Most news coverage on new PSA testing study acknowledged costly trade-offs

Cassels is a drug policy researcher. He frequently reports on consumer drug issues and has authored and co-authored several books. He tweets as @AKECassels.

The major health news story last week was on a re-assessment of prostate cancer screening published in the Annals of Internal Medicine. The study’s release generated stories in the New York Times, TIME, Consumer Reports, STAT,  ABS-CBN News (Philippines) and the LA Times, among others.

The reporting confirms what we already know about the PSA test — that it involves trade-offs including financial costs and long-term harms from screening in exchange for what appears to be very few numbers of lives saved. It is in knowing and understanding these trade-offs that men can be sure that they are going into a PSA test with their eyes wide open.  

According to the American Cancer Society, in 2014 (the most recent data available) 28,343 men in the United States died from prostate cancer. As the second leading cause of cancer death in men, examining and re-examining a test that might prevent those deaths never fails to generate headlines. The PSA test has, over time, become more closely dissected and debated than many screening tests, which is a good thing, seeing as it’s offered to otherwise healthy people (and thus possesses the potential to make those people less healthy via harms of over-treatment.)

Prostate cancer screening is not a new topic to We’ve discussed mass prostate cancer screening events, the problem of using anecdotes, and praised high quality reporting of a study of treating vs. not treating prostate cancer patients. In this recent batch of headlines on the Annals of Internal Medicine study, there are enthusiasts and detractors on the “significance” of the new findings, and overall the coverage gets a pretty good passing grade. Some stories drilled down into the problems with this type of research and why it is difficult to draw firm conclusions.

Some observations on the news coverage

What kind of study was it? In this case, it was a mathematical modeling study parsing out what happens when men randomized to be in the “non-screened” group don’t follow instructions and get screened anyway, and how this affects the interpretation of results. Few stories commented on this aspect of the study except the New York Times which quoted Kenneth Lin, MD saying that “results from models are less convincing than data from actual clinical trials.”

What other factors could be in play? The STAT article mentioned what is known as the presence of “healthy user bias,” which is often overlooked when examining two groups (for example, drug takers versus non-drug-takers and screened men versus non-screened men). Basically, this means that those who choose to undergo screening or treatments might be a healthier cohort of patients at the start. Having better health introduces a bias that may not be accounted for when there are differences between groups.

Did the journalists follow the money? None of these stories reminded us about how big the PSA enterprise is. We need to be reminded of the costs of PSA screening programs and the great bolus of medical services they can generate, including more biopsies, tests, drug therapies, surgeries, hospital time, and the impact on the patient’s earning power as he’s recovering. Urologists make a significant amount of their income from treating the results of PSA tests. We found Etzioni, the lead researcher of this study, does have a potential financial conflict (which was disclosed in the study) but which was not mentioned in any of the stories we reviewed. (In’s systematic reviews of the Los Angeles Times and STAT News stories, reviewers pointed out that neither story raised the issue of financial conflicts.)

Implications for patients

What is the bottom line for the patient? Lots of expert commentary might be enlightening for some, but aggravating for others. At the end of the day a news story should lay out in clear terms the benefits for men who undergo PSA testing, as well as the not inconsequential trade-offs. This infographic produced by the US Preventative Services Task Force lays out the benefits and risks in simple terms.

How long will you survive if you have prostate cancer? reports statistics on 5-year survival rates related to prostate cancer, and states that “The 5-year survival rate for most men with local or regional prostate cancer is almost 100%. Ninety-eight percent (98%) are alive after 10 years, and 96% live for at least 15 years. For men diagnosed with prostate cancer that has spread to other parts of the body, the 5-year survival rate is 29%.”

Last tip: Should we avoid using the  term “life-saving” when we really mean “death-delaying?” No drug or screening program has ever saved a life. At best, it has delayed a type of death. 

You might also like

Comments (9)

We Welcome Comments. But please note: We will delete comments left by anyone who doesn’t leave an actual first and last name and an actual email address.

We will delete comments that include personal attacks, unfounded allegations, unverified facts, product pitches, or profanity. We will also end any thread of repetitive comments. Comments should primarily discuss the quality (or lack thereof) in journalism or other media messages about health and medicine. This is not intended to be a forum for definitive discussions about medicine or science. Nor is it a forum to share your personal story about a disease or treatment -- your comment must relate to media messages about health care. If your comment doesn't adhere to these policies, we won't post it. Questions? Please see more on our comments policy.

