Geographic impact on access to care and survival for non-curative esophagogastric cancer: a population-based study

Abstract

Among patients not undergoing curative-intent therapy for esophagogastric cancer, access to care may vary. We examined the geographic distribution of care delivery and survival and their relationship with distance to cancer centres for non-curative esophagogastric cancer, hypothesising that patients living further from cancer centres have worse outcomes. We conducted a population-based analysis of adults with non-curative esophagogastric cancer from 2005 to 2017 using linked administrative healthcare datasets in Ontario, Canada. Outcomes were medical oncology consultation, receipt of chemotherapy, and overall survival. Using geographic information system analysis, we mapped locations of cancer centres and outcomes across census divisions. Bivariate choropleth maps identified regional outcome discordances. Multivariable regression models assessed the relationship between distance from patient residence to the nearest cancer centre and outcomes, adjusting for demographic, clinical, and socioeconomic factors. Of 10,228 patients surviving a median 5.1 months (IQR: 2.0–12.0), 68.5% had medical oncology consultation and 32.2% received chemotherapy. Certain distances (reference\u2009≤\u200910 km) were associated with lower consultation [relative risk 0.79 (95% CI 0.63–0.97) for\u2009≥\u2009101 km], chemotherapy receipt [relative risk 0.67 (95% CI 0.53–0.85) for\u2009≥\u2009101 km], and overall survival [hazard ratio 1.07 (95% CI 1.02–1.13) for 11–50 km, hazard ratio 1.13 (95% CI 1.04–1.23) for 51–100 km]. A third of patients did not see medical oncology and most did not receive chemotherapy. Outcomes exhibited high geographic variability. Location of residence influenced outcomes, with inferior outcomes at certain distances\u2009>\u200910 km from cancer centres. These findings are important for designing interventions to reduce access disparities for non-curative esophagogastric cancer care.