This spring, Dr. Johannes Bohannon and a team of German scientists discovered that people on low-carbohydrate diets could lose weight faster if they used one weird trick: Eat a bar of chocolate every day.
Newsrooms around the world responded eagerly to Bohannon’s findings.
“Excellent News: Chocolate Can Help You Lose Weight!” Huffington Post India declared in a report.
The U.K.’s Daily Mail blared in a headline, “Pass the Easter Egg! New study reveals that eating chocolate doesn’t affect your Body Mass Index…and can even help you LOSE weight!”
In the United States, Modern Healthcare wrote, “Dieting? Don’t forget the chocolate.”
The story continued to grow, with news of the sweet discovery spreading from the Internet to print and television. Even Europe’s highest-circulation newspaper, Bild, got in on the action, publishing a report titled “Slim by Chocolate!”
Journalists and readers looked past the too-good-to-be-true nature of the findings and devoured the story wholesale.
But Bohannon’s research was a hoax.
The health study was deliberately faked to test the hypothesis that scientists and reporters rarely detect junk science. No one caught on to this ruse.
“Our point was not that journalists could be tricked by fakers, but rather that scientists themselves in this field and other fields are making the kinds of mistakes that we made on purpose,” said Bohannon, a journalist whose real first name is John and who holds a Ph.D. in molecular biology. “This whole area of science has become kind of corrupted by really poor standards between scientists and journalists.”
He told the Washington Examiner that his interest in the project was rooted in personal experience. His mother was duped by a scientifically dubious fad diet, which left her kidneys badly damaged.
Dismayed that no scientists or reporters had questioned the scientific claims of the diet, Bohannon began wondering how many other studies might not have been vetted. So when a German television reporter named Peter Onneken approached him with an idea to find out how easy it might be to advance bad science in the marketplace, he was all in on the idea.
“There are smart people out there who are getting fooled by this stuff because they think scientists know what they’re doing,” he told the Examiner.
Operating under a fake organization called the “Institute of Diet and Health,” Bohannon and his co-conspirators contacted actual test subjects and performed experiments. They even crafted clever and convincing, but bogus, press releases.
Then came the real test.
Bohannon and his collaborators purposely falsified some of their data and left crucial details out of their press release to see how many in the media would notice. It turned out to be none.
Not a single person double-checked his research, he said. No one sought comment from independent experts. No one asked him about possible inaccuracies in his work.
“I was kind of shocked at how bad the reporting is,” he said. “I didn’t realize how bad people who call themselves proper journalists are at covering this beat.”
The problem wasn’t unique to online publications eager for a few quick clicks, Bohannon said. Even reputable publications that employ fact-checkers skimmed over the details of his research.
“Right now, there’s absolutely no accountability,” he said. “The bulls— is just flooding. And it’s flooding out of these media venues and no one gets any push back.”
State of media
Next to Congress, the least-trusted institutions in the United States include newspapers, television and Internet news, according to Gallup.
Americans barely trust their own local news affiliates, let alone the word of a network anchor hundreds or thousands of miles away in New York City, according to Pew polling data. The media credibility gap is constantly widening, with public opinion of the press continually slipping to new lows.
But Americans have a generally favorable view of science.
It should come as no surprise, then, that the press would latch on to science to help it reestablish trust with viewers, which may explain the rise of self-proclaimed “nerds” and “wonks” in media.
NBC News’ Chuck Todd on Sundays likes to consult the “nerd screen.” MSNBC’s Melissa Harris-Perry likes to refer to her weekend show as “#nerdland.” Ezra Klein, who used to head the Washington Post’s “wonkblog,” now runs a website that promises to explain the complexities of the world, one Excel spreadsheet at a time.
Headlines increasingly come with the promise of explaining any given topic with “science.”
“Science explains why your brain wants you to go on Facebook for a break from work,” reads one headline. Another reads, “Science explains how time spent outdoors colors your view of #thedress,” while yet another reads, “Science finally explains why men exist.”
Yet, as Bohannon concluded from his experiment, most reporters aren’t interested in the finer points of scientific research. They want a sexy story, and if it can be based on a catch-all authority, even better.
“There has been an undeniable decline of science journalism,” he said, noting the small number of reporters who cover this beat. “Which is interesting as we’re talking about the rise of the nerds. It’s weird, because don’t you think that now would be the time for science journalists?”
The consequences of junk science include more than just the spread of bad information or embarrassment for media outlets. Members of the public who absorb news reports disseminating bad science can suffer ill health effects, like Bohannon’s mother.
There are other consequences. One is the adoption of bad policies.
