• “A prostate screening picture worth a thousand words”

    That’s the title of a post by Carey Goldberg on the CommonHealth blog. I agree. Click over to see the picture.

    My comment: This is, of course, controversial. I’d like to see the “other side’s” version. Then I’d like to hear a reasoned, respectful debate over the two, one that is accessible to a lay audience not familiar with the relevant studies. If anyone could produce such an event, I’d be impressed and grateful.


    Comments closed
    • Now, if only the attorneys would see this and act appropriately.

      But, they still have their heads in their wallets, if twisting metaphors is permissable.

    • Dear Austin —

      I applaud your emphasis on balance and mutual respect.

      The big problem with the PSA debate is that there isn’t really an alternative to the test. As a result, arguing that the test isn’t very informative is essentially claiming that we just can’t do anything proactive about prostate cancer, and a lot of people feel very uncomfortable about this. Patients, of course, but also doctors — the culture of medicine emphasizes taking all possible action in favor of the patient, and it’s difficult for doctors to say, “Well, there’s nothing that I can do!”

      The graph is based on the estimates from the PLCO study, which showed absolutely no difference between treatment and control groups. But if estimates from the ERSPC study had been used, there would have been a small benefit for screening. So the choice of study really makes a difference.

      Both prostate and breast cancer screening have been extensively studied for many many years, and there is universal agreement that if these interventions have an effect, it is relatively small, on the order of 1 out of 1,000. If we can’t get a clear agreement on this, then what sort of impact should we anticipate for comparative effectiveness studies in general?

      • Good question. I think that with good CER we will be able to give ranges. We can use both studies and show the range of cost effectiveness. If you are correct, we can weigh the cost of a single positive outcome out of 1,000 vs the costs of the test and the false positives.


      • As Mr. Whitfield points out, the picture presented rests exclusively on the U.S. study. The picture is different if one looks at the European study.

        This is why it is puzzling that the USPSTF can give a recommendation, which means there is “moderate or high certainty” that there are no net benefits of PSA screening or that harms exceed benefits. This statement can only logically follow from the scientific evidence if one believes it has been scientifically established from the scientific literature that the European study is wrong and the U.S. study is right. I don’t think there is any such consensus.

        Furthermore, what the picture obscures to some degree is that the benefit-cost analysis of screening is really dominated by the ratio of side-effects of treatment to lives saved (or years of life saved). The screening itself is relatively cheap and has no side-effects. The biopsies are more expensive and have some side-effects. However, under reasonable assumptions, the biopsy financial and side-effect costs are small relative to the benefits and harms from definitive treatment. A better picture would somehow incorporate the magnitude of the benefits and costs associated with different outcomes.