Health News Review

Knowledge Comes and Knowledge Goes

A guest post by Arnon Krongrad, MD, urologist and coauthor of the selenium study cited below.

Once, the earth was flat and everybody knew it. Today, the earth is round and everybody knows it. Knowledge comes and knowledge goes.

So it is with health knowledge. Consider selenium, found in soil in inverse relationship to the incidence of skin cancer. Someone hypothesized that adding selenium to diet would reduce the incidence of skin cancer. Thus was hatched a multi-center, randomized, placebo-controlled trial that tested and found that selenium made no difference to the incidence of skin cancer.

As a secondary endpoint, the trial found that prostate cancer incidence was substantially lower in the selenium arm. There was a problem: This was a patterns-of-care study. This meant that, contrasted with designs of future studies of such drugs as finasteride, the selenium study did not require prostate biopsy: It was possible that selenium concealed – not prevented – prostate cancer.

No matter. In the wake of that study, selenium flew off the shelves. The people selling selenium knew what they wanted to know, regardless of primary data.

Then a new trial, SELECT, was instituted to better look at selenium and prostate cancer. This better study found that selenium makes no difference to prostate cancer incidence and it was summarily terminated. No matter: Various interests continued to promote selenium in prostate health, even in the face of overwhelming evidence that such promotion was contradicted by science.

Journalists still perpetuate the distribution of false knowledge in regards to selenium and prostate cancer. Witness a recent article from Staten Island Live. In misrepresenting current scientific knowledge about selenium and prostate cancer the article is broadly and consistently wrong. For example, it incorrectly identifies Bill Fair as the man behind the selenium story; years ago Dr. Fair did study zinc in relationship to infections, not selenium in relationship to cancer. The article omits the critical fact that SELECT negated the original hypothesis. How can false health knowledge be discarded if health reporters steadily ignore the facts?

Sometimes false knowledge lingers not from ignorance but elusive, other reasons. Consider antibiotics and chronic prostatitis, a condition associated with organ-specific pain and non-organ-specific complaints such as fatigue, irritable bowel, and social isolation. We know that antibiotics usually work no better than placebo, yet we prescribe antibiotics in wild excess. We know the knowledge is wrong but do not change our behavior to accommodate this.

Why are we fixed in old, false knowledge when we know it? As in the case of chronic prostatitis, one common problem may be the absence of effective alternatives. And so antibiotics, alpha-blockers, prostate massage, all minimally effective, are commonly prescribed. And when patients do not get better, they’re shuffled off to the next consultant as if modern-day lepers belonging to an international, itinerant colony. They disappear from view, which makes it easier to keep on keeping on with the old and ineffective.

At the nexus of old and new knowledge is scientific research. As we seek to unring the bells of selenium and antibiotics, we look at fish oil for prostate cancer prevention, laparoscopic surgery for chronic prostatitis treatment, and many other proposed new solutions. We start with the clues found in anecdote and, guided by hypothesis, grow them into case series and clinical trials and broad reproduction until maybe, one day, we arrive at knowledge that can endure for the ages, but not always.

Once, gastric ulcer was a surgical disease and everybody knew it. Today, gastric ulcer is an infectious disease and everybody knows it. Knowledge comes and knowledge goes. Health journalists should faithfully cover knowledge coming and knowledge going.

Copyright © 2010 Arnon Krongrad, MD

Publisher’s note: We don’t often turn over our blog to guest bloggers. But when someone makes a convincing case about a problem in the way health news and information is communicated, I will consider it. I don’t necessarily agree with Dr. Krongrad’s characterization of the article in question as “journalism.” It is, after all, written by another urologist who appears to be a columnist for a website. That doesn’t make it journalism. But that also doesn’t make it irrelevant. As more people turn to the Web for their health information, they come across many sources. And it behooves all of us to consider the source – whether it be Dr. Krongrad or Dr. Motta or me.


Elaine Schattner, M.D. posted on September 22, 2010 at 12:11 pm

Great post! Medical knowledge evolves; people read and believe what they choose.

Alan posted on September 23, 2010 at 1:25 pm

All the above may be true but I don’t think we should throw out the baby with the bathwater. I’m surprised to find this coverage here on this blog. The SELECT study found that selenium supplementation made no difference in prostate cancer incidence. For whom, one wonders? And it’s a “better study”? Really? The post above gives the impression that SELECT settled the matter and there is a scientific consensus on the issue.
In North America dietary intake of selenium is quite high. The population in the SELECT study had a fairly healthy level of selenium intake to begin with. Isn’t that a problem? Check out the medical literature and you’ll see there’s plenty of criticism of SELECT on these grounds. e.g.
“SELECT found no decline in prostate cancer….the relatively high initial levels of selenium in the enrolled men may have contributed to this outcome.”
Hatfield, Dolph L, and Vadim N Gladyshev. 2009. “The Outcome of Selenium and Vitamin E Cancer Prevention Trial (SELECT) reveals the need for better understanding of selenium biology.” Molecular Interventions 9:18-21.
Here’s the results from another study that takes baseline levels of selenium into account:
“In this large prospective study based on a representative sample of the US population, we identified a nonlinear association of serum selenium with all-cause and cancer mortality. At serum selenium levels of less than 130 ng/mL, increases in serum selenium level were associated with a reduced risk of all-cause and cancer mortality. The association was essentially flat at serum selenium levels from 130 to 150 ng/mL. At levels greater than 150 ng/mL, there was a small positive association between serum selenium levels and all-cause and cancer mortality.”
Joachim Bleys; Ana Navas-Acien; Eliseo Guallar. Serum Selenium Levels and All-Cause, Cancer, and Cardiovascular Mortality Among US Adults. Arch Intern Med. 2008;168(4):404-410.
As the first article states, there’s more to be discovered about the selenium and its role in human health, but there’s plenty of evidence already to suggest that if you have a low intake of selenium taking a supplement is probably a good idea. For example, it might be a good idea if you live in the UK where the dietary intake is about 1/3 or 1/4 of the intake in North America and well below recommended intake levels.

