American Urological Association (AUA) 2011 Annual Meeting

May 14-19, 2011; Washington, DC

Rates of Prostate-Specific Antigen Testing Among Elderly Patients Remain High Despite Concerns About Necessity

An examination of the annual rates of prostate-specific antigen (PSA) testing administered by primary care providers (PCPs) in the United States from 1997 through 2008 found that the rate of PSA testing more than doubled for men age ≥40 years and that many men age ≥75 years continue to be tested despite mounting evidence that they are unlikely to benefit from prostate cancer therapy. The study was presented during a podium session at the recent AUA annual meeting. Lead author Sandip Prasad, MD, Urologic Oncology Research Fellow, University of Chicago Medical Center, IL, discussed the data and their implications with Clinical Geriatrics.

To determine the frequency of PSA testing among men age ≥40 years, the authors used data from the National Ambulatory Medical Care Survey for the years 1997 through 2008. They limited the study to men without prostate cancer, most of whom received testing at a primary care visit.

In 1997, approximately 5.3 million PSA tests were administered at 5.7% of PCP visits. More than twice as many PSA tests were performed during 2008, with an estimated 12.4 million tests administered at 11.5% of office visits. This indicated increases of nearly 6.5% annually (P <.01), a finding that surprised the authors.

“Given that prostate cancer incidence declined during our study period and that multiple organizations issued recommendations against routine PSA screening between 1997 and 2008, we [had] expected to find that both PCPs and urologists would have lower rates of PSA test utilization,” Prasad explained.

Although the proportion of PSA tests administered by urologists was not provided in the study abstract, Prasad said that testing rates were “essentially identical” between urologists and PCPs. In addition, both PCPs and urologists were significantly more likely to give PSA tests to non-Hispanic (black and non-Hispanic white) men than to Hispanic men (P = .01 and P = .04, respectively). “This may reflect practice patterns or perceptions regarding the higher rates of prostate cancer incidence and mortality in non-Hispanic men,” he said.

Men ages 40 to 49 years were least likely to undergo PSA testing, which was carried out at 5.1% of PCP visits. This compared with 9.6% of visits for men ages 50 to 59 years, 11.0% for men ages 60 to 69 years, 8.5% for men ages 70 to 74 years, 6.9% for men ages 75 to 85 years, and 5.3% for men age >85 years.

In the study’s latter years, data and guidelines started to emerge that questioned the value of PSA testing for men ≥75 years. Despite this, Prasad said that the study team did not see a decline in testing for this demographic. “In fact, it continued to increase,” he noted. Individuals with Medicare and Medicaid were less likely than those with private insurance to undergo PSA testing at the primary care office. Findings were consistent across all geographic regions. Nor was an association identified between any comorbidity and the likelihood of undergoing PSA screening.

Prasad said that the relationships between PSA screening rates, prostate cancer incidence, and prostate cancer deaths are not clear. Whereas one might expect that an increase in PSA screening would correlate with more diagnoses of prostate cancer, he pointed out that prostate cancer diagnoses had actually declined during the study period. Prasad proposed several factors that might account for this discordance, including the possibility of a decrease in biopsy rates due to demographic or clinical elements; fewer men undergoing invasive digital rectal examinations, a procedure that can also suggest the need for a biopsy; and a “culling effect” that occurs when men with suspicious PSA levels are found to have prostate cancer on biopsy and are then removed from the PSA screening pool, thus reducing the likelihood that repeated PSA screening among the remaining men will result in a diagnosis of prostate cancer.

The study has important implications for clinicians who treat geriatric patients. “PSA testing continues to be performed at unnecessarily high rates in older men with limited life expectancy, who are least likely to benefit from screening efforts,” said Prasad. He cautioned against using “age alone…as a surrogate for life-expectancy,” however. “Geriatric patients [should] be assessed for concomitant comorbidities so that those individuals with greater than 10-year life expectancy are offered screening.”