PROSTATE-SPECIFIC ANTIGEN TESTING IN THE MEDICARE POPULATION: ARE WE TOO AGGRESSIVE?

Abstract

Objective Prostate-specific antigen (PSA) testing for prostate cancer is controversial. Although prostate cancer-specific mortality is probably reduced, the all-cause mortality benefit of aggressive prostate cancer treatment is unproven and probably small at best. There is a substantial negative impact of treatment on men9s quality of life. As a result, PSA screening should be discussed with all men but generally reserved for men who are likely to live at least 10 years, as reflected in recommendations from national organizations. Since functional limitation generally indicates poorer prognosis, we would expect that PSA testing should decrease with the presence of impairment in activities of daily living (ADL) and instrumental ADL (IADL) in Medicare beneficiaries. Method We performed a retrospective cross-sectional analysis of the Medicare Current Beneficiary Survey (MCBS), a nationally representative sample of noninstitutionalized Medicare beneficiaries (age ≥ 65) from the 2002 wave of the MCBS. Using logistic regression, we explored the relationship of PSA to age, race, HMO status, each of eight comorbidities, and any ADL or IADL impairment. Results Of 3,729 surveyed Medicare beneficiaries, 70% reported having undergone PSA testing in the past year. With age 65, white race, and no impairment or comorbidity as the reference group, factors associated with significantly higher odds of PSA testing included cardiovascular disease (odds ratio [OR] =1.24, 95% CI 1.05-1.45), hypertension (OR = 1.60, 1.38-1.88), cancer (OR = 1.60, 1.23-2.08), and arthritis (OR = 1.30, 1.12-1.52). Significantly lower than reference odds for PSA testing were seen for African American race (OR = 0.66, 0.51-0.87), IADL impairment (OR = 0.89, 0.84-0.96), and stroke (OR = 0.79, 0.63-0.99). No significant difference was seen for age, ADL impairment, HMO membership, diabetes, COPD, or psychiatric comorbidity. Conclusion There is no evidence of reduction in PSA testing with ADL impairment and little for IADL impairment in Medicare beneficiaries in the 2002 MCBS survey. There may be an opportunity to reduce the adverse quality of life impact of PSA testing and subsequent treatment for men who are unlikely to derive a longevity benefit, specifically those men who are unlikely to live 10 years. The study is limited in that we cannot determine if PSA testing was used as a screening or diagnostic test in response to urinary symptoms and that functional status is an imperfect predictor of prognosis.