Using community data to inform medical school admissions: a cohort study of medical students in Ottawa, Canada

Sierra A. Land¹, Jordyn N. Linders², Kady Carr³, Bradley MacCosham⁴, Geneviève Lemay⁵, and Claire E. Kendall⁶

⁴Senior Manager, Accreditation and Program Evaluation Faculty of Medicine University of Ottawa Ottawa, ON, Canada

⁵Assistant Dean of Admissions Associate Professor of Medicine Faculty of Medicine University of Ottawa / Institut Savoir Montfort Ottawa, ON, Canada

⁶Associate Dean, Social Accountability Professor, Department of Family Medicine Faculty of Medicine, University of Ottawa Scientist, Bruyère Research Institute Ottawa, ON, Canada

Date submitted: 23-June-2025

Email: Sierra A. Land (sland040@uottawa.ca)

This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-Non Commercial-Share Alike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Citation: Land S, Linders J, Carr K, MacCosham B, Lemay G, and Kendall C. Using community data to inform medical school admissions: a cohort study of medical students in Ottawa, Canada. Educ Health 2025;38:361-366

Online access: www.educationforhealthjournal.org
DOI: 10.62694/efh.2025.378

Published by The Network: Towards Unity for Health

Background

Equitable health care relies on a diverse physician workforce, yet medical school admissions often fail to reflect the diversity of the populations they serve.¹ Research shows that physicians often prefer to provide care to populations with whom they share similar life experiences, and there is evidence that this concordance fosters comfort and familiarity and can improve population health outcomes.^2–6 A systematic review of studies of sociodemographic and training factors and their association with practice patterns in the United States further suggests that recruitment of students from underserved areas could be an effective strategy for increasing the physician workforce in those regions.⁷

Recognizing these challenges, the Association of Faculties of Medicine of Canada emphasizes the importance of diversifying medical school admissions to better reflect the Canadian populace.⁸ This approach also aligns with Canadian medical schools' social accountability mandates, which aim to recruit students from culturally diverse backgrounds to better represent the populations within their service regions.^9,10 However, there is limited research examining how these policies contribute to representative medical student cohorts.¹¹

To address this gap, The University of Ottawa (uOttawa) Faculty of Medicine established a Regional Profiles Working Group, composed of faculty, community members and students, to advance social accountability in our admissions processes. This study will analyze the sociodemographic profiles of first-year students admitted to uOttawa's medical doctorate (MD) program, comparing them to those of the Ottawa region. By identifying disparities, the study aims to inform admissions strategies that enhance equity and representation within the physician workforce.

Methods

Study Design

We conducted a cohort study of all first-year MD students admitted to the uOttawa Faculty of Medicine during the 2023 application cycle. uOttawa’s bilingual undergraduate medical education (UGME) program includes parallel streams in English and French. We identified our catchment area, also known as our primary region, based on provincially-defined boundaries for the Champlain Local Integrated Health Network, which until 2021 had a mandate to plan, fund and coordinate health services in our region. This study met research ethics board exemption criteria.

Data sources

We obtained survey data from Ontario Medical Student Applicant Service (OMSAS). The survey is mandatory for all Ontario medical school applicants and includes questions pertaining to first language, equity, diversity and inclusion (EDI) factors, and the postal code of residence during high school if the applicant completed secondary education in Canada. OMSAS participants provide informed consent during the application process, permitting the collection and institutional use of their personal information in compliance with privacy regulations. Applicants had the option to select “prefer not to answer” for all survey questions. uOttawa received these data from OMSAS in aggregate form, ensuring individual identities could not be reconstructed.

For regional comparison, we used data from Statistics Canada’s 2021 Census of Population, held at the Ottawa Neighbourhood Study (ONS). ONS is a uOttawa research initiative established in 2006 that profiles natural neighbourhoods in Ottawa by collecting, mapping, and analyzing neighbourhood-level data on the social determinants of health and health outcomes.¹²

Variables

We selected variables based on the OMSAS questions and their respective response categories, including: first official language spoken, gender identity, race, generation status, education of parent(s) or guardian(s), occupation of parent(s) or guardian(s), parents’ or guardians’ combined gross income, disability status, and sexual orientation. Some OMSAS variables included a “prefer not to answer” option, and some were not mutually exclusive, meaning that respondents could indicate more than one response.

For regional comparisons, we aligned these variables with data from the ONS for our entire priority region by deriving indicators that best matched the OMSAS response categories. We collapsed variables if required for meaningful comparisons, and when required to avoid reporting six individuals or less. Due to limitations in data availability, we were not able to directly compare some OMSAS variables to our priority region. For example, we used national-level data to estimate comparisons for sexual orientation.

We used descriptive statistics (counts and proportions) to describe and compare our cohort with the regional population. As population estimates varied, we did not report denominators for regional comparisons. Due to the study's focus on descriptive comparison and the absence of inferential statistical testing for hypothesis generation, statistical tests were not conducted. Instead, observed differences between the cohort and regional population were described narratively. We created tornado charts to display side-by-side comparisons, highlighting key disparities in sociodemographic factors. These visualizations provided an accessible and intuitive representation of the observed differences.

