Authors: Keating NL, Landrum MB, Klabunde CN, Fletcher RH, Rogers SO, Doucette WR, Tisnado D, Clauser S, Kahn KL
Title: Adjuvant chemotherapy for stage III colon cancer: do physicians agree about the importance of patient age and comorbidity?
Journal: J Clin Oncol 26(15):2532-7
Date: 2008 May 20
Abstract: PURPOSE: We surveyed cancer physicians to understand how patients' age and comorbidity influence adjuvant chemotherapy recommendations and whether physician or practice characteristics also affect these recommendations. METHODS: National survey of surgeons and medical oncologists caring for patients with colorectal cancer regarding whether they would recommend adjuvant chemotherapy for hypothetical patients with stage III colon cancer who varied by age (55 v 80 years) and comorbid illness (none, moderate, severe congestive heart failure [CHF]). Repeated measures logistic regression was used to assess the influence of patient, physician, and practice characteristics on chemotherapy recommendations. RESULTS: Of 1,096 physicians, nearly all recommended chemotherapy for patients who were 55 years old with no comorbidity (99.0%), 55-years old with moderate CHF (88.6%), or 80 years old with no comorbidity (92.6%); many fewer recommended chemotherapy for 55-year-old patients with severe CHF (24.9%) or 80-year-old patients with moderate (47.2%) or severe (9.0%) CHF (P < .001). Younger physicians (P < .001) were more likely than others to recommend adjuvant chemotherapy overall, although physician factors explained little of the variability in recommendations. CONCLUSION: Physicians agree with guidelines recommending adjuvant chemotherapy for young, healthy patients with stage III colon cancer but differ widely on recommendations for patients who are older and sicker. Few physician or practice characteristics were associated with recommendations. For older and sicker patients, the individual physicians seen may have a substantial impact on the likelihood of receiving chemotherapy. Understanding better the sources of variation not explained by patients' clinical characteristics may allow improved tailoring of therapy to patients most likely to benefit.