How geography affects end-of-life continuity of care

The Geography and End-of-Life Continuity of Care project, supported by Michelle Howard’s CIHR funded research program in end of life continuity of care. This project is led by Shuaib Hafid, Statistical Analyst and part-time PhD student. The study looks at whether how far people live from their family doctor affects their continuity and health care use in the last year of life. More details about how the study was designed and analyzed are available in our published protocol in PLOS ONE (Hafid et al., 2025; DOI: 10.1371/journal.pone.0336790).

To measure how close people live to their doctor, we used mapping software (ArcGIS Network Analyst) to calculate the shortest driving distance and travel time, as well as distances by transit and walking. We also calculated straight-line (Euclidean) distance for comparison. We then looked out how these distances were related to different measures of continuity of care (e.g., the Usual Provider of Care index, the modified Bice-Boxerman index, and the Relative Variance Index).

We found that people who live farther from their family doctor tend to have lower relational continuity. For example, the Usual Provider of Care (UPC) index was 0.35 for those living within 2 km of their doctor, but dropped to 0.28 for those living more than 14 km away. People who lived farther away also had fewer family doctor visits. Team-based continuity (measured by the modified Bice-Boxerman index) stayed about the same regardless of distance. After adjusting for other factors, our statistical models showed that greater distance was consistently linked to lower continuity.

The next phase of our work will look at whether distance from a family doctor also affects how often people use acute care services near the end of life.

Project Lead: Shuaib Hafid, Michelle Howard
Contact: Shuaib Hafid (hafids3@mcmaster.ca)