Abstract
OBJECTIVES: Language-concordant care, or healthcare in one's preferred language, is important both for health equity and for improving health outcomes. Linguistic minorities, like Francophones in Ontario, Canada, are at risk of poorer clinical outcomes if they receive non-language-concordant primary care. However, common ratio-based access measures can provide misleading views of minorities' actual access levels. This cross-sectional geospatial study demonstrates a new way to measure primary care access using average travel time to the nearest five English- and French-speaking family physicians. We also introduce the concept of primary care access fragility, where a region's primary care access may depend on one or a few local family physicians. Our research question is: are there differences in travel burden and access fragility for census subdivisions (CSDs) across language (English/French), rurality (urban/rural) and region (north/south) in the province of Ontario, Canada? DESIGN: We conducted a cross-sectional geospatial analysis to estimate English-language and French-language primary care travel burdens and access fragility in Ontario, Canada. We used population and boundary data from Statistics Canada's 2021 census, road-network data from OpenStreetMaps, and family physician practice locations and language abilities from the College of Physicians and Surgeons of Ontario. We measured travel burden using Valhalla, an open-source road-network analysis platform. SETTING: We conducted our analysis for Ontario, Canada's 577 CSDs, which correspond roughly to municipalities and with populations ranging from 5 inhabitants in Rainy Lake 17B to a high of 2 794 356 in Toronto. PARTICIPANTS: Using public data from January 2026, we identified 15 762 family physicians practising in Ontario, of whom 11.0% reported speaking French. Patient data were obtained from the most recent 2021 census. PRIMARY AND SECONDARY OUTCOME MEASURES: Our first primary outcome measures were CSD-level mean travel time to the nearest five English-speaking family physicians, and CSD-level mean travel time to the nearest five French-speaking family physicians, which we compared to explore regional inequities in travel burden. Our secondary outcome measures were based on a novel notion of the travel burden component of 'primary care access fragility'. This metric indicates how dependent a region's access is on a small number of local physicians and is defined as the difference between the CSD-level mean travel time to the nearest one physician and to the nearest five physicians. As the difference in travel times grows, so too does access fragility. RESULTS: Median differences in French-language and English-language travel burdens were strongly significant across rurality, regions and overall (median difference 13.4 min, p<0.001, IQR 4.9-29.3 min), as was primary care access fragility (median difference 5.2 min, p<0.001, IQR 0.8-18.4), with larger travel burdens and fragilities in northern and rural communities. CONCLUSIONS: Compared with the general public, Ontario's French-speakers face higher travel burdens to language-concordant family physicians and higher access fragility, especially in rural and northern regions. Our results are of interest to policymakers and health-system planners, and our methods are applicable to other populations and regions.