Once more, screening for prostate cancer
It’s pretty much conventional wisdom among the experts that routine prostate cancer screening for the prostate-specific antigen can cause more harm than good. The recent results of the European study of routine PSA screening after 13 years of follow-up, which The Lancet published early this month, also viewed population PSA screening with a jaundiced eye–even though its data showed that screening reduced the death rate from prostate cancer by about 20%.
How can your average aging man cope with that kind of seeming contradiction–and from the so-called experts too? Can we blame him for taking a chance on the considerable risks of PSA screening, including unnecessary surgery, for the sake of being among those whose lives are saved? For some, even the risk of urinary tract damage and impotence could seem like a reasonable trade-off.
A man might make a different decision, though, if he was told that a 20% reduction in the death rate means that the average middle-aged man’s risk would drop from about 3% to about 2.4%, which is what Tara Parker-Pope says at Well. The absolute figure quoted in The Conversation post about the study was just one man’s life saved out of 780.
Looking closely at the European research
At her blog Patient POV, Laura Newman delves into the research presented in that paper, consults other experts, and comes to conclusions that call into question even the finding about the 20% death rate reduction.
The study covered eight countries in Europe, but screening benefits were seen in only two: Sweden and Holland–and the results in Holland were on the cusp of not being statistically significant. Newman quotes Anthony Zietman, a prominent radiation oncologist at Harvard Medical School, as saying “Explain that if you can! I know I can’t.”
Finland, which has a prostate cancer rate comparable to Sweden’s, which is as high as the rate among African-Americans, contributed the largest number of patients to the study. Still, there were no benefits to PSA screening in Finland. Peter Albertsen, surgeon at the University of Connecticut Health Center, told Newman that in the other five countries, sample sizes were too small to have sufficient statistical power.
The researchers weren’t convinced either
There were other methodological issues with the research as well. It’s easy to see why the authors of a study reporting a 20% reduction in the death rate from prostate cancer associated with PSA screening still ended up unenthusiastic about it. Sounds as if they weren’t convinced either.
Albertsen told Newman, “So screening works for some cancers, but not for others. Now the problem is how to tell these two groups apart.” Newman concludes, “There’s plenty more that has to be done if doctors and patients are going to get on the same page.”
The Conversation post takes off from The Lancet paper but is really an explainer about PSA testing, written by Alexandra Miller and Reema Rattan. It quotes Dragan Illic, an epidemiologist at Monash University, thus: “The problem with the PSA test is that, although it’s prostate specific, it’s not prostate cancer specific.”
You can almost hear Incidental Economist Aaron Carroll sigh as he writes “The overscreening never seems to end.” He’s talking about a new study from JAMA Internal Medicine showing that routine screening for assorted medical conditions persists pretty much to the end of life–far beyond the point when even a useful test, like the one for cervical cancer, can do a patient any good.
The study zeroed in on patients who were highly likely to die within 9 years and found, for instance, that 55% of the men were being screened for prostate cancer (and the rate was even quite high in the group likely to die within 5 years.) My favorite datum is the PAP smear screening rate–from 34% to 56% among women who had no uterus.
Of course, the fact that the testing does people other than the patients good–the prescribing doc, the testing labs, the test manufacturers–may not be irrelevant. But Carroll declines to engage in this sort of rough stuff. He concludes merely, “Bottom line is that we’re screening a huge number of people who are incredibly unlikely to receive a benefit. Why? It costs a ton of money, and it can lead to harm.”
Howard LeWine, writing at the Harvard Health Blog, points out that the group footing a large part of these unnecessary bills is made up of taxpayers, since the screening tests are covered by Medicare. LeWine is kind about the motives of physicians, assuming that they don’t want to be in the position of making decisions for their patients. I’m sure that’s true of many.
Barak Gaster, a doc, is also kind to his peers at the Well blog, in a post wrestling with the PSA testing dilemma. His patients often ask him what he would do. His answer strikes me as a cop-out. He says he tells patients desperate for advice that it’s an individual decision. Sure it is, but I hope he at least lays out for his patients some of the questions they have to answer for themselves. Such as, could they (and their partners) live with impotence?
The human microbiome, hyped
I am already feeling nostalgic about the blog version of the Knight Science Journalism Tracker, which I wrote about here at On Science Blogs last week because it is going out of business next week. So here’s a farewell comment on a Tracker blog post on a topic dear to my heart, the human microbiome.
The specific subject is the hype surrounding our resident microbes, which I wrote about here at On Science Blogs in May. That post described microbiologist Jonathan Eisen’s continuing campaign against overselling the microbiome.
The new Tracker post is from Paul Raeburn, and he speaks of a Nature commentary about microbiome hype. The author, William P. Hanage, argues that those interpreting research on the body’s microscopic communities, science journalists for example, should ask five questions to guard against hype.
Raeburn discusses one of Hanage’s questions, and another is one that I want to keep harping on because it’s a continuing error in reporting on science: whether a study shows that the factor under study causes a condition or only correlates with it.
Skeptical OB Amy Tuteur is even more emphatic in a post called “Why you shouldn’t believe anything you read about the microbiome.” That’s a high level of skepticism indeed, and I doubt even Eisen would go that far. She does, however, make an intriguing point that is not much discussed, noting that the human microbiome has its own microbiome, which she calls the virome. “[I]t is the interaction between the bacteria and the viruses that prey on them and on human beings, that determines health and disease.”
Microbes as puppet masters
Still, microbial science marches on and so does reporting on it. Not to mention speculation.
Meredith Knight, at the Genetic Literacy Project (disclosure: which I also write for), discussed the new study showing that “giving mice antibiotics early in life shifted the bacterial balance of their guts enough to make them twice as likely to have obesity as adults.” One of the potential links here is to the near-universal practice of feeding livestock antibiotics to make them grow bigger faster. Another is the possibility that normal treatment of an infection in infancy might result in a fat adult.
And then there is Carl Zimmer’s speculation about whether our microbes are manipulating our behavior to suit their own ends. Here’s his brief post on his blog The Loom, which links to his New York Times column on microbes as puppet masters. An exasperated commenter takes him to task for anthropomorphizing our microbes.
Regular Zimmer readers will recall that he also views our parasites, for example Toxoplasma gondii, as puppet masters. Here’s an intriguing question that I hope Zimmer will tackle next: are our parasites and our microbes battling it out for control of our behavior? Does that internal war dispose of our last claim to having free will?
Nevertheless I will assert what remains of my free will and take next Friday (August 29) off. It’s the beginning of Labor Day weekend, and so I shall do no labor. Back here September 5.