Association of cancer characteristics with disability status and timing of onset among working age cancer survivors.

BACKGROUND: Numerous studies examine the impact of cancer and disability on healthcare use, financial toxicity, and health outcomes, but not the relationship between cancer and disability. This study investigates whether and how cancer characteristics are associated with disability status (Social Security Disability Insurance) and timing of onset among prime working age cancer survivors. METHODS: We used the 1999-2022 Surveillance, Epidemiology, and End Results file linked with Medicare data. Outcome was a categorical variable indicating no disability, disability onset ≥2 years and 0-<2 years before cancer diagnosis, and 0-<2 years and ≥2 years after diagnosis. Primary independent variables were 16 categories of cancer sites and stage at diagnosis. We employed a multivariable multinomial probit regression model, adjusting for patient and neighborhood characteristics, overall and by 5-year survival. RESULTS: Of the total sample (n = 97,642), 62.2%, 22.4% and 15.4% had no, pre-cancer, and post-cancer disability, respectively. Likelihood of disability and timing of disability onset differed by cancer sites. Compared to prostate cancer survivors (lowest prevalence of disability), adjusted predicted probability of having disability ≥2 years before diagnosis was higher by 0.46 (95% CI 0.43-0.49) among liver cancer survivors while having disability <2 years after diagnosis was higher by 0.29 (95% CI 0.27-0.32) among brain cancer survivors. Nearly all <5-year cancer survivors had disability and the patterns of association varied significantly by 5-year survival. CONCLUSIONS: Clinically assessed disability is strongly associated with specific cancer sites. These findings may point to strategies to improve efficiency of cancer prevention and care, reducing untoward impacts on survival.