Politics, pesticides and power
In April 2013, the European Commission voted to ban the three main products in a specific class of pesticide known as “neonicotinoids.” News media cheered the outlawing of the pesticides as a step in the right direction to save Europe’s honeybee population.
Two years later, the E.U. is considering whether to scrap the ban. In salivating over the sensational story of the supposedly deadly chemicals, the press failed to question whether the ban was even necessary in the first place.
Not only have the number of hives in Europe increased since the ban was enacted, but the number of bees lost annually is mostly unchanged, according to the Food and Agricultural Organization of the United Nations.
Yet farmers have also seen an increase in infestations and a decrease in crop yields due to the ban.
The ban may not have ever been put into place if the Fourth Estate had been doing its job.
In addition to ignoring the all-important question of whether bee numbers were actually declining, the press missed a red flag signaling that politics, rather than science, was driving the ban.
Public support in the European Union for a ban of neonicotinoids began building in January 2013 following extensive media coverage of a European Food Safety Authority report.
But the findings of the regulatory body’s risk assessment of the chemicals were badly misreported. As is common with lengthy and technical studies, the risk assessment was made available with an accompanying press release.
“The press release, which is all most people would have seen, misrepresented the [European Food Safety Authority] report,” said E.U. Referendum’s Richard North.
The press release claimed that the study found evidence that three chemicals posed risks to bees. For thiamethoxam, this wasn’t true. For the other two chemicals, clothianidin and imidacloprid, it was an overstatement.
Further, North said, the release went “far beyond the terms of the report, and indeed the agency’s remit. Its role was supposed to be limited to reviewing the evidence, not to define the acceptability of risks — that was the responsibility of the commission and member states.”
Media hardly noticed.
Reporters focused more on the press release’s mischaracterization of the study than on what the document said. The press also ignored the crucial point that the risk assessment lacked the necessary data to come to a conclusion, data which later undermined a scientific case for a ban.
In addition to lacking key data, critics of the press’ handling of the report also pointed out that the assessment failed to account for several key studies that exonerated the chemicals in question.
It didn’t matter. Journalists rushed to report that the chemicals were bad for pollinators.
The inaccuracies stuck, the public picked a side and the Union-wide ban was put in place, despite a lack of scientific justification and a bevy of likely economic repercussions.
A German research group said in a study that the economic consequences of the ban could be severe. “[I]f neonicotinoid seed treatments were no longer available in Europe, there would be a significant reduction of food production, dramatically altering the commodities trade balance,” the Humboldt Forum for Food and Agriculture reported.
If the ban is left in place over a five-year period, it could cost the European Union up to $19 billion, farmer income would decline by 5 percent and at least 50,000 jobs would be lost, the group added.
Gas on the fire
Once media outlets latch on to a tantalizing piece of garbage research, it’s practically impossible to stop the flow of bad information.
“If junk science existed in a vacuum, it wouldn’t be so harmful,” says Junkscience.com founder Steve Milloy, who has followed the effect of bad science on the public and regulatory policy for more than 25 years.
“The digital age has made it easier to spread junk science. The bad news is that it’s really easy to get it out there,” he told the Examiner. “It’s a real battle out there. It’s always hard to turn the ship around.”
Milloy listed examples of bad data that fooled reporters, scientists or both.
He has helped debunk, for example, dietary scares about cholesterol (“We stopped eating eggs in the 70s!”), reports that dietary fiber prevents colon cancer (debunked in 2013), claims that electric power lines cause cancer and, more recently, claims that vaccinations cause autism.
The vaccination scare has been linked directly to a recent measles outbreak in California.
“I can’t think of a single major environmental, dietary or public health story in my career that has not been driven in some way by junk science,” he said. “But we are still awash in junk science that affects our health or costs us money, peace of mind and our liberties.”
Greg Conko is from the Competitive Enterprise Institute, a free-market think tank dedicated to fighting bad and unnecessary regulation. He said junk science is “much more prevalent today than it was 20 years ago. There are more media outlets and there are more people putting out content.”
Garbage science is quickly amplified and distributed far and wide by a media that can be uncritical in the spread of dubious and, sometimes, outright false information.
In a digital news age, when websites are fighting furiously for clicks and traffic, reporters are often required to operate under harsh deadlines, produce content quickly and move on to the next story.
“We’re perpetually seeing stories about ‘X’ will cure cancer or ‘Y’ will give you cancer,” Conko said. “It’s all junk science. But it’s exciting news that sells newspapers and attracts eyes to television shows and the web.”
Reporters “will take media releases from journals and institutions and basically regurgitate them,” Milloy said.
