Why a Resource Like NutritionFacts.org is So Necessary

In a study of the dietary advice given by newspapers in the United Kingdom, “no credible scientific basis” was found for most claims. Indeed, “[m]isreporting of dietary advice…is widespread and may contribute to public misconceptions about food and health”—and potentially not only the public.

Scientists like to think they are not influenced by popular media. One study decided to put it to the test. The New York Times reports on scientific research each week, and researchers found that the studies covered by the Times end up being cited in the New England Journal of Medicine more than those that don’t. Seems like the popular press does indeed have an impact on science? Not so fast. That’s just one potential explanation. Perhaps outstanding studies are more likely to be picked up by the media and, independently, more likely to be cited. It’s possible the Times was just earmarking important science and its publicizing of that research didn’t have any effect on how often it was cited in future studies.

How can we disentangle the two? In 1978, there was a three-month strike during which the Times continued to print copies but couldn’t sell them to the public. So, a natural experiment was set up. Researchers compared the number of citations of Journal articles published during the strike with the number published when the paper wasn’t on strike to “discover whether publicity in the popular press truly amplifies the transmission of scientific findings to the medical community.” If the paper were just earmarking important articles, then the strike would have no effect on the studies’ future impact, but that’s not what happened. As you can see from a graph shown in my video Spin Doctors: How the Media Reports on Medicine, the studies covered by the Times during the strike when no one could read them appeared to have no impact on the medical community.

The next question, of course, is whether the press is simply amplifying the medical information to the scientific community or distorting it as well? “[S]ystematic studies suggest that many stories about new medicines tend to overstate benefits, understate risks and costs, and fail to disclose relevant financial ties.” What’s more, “[o]verly rosy coverage of drugs may also result from the direct and indirect relations between journalists and drug companies”—that is, the financial ties between the reporters and Big Pharma with all its perks.

Scientists and physicians often blame the press for the public being “poorly served” by the media’s coverage of medical science. In fact, the famous physician William Osler was quoted as saying, “Believe nothing that you see in the newspapers…if you see anything in them that you know is true, begin to doubt it at once.” Both parties, however, share the blame. Reporters may only have an hour or two to put together a story, so they may rely on press releases. It’s not hard to imagine how drug company press releases might be biased. But, surely, press releases from the scientists themselves and their institutions would “present the facts fairly, unambiguously, and without spin,” right?

Researchers decided to put it to the test. Critics may blame the media, but where do you think the media gets its information? “One might assume” that press releases from prestigious academic medical centers would be “measured and unexaggerated,” but researchers found they suffered from the same problems: downplaying side effects, having conflicts of interest and study limitations, and “promot[ing] research that has uncertain relevance to human health…”

For example, most “animal or laboratory studies…explicitly claimed relevance to human health, yet 90% lacked caveats about extrapolating results to people.” Indeed, “a release about a study of ultrasonography [ultrasound] reducing tumors in mice, titled ‘Researchers study the use of ultrasound for treatment of cancer,’” failed to add “for your pet mouse.”

“For animal research, it is estimated that less than 10% of non-human investigations ever succeed in being translated to human clinical use. Over-selling the results of non-human [lab animal] studies as a promised cure potentially confuses readers and might contribute to disillusionment with science.”

Although it is common to blame the media for exaggerations, most times, they don’t just make it up—it is what the research institutions are sending out themselves. Researchers found that “most of the inflation detected in our study…was already present in the text of the in their own press releases produced by academics and their establishments.” Medical journals, too. Indeed, sometimes medical journal press releases do more harm than good. An analysis of press releases from some of the most prestigious medical journals found the same litany of problems. I don’t think most people realize that journals sell reprints, which are official-looking copies of the articles they print, to drug companies and others. Reprints can bring in big bucks. Drug companies may buy a million copies of a favorable article. Indeed, they “usually buy reprints of studies that they have funded themselves. Unsurprisingly, they buy them only when the results are positive for their drugs, and they use these reprints as a form of marketing.” What’s more, sometimes a company will submit an article and promise to buy a certain number of reprints if it’s accepted, which “is effectively a bribe…” A long-time editor-in-chief at the prestigious British Medical Journal recalled that a woman from a public relations company called him, offered to take him to a restaurant of his choice, “and stopped just short of saying she would go to bed with me if we took the paper.”

“Another conflict of interest for editors relates to advertising—a major source of income for many journals. Most of the advertising comes from pharmaceutical companies.” If they don’t like a study, they can threaten to withdraw their advertising if it’s published. This potentially leaves editors “faced with the stark choice of agreeing not to publish a particular piece or seeing their journal die.”

