Prostate specific antigen (PSA) test is potentially both a beneficial and a harmful screening tool for prostate cancer. Evidence from randomized clinical trials (RCTs) show that in men 50 to 69 years of age, PSA-based screening may have benefit in preventing 1.3 deaths per 1,000 men screened over 13 years. However, in men older than 70 years, evidence from RCTs consistently provides no benefit of PSA-based screening for averting deaths from prostate cancer. Therefore, current practice guidelines from various authorities, such as the American College of Physicians, American Urological Association, and U.S. Preventive Services Task Force (USPSTF), recommend that PSA-based screening not be used in men who are either older than 69, who may not have been in good health, and in those who have a life expectancy of less than 10 to 15 years, unless men prefer after being informed about the benefits and risks. The USPSTF concludes with moderate certainty that the potential benefits of PSA-screening in men 70 years and older does not outweigh its expected harms from overdiagnosis. This is because the potential benefits of screening from treatment can occur over decades, and the progression of screen detected asymptomic prostate cancer to symptomatic prostate cancer or death may take years to manifest. Localized prostate cancer is slow growing and many men die of causes other than prostate cancer.
Despite these recommendations, the rate at which primary care providers (PCPs) perform PSA screening in men 75 years and older varies significantly. Simply seeing a PCP who conducts PSA screening more often accounts for just over a quarter of the variance in PSA use in men older than 75. In addition to harms of screening stemming from unnecessary treatments and harms of treatments, e.g. erectile dysfunction, urinary incontinence, bowel symptoms, there is a significant cost issue from PSA screening and with additional downstream tests and treatments that follow a false-positive PSA result. Because of this, a research team led by Dr. Preeti Zanwar, instructional assistant professor in the department of epidemiology and biostatistics at the Texas A&M School of Public Health, conducted research that estimated PSA testing rates, tests, treatments and doctor visits and associated Medicare payments in patients cared for by two groups of PCPs in Texas – those with high vs. low PSA ordering rates for men older than 75.
Dr. Zanwar and her colleagues first examined the rates of PSA screening use in men older than 75 in 2009. They selected PCPs who had PSA screening rates that were either significantly higher or significantly lower than the average. They then divided a cohort of male patients older than 75, which they obtained from Texas Medicare data, into high-testing PCP and low-testing PCP groups. The researchers also measured the number and proportion of men in both groups who had received PSA screening, other tests such as biopsies and ultrasounds and cancer treatments such as radiation therapy, androgen deprivation therapy and prostatectomy.
Their analysis found that about 25 percent of the PCPs in the study performed PSA screenings at rates significantly higher than the mean rate for all PCPs and nearly 30 percent had PSA screening rates significantly lower than the mean rate. Patients cared for by high testing PCPs had significantly higher rates of PSA testing ordered by PCPs, with nearly 72 percent of PSA testing ordered by PCPs in high testing PCPs vs 33 percent of PSA testing ordered by PCPs in low testing PCPs. A further statistical analysis adjusted for patient characteristics found that men seeing high testing PCPs had significantly higher rates of receiving PSA screenings, prostate biopsies, ultrasounds, radiation treatment, higher number of evaluation and management services for prostate cancer, and higher prostate cancer associated Medicare payments.
These findings indicate that men seeing high testing PCPs undergo more diagnostic and treatment procedures and incur higher health care costs than men with low testing PCPs. For every 10,000 patients with high testing rate PCPs, 2,495 more men underwent PSA screening, with an additional 31 having prostate biopsies, 51 receiving ultrasounds and 18 undergoing radiation treatments.
“If these findings are extrapolated to all men older than 75 in Texas, then having these men see PCPs with low PSA testing rates instead of high testing PCPs could reduce health care expenses by three million dollars per year,” said Dr. Zanwar.
The authors caution about generalizing this study’s findings to the entire population as there were a few limitations. First, although the authors excluded men with a history of prostate cancer diagnosis and treatment in the prior three years, because of the high prostate cancer survival rate, it is possible that the study could have included men who completed prostate cancer treatment before this study period. Second, the study made the assumption that the PSA testing was used for screening rather than diagnostic purpose. However, it is possible that PSA test could have been administered to evaluate other conditions such as urinary obstructions or blood in the urine. Additionally, the study results might not be generalizable to states other than Texas, in younger men, or in those with insurance coverage other than Medicare.
Despite these limitations, Dr. Zanwar and her colleagues have highlighted large variations in prostate screening and prostate cancer associated tests and treatments based on PCP’s PSA ordering rate. The authors state that future work is needed to examine the differences between these two groups of PCPs and how they interact with other providers such as urologists. Additionally, payment reforms such as bundled payments could affect how providers inform patients about potential benefits and risks of health care services and involve patients in shared decision making, which is another area ripe for future study. Having a better understanding of what drives the health care decision making process in older men and in men with multiple co-morbidities will likely help inform recommendations on balancing the benefits and harms of potentially low-value clinical preventative services, such as PSA-screenings for diseases like prostate cancer.