Though some newsrooms occasionally apologize for publishing what later turn out to be false scientific reports, they will never apologize for pouncing on a juicy headline.
Junk science often gives them exactly that. But long after the outlet moves on to the next headline, misinformation lingers.
The death of the bee
Though the European Union is considering lifting its ban on neonicotinoids, the U.S. Environmental Protection Agency is under pressure to restrict the same chemicals for the same reason: to save bees.
After years of headlines about a pending “beemaggedon,” the Obama administration announced a strategy in May to stem what it characterized as an unprecedented decline in the number of America’s pollinators, particularly the honeybee.
Part of the plan includes speeding up EPA’s scheduled review of neonicotinoids.
The press hailed the strategy as a moment of redemption for threatened bee populations.
“After the sting of vanishing bees, White House pollinates protection plan,” CNN reported, discussing the “effort to help the declining bee and butterfly populations.”
“After years of devastation, the American honey bee finally has the White House’s attention,” Quartz reported May 19, stating in a separate article that “the world is finally trying to save the bees.”
Many other outlets welcomed the decision, including the New York Times, National Public Radio and the Wall Street Journal, which reported that there has been a “surge in honeybee deaths.”
As newsrooms reported on the White House’s announcement, few — if any — asked whether there has been an actual decline in honeybees.
“The whole ‘mass death’ thing is off,” biologist and beekeeper Randy Oliver told the Examiner, claiming that media is purposely confusing the issue by not giving full context.
“In the United States, the number of colonies is increasing. Simply look at the number of colonies available for almond pollination each year,” he said. “The acreage of almonds is increasing each year, so the demand for colonies is increasing each year. And it’s all across the world. African countries, Canada, many European countries are increasing their numbers [of colonies].”
Beekeeper and biomedical researcher Peter Borst said the numbers are much better than people are led to believe.
“The number of managed bee hives in the world [have risen] from 50 million in 1960 to more than 80 million today. But this figure only reflects managed colonies, not wild colonies. It is hard to know the real number of ‘unkept’ honeybee colonies in the world,” he wrote in the American Bee Journal, suggesting that Africa has at least 310 million.
Borst told the Examiner, “In most areas where honeybees are kept, the numbers are going up, not down.”
He and Oliver cited several reasonable and non-shocking explanations for past fluctuations in bee numbers, including the drop-off a few decades ago in the number of recreational beekeepers.
“It’s a cyclical thing. People lost interest in [beekeeping] in the ’80s and ’90s, especially when it got to be harder to take care of bees,” Borst said. “Now there’s a huge resurgence in beekeeping as a hobby, because people are reading about it in the papers and now they want to be part of the solution.”
This is not exactly new, he said.
A decline of bees and wasps in England, for example, has been going on for at least a century, Smithsonian’s Sarah Zielinski reported in December.
“Changes in agricultural practices since the 19th century may be a major culprit in the pollinators’ decline,” she wrote in an article titled, “Bees and Wasps in Britain Have Been Disappearing For More Than a Century.”
The same issue of changing agricultural practices holds true in the United States, an important bit of context that the White House fails to account for in its representation of honeybee populations as massively failing.
By comparing current hive numbers to those of the 1940s, the White House claims that bee populations are in a precipitous decline.
Left out of this picture, however, is the fact that the number of farmers, many of whom kept bees, has also declined since the ’40s, as post-war agricultural practices trended toward larger farms, University of Missouri economics professor John Ikerd wrote in Small Farm Today Magazine.
Since the mid-’90s, when the supposedly harmful neonicotinoids hit the market, there has not been a massive drop in the number of honey-producing hives.
Furthermore, recent Department of Agriculture statistics show there were 2.74 million honey-producing hives in the United States in 2014, an increase of 4 percent from 2013.
Honeybee numbers in the United States are at a 20-year high, according to Agriculture statistics.
Separately, the European Academies Science Advisory Council said in a report analyzing Europe’s pollinators that drawing any conclusions about trends from honeybee data “requires a differentiation between ‘losses’ and ‘declines.’ “
“Losses are the deaths of colonies which may occur in the temperate regions especially over winter,” the report reads. “However, declines may occur both in the number of beekeepers or in the numbers of colonies maintained by each beekeeper. The latter are particularly heavily influenced by socioeconomic factors, by the price of honey, the presence or absence of subsidies, or the popularity of beekeeping as a hobby.”
From Oliver’s point of view, reporters don’t appear interested in getting to the bottom of these nuanced and non-sexy details.
Oliver said a cable news correspondent once called him for information on reports that bees were dying off in record numbers.
