Age and Clinical Decision Making in Oncology Patients

BACKGROUND: Ageism is a cultural bias that might inappropriately steer oncologists away from recommending aggressive treatments for older patients. The extent to which older patients might prefer less aggressive cancer therapies is unknown. Our lack of knowledge about patients' personal preferences for therapy may perpetuate this bias. PURPOSE: We conducted a study to determine 1) if age influences patient acceptance of cancer therapy and 2) if the older patients would be more or less likely to trade increased survival for maintaining quality of life than their younger counterparts. METHODS: Using an interview format, 244 cancer patients of all ages treated at a tertiary care cancer center read two sets of hypothetical vignettes. The first set consisted of four vignettes that varied in terms of stage of disease and treatment toxicity. Patients were asked to make hypothetical decisions about treatment given with respect to varying levels of either increasing cure or extending survival. The second set of vignettes presumed acceptance of cancer therapy. Within each vignette, two hypothetical treatments (mild versus severe) with different probabilities of 1-year survival were contrasted. The point at which patients shifted preferences from a treatment with mild versus severe side effects was the dependent measure. Mixed analysis of variance (ANOVA) procedures (F test) assessed the impact of age ( or = 65 years) and patient disease stage (early versus advanced) on hypothetical decisions about treatment. All P values are two sided. RESULTS: In the treatment-preference vignettes, there was no effect of either age F(1,239) = 2.14; P = .14 or patient stage F(1,239) = .40; P = .53 on treatment acceptance. Older adults were as likely as their younger counterparts to agree to chemotherapy for both curative and control purposes. In the switch-point vignettes, younger adults switched to a more toxic treatment to gain survival advantage at an earlier point than the older patients in both the early-disease vignette F(1,232) = 3.88; P = .05 and the advanced-disease vignette F(1,232) = 4.43; P = .036. There was neither an effect of disease stage on treatment decisions nor an interaction between disease stage and age. CONCLUSIONS AND IMPLICATIONS: In a tertiary care setting, older adults do not differ from their younger counterparts in terms of acceptance of chemotherapy. However, when treatment is presumed, they differ in terms of willingness to trade survival for current quality of life. Generalization of findings is limited by the relatively small sample of older adults (n = 43) and the referral population from which the sample was drawn. Replication with a larger older adult sample in a community setting is needed.