George Ritter MD

September 12, 2017 at 6:56 pm

Some urologists SAY that as a result of a PSA and surgery (or radiation) the cancer is “cured” Is this not a life saved or death delayed) which for many people is a good thing.We need to klnow the percentage of “cures” given the various forms of cancer.We need those statistics.
George Ritter MD,


Tim Bartik

September 15, 2017 at 9:06 pm

You might want to ask if you are providing your readers with the ratio right of harms to premature deaths from prostate cancer.

Etzioni’s analysis, in a number of articles, argues that depending upon the exact threshold used for doing biopsies, and how aggressive treatment is, the ratio of “overtreatment” to “reduced prostate cancer deaths” might be about 5 to 1. Or it can be lower if higher biopsy thresholds are used and active surveillance is used more.

You use an Infographic from USPSTF that appears to claim that the ratio is either 80 to 1 or 40 to 1.

There is a substantive disagreement here. You might consider whether or not you are guiding readers in an accurate direction by pointing them to a USPSTF graphic that is clearly viewed by many as seriously misleading. At the very least, you should note that the infographic’s accuracy is not generally accepted.

Among other things; you might consider that the USPSTF infographic assumes that NO ONE who is unscreened is EVER treated for prostate cancer and suffers harms from the treatment.

In other words, you need to be consistent: if the ratio of “lives saved” is based on the treatment group compared to the control group, the ratio of “harms” needs to be based on the same metric. The USPSTF does not appear to have done so.

I don’t think you are providing a balanced review of the health news in this area.


    September 18, 2017 at 12:05 pm

    Tim, thank you for the comments. Your key issue seems to be the accuracy of the USPSTF infograghic, rather than the reporting of this recent study. When you say “you might consider that the USPSTF infographic assumes that NO ONE who is unscreened is EVER treated for prostate cancer and suffers harms from the treatment,” I think that is a bit misleading.

    A lot of men with prostate symptoms will be treated for prostate cancer, which is a completely different cohort of people than those men who are otherwise minding their own business but then get off their couches and go get screened because they are told it would “save your life.” There is a difference between “diagnostic screening” (men with symptoms) versus good old “medical screening” (symptomless men). The former might derive some benefit yet the latter will be statistically more likely harmed. I also wonder where all the opposition to this “seriously misleading” USPSTF infographic is? If anything, the controversy should be around whether the PSA test saves any lives whatsoever. As for overall lives saved, we’ve covered this ground before and for those readers interested in our past discussions of whether PSA testing improves overall mortality, please check out one of our past discussions –link here: which went into this issue in great detail.

    My feeling is that those healthy men given a PSA test and then offered “active surveillance” have already been harmed, psychically damaged probably for no benefit, and that psychic harm is not represented anywhere in the USPSTF’s calculation. So let’s agree, the USPSTF infographic needs work and if it were wholly accurate, it would also account for the great but unmeasured suffering of men who now live with the sword of Damocles over their heads. As one doctor once told me: “The PSA test won’t make you live longer, but it will make your life feel longer.”


      Tim Bartik

      September 19, 2017 at 8:16 am

      My point is the following: Etzioni’s work clearly disagrees with the USPSTF work. They CANNOT both be right. One says the ratio of men treated to prostate cancer deaths averted is around 5 to 1. The other says it is 40 to 1. There is a contradiction here. This is why people are disagreeing. If PSA screening leads to treatment that has a 1 in 5 chance of avoiding a prostate cancer death, with the other 4 cases having a 50% chance of unnecessary side-effects, that is one scenario. If PSA screening leads to treatment that has a 1 in 40 chance of avoiding a prostate cancer death, and then in the other 39 cases I receive treatment I didn’t really need with a 50% or more chance of serious side effects, that is a very different scenario.

      In other words, the news coverage needs to reflect that there is a serious factual dispute here, that leads to potentially quite different views of what individual patients should be doing, and what the health care system should be doing. Let’s not paper over the dispute and pretend there is a consensus here. There isn’t.

      In reviewing “news”, we need to be clear that Etzioni’s work on this topic is “news” precisely because it is implicitly saying that the USPSTF’s review of the research was incorrect. So, in my view, if you’re going to refer readers to the USPSTF infographic, you need to say: “At least some researchers, such as Etzioni, have done work that disagrees with the USPSTF ratios of harms to benefits from PSA screening and prostate cancer treatment.”

