worth and value of all human beings” [18]. Given the focus of municipal Vision Zero policies on protecting VRUs, the innovation, as a whole, maps onto this subconstruct. For example, “life and health should never be placed at risk in exchange for other benefits” [59]; however, participants in this study focused more so on the need for local buy-in and understanding of road safety goals and strategies. The remainder of our discussion will more closely examine the relationships between the CFIR constructs onto which our data is mapped and illuminate relevant connections to the SSA.

5.2. Collaboration and Complexity

Several data mapped onto the Partnerships and Connections constructs within the Outer Setting domain, demonstrating how participants viewed collaboration across sectors and partner groups as facilitators to the uptake and implementation of Vision Zero. This aligns with an integral component of a SSA, which is sharing responsibility. As stated by Elvik and Nævestad, “road safety management operates through a partnership model that is adaptive and accountable” [83]. Data also mapped onto the Innovation Complexity construct within the Innovation domain, which is “reflected by its scope and the nature and number of connections and steps” [18], and illustrates how Vision Zero, as a multifaceted systems-based approach, is perceived as complex by participants. According to the Government of Sweden, “cooperation is essential to successful safety improvement” [25], and coordination across sectors involved in health and safety highlights the “nexus between transport and health practitioners and experts who have a responsibility of shaping the system” [26]. Such complexity is also acknowledged in the Montréal Vision Zero plan: “road safety in urban settings is complex and a major challenge that requires continuous collaborative effort by Montréal and its partners” [62]. Further, complexity is identified in the fourth SSA pillar, “all parts of the system must be strengthened in combination to multiply their effects, and to ensure that road users are still protected if one part of the system fails” [41], which also emphasizes the high importance of ensuring effective implementation for the safety of road users.

Parachute, the national lead on Vision Zero in Canada, also stresses the importance of cross-sectoral collaboration as necessary for successful SSA programing [84] and works to “connect key road safety stakeholders to increase the overall awareness and effectiveness of the Vision Zero approach” [9]. Further, Fuselli explains that, in Canada, each level of government has unique road safety responsibilities and that when implementing Vision Zero, “it is important to understand the governmental jurisdictional responsibilities and address each one, taking a collaborative approach to Vision Zero and road safety as a whole” [57]. As previously noted, cross-sectoral collaboration has been identified as a facilitator for BE change in other road safety work in the Canadian context [17]. Our results support and align with the Parachute mission, while also highlighting how injury prevention and transportation professionals perceive the multi-faceted and complex nature of Vision Zero and a SSA. The Montréal 2022-2024 Vision Zero Deaths and Serious Injuries Action Plan also emphasizes the importance of partnerships and connections:

The experience acquired over the recent years has underlined the need to improve cooperation among the partners involved in the Vision Zero effort. A variety of actions could help improve decision-making and implementation processes, including establishing new committees or adding partners to existing committees and working groups [62].

As noted by participants, Peel Region has a dedicated Vision Zero Team representing multiple sectors and stakeholders involved in road safety work. Thus, not only does Vision Zero require that “a well-established interdepartmental and multisectoral partnership administers plans, strategies, and responsibilities” [83], but its adoption at the municipal level can also facilitate increased connection and cooperation in the Canadian context.

5.3. Local Attitudes, Conditions, and Evidence

Participants commented on the importance of local acceptance and understanding of Vision Zero goals, as well as the need for evidence-informed changes to guide Vision Zero interventions and programming. According to Damschroder and colleagues, Local Attitudes refer to “sociocultural values (e.g., shared responsibility in helping residents) and beliefs (e.g., convictions about the worthiness of recipients)” [18], which are influenced by the Local Conditions (e.g., economic, environmental, political, and/or technological). Regarding local acceptance, “cultural practices, health inequities, social influences, emotions, and attitudes all come into play when we choose our modes of travel” [8], and increasing local community understanding of Vision Zero goals is required for successful implementation and acceptance, which is another helpful insight for municipalities interested in adopting Vision Zero.

Given the necessity of partnerships and connections emphasized by participants, the priority of VRUs, and the importance of local acceptance of Vision Zero programs, centering equity in road safety work is pivotal [63, 85]. According to Damschroder and colleagues, it is important for implementation researchers and practitioners to recognize their own positionality (i.e., who we are) in relation to the local communities the innovation is aimed at serving, and the CFIR provides constructs that help users account for these important dimensions. In other words, “being deliberate in collaborating with communities and deeply knowledgeable equity researchers” [18]. Participants in this study did not identify collaboration activities with specific equity-deserving [67] groups; however, the B.C. Road Safety Strategy specifically outlines efforts to engage Indigenous communities in their collaborative road safety work: “we recently expanded our Steering Committee membership to include representation from the Ministry of Indigenous Relations and Reconciliation in recognition of the unique road safety challenges faced by Indigenous communities” [64]. This provides context for the contributions from Vancouver participants who are working not only in a municipal setting that has adopted Vision Zero, but provincially as well.

5.4. Lessons for Municipalities

According to participants, the adoption of Vision Zero has been a facilitator for changes to road transport systems that prioritize the safety of VRUs. In addition, participants’ contributions identify key barriers and facilitators to the implementation of Vision Zero as a road safety innovation (Table 5), which can help municipalities in their plans to adopt and implement Vision Zero as a road safety strategy. Previous research on Vision Zero in the Canadian context [56, 84] is sparse; however, the significance of collaboration and coordination across sectors and partners is consistent. To recall, Vision Zero is an example of a SSA, a systems-level approach to change which necessarily involves multiple organizations, agencies, and communities [3, 6, 32, 83]. Thus, the cultivation and maintenance of partnerships across these entities is crucial for the success of Vision Zero, aligning with the central tenants of the SSA, as well as the contributions from our participants.

In addition, our participants emphasized the importance of local buy-in and knowledge about Vision Zero with regard to successful implementation, another important takeaway for municipalities. Engaging in meaningful dialogue with local communities is an important facilitator [53, 86-89], as well as ensuring that messaging about the Vision Zero approach and associated projects is clear and understood. As noted by our participants, Local Attitudes and Local Conditions can function as both barriers to and facilitators of Vision Zero acceptance. Thus, it is important for municipalities to communicate the goals and intentions of Vision Zero, broadly, and of particular projects, specifically to help ensure local community acceptance and support.