EDI evaluation criteria

All Canadian PGME websites were evaluated based on EDI criteria adapted from the Diversity Audit Tool (DAT) [22]. Initially created by the Ryerson University Diversity Institute and the Canadian Advanced Alliance Women in Technology to boost women’s participation in the information and communications technology sector, the DAT was later refined to include education and healthcare sectors [22]. The DAT examines how organizations leverage EDI across their value chain through 58 dichotomous assessment criteria. Compared to six other tools, including the Canadian Bar association equity and diversity guide, and the chamber of minerals and energy Australia gender diversity audit tool, the DAT distinguishes itself for its comprehensiveness and macro-level approach, enhancing its adaptability across various industries and sectors [22].

When relevant, assessment criteria language was adjusted to align with the study context. For instance, the inquiry "Has the business case for diversity been developed and widely communicated?" was refined to "Are the benefits of diversity communicated anywhere on the PGME website?" Similarly, elements of the DAT that could not be fairly assessed on webpages, such as "Does succession planning take into account diversity targets?" were excluded. The reasons for exclusion, detailed in S1 Table, include assessment challenges, limited information disclosure, and lack of applicability. These exclusions were considered in the broader context of literature reviews on program discipline EDI reviews, ensuring relevance to medical education and the Canadian landscape [16–20, 23]. The 20 retained EDI criteria were subsequently grouped into five distinct themes, as listed in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Description of EDI criteria assessed on PGME webpages.

https://doi.org/10.1371/journal.pone.0307584.t001

Each criteria was assigned one point. For inclusion, information had to be available directly on the PGME website, through hyperlinks on the landing page, or in publicly accessible linked documents (S2 Table). Two investigators (MSB and MB) independently evaluated the presence or absence of EDI criteria, and discrepancies were resolved by consensus. French websites were evaluated by a bilingual reviewer (LM), then electronically translated to English for evaluation by a second reviewer (MSB).

Subgroup analysis

Subgroup analysis comparing EDI scores based on geographic location and municipal population diversity was conducted. For geographical location, programs were grouped into six regions based on Statistics Canada standardized segmentation: Atlantic (Nova Scotia, Newfoundland and Labrador); Quebec; Ontario; Prairies (Saskatchewan and Manitoba); and Western Provinces (Alberta and British Columbia) [24]. Municipal population diversity was assessed using the 2022 Canadian Census data on ethnocultural diversity of each program’s respective metropolitan area, based on their listed corresponding address [25].

Statistical analysis

Descriptive statistics were used to report website EDI scores. Kruskal-Wallis tests were conducted to assess for differences in mean EDI scores based on municipal population diversity and geographic location segmentation. Post hoc tests for pairwise comparisons were performed using Dunn’s post-hoc test with Bonferroni correction. Statistical analyses were performed using SPSS Statistics version 28 (IBM Corporation). P values less than 0.05 were considered statistically significant.

Leadership & governance

While PGME programs have integrated EDI into their mission statements and developed robust harassment/discrimination policies, there is room for improvement, particularly in securing written commitments to EDI from senior leadership and establishing dedicated roles focused on EDI. Additionally, less than half of the programs communicate explicit diversity goals and policies. Commitment to EDI, along with actionable steps, are essential for driving organizational change, and measuring performance is critical for assessing the effectiveness of interventions [26, 27]. Therefore, we suggest that programs include explicit EDI goals in their strategic planning reports and share both qualitative and quantitative measures of progress toward these goals.

Transparent recruitment practices

Despite most PGME websites advocating for bias-free interviewing, there remains gaps in targeting underrepresented groups in recruitment efforts and a limited call for diverse representation on interviewing committees. Research demonstrates biases throughout application processes, which include linguistic differences in personal statements based on gender and ethnic groups, to interviewers favoring candidates who resemble them in appearance and experiences [28–30]. Additionally, the transition to virtual residency interviews may exacerbate bias, disproportionately impacting underrepresented groups [31].

Some individual programs have implemented equity-driven recruitment and selection processes, such as those for Indigenous applicants in various provinces or for Black applicants to internal medicine at the University of Toronto. However, these efforts remain limited and are often restricted to specific specialties [32, 33], and are fewer in number than those in place for medical school admissions [5–7]. PGME programs should encourage all specialties to adopt key strategies for recruitment, emphasizing explicit targeting of underrepresented groups and integrating diverse representation on recruitment panels in accordance with Governmental best practices [34]. Programs may benefit from consulting the University of Calgary’s "PGME Anti-Racism Task Force Guidance for Selection Interviews" for an example of implementation of these principles [35].

Program accommodations

All programs mentioned access to counselling and mental health services, a crucial support given the high rates of burnout among medical trainees [36]. According to a 2021 survey by the Canadian Medical Association, burnout rates among medical trainees reached as high as 53% [37]. While most programs had a flexible work policy, the specific circumstances for granting part-time equivalent positions were often unclear. To provide context, an estimated 15% of trainees under the British Medical Association (BMA) are working less than full time equivalent, while only 5.1% of Canadian trainees reported working less than 50 hours per week [36, 38]. Therefore, Canadian PGME and trainee unions may benefit from implementing policies developed by BMA for guidance on flexible work arrangements [38].

Furthermore, despite parents making up about 25% of trainees [39], only a single institution mentioned having emergency daycare onsite. Research suggests that increased flexibility may decrease work-to-family conflict [40], yet only two institutions had existing family-friendly policies, including the provision of family emergency days.

Community engagement

Community engagement is a cornerstone of medical institutions’ commitment to social accountability [41]. Research has highlighted the profound influence of behavioural, social, and environmental factors on population health [42, 43]. Community engagement enables the creation of effective clinical and community programs and serves as an integral part of allyship by connecting with diverse communities and promoting inclusivity beyond institutional boundaries [44–47]. Despite these benefits, our analysis revealed that most PGME program websites lack publicly available EDI-promoting newsletters and communication regarding diversity-centered community engagement activities. However, internal newsletters with access restrictions requiring institutional login limited the extent to which community engagement efforts could be evaluated.

Pathways to entry

While most PGME program websites explicitly mentioned transitional or re-entry programs, detailed qualification requirements and related application procedures were rarely addressed within available policies. Only two PGME websites highlighted their engagement with associations and professional organizations to promote diversity. To strengthen diversity initiatives, PGME may consider developing additional residency pathways, such as the Black applicant pathway to internal medicine at the University of Toronto [32]. These pathways can be implemented without changes to the existing application process administered through the Canadian Resident Matching System (CaRMS). For instance, in Toronto’s internal medicine Black and Indigenous student pathways, candidates self-identify their personal letter and are encouraged to reflect on how their identity influences their view of medicine and career goals [32, 33]. Additionally, establishing networking opportunities for visiting students from underrepresented groups may be beneficial [48].

Limitations

Our study sample size was constrained by the number of Canadian Postgraduate Medical Education Offices, although it is representative of all medical schools in Canada. Data collection was restricted to publicly accessible websites, excluding alternative sources such as private forums, emails, or direct interaction with trainees or staff which prospective residents might utilize. Website assessments were conducted with a degree of subjectivity, employing a binary approach to evaluate the presence or absence of content. A comprehensive analysis and critique of available content fell beyond the scope of this study. This study focused on assessing PGME at an overarching level and did not delve into individual program activities, which may vary from the presented findings. It is important to note that more nuanced information may be available on individual residency program websites.