      Kevin Lomangino

      September 19, 2017 at 9:09 am


      Thanks for commenting. I think Alan’s post establishes that Etzioni’s study is one more piece of information that will guide and inform the scientific debate on this topic. It also discusses the reasons why such a study may not end up moving the needle very much, if at all, on the numbers that drive screening recommendations and individual decisions about screening. We’ve seen time and again how the flaws/limitations in new research take time to be exposed and accounted for. We will not be the ones to rush to declare a change in recommendations or practice based on the latest study. Nor will we cast doubt on a robust analysis of a large body of research by independent experts based on one new mathematical modeling study. It takes time to sort out how new research impacts clinical practice. We won’t jump the gun or react to every lurch of the news cycle.

      Kevin Lomangino
      Managing Editor

Robert Jaffe

September 18, 2017 at 1:17 pm

The arguments that I’ve seen all seem to start from a group of men roped into stand-alone PSA testing by television commercials or “testing fairs” at some kind of public event. That in itself stacks the psychological deck against such testing. I would think it probable that any such group is smaller than the number of men who, at regular check-ups with their family doctor, have blood panels taken for which PSA tests are automatically included.

Are there any statistics that could settle that issue? Which group is bigger? What are the family doctor/ internist/GP professional associations recommending to their physician members? It probably would be cost-prohibitive to arrange for separate, pre-bloodwork physician/patient appointments for discussion of the PSA issue alone.

Then, instead of the issue being posed in terms of the asymptomatic, couch-potato patient thinking about whether or not to have PSA testing as a special medical event, it’s really posed in a doctor-patient interview with a specific PSA result on the table between them. How easy would it be, then, for the physician to tell the patient he can/should/ought to ignore what he’s just been told is a PSA level two or three hundred percent of normal?

I write as a prostate cancer patient en route to my second biopsy to find out if the cancer has spread, initially recommended to a urologist by my family doctor after an annual checkup because of an elevated PSA level. The push to minimize PSA testing seems to me oriented much more towards societal costs than individual welfare. My family physician agrees. To the extent that governmental bodies take society’s side in zero-sum society vs. individual issues, “authorities,” whether governmental, scientific or medical, run a very serious risk of exacerbating current distrust of “authorities,” including scientists and doctors. This is NOT separate from the fears of the anti-vaxxers and conspiricists (whom I despise) and probably adds to the credibility of their (incredible) claims.


    Kevin Lomangino

    September 18, 2017 at 1:45 pm


    Thanks for your comment and I hope your biopsy brings good news. You wrote:

    The push to minimize PSA testing seems to me oriented much more towards societal costs than individual welfare.

    I could not disagree with this statement more. We are continually told by individual patients who’ve had PSA testing done — sometimes without their consent — that they regret their prostate cancer screening and the treatment it often leads to. These men have often been harmed both physically and psychologically in ways that they are reluctant to speak about. It is the welfare of such individual men that motivates our emphasis on informed decision-making in this area. We don’t “push to minimize” screening but we do push for men to be fully informed of the implications of a decision to get tested.

    Kevin Lomangino
    Managing Editor


      Robert Jaffe

      September 18, 2017 at 2:12 pm

      This is in reply to Kevin’s reply to my post of earlier today.

      Kevin: I have no doubt that you know far more about this entire issue than I do. I’m surprised, though, at what you say. I’ve known, solely on a personal level, quite a number of men, and their families, who have been through (and some died from) prostate cancer and its various treatments. The large majority of survivors seem grateful for their treatment, while the families of those who died regret not having sought and obtained effective treatment in time. Of course, I’m talking anecdotally. Are you? What percentage of treated individuals have been unhappy with their treatments and unhappy to what extent? Cancer treatment is no fun in the best of cases. We must weigh the admittedly subjective unhappiness of those who regret treatment with the (again, admittedly subjective) relief of those who believe such treatment cured them. You give the impression (correct me if I’m wrong) of ignoring the very real value of being told one is “cancer-free” (for what that’s worth). I’m not convinced that the entire issue isn’t a bean-counter issue.

      Kevin Lomangino

      September 18, 2017 at 3:17 pm


      I am speaking about the anecdotes I’ve been made aware of. But there is certainly data to corroborate the fact that many men experience serious harm from prostate cancer treatment. And as to the value of being told one is “cancer-free,” I think this is deceptively complicated. Most men with prostate cancer would never even know that they had cancer — and would be none the worse off — if they hadn’t been PSA tested in the first place. So how do you value the “relief” of no longer having a cancer that in most cases wouldn’t have done you any harm and you wouldn’t even have known about? That’s why I think discussions about the test need to happen before one is tested. Some men may look at the numbers and decide testing makes sense, others definitely will not. As for bean-counters, there are plenty of those at urology practices and hospitals who understand that PSA is a money-maker. You may find this Men’s Health article enlightening on that front:

      Kevin Lomangino
      Managing Editor