Our survey of Canadian men found a comparable level of PSA testing to levels previously reported, with 49% of men aged 41 to 80 reporting having undergone a PSA test in the past 2 years. Despite the lack of convincing evidence for routine PSA testing in asymptomatic men, the patients surveyed in our study had favourable opinions of PSA screening and were unaware of any risk associated with its use.
Results from the ERSPC suggest that, at best, PSA screening prevents 1 death for every 1410 men between the ages of 55 and 69 screened during a 10-year period.6
Unfortunately, this modest benefit, if accurate, is accompanied by substantial morbidity due to overdiagnosis and overtreatment, with an estimated 48 men receiving treatment to prevent the 1 prostate cancer–related death. While the PLCO trial found no mortality benefit among the screening group, the control group of this study was biased by the fact that 52% of the men in this group received some type of screening as part of “usual care,”10
leaving a mild undetected benefit of PSA screening a possibility, which would agree with the ERSPC trial. However, a meta-analysis of 6 different randomized controlled trials (including the ERSPC and PLCO trials) and 387 286 participants concluded that the current evidence does not support routine screening for prostate cancer with the PSA test.11
In our study, we found that men in the potential screening age category defined in the ERSPC had a high self-reported rate of PSA screening (66% in the past 2 years), believed that the PSA test was not associated with any risk (97%), and believed that the PSA blood test was important or very important to their health (84%). The very high level of men who believe there is no risk associated with the PSA blood test is surprising and worrying, especially considering that the known adverse events related to PSA testing were included on our survey as a cue (Box 1
). Men in our study do report discussing the PSA blood test with their physicians (75% in the 51- to 70-year-old age group) but the poor performance of the test and the high burden of adverse events are clearly not being communicated in an enduring manner.
The controversy in the medical literature notwithstanding, patient opinion seems to strongly favour the continued use of the PSA blood test. This finding is consistent with previous studies.9
One possible explanation for the ongoing favourable patient impression of PSA testing is what we will call an existential bias
toward a medical test or procedure. The root cognitive error of this patient bias is as follows: this test exists, and is done all the time, therefore it must be beneficial. Most physicians are presumably aware of the many potential risks of testing, but even physician personal behaviour seems to predict the ongoing use of the PSA test; a recent study showed that 78% of primary care doctors older than 50 years in the United States had themselves undergone PSA testing.12
Men in our study were homogeneous in their belief that PSA testing was not risky, and that it was both effective and important to their health. Our findings suggest that the ongoing debate about PSA testing has not trickled down from medical journals to the general population in a meaningful way. Of course, there are many who believe PSA testing is beneficial (including physicians, lay people, and various organizations) who continue to publicly promote its use. However, it is worrisome that the known risks associated with diagnostic workup and treatment of PSA testing are unknown to most patients.
Our study limitations include a relatively small sample size and the method of convenience sampling. It was not possible to calculate a response rate because respondents either completed surveys voluntarily in the waiting room or were approached by the lead author. Our survey was designed to be concise and to measure patient perceptions of the risks and importance of PSA testing. Our survey tool did not undergo formal psychometric testing. As noted, the high level of college or university education (67%) in our sample is inconsistent with the general level in the Canadian population, but not grossly so; a study using 2004 Statistics Canada data reported the proportion of Canadians with a university degree or postsecondary certificate to be 59%.13
In our study, there also exists the potential for selection bias, with men having a more favourable impression of PSA screening being more likely to want to fill out surveys. Despite the aforementioned limitations, our findings agree closely with previous studies regarding the self-reported rate of PSA testing,4,5
and the general belief among men that PSA testing is beneficial to their health.9
A striking number of the men surveyed in our study believe that PSA screening does not carry any risk of adverse events. To our knowledge, our study is the first to show that patients perceive PSA testing to be a risk-free enterprise. This might be explained by the fact that, on the surface, the act of having a simple blood test seems risk-free. As physicians, we know that the risks are implied in the downstream diagnostic procedures and treatment that necessarily follow a positive PSA test result. Communicating these risks, along with the potential benefits of the test, in a clear and concise way to our patients remains a challenge for primary care physicians.