Even if journalists, as they’re writing an article, have the time to skip the press releases and go directly to the source to read the studies themselves, they may find them “incomprehensible; utter gobbledygook.” Yet even if they do understand the studies, scientific articles are not simply reports of facts. Authors have many opportunities to add spin to their scientific reports, with “spin” defined as distorting the interpretation of results and misleading readers, either unconsciously or with a willful intent to deceive.  Researchers looked at randomized controlled trials with statistically nonsignificant results, meaning, for example, a drug was compared to a sugar pill and the difference between the two was essentially nonexistent. Would the researchers just lay out the truth and report that they spent time and money, but, in terms of their primary outcome, got nothing? Or would they try to spin it? In 68 percent of cases, they spun it. There was spin in the abstract, the article summary, which is particularly alarming because the abstract is often “the only part of an article [people] actually read.”

Given all of this, it’s no wonder the media often gets it wrong. Spin in the abstracts can turn into spin in the press releases and result in spin in the news. “Therefore, even if journalists [do their due diligence and] are using the original abstract conclusion in good faith, they still run the risk of deceiving their readers.” Researchers presenting new findings can always be careful to stress how preliminary the findings may be. “But let’s be serious. Powerful and reinforcing self-interests” may prevail.

I think the biggest problem with the way the media reports on medicine, though, is the choice of which stories are covered. In 2003, for instance, SARS and bioterrorism killed less than a dozen people, yet generated over a hundred thousand media reports, which is far more than those covering the actual greatest threats to our lives and health. In fact, ironically, “the more commonplace the cause of death, the less likely it is to be covered by the mass media.” Our leading killer is heart disease, yet it can be prevented, treated, and even reversed with diet and lifestyle changes—now that’s what should be front page news.


If we can’t trust the medical literature on its face, where can we turn? We’re talking life-or-death information here. What we need is someone who will dig deep into the data and translate the gobbledygook into actionable tips on keeping us and our families healthy. If only there was a website we could trust to tell us the unbiased truth…

If you appreciate the work we do, please consider supporting us. NutritionFacts.org relies solely on individual donations from users like you!

If you think just a little spin is bad, there is a much deeper rot in the medical literature. For more on this critical topic, see:

Interested in some specific examples of the spin and conflicts of interest we’ve been discussing? See:

It’s no wonder Physicians May Be Missing Their Most Important Tool.

In health,
Michael Greger, M.D.

PS: If you haven’t yet, you can subscribe to my free videos here and watch my live, year-in-review presentations:

Should You Be Concerned about Bovine Leukemia Virus in Milk?

Decades ago, concern was raised that the milk of dairy cows frequently contains a leukemia-causing virus—more specifically, bovine leukemia virus (BLV), the leading cancer killer among dairy cattle. Most U.S. dairy herds are infected with the cancer virus. “Thus the question of whether dairy cows naturally infected with BLV release infectious virus into milk is an important public health consideration” and the subject of my video Is Bovine Leukemia Virus in Milk Infectious?.

Researchers at the University of Pennsylvania decided to put it to the test. And indeed, infectious virus was demonstrated in the milk of 17 of the 24 cows tested, indicating that “humans are often orally exposed to BLV.” Just because we’re exposed to it doesn’t mean it’s causing human disease, though. How do we know BLV can even infect human cells? We didn’t until 1976 when it was discovered that BLV can indeed infect human, chimpanzee, and rhesus monkey cells. Nevertheless, that still doesn’t mean BLV necessarily causes cancer in other species.

Researchers can’t lock human infants in a cage and feed them infected milk, but they can cage infant chimpanzees. Chimps Bois and Roger were fed infected milk, developed leukemia, and died. Until then, we didn’t even know chimps could get leukemia. The fact that BLV-infected milk appeared to transmit or induce leukemia in our closest living relatives certainly did raise the stakes, but human beings are not chimpanzees. Yes, our DNA may be 98 percent identical, but we may share 60 percent of our DNA with a banana. We need human studies.

We can’t do interventional trials in this case, thanks to those pesky Nuremberg principles, but what about observational studies? Do cattle farmers have higher rates of cancer? Apparently so. This finding led some to suggest that “milk- and egg-borne viruses may be highly important in the pathogenesis [or development] of human leukemia and lymphoma,” but farmers may be exposed to all sorts of potential carcinogens, such as pesticides. Large animal veterinarians may also have more leukemia and lymphoma, but some are also “particularly lax in the use of X-ray protective equipment,” so it didn’t necessarily have anything to do with viruses.

We needed so-called serology studies, testing people’s blood for antibodies against the virus, which would prove human exposure, and we got them. Ten different studies looked for BLV antibodies in cancer patients and non-cancer patients, creamery employees versus office employees, veterinarians, unpasteurized milk drinkers, and more. “Not one of these studies found a single individual with antibodies to BLV…” As a result, in 1981, the case was closed: “Therefore, there is strong serological evidence to indicate that BLV is not transmissible to man.” However, the strength of the evidence is only as strong as the strength of the test. Chimpanzees Bois and Roger didn’t develop detectable antibodies either, and they died from BLV.

