In lung cancer patients, age, race-ethnicity, gender and smoking predict adverse comorbidity, which in turn predicts treatment and survival

This study evaluates the relationship between sociodemographic/exposure factors and comorbidity, and their impact on lung cancer treatment and survival. Data for 1,155 patients were abstracted from the Josephine Ford Cancer Center Tumor Registry and medical records. Associations were analyzed by linear, logistic, and Cox regression. Approximately 88% of patients had ⩾1 of 56 comorbidities assessed. In multivariate analysis, comorbidity count was associated with older age, pack-years smoked, heavy alcohol use, lower socioeconomic status (SES), and female gender. Approximately 63% of patients had ⩾1 of 18 adverse prognostic comorbidities (AC), and significant independent predictors of AC were age, pack-years, African–American race/ethnicity, and gender. In multivariate analysis, comorbidity count and AC predicted nonreceipt of surgery in localized disease (OR ⩾1 vs. 0 AC = 0.38, 95% 0.18, 0.81) and chemotherapy in advanced disease (OR ⩾1 vs. 0 AC = 0.72, 95% 0.51, 1.00). In adjusted analysis, comorbidity predicted survival in localized (hazard ratio (HR) ⩾2 vs. 0 AC = 2.99, 95% CI 1.75, 5.10) and advanced lung cancer (HR ⩾2 vs. 0 AC = 1.56, 95% CI 1.25, 1.94). Comorbidity has important deleterious effects on lung cancer outcomes and significant predictors of comorbidity included age, smoking, race/ethnicity, SES, alcohol, and gender.