| In this issue, Young et al. present a six-point framework for pipeline and program development, intended to increase diversity in Canadian medical schools [1]. Using a framework of knowledge translation, the authors emphasize the key role of collecting and monitoring longitudinal diversity-related data in designing, implementing and evaluating diversity-related initiatives. Young et al. argue the use of such data will not only improve best practice, but will also help build more tailored structures for lasting change. We wholeheartedly agree that such data are essential. However, Young et al.’s focus on processes, new initiatives and structural change positions the article at the level of organizational and systems-level change. This is potentially problematic as, while systems-level changes may be an important initial step towards reducing functional barriers (to, in this case, increasing diversity within medicine), unless these barriers are also addressed at a cultural level, it is unlikely change will be effective, lasting or genuine [2, 3]. Institutional culture is loosely understood to be the shared assumptions, meanings, beliefs, understandings and ideas held within an organization, school or team – with a focus on values and judgements, rather than procedures and practices [4]. In established institutions, culture often errs towards stability and the status quo, largely allowing people to stay within their comfort zones and use established approaches rather than challenging these with innovation and growth [5]. Furthermore, medicine, and selection into medical school, are influenced by a pervasive culture based on meritocracy, where the prior academic achievements of applicants may be over-emphasized. Therefore, although an admissions system may be perceived to be effective by those in the institution, (for example, because it processes large numbers of applications efficiently or admits students who historically have low dropout rates), it may not be conducive to achieving ‘newer’ goals, such as those of increased diversity. For example, and drawing on data from medical schools represented in the Young et al. study, Razack and colleagues [6–8] identified a potential conflict between medical schools’ definition of excellence and a culture that would permit goals for increased diversity (referred to here as widening access) to be met. As a group, medical schools’ discourses reflected that they valued ‘excellence in scholarship’ above other, potentially broader and more inclusive, notions of excellence [8]. Similarly, Alexander et al.’s [9] recent discourse analysis of UK medical school websites identified that, while the discourse of widening access for social mobility through academic meritocracy was dominant, a counter discourse for improving the workforce through increased diversity was marginalized. In neither study was widening access displayed as a strength, implying that while institutions may, at one level, acknowledge the need to widen access, and claim to put systems in place which support this, these systems may not be accompanied by a cultural shift towards truly embracing the value of diversity in medicine. Admissions practices, policies and their institutional interpretations may instead act as ‘filters’ [10] to intercept, moderate and even halt the effective implementation of widening access. |