Gary Schwitzer posted on September 23, 2010 at 1:59 pm

Thanks for your note.
I don’t understand why you’re “surprised to find this coverage here on this blog.”
I have always supported the open marketplace of ideas and the exchange of opinions, just as I accept yours by posting your comment.
As noted, when someone raises a compelling and well thought-out concern about health care news and information, I will consider posting it.
Which is all I did in this case.
Thanks for your continued interest.

Arnon Krongrad, MD posted on September 23, 2010 at 3:57 pm

You raise a good point: Just because something is not therapeutic today in this group of patients does not mean that it will not be therapeutic tomorrow in another group of patients with the same disease or in a group of patients with a different disease. In treatment as in riflery, targeting matters.
The corollary point is that dosing also matters. The issue of quantitative characterization and optimization of the intervention, like the issue of treatment targeting, is routinely glossed over by scientists and journalists alike. Given that in the wrong doses just about anything — water, oxygen, aspirin -– can harm, this is a potentially serious flaw in our discourse.
Which “baby” is selenium: Lovable retriever or fratricidal hyena?
The SELECT trial does not specifically answer this question. However, it does note an increased risk of diabetes in men taking on selenium and more prostate cancer in men taking only vitamin E ( Ironic, if not conclusive, and a reminder to be cautious about assumptions of safety.
Perhaps more directly, another study makes the point that all is not what it may seem with selenium ( Nested within the National Institutes of Health (NIH)-AARP Diet and Health Study was the finding of increased risk of aggressive prostate cancer with high multivitamin consumption relative to no vitamin consumption. The risk was greatest in men who took individual micronutrients, including selenium. Yet another reminder that selenium supplementation may be dangerous.
Overall, to the best of our aggregated apparent knowledge, there are no data to show that selenium supplementation prevents prostate cancer. There are data to show that supplementation as dosed and targeted in the community and in clinical trials actually promotes the development and growth of prostate cancer – including aggressive and fatal prostate cancer. Perhaps we will one day identify a population – not North American, not veterans, low level of dietary selenium intake – that will benefit from selenium supplementation. We have not done so yet.
This is perhaps a good time to reflect on the differences between epidemiological observations and intervention trials. If epidemiologically generated hypothesis – selenium prevents cancer because there is no cancer where diet contains a lot of selenium – was sufficient, we could stop doing trials (see “The Ghost of Larry Clark” ). And to reflect that over-the-counter supplements are potentially deadly (see “Death by Supplements” ).
Correct. The epidemiological data did “suggest that if you have a low intake of selenium taking a supplement is probably a good idea.” With reference to prostate cancer prevention, at the doses used and on the populations targeted, this suggestion has been tested and refuted.

Alan posted on September 23, 2010 at 7:41 pm

I’m a fan of this blog! In this case I just thought there was a lack of context, one of the main problems in much of the journalism rightly criticized here.
Thanks for your response. I think we are largely in agreement. Selenium supplements may have no impact on health or even a negative impact. For some people, however, boosting selenium levels may have significant health benefits. Who exactly those people are and what those benefits might be exactly is still being defined.
More on prostate cancer risk and selenium:
The conclusion from the NPC trial: “To the end of the blinded treatment the NPC trial continued to show a significant protective effect of SS on the overall incidence of prostate cancer, although the effect was restricted to those with lower baseline PSA and plasma selenium concentrations.”
Duffield-Lillico AJ, Dalkin BL, Reid ME, Turnbull BW, Slate EH, Jacobs ET, Marshall JR, Clark LC; Nutritional Prevention of Cancer Study Group. Selenium supplementation, baseline plasma selenium status and incidence of prostate cancer: an analysis of the complete treatment period of the Nutritional Prevention of Cancer Trial. BJU Int. 2003 May;91(7):608-12.
See also discussion here:
Bekaert B, Rayman MP. Plasma selenium concentration and prostate cancer risk. Am J Clin Nutr. 2009 Apr;89(4):1276-7
There’s also interesting work being done on the impact of polymorphisms in selenoprotein genes on prostate cancer.
Penney KL, Schumacher FR, Li H, Kraft P, Morris JS, Kurth T, Mucci LA, Hunter DJ, Kantoff PW, Stampfer MJ, Ma J. A large prospective study of SEP15 genetic variation, interaction with plasma selenium levels, and prostate cancer risk and survival. Cancer Prev Res (Phila). 2010 May;3(5):604-10.

gzuckier posted on September 28, 2010 at 2:43 pm

This is another example of our tendency to adopt wide-ranging assumptions based on downstream evidence when we have no real model underlying the upstream disease process. Prostate cancer associated with low selenium levels? Then raise selenium levels to reduce prostate cancer! Alzheimer’s characterized by amyloid plaques? Then inhibit amyloid production, and inhibit Alzheimer’s! Chronic cerebrospinal venous insufficiency associated with MS? Then stent the jugular to ameliorate MS!
Not that Hail Mary plays are totally inappropriate for diseases where there isn’t any other hope, but it should be recognized that these are at the level of banging the TV when it won’t turn on, and shouldn’t be allowed to divert research efforts from a real fix based on an understanding of the underlying process/defects.