The tests available a handful of decades ago were not really sensitive. “Clearly, the question of whether BLV poses a public health hazard deserves thorough investigation” using highly sensitive molecular probes. It would take a few decades for us to get such an examination, and I discuss those landmark findings in my videos The Role of Bovine Leukemia in Breast Cancer and Industry Response to Bovine Leukemia Virus in Breast Cancer.


Thankfully, feline leukemia virus does not appear to be transmissible. See Pets and Human Lymphoma.

In health,
Michael Greger, M.D.

PS: If you haven’t yet, you can subscribe to my free videos here and watch my live, year-in-review presentations:

The Healthiest Way to Eat Paleo

There have been about a half dozen studies published on Paleo-type diets, starting around 20 years ago. For example, in what sounds like a reality TV show: ten diabetic Australian Aborigines were dropped off in a remote location to fend for themselves, hunting and gathering foods like figs and crocodiles.

In Modern Meat Not Ahead of the Game, my video on wild game, I showed that kangaroo meat causes a significantly smaller spike of inflammation compared to retail meat like beef. Of course, ideally we’d eat anti-inflammatory foods, but wild game is so low in fat that you can design a game-based diet with under 7 percent of calories from fat. Skinless chicken breast, in comparison, has 14 times more fat than kangaroo meat. So you can eat curried kangaroo with your cantaloupe (as they did in the study) and drop your cholesterol almost as much as eating vegetarian.

So, how did the “contestants” do? Well, nearly anything would have been preferable to the diet they were eating before, which was centered on refined carbs, soda, beer, milk, and cheap fatty meat. They did pretty well, though, showing a significantly better blood sugar response—but it was due to a ton of weight loss because they were starving. Evidently, they couldn’t catch enough kangaroos, so even if they had been running around the desert for seven weeks on 1,200 daily calories of their original junky diet, they may have done just as well. We’ll never know, though, because there was no control group.

Some of the other Paleo studies have the same problem: They’re small and short with no control groups, yet still report favorable results. The findings of one such study are no surprise, given that subjects cut their saturated fat intake in half, presumably because they cut out so much cheese, sausage, or ice cream. In another study, nine people went Paleo for ten days. They halved their saturated fat and salt intake, and, as one might expect, their cholesterol and blood pressure dropped.

The longest Paleo study had been only 3 months in duration, until a 15-month study was conducted—but it was done on pigs. The pigs did better because they gained less weight on the Paleo diet. Why? Because they fed the Paleo group 20 percent fewer calories. The improvement in insulin sensitivity in pigs was not reproduced in a study on people, however. Although, there were some benefits like improved glucose tolerance, thanks to these dietary changes: The Paleo group ate less dairy, cereals, oil, and margarine, and ate more fruits and nuts, with no significant change in meat consumption.

A follow-up study also failed to find improved glucose tolerance in the Paleo group over the control group, but did show other risk factor benefits. And no wonder! Any diet cutting out dairy, doughnuts, oil, sugar, candy, soda, beer, and salt is likely to make people healthier and feel better. In my video Paleo Diet Studies Show Benefits, you can see a day’s worth of food on the Standard American Diet, filled with pizza, soda, burgers, processed foods, and sweets, versus a Paleo diet, which, surprisingly, has lots of foods that actually grew out of the ground.

But the Paleo diet also prohibits beans. Should we really be telling people to stop eating beans? Well, it seems hardly anyone eats them anyway. Only about 1 in 200 middle-aged American women get enough, with more than 96 percent of Americans not even reaching the minimum recommended amount. So telling people to stop isn’t going to change their diet very much. I’m all for condemning the Standard American Diet’s refined carbs, “nonhuman mammalian milk”, and junk foods, but proscribing legumes is a mistake. As I’ve noted before, beans, split peas, chickpeas, and lentils may be the most important dietary predictor of survival. Beans and whole grains are the dietary cornerstones of the longest living populations on Earth. Plant-based diets in general and legumes in particular are a common thread among longevity blue zones around the world.

The bottom line may be that reaching for a serving of kangaroo may be better than a cheese danish, “but foraging for…[an] apple might prove to be the most therapeutic of all.”


I’ve reported previously on Paleo’s disappointing results in Paleo Diets May Negate Benefits of Exercise.

The underlying philosophy behind “caveman” diets may be flawed in the first place. See:

So, What’s the Natural Human Diet? Watch the video!

The wild game video I mentioned is Modern Meat Not Ahead of the Game. Kangaroo is kind of the Australian version of venison. Note that it also matters how the animals are killed. See Filled Full of Lead and Lead Contamination in Fish and Game.

And, for more on the musical fruit, see:

In health,
Michael Greger, M.D.

PS: If you haven’t yet, you can subscribe to my free videos here and watch my live, year-in-review presentations: