Mediators of disparities in survival among patients with non-small cell lung cancer

Document Type

Conference Proceeding

Publication Date


Publication Title

J Gen Intern Med


BACKGROUND: Non-small-cell lung cancer (NSCLC) patients who are non-white,have low incomes, Medicaid or no insurance have worse survival than other patients.However, it is unclear how clinical and treatment differences contribute to the survival differences. We assessed whether racial/ethnic or socioeconomic disparities in survival for patients with NSCLC were attributable to sociodemographic factors, clinical characteristics at diagnosis, or treatments received.

METHODS: Data were collected as a part of the Cancer Care Outcomes Research and Surveillance (CanCORS) Consortium, a multi-regional observational study of newly-diagnosed lung and colorectal cancer patients. Among4071 patients diagnosed with incident NSCLC in 2003–2005, we excluded 984patients from one study site with <5 >% non-white patients and 32 without vital status data for a final cohort of 3250 patients. Vital status was followed through 2012. We used unadjusted Cox proportional hazard analyses to estimate risk of death associated with race/ethnicity, annual income, education attainment, and insurance status. We used multivariable proportional hazard models adjusting sequentially for sociodemographic characteristics (age, race/ethnicity, sex, marital status, income, education, insurance,smoking status, integrated health system); clinical characteristics at diagnosis (stage, number of comorbid conditions); and treatments received (radiation, surgery, chemotherapy). All models adjusted standard errors for clustering by study site.

RESULTS: Overall, 65 % of patients were white, 16% black, 7% Hispanic, 7% Asian, and 5% other race/ethnicity; 36% of patients had incomes <$20,000/year; 23 % of patients had not completed high school; 2% were uninsured and 4% had Medicaid insurance. The median survival time was494 days after diagnosis. In unadjusted analyses, mortality was higher for patients with lower income, less education, and no insurance, Medicaid, or Medicare versus private insurance (Table, all P$60,000/year,non-high school graduates vs. some college or more, and Medicaid or Medicare only vs. private insurance (Table). After adjustment for clinical characteristics (stage and comorbidity), the survival disadvantage for His-panic ethnicity and education were no longer evident, although income<$40,000/year, uninsurance, and Medicaid insurance remained associated with worse survival. Additional adjustment for treatment explained much of the uninsurance difference, but a survival disadvantage among patients with incomes <$20,000/year and those with Medicaid insurance persisted.After adjustment for treatment, black and Asian patients had better survival than whites (Table).

CONCLUSIONS:Survival disparities in NSCLC exist for patients with low income and Medicaid insurance, even after adjustment for sociodemographic,clinical, and treatment factors. Clinical characteristics at diagnosis account formost of the survival disparity by education, but not income. Differences in treatments received explained much of the survival disparity between uninsured and insured patients. Additional efforts are needed to assure timely diagnosis and use of effective treatment as well as improved post-treatment care to lessen these disparities.




2 Suppl

First Page


Last Page


This document is currently not available here.