Results

There were 183 MD students in our cohort, all of whom responded to the OMSAS survey. A lower percentage of students reported English as their first official language compared to regional residents (61.7% vs 80.8%). While a similar proportion of students and regional residents spoke French as their first language (18.6% vs 16.0%), a far greater percentage of students reported a first language other than English or French (19.7% vs 1.1%).

We found differences in the self-reported race of our student cohort compared to our region’s residents. The student cohort had higher proportions of East Asian (14.2% vs 3.9%), Middle Eastern (22.4% vs 5.2%), and South Asian (16.4% vs 4.5%) racial identities, compared to the region. There were lower proportions of Black students compared to our region’s residents (4.4% vs 6.4%). Further, our proportion of Indigenous students was slightly greater than the general population (4.4% vs 3.4%), however we acknowledge that general population statistics regarding Indigenous status is underreported in census data.¹³, ¹⁴ While the proportions were similar for first generation status of our cohort and the region (26.8% vs 23.7%), our cohort included a much higher percentage of second-generation (43.7% vs 17.9%) and third-generation (24.6% vs 58.4%) individuals, compared to our region.

Sexual orientation among MD students closely reflected national estimates, with 12.0% of students identifying as 2SLGBTQIA+, 76.0% as straight/heterosexual, and 12.0% preferring not to answer. Similarly, gender identity among medical students was reflective of the general population within the city of Ottawa, with the exception of a lower reported proportion of cis men within the medical school cohort (38.8% vs 49.7%). A much lower proportion of students self-reported living with a disability compared to the Ontario population (4.3% vs 28.0%).

We found overrepresentation of students with parents having completed higher education compared to the region, including completion of a bachelor’s degree (51.4% vs 25.0%), masters (29.0% vs 10.9%), doctorate (8.7% vs 1.9%) and professional degrees (29.0% vs 0.9%). Parental occupations also differed from the Ottawa region, as parents or guardians of students were overrepresented in business, finance, and administrative (27.3% vs. 18.7%); education, law, social, and community service (30.1% vs. 16.5%); and health occupations (15.8% vs. 7.5%). Conversely, they were underrepresented in sales and service occupations (12.6% vs. 21.2%) compared to the Ottawa region. Our students were also disproportionately likely to come from the highest income households (32.8% vs 13.2%) and less likely to come from the lowest income households (5.8% vs 15.3%).

Discussion

This study comparing uOttawa MD students and the residents in the medical school priority region highlights potential disparities across several sociodemographic factors. Our bilingual program with parallel English and French streams successfully facilitated a representative Francophone cohort, aligning with our mandate. Further, our student cohort had greater diversity in terms of non-official first languages, second and third generation status, and non-white self-reported race. Despite having specific admissions streams for Black and Indigenous students, we found that Black students were underrepresented, and Indigenous identities were likely underrepresented, given they are already known to be underreported in census data and in the current physician workforce, as were people reporting disabilities.¹³ Trends related to Black and Indigenous groups prevailed despite concerted efforts to recruit via community consultation, representation of Black and Indigenous individuals in admissions committees and interview panels, and targeted mentorship programs in partnership with the Indigenous Program Office and the Black Medical Student Association. We found an overrepresentation of students from more educationally and economically privileged backgrounds, particularly those with parents in education and health professions.

The greater racial diversity in our cohort compared to the regional population was most pronounced among East Asian, Middle Eastern and South Asian groups. This diversity was also reflected in a greater number of students with second and third generation status and may reflect general EDI efforts in the admissions process or a greater interest in medical careers among these communities.¹⁵, ¹⁶ However, underrepresentation of Black and Indigenous students likely arises from persistent systemic educational disparities, disproportionate socioeconomic challenges and a persistent lack of mentorship, which limit the effectiveness of targeted admissions initiatives.^17–19 These findings align with existing literature suggesting that students from higher socioeconomic backgrounds are more likely to pursue medical education due to financial resources and educational support they receive.¹⁷ To address these barriers, some medical schools have implemented initiatives aimed at increasing accessibility for these students. For example, the University of Toronto’s Community of Support program provides mentorship, financial guidance and academic support to those from underrepresented backgrounds, including Indigenous, Black and economically disadvantaged students.²⁰

The proportion of 2SLGBTQIA+ students aligns with national estimates, though the "prefer not to answer" category in the OMSAS survey may obscure true representation, possibly due to discomfort or privacy concerns in disclosing sexual orientation.²¹ Gender identity of students was largely proportional to that of the City of Ottawa, although there were fewer reported cis men and a proportion of students opted out of answering. Self-reported disability status within our cohort was significantly lower than that of Ontario, which likely reflects differences in average age between the populations, however this could indicate the need to increase accommodation during admissions processes.²², ²³

Limitations

There are several potential limitations to our findings. First, OMSAS relies on self-reported data which may be subject to biases, including 35% of students choosing “I do not know” or “Prefer not to answer” for parental income and 8.9% selecting “Prefer not to answer” for disability status, possibly due to concerns regarding discrimination in the admissions process. Second, comparisons to regional sociodemographics required nuanced synthesis and extrapolation from provincial or national data, limiting cohort comparability, especially for non-mutually exclusive responses. Finally, our data is from a single cohort at one medical school, which may make it less generalizable to other settings.

