Influence of Age on Guideline-Concordant Cancer Care for Elderly Patients in the United States

Purpose: To examine the frequency of guideline-concordant cancer care in elderly patients, including "older" elderly (age >= 80 years). Methods and Materials: Using the Surveillance, Epidemiology and End Results-Medicare dataset in patients aged >= 66 years diagnosed with nonmetastatic breast cancer (n=55,094), non-small cell lung (NSCLC) (n=36,203), or prostate cancer (n=86,544) from 2006 to 2011, chemotherapy, surgery, and radiation (RT) treatments were identified using claims. Pearson chi(2) tested associations between age and guideline concordance. Results: Older patients were less likely to receive guideline-concordant curative treatment: in stage III breast cancer, receipt of postmastectomy RT (70%, 46%, and 21% in patients aged 66-79, 80-89, and >= 90 years, respectively; P<.0001); in stage I NSCLC, RT or surgery (89%, 80%, and 64% in age 66-79, 80-89, and >= 90 years; P<.0001); in stage III NSCLC, RT or surgery plus chemotherapy (79%, 58%, and 27% in age 66-79, 80-89, and >= 90 years; P<.0001); and in intermediate/high-risk prostate cancer, RT or prostatectomy (projected life expectancy >10 years: 85% and 82% in age 66-69 and 70-75 years; and <= 10 years: 70%, 42%, and 9% in age 76-79, 80-89, and >= 90 years; P<.0001). However, older patients were more likely to receive guideline-concordant de-intensified treatment: in stage I to II node-negative breast cancer, hypofractionated postlumpectomy RT (9%, 16%, and 23% in age 66-79, 80-89, and >= 90 years; P<.0001); in stage I estrogen receptorepositive breast cancer, observation after lumpectomy (12%, 42%, and 84% in age 66-79, 80-89, and >= 90 years; P<.0001); in stage I NSCLC, stereotactic body RT instead of surgery (7%, 16%, and 25% in age 66-79, 80-89, and >= 90 years; P<.0001); and in lower-risk prostate cancer, no active treatment (25%, 54%, and 68% in age 66-79, 80-89, and >= 90 years; P<.0001). Conclusion: Actual treatment of older elderly cancer patients frequently diverged from guidelines, especially in curative treatment of advanced disease. Results suggest a need for better metrics than existing guidelines alone to evaluate quality and appropriateness of care in this population. (C) 2017 Elsevier Inc. All rights reserved.