Bad science reporting, and “disease of the day”

Here we go again:

According to a new study released by researchers at Rutgers University in New Jersey, a firm link has been found between very common moisturizers and the development of skin cancer.

Researchers were led by Dr. Allan Conney from Rutgers University, as they ran a test on mice to see what type of effect these moisturizers had on their bodies.

They tested four common types, Dermavan, Eucerin, Vanicream, and Dermabase. What they found was that all four were linked to the development of skin cancer tumors on the mice tested in the study.

A “firm link,” huh? Before panicing, let’s Google up a better explanation of how the study actually worked^[1]:

For the study, Conney's team exposed hairless mice to an extended period of UV radiation, which induced non-melanoma skin cancer. After stopping UV treatment, they applied four different common brands of skin moisturizers to the animals’ skin five days a week for 17 weeks.

The researchers found that mice treated with skin moisturizers showed an increased rate of tumor formation. In addition, there were more tumors on the animals treated with moisturizers than on the mice that were only given UV radiation.

So it appears (to this layman) that, based on this study’s findings, we can conclude:

• If you already have skin cancer (or, perhaps, pre-cancerous UV-induced skin damage),
• some of the ingredients in moisturizers promote tumor growth, or accelerate tumor development,
  
• provided you’re a hairless mouse.

So far, that’s it. But it was instructive to look at how this (very preliminary) study has been covered in the press. For my findings, read on...


The dBTechno post quoted above (167 words, written by “admin”), besides carrying the misleading headline “Firm Link Found Between Moisturizers, Skin Cancer,” misunderstands how the study worked, and goes for sensation at the expense of accuracy:

What they did was to expose hairless mice to UV rays, similar to what they would get from sun exposure.

They then applied the moisturizers to the mice.

They found that the moisturizers caused non-melanoma skin cancer.

Which misses the fact that the mice were already cancerous (or pre-cancerous) before the moisturizers were applied. Oops!

Over at Fox News, we get a slightly-less deceptive headline (“Study: Popular Moisturizers May Cause Skin Cancer”) and a better description of the study, but a botch of its conclusion:

Conney and colleagues stumbled across the findings after they exposed hairless mice to ultraviolet radiation to mimic sun exposure. Afterward, they applied four popular moisturizers to the mice.

What they found was that all four — Dermabase, Dermavan, Eucerin and Vanicream – caused tumors to grow on the mice.

Well, erm, not exactly. Betcha 99% of the readers interpret that sentence as “all four caused tumors.”  Could have been fixed by adding the word “existing,” as in â€œcaused existing tumors to grow,”  which would still not completely accurate, but would be closer to the study’s conclusion– and less alarmist.

CBS (posting a report originating at WebMD) did better, provided you drill down past the sensation and the cries for more regulation. The story leads with:

Four commonly used moisturizers promoted skin cancers in mouse studies.

...again leaving out the UV. Then, after 12 paragraphs (including one that frets: “Moisturizers are classified as cosmetics by the FDA, which does not require that they undergo the same safety and efficacy tests required for drugs.”), we get the real story:

The moisturizers did not cause cancer in the mice. That came from their early-life radiation exposure. But the creams did make skin cancers grow faster and more readily.

WebMD’s reporter does make an stab at being non-alarmist. There is this warning against premature panic from skin cancer expert Dr. Kevyan Nouri^[2]:

“As we get older, our skin gets drier,” he says. “We need to moisturize, otherwise our skin gets dry and we get eczema, dermatitis, rashes, and so on. It is too soon to say from this study people should stop moisturizing.”

Unfortunately the quote is three screens down, just before the disclaimers from the creams’ manufacturers.^[3] And most people would walk away from the article believing that cosmetics aren’t tested for safety.^[4]

The print press does it better. This Tulsa World article (actually a reprint from the Hackensack Record^[5]) leads off with:

Moisturizers may speed skin cancer in reformed sunbathers, even years after giving up on tanning, according to a Rutgers University study.

...and actually finds an authority who is willing to raise questions on-the-record:

The chief of dermatology at Hackensack University Medical Center said that many doctors recommend moisturizers and would question the study.

“I disagree with the article,” Dr. Margaret Ravits said Thursday. “I’ve been in practice 30 years, and we don’t find any problem with using moisturizers.”

Ravits said she was “kind of surprised” that a respected research journal would publish the study.

What about elsewhere on the web? HealthScout carries a well-balanced report from HealthDay:

Common moisturizing creams helped skin cancers spread and tumors grow in mice exposed to UV radiation, researchers at Rutgers University reported Thursday.

“These creams we tested have tumorigenic [tumor-causing capability] activities,” said lead researcher Allan H. Conney, from the university's Susan Lehman Cullman Laboratory for Cancer Research.

But, he added, “I need to emphasize that what we have done is only in mice. We don't know what the implications are for humans. But it does raise a red flag...”

Right there in paragraph 3, a caution from the study’s author! Very good!^[6]

The best electronic-media report I found was this one from Australian Broadcasting. In addition to finding outside experts to put things in balance:

ASHLEY HALL [presenter]: Gavin Greenoak is the managing and scientific director of the Australian Photobiology Testing Facility at the University of Sydney, where researchers test the effects of light on skin, and evaluate the protective capabilities of sunscreens and cosmetics.

GAVIN GREENOAK: One finds considerable inadequacies in, I believe, the way these studies have been undertaken. And the more one looks through it, the meaning that one might conclude from the data dissolves in those inadequacies.

...it features this straight-up question-and-answer:

ASHLEY HALL: So should moisturiser users be concerned?

ALLAN CONNEY: I guess the basic question is, “can we apply our data to humans?” and we don’t know the answer to that...

What’s the significance of all this? It’s obvious: Cancer scares people. Inaccurate or sensational reports scare people unnecessarily. Frightened people not only make foolish choices,^[7] but they waste their lives worrying over nothing. As C.G. Hill notes:

A coworker seemed alarmed while reading this over my shoulder, as it were; I pointed out that she was a black woman, not a white mouse, and therefore the situation might not apply.

I also suggested that everyone gets cancer, and that those who don't die from it died from something else first.

Neither of these observations made her feel any better.

Via: C.G.Hill

-----

Posted by: Old Grouch in Rants at 21:36:29 GMT | No Comments | Add Comment
Post contains 1248 words, total size 14 kb.