A win for readers: coverage of new prostate cancer screening guidelines focuses on shared decision making

Richard Hoffman, MD, MPH is Professor of Internal Medicine and Epidemiology and Director of the Division of General Internal Medicine for the University of Iowa Carver College of Medicine/Iowa City VA Medical Center. He is a long-time HealthNewsReview.org contributor and has written many times about news coverage of cancer screening.

When the United States Preventive Services Task Force (USPSTF) issued a statement recommending against prostate cancer screening in 2012, its guidance was met with often vituperative attacks from professional organizations and advocacy groups. The muted response to new recommendations released this past Tuesday supporting personalized decision making for men 55 to 69 must have come as a welcome relief to the Task Force. The response was not surprising; the Task Force had issued a draft recommendation last year that received a similar reception.

The Task Force’s qualified support for screening disarmed some of its most vehement critics, including the American Urological Association (AUA). The AUA commended the Task Force for now being aligned with practice guidelines issued by the AUA and other major physician groups.

With a few exceptions, news stories about the recommendations appropriately recognized the complexity of screening decisions, as reflected in these headlines:

CNN’s headline – “Prostate cancer screening recommendations get rolled back by US task force” – didn’t capture the incremental nature of the guideline revisions. The evidence rating for the recommendations went from a “D” to a “C” grade and the focus on shared decision making does not amount to a “roll back” of previous guidance.

In addition, the AP’s coverage includes some misleading statements from an American Academy of Family Physicians representative, who states that African Americans and men with a family history of prostate cancer are “the best candidates for screening.” In fact, the Task Force recommendations apply to all men and do not indicate that high-risk men would be the best candidates for screening. The Task Force specifically points out that earlier screening of high-risk men is not supported by the evidence—men of African ancestry were not included in large numbers in the supporting studies and those studies did not identify men with family histories.

What did the Task Force say and why?

The Task Force emphasized that men could understandably have markedly different perspectives about the same data on benefits and harms, which is why the group supports individualized decision making. The ideal scenario is shared decision making, a collaborative process between patients and clinicians based upon the best available clinical evidence and the values and preferences of an informed patient.

Why did the Task Force revise its 2012 recommendation against screening to this more nuanced, individualized approach? News coverage appropriately highlighted the influence of longer-term follow up of men in the European Randomized Study of Screening for Prostate Cancer (ERSPC). Data published in recent years showed increasing benefits for screening over time in preventing prostate cancer deaths and metastatic spread. News coverage also recognized that the harms of screening related to overdiagnosis and overtreatment were being partly mitigated by the increased uptake of active surveillance—a monitoring strategy for men with a low-risk prostate cancer that defers active treatment (surgery or radiation) in the absence of cancer progression.

Additional context and active surveillance

The stories, though, overlooked a few important details in the Task Force recommendation. Not only did the Task Force factor in the increasingly positive ERPSC results, but it also discounted the negative results of the large American trial, the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO). Previously, the Task Force equally weighted results from both studies, but the PLCO is now widely recognized as a flawed evaluation of screening. The study was supposed to compare the effects of annual PSA testing vs. no testing on prostate cancer mortality. However, it was later shown that over 80% of the men in the control arm received PSA testing either during or before the study. This helped bias the study towards finding no benefit for screening.

Another issue is that active surveillance is primarily described as sparing patients from the harms caused by surgery or radiation. However, the evidence offers an even more compelling rationale for this strategy. The Prostate Testing for Cancer and Treatment (ProtecT) trial, conducted in the United Kingdom, recently showed that men with early-stage prostate cancers who were randomly assigned to surgery, radiation therapy, or active monitoring had the same low risk of dying from prostate cancer after 10 years of follow up. Active surveillance thus prevents treatment harms without increasing the risk of dying from prostate cancer.

More support needed for shared decision making

Finally, coverage about the new guidelines seems to assume that patients can make good decisions if they just talk to their primary care clinicians. This is overly simplistic. Studies have consistently shown that clinicians lack the time and often the expertise to conduct shared decision making visits. When screening is discussed, providers often fail to present both the pros and cons of screening or elicit patient preferences. Men are often poorly informed about the screening decision, markedly overestimating the risks of developing cancer and dying from it. They also tend to overestimate the benefits of screening.

An important missing ingredient from the news coverage—and the Task Force recommendation—is the use of decision aids. These are educational tools that can consistently provide patients with comprehensive and objective information about prostate cancer, the pros and cons of screening, and help patients clarify their values and prepare to discuss screening with clinicians. These tools are readily available, and patients should be encouraged to take responsibility for these decisions, coming into a visit informed about the disease and their options. This preparation can lead to much more efficient and productive discussions.  Most people wouldn’t buy a car or a flat-screen TV without first doing some homework, and it should be the same for making cancer screening decisions—which have much higher stakes.

The USPSTF’s critical contribution

While the Task Force’s prostate cancer screening recommendations have been controversial over the past few decades, they have consistently been based on rigorous evidence syntheses—unlike organizations such as the AUA which did not embrace this approach until 2013.  Importantly, these guidelines have helped Americans recognize that there are potential benefits and harms from all screening tests, and that patients need to actively participate in screening decisions.

You might also like

Comments (1)

Please note, comments are no longer published through this website. All previously made comments are still archived and available for viewing through select posts.

Chris O'Neill

May 12, 2018 at 8:40 am

“News coverage appropriately highlighted the influence of longer-term follow up of men in the European Randomized Study of Screening for Prostate Cancer (ERSPC).”

The statistical analysis in the ERSPC is still based on the fundamental error of assuming the 7 trials involved all had the same prostate cancer mortality rate ratio between the intervention and control arms (null hypothesis of homogeneity). This null hypothesis was tested and not rejected but you can’t prove the null hypothesis by not rejecting it.

This means you can’t lump all the 7 trials’ results together as if their populations were similar and you can’t get a general prostate cancer mortality risk ratio that applies to all 7 trials by applying such a process.

The ERSPC analysis has made a truly huge mistake. I can only guess this has happened because it wasn’t written by sufficiently competent statisticians.