Earlier this summer, for sale three of my patients booked visits with me to discuss PSA testing results. Those tests weren’t ordered by me; they came about because my patients had attended a local hospital’s “prostate cancer awareness event”, which offered free testing to any man older than 40.
The conversations were a bit uncomfortable; I had to explain to each of these men that our best research shows that normal risk men (meaning those without a family history of prostate cancer) don’t gain any health benefit from PSA tests. More specifically, a man will not live any longer if he participates in regular PSA testing than if he chooses not to. While one of the three had heard the story before, the other two were taken aback by what I told them.
That raises an important question: How is it that we’re failing to heed the clear evidence on PSA testing and men’s health?
To answer, we need to review the findings and then look at the way that they can be misunderstood or distorted.
Let’s start with what we know:
- An analysis of six trials that included nearly 400,000 men that was released in September 2010 showed that PSA testing caused no change in the risk of dying
- Ongoing studies of 75,000 US men and 180,000 European men also reported in winter 2012 that PSA testing did not lower overall death rates
- Further follow-up reports from both these studies continue to show the same result: PSA testing does not help men live longer
Based on these studies and others, the US Preventive Services Task Force (USPSTF) in 2012 and the Canadian Task Force on Preventive Health Care (CTFPHC) in 2014 have both recommended against PSA testing.
Seems straightforward enough. So how can anyone argue that PSA testing is worthwhile for a normal risk man?
Because it’s easy for anyone, including doctors, to confuse two similar but distinct statistical measures of disease and death.
The first of the two is “disease-specific mortality”, which means death caused by a particular disease. In the case of prostate cancer, we would be looking at changes in the likelihood of a man dying from prostate cancer and only from prostate cancer; we ignore all other causes of death.
The second statistic is “all-cause mortality”, which counts death of any kind. For men with prostate cancer that means we would include deaths from heart disease, stroke, other cancers, accidents etc.
The problem is that when we hear or read things like “PSA testing lowers your risk of dying from prostate cancer” we don’t normally think to ask whether or not that lower risk might be at the expense of raising risk for other causes of death. After all, what harm could there possibly be in a simple PSA blood test?
Of course, the test itself poses no immediate risk. But the results of the test are what can start a dangerous cascade of events.
That’s because we now know that about 1/3 of prostate cancers pose no risk to the future health of the man who has the cancer. The problem is that present science hasn’t found a way to distinguish between a dangerous cancer and one that we can ignore.
The result is that one out of every three men who we diagnose and treat for prostate cancer don’t actually need the treatment. And they certainly don’t need the side effects and complications (including death) that sometimes come with treatment.
Returning to death rates with and without PSA testing, it turns out that PSA testing slightly reduces the risk of dying from prostate cancer. However, it does NOT reduce the overall chance of dying. That must mean that interventions and treatment somehow raise a man’s likelihood of dying from other causes, and that those extra deaths are enough to completely offset the reduced deaths from prostate cancer. What we don’t yet know is how the prostate cancer diagnosis and treatment process creates those deaths.
The bottom line is this: PSA testing will reduce a man’s chance of dying from prostate cancer but INCREASE the man’s risk of dying from other causes by an equal amount. That means that PSA testing won’t extend a man’s life. Worse, it might cause needless suffering from tests and treatments that he doesn’t need.
Many people bristle at the idea that PSA testing isn’t beneficial, particularly if they or someone close to them has been diagnosed and treated. But we shouldn’t confuse a fight against misleading statistics with lack of sympathy for those afflicted with prostate cancer. Instead, we have to focus on funding research that will provide us with better diagnosis and treatments.
Until that happens, normal risk men should just say no to PSA testing.