Lung carcinoma symptoms--an independent predictor of survival and an important mediator of African-American disparity in survival

Cancer. 2004 Oct 1;101(7):1655-63. doi: 10.1002/cncr.20547.


Background: The extent of disease in patients with lung carcinoma is reflected morphologically by stage and pathophysiologically by sign/symptoms. This study evaluates the associations between symptoms and stage, the independent impact of symptoms on survival, predictors of symptoms, and the extent to which symptoms mediate survival disparities.

Methods: Data from 1154 patients with lung carcinoma were collected from the authors' tumor registry and by abstraction of medical records. Associations were evaluated by logistic and Cox regression analyses.

Results: Symptomatic diagnoses were associated with advanced disease stage (odds ratio [OR], 4.53; 95% confidence interval [95%CI], 3.17-6.48). Hoarseness, hemoptysis, dyspnea, noncardiac chest pain, extrathoracic pain, neurologic symptoms, weight loss, and weakness/fatigue (adverse symptoms) were associated independently with relatively higher/advanced stage and/or reduced survival. Adverse symptoms (> or = 1 vs. 0) predicted reduced survival independently of stage and other prognosticators (hazard ratio [HR], 1.84; 95%CI, 1.52-2.21). Independent predictors of adverse symptoms included gender (OR(male vs. female), 1.50; 95%CI, 1.11-2.01), race/ethnicity (OR(black vs. white), 1.62; 95%CI, 1.18-2.21), and marital status (OR(spouseless vs. not), 1.79; 95%CI 1.31-2.45). The hazard ratios (HR; black vs white), univariate, adjusted for stage, and adjusted for stage and adverse symptoms, was 1.206 (95%CI, 1.05-1.38), 1.165 (95%CI, 1.01-1.34), and 1.075 (95%CI, 0.94-1.26), respectively. Adverse symptoms explained 43% of race/ethnic survival disparity beyond stage.

Conclusions: Symptoms were associated with disease stage, yet both were important, independent predictors of survival; and symptoms explained an important amount of race/ethnic disparity in the survival of patients with lung carcinoma. Symptomatology needs to be incorporated into cancer clinical trials and into outcomes and disparities research.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • African Americans
  • Ethnic Groups
  • Female
  • Humans
  • Lung Neoplasms / mortality*
  • Lung Neoplasms / physiopathology*
  • Male
  • Marital Status
  • Prognosis
  • Socioeconomic Factors