Health care system factors associated with receipt of treatment and treatment intent in stage III non-small cell lung cancer: A population-based study in Ontario

Abstract

PurposeStage III non-small cell lung cancer (NSCLC) is a heterogeneous disease, with a spectrum of anatomic extent, health status, and treatment approaches. Receipt of treatment and its intent should be independent of health system factors where care quality is optimal. We investigated the degree that modifiable health system factors are associated with receipt of treatment and treatment intent in stage III NSCLC in a large, universal health system. MethodsThis was a population-based, retrospective cohort study with health administrative data from Ontario, Canada, 2010–2018 for those aged ≥ 20 years, with AJCC 7 or 8 stage III NSCLC. We explored system factors associated with NSCLC treatment: region of residence, diagnostic interval, travel distance, advanced radiation (e.g. IMRT, VMAT) and systemic therapy treatment volumes and year of treatment. The relative risk (RR) of (1) any treatment versus no treatment, and (2) palliative versus non-palliative treatment was determined, using multivariable stepwise Poisson regression models. We adjusted for patient, disease and treatment factors. ResultsWe identified 7,093 people with stage III NSCLC between 2010 and 2018. There were no system factors associated with receipt of treatment versus no treatment in adjusted analysis. The major system factor associated with palliative intent was region of residence (RR: Region ranges from 0.88 to 1.67, p < 0.001). Stratifying by era (2010–2012 vs. 2013–2015 vs. 2016–2018), there was an increase in receipt of curative treatment and use of advanced radiotherapy techniques and immunotherapy over time, but regional variation of treatment intent was similar. ConclusionsRegion of residence emerged as the major health system factor associated with treatment intent for stage III NSCLC. This variation remained, even as advances in radiotherapy and systemic therapy were adopted. Our study suggests possible opportunities to improve care outcomes by addressing unexplained regional variation in care.