Longitudinal Changes in Depression Symptoms and Survival Among Patients With Lung Cancer: A National Cohort Assessment

Abstract

Purpose Depression symptoms are common among patients with lung cancer; however, longitudinal changes and their impact on survival are understudied. Methods This was a prospective, observational study from the Cancer Care Outcomes Research and Surveillance Consortium from five US geographically defined regions from September 2003 through December 2005. Patients enrolled within 3 months of their lung cancer diagnosis were eligible. The eight-item Center for Epidemiologic Studies Depression scale was administered at diagnosis and 12 months' follow-up. The main outcome was survival, which was evaluated using Kaplan-Meyer curves and adjusted Cox proportional hazards modeling. Results Among 1,790 participants, 681 (38%) had depression symptoms at baseline and an additional 105 (14%) developed new-onset depression symptoms during treatment. At baseline, depression symptoms were associated with increased mortality (hazard ratio [HR], 1.17; 95% CI, 1.03 to 1.32; P = .01). Participants were classified into the following four groups based on longitudinal changes in depression symptoms from baseline to follow-up: never depression symptoms (n = 640), new-onset depression symptoms (n = 105), depression symptom remission (n = 156), and persistent depression symptoms (n = 254) and HRs were calculated. Using the never-depression symptoms group as a reference group, HRs were as follows: new-onset depression symptoms, 1.50 (95% CI, 1.12 to 2.01; P = .006); depression symptom remission, 1.02 (95% CI, 0.79 to 1.31; P = .89), and persistent depression symptoms, 1.42 (95% CI, 1.15 to 1.75; P = .001). At baseline, depression symptoms were associated with increased mortality among participants with early-stage disease (stages I and II; HR, 1.61; 95% CI, 1.26 to 2.04), but not late-stage disease (stages III and IV; HR, 1.05; 95% CI, 0.91 to 1.22). At follow-up, depression symptoms were associated with increased mortality among participants with early-stage disease (HR, 1.71; 95% CI, 1.27 to 2.31) and those with late-stage disease (HR, 1.32; 95% CI, 1.04 to 1.69). Conclusion Among patients with lung cancer, longitudinal changes in depression symptoms are associated with differences in mortality, particularly among patients with early-stage disease. Symptom remission is associated with a similar mortality rate as never having had depression.

Conflict of interest statement

Authors’ disclosures of potential conflicts of interest are found in the article online at www.jco.org. Author contributions are found at the end of this article.

Figures

Fig 1.

CONSORT diagram showing the flow of patients from enrollment to baseline and follow-up survey based on depression symptoms in the CanCORS study. CanCORS, Cancer Care Outcomes Research and Surveillance Consortium.

Fig 2.

Kaplan-Meier estimates of survival based on depression status. Survival was calculated from baseline survey to the time of death, if it occurred during the study period, or to the time of data censoring. The difference in median estimates of survival between the no depression symptoms group and the depression symptoms group was 193 days. Depression symptoms were a significant predictor of mortality (hazard ratio for death in the depression symptoms group, 1.17; 95% CI, 1.03 to 1.32) after adjustment for age, sex, race/ethnicity, lung cancer stage and histology at diagnosis, income, education, marital status, smoking and alcohol use, and Adult Comorbidity Evaluation-27 (ACE-27) index. (Insert): Survival was calculated from follow-up survey to the time of death, if it occurred during the study period, or to the time of data censoring. Median estimates of survival from cancer diagnosis were 896 days in the never depression symptoms group, 889 days in the remission of depression symptoms group, 746 days in the new-onset depression symptoms group, and 778 days in the persistent depression symptoms group. At follow-up, the difference in median survival between depression symptoms (the new-onset and persistent groups) and no depression symptoms groups (includes the never and remission groups) was 130 days. Persistent depression symptoms (hazard ratio for death 1.42; 95% CI, 1.15 to 1.75) and new-onset depression symptoms at follow-up (hazard ratio for death, 1.50; 95% CI, 1.12 to 2.01) were significant predictors of mortality after adjustment for age, sex, race/ethnicity, income, education, marital status, smoking and alcohol use, cancer stage and histology, and ACE-27 index.