The U.S. Preventive Services Task Force is changing their guidelines about prostate cancer screening. In 2012, the Task Force gave it a D grade – meaning they recommended against prostate specific antigen (PSA) screening because its harms exceeded its benefits. Last week, in their draft recommendation, the Task Force upped it to a C grade for men ages 55 to 69 – meaning the decision to screen should be based on professional judgment and patient preference.
Is this just another example of grade inflation?
No. Instead, the science of health care delivery – like all science – is dynamic. As the available information changes, so can the recommendations.
What’s changed? Well there is a little more confidence from a long-term follow-up of the European randomized trial that PSA screening can lower the death rate from prostate cancer. It’s not a big benefit, mind you – in the order of one prostate cancer death averted per 1000 screened for 10 years – but probably a real one.
The big change relates to the harms of screening: overdiagnosis and overtreatment. Overdiagnosis refers to the detection of prostate cancers that were otherwise not destined to ever cause problems for men. Overdiagnosis is a big problem in prostate cancer, a colleague’s and my 2009 estimate was that over a million American men had been overdiagnosed since the advent of screening. Once detected anything labeled “cancer” tends to be treated. Prostate cancer treatment can’t help an overdiagnosed patient, because there is nothing to fix. But treatment can cause real problems, most commonly incontinence and impotence.
What’s changing is the practice of urologists – the doctors who treat prostate cancer. They are now more attuned to the harms of overdiagnosis and overtreatment than they were in the past. More men with prostate cancer are being offered – and are choosing – not to undergo immediate treatment, but instead active surveillance: regular monitoring to determine whether or not the cancer is likely to cause problems. It's a strategy that reduces overtreatment by making use of the diagnostic value of time.
Furthermore, fewer men are having their prostate biopsied. This may reflect more judicious use of biopsy by urologists (e.g., avoiding biopsy in men with trivial PSA elevations) or more judicious decisions about when to refer men to urology by primary care practitioners. Fewer biopsies in low risk men reduces the problem of overdiagnosis.
On the other hand, fewer biopsies may simply reflect the fact that fewer men are being screened since the Task Force’s 2012 recommendation against screening.
In short, the Task Force re-graded PSA screening for middle aged men because they had a little more confidence that screening provided some benefit and a lot more confidence that current practice was changing to reduce its harms. They are clear about the screening trade-off: for every 1 to 2 men who avoid a prostate cancer death, 20 to 50 are overdiagnosed. But they also acknowledge that different men could view this trade-off differently and that the right recommendation is to make screening a choice.
Don’t be surprised if they reach the same conclusion about breast cancer screening in the future – because the truth is the two are not that different.
None of this changes my decision. I’m a 62 year-old male. I’m not interested in pursuing a tiny chance of avoiding a prostate cancer death at the cost of a more common chance of a variety of harms: incontinence and impotence from a surgery, infection and bleeding from a biopsy, and the uncertainty and worry from a slightly abnormal test. Not to mention the entire exercise could cost me a lot of money. I choose not to be screened.
There is no one right answer here, and others may rationally choose to be screened. Those that do are best served if they can take their time and tolerate uncertainty. Don’t rush to biopsy for trivial PSA elevations; don’t rush to treatment for low risk prostate cancer.
Yet we all need to prepare for more trade-offs like this. Testing healthy people for early signs of disease is a great market, one that can involve millions of customers. That’s why we can expect to hear more about population-wide genetic testing, biomarkers, immunosignatures, nanocytology, and liquid biopsies.
Overdiagnosis turns healthy people into patients – diagnosing disease not otherwise destined to bother them. And overdiagnosis is not the only harm of population-wide testing. Abnormal tests make healthy people worry about whether they are deathly sick. Although most turn out to be false alarms, this is only known after a cascade of additional testing, which may involve invasive procedures and complications.
Whether testing actually benefits healthy people, is harder to know. Many must be tested to potentially help a few. Studies require tens of thousands of people and years to complete, and researchers have been studying PSA screening for a quarter century.
And even if a few are genuinely helped, the effect on many others may be negative. Testing healthy people doesn’t promote health; it promotes disease. The emphasis on testing can distract people from behaviors much more important to their health – good food, regular movement, and finding purpose in life.
That’s why screening is a choice.
H. Gilbert Welch is a professor of medicine at the Dartmouth Institute for Health Policy and Clinical Practice and the author of “Less Medicine, More Health – 7 Assumptions that Drive Too Much Medical Care”.
