Intensity of Medical Interventions between Diagnosis and Death in Patients with Advanced Lung and Colorectal Cancer: A CanCORS Analysis

Abstract

Background: Medical interventions are an important component of the illness experience in advanced cancer.

Objective: To describe the use of medical interventions between diagnosis and death in decedents with metastatic lung and colorectal cancer.

Design: Retrospective analysis of a prospective cohort study.

Setting/subjects: We studied 1,840 decedents from the Cancer Care Outcomes Research and Surveillance (CanCORS) study. Subjects had been diagnosed with stage IV lung or colorectal cancer between 2003 and 2005.

Measurements: Hospitalizations, surgeries, radiation therapy treatments, chemotherapy treatments, and end-of-life care, reported by tertile of overall survival time.

Results: Median survival in the bottom, middle, and top tertiles of survival was 1.2, 5.3, and 15.3 months for lung cancer, and 3.0, 18.0, and 44.4 months for colorectal cancer. Hospitalizations, chemotherapy receipt, and hospice enrollment increased with increasing survival. The median duration of chemotherapy in the top survival tertile was 149 days for lung cancer and 498 days for colorectal cancer. A minority of decedents used any hospice services, and the median duration of hospice enrollment exceeded 30 days only for enrollees in the top survival tertile (lung cancer, 35 days; colorectal cancer, 66 days).

Conclusions: For patients with metastatic lung and colorectal cancer, longer survival is associated with increased intensity of medical care, characterized by greater use of chemotherapy and acute hospital care. Hospice utilization was uniformly low, and most hospice enrollees were referred to hospice in the last 30 days of life.

Figures

FIG. 1.

Cohort assembly diagram. aDeceased according to query of plan records for managed care sites (CRN and VA) or to linkage to national death records (all other sites). CanCORS, Cancer Care Outcomes Research and Surveillance; CRN, Cancer Research Network; VA, Veterans Affairs Medical Center.

FIG. 2.

Frequency and timing of medical interventions in patients with lung cancer. A–C: Outcomes for patients in the bottom, middle, and top tertile of survival, respectively, are shown. Timing (with respect to cancer diagnosis) is shown for events that occurred in at least 10% of patients. Percentages indicate the proportion of patients experiencing each event. Circles connected by bars indicate the median start date and median duration for chemotherapy treatment lines, radiation therapy courses, and hospitalizations. Diamonds indicate the median time to occurrence for other discrete events, and the median time to death is marked with an ‘X’. Due to scale, only the start date of hospitalization is indicated in (C). RT, radiation therapy; EOL, end-of-life; DNR, do-not-resuscitate order; ICU, intensive care unit admission.

FIG. 3.

Frequency and timing of medical interventions in patients with colorectal cancer. A–C: Outcomes for patients in the bottom, middle, and top tertile of survival, respectively, are shown. Timing (with respect to cancer diagnosis) is shown for events that occurred in at least 10% of patients. Percentages indicate the proportion of patients experiencing each event. Circles connected by bars indicate the median start date and median duration for chemotherapy treatment lines and hospitalizations. Diamonds indicate the median time to occurrence for other discrete events, and the median time to death is marked with an ‘X’. Due to scale, only the start date of hospitalization is indicated in (B) and (C). EOL, end-of-life; DNR, do-not-resuscitate order; ICU, intensive care unit admission.