Conclusions

This study underscores the influence of socioeconomic and cultural factors on medical school admissions and the need for diversity initiatives and further outreach and recruitment to underserved groups to enhance equity, diversity, inclusion, and accessibility in medical education. Our methodology can serve as an approach for other medical schools who want to use regional data as a benchmark for representation with the aim of graduating a representative cohort of physicians who will serve their communities. Further, to progress from a state of responsiveness to community needs and shift towards responsibility and further accountability, we need a comprehensive approach to obtain additional qualitative data directly from the community to fully grasp the underlying causation of trends in student admissions profiles.²⁴, ²⁵

2. Heng D, Pong RW, Chan BTB, Degani N, Crichton T, Goertzen J, McCready W, Tannenbaum D, Brown JB. Graduates of northern Ontario family medicine residency programs practise where they train. Canadian Journal of Rural Medicine. 2007;12(3):146–153
PubMed

3. Laven G, Wilkinson D. Rural doctors and rural backgrounds: how strong is the evidence? A systematic review. Australian Journal of Rural Health. 2003;11(6):277–284
Crossref PubMed

4. Rabinowitz HK, Diamond JJ, Markham FW, Santana AJ. The relationship between entering medical students' backgrounds and career plans and their rural practice outcomes three decades later. Academic Medicine. 2012;87(4):493–497. https://doi.org/10.1097/ACM.0b013e318248e698
Crossref

5. Ray RA, Young L, Lindsay DB. The influences of background on beginning medical students' perceptions of rural medical practice. BMC Medical Education. 2015;15:58. https://doi.org/10.1186/s12909-015-0339-4
Crossref

6. Snyder JE, Upton RD, Harewood R, Holmes GM. Black representation in the primary care physician workforce and its association with population life expectancy and mortality rates in the US. JAMA Network Open. 2023;6(1):e2253732. https://doi.org/10.1001/jamanetworkopen.2022.53732
Crossref

7. Goodfellow A, Ulloa JG, Dowling PT, Talamantes E, Chheda S, Bone C, Moreno G. Predictors of primary care physician practice location in underserved urban or rural areas in the United States: a systematic literature review. Academic Medicine. 2016;91(9):1313–1321. https://doi.org/10.1097/ACM.0000000000001203
Crossref PubMed PMC

8. Committee on Accreditation of Canadian Medical Schools (CACMS). Data Collection Instrument (DCI): For schools with visits in 2025–2026. Available from https://cacms-cafmc.ca. Retrieved June 15, 2025

12. Ottawa Neighbourhood Study (ONS). Welcome to Ottawa Neighbourhood Study. Available from https://www.neighbourhoodstudy.ca/. Retrieved June 15, 2025

13. Smylie J, Firestone M. Back to the basics: identifying and addressing underlying challenges in achieving high quality and relevant health statistics for Indigenous populations in Canada. Statistical Journal of the International Association for Official Statistics. 2015;31(1):67–87. https://doi.org/10.3233/SJI-150864
Crossref

15. Nguemeni Tiako MJ, Ray V, South EC. Medical schools as racialized organizations: how race-neutral structures sustain racial inequality in medical education-a narrative review. Journal of General Internal Medicine. 2022;37(9):2259–2266. https://doi.org/10.1007/s11606-022-07500-w
Crossref PubMed PMC

17. Griffin B, Hu W. The influence of socioeconomic status on medical school admissions: a qualitative study. BMC Medical Education. 2019;19(1):1–8. https://doi.org/10.1186/s12909-019-1732-0

18. Henry F, Dua E, Kobayashi A, James C, Li P, Ramos H, Smith M. The Equity Myth: Racialization and Indigeneity at Canadian Universities. Vancouver: UBC Press; 2022

19. Curtis E, Wikaire E, Jiang Y, McMillan L, Loto R, Fonua S, Reid P. The academic pipeline and minority faculty representation at medical schools. Academic Medicine. 2022;97(2):190–197. https://doi.org/10.1097/ACM.0000000000004465

22. Gertsman S, Dini Y, Wilton D, Neilson S. Tackling barriers in Canadian medical school admissions for students with disabilities. Canadian Medical Association Journal. 2023;195(44):E1512-E1516. https://doi.org/10.1503/cmaj.230734
Crossref PubMed PMC

24. Boelen C, Gibbs T, Dharamsi S. The social accountability of medical schools and its indicators. Education for Health; Abingdon. 2012; 25(3), 180–194. https://doi.org/10.4103/1357-6283.109785
Crossref

Original Research Paper