“I asked him if he wanted the facts or if he just wanted some printable sound bites to makes a sensational story. The reporter pretty much said he wanted the second. The conversation ended after that,” he said.
Weird science: GMOs and activists
Last year, General Mills announced that it would drop ingredients with genetically modified organisms from its products.
The decision was hailed as a step in the right direction, a signal that corporations were at long last becoming more health conscious. The press reported that Cheerios would not have the ingredients, with little analysis of whether the ingredients are actually harmful.
One year later, Chipotle announced that it, too, would go GMO-free. This announcement, however, was met with far more derision.
“By Feeding Bogus GMO Fears, Chipotle Treats Customers Like Idiots,” ready one headline from Reason.
NPR followed up with an article headlined “Why We Can’t Take Chipotle’s GMO Announcement All That Seriously,” arguing that the popular restaurant was merely trying to pander to supposedly health-conscious customers.
Vox, a so-called “explainer” website, responded to Chipotle’s announcement with a story headlined “Chipotle will stop serving GMO foods — despite zero evidence they’re harmful to eat.”
In just a few months, something had changed dramatically. Going GMO-free was no longer laudable. It was now an object of mockery.
For experts such as Kevin Folta, professor of horticulture at the University of Florida, Chipotle’s anti-GMO stance represents the worst of the effect of junk science and sloppy media reporting.
“I’ve been involved in this for 30 years in the genetic engineering business,” he told the Examiner. “What I see is an increased visibility in a sort of tribal response to this. Chipotle is using fear. They’re not telling you what they’re using. They’re telling you what’s not in their product and they’re using it as a marketing weapon against other companies that have chosen to use perfectly safe ingredients.”
The dramatic difference between Cheerios’ announcement and Chipotle’s announcement shows that more Americans are paying attention to the debate.
This is a good thing, Bohannon said, adding that, “We definitely defer too much to scientific authority.”
There’s still a lot of bad information out there. Too much, Folta said.
“These technologies that have been safely implemented for almost 20 years now, without one single case of so much as a sniffle from genetically modified food,” he said. “The safest, most tested products on the market are GMO.”
This can be attributed to media giving airtime to anti-GMO sentiment that doesn’t appear to be grounded in sound science.
French scientist Gilles-√âric S√©ralini, for example, claimed in 2012 that pesticide-treated, genetically-modified corn was causing cancer in lab rats.
“His methodology was nonsense,” Conko said. “S√©ralini used a rodent variety that was specifically bred to develop tumors. The results were all over the map. He had to specifically cherry pick the one result from the study to highlight and put in the abstract.”
S√©ralini’s study was quickly retracted by the journal that published it. Undeterred, the French scientist, who also founded and works for an anti-GMO advocacy organization, had his work republished elsewhere, in the Environmental Sciences Europe.
Despite the many questions surrounding S√©ralini’s work, and the fact that it was retracted by the first journal that published his research, the notion that GMOs cause health problems remains enshrined in popular consciousness.
Reporters trying to assess the latest scientific study may be hindered by a crisis in how science is produced and verified.
“[S]omething has gone fundamentally wrong with one of our greatest human creations,” Dr. Richard Horton, editor-in-chief of prestigious medical journal Lancet, lamented this year. “Much of the scientific literature, perhaps half, may simply be untrue. Afflicted by studies with small sample sizes, tiny effects, invalid exploratory analyses and flagrant conflicts of interest, together with an obsession for pursuing fashionable trends of dubious importance, science has taken a turn towards darkness.”
Horton reported on a symposium exploring the reproducibility and reliability of biomedical research. The conclusions were not encouraging. “A lot of what is published is incorrect,” as Horton quoted an attendee.
Top scientists at the National Institutes of Health reported similar findings in Nature last year.
“The checks and balances that once ensured scientific fidelity have been hobbled. This has compromised the ability of today’s researchers to reproduce others’ findings,” wrote Director Francis Collins and Principal Deputy Director Lawrence Tabak.
It’s easy, for example, for a determined researcher to subvert the all-important peer-review process, all while earning the coveted “peer-reviewed” stamp of approval.
“The big problems we have are predatory journals and journals with very soft editorial and review standards,” Folta told the Examiner. Some of these will publish “just about anything.”
In 2005, two computer scientists, David Mazi√®res and Eddie Kohler, tested the peer-review process by submitting a hoax study titled “Get Me Off Your F—ing Mailing List.”
They made their point: The impressive-sounding International Journal of Advanced Computer Technology published their paper last year.
Despite the fact that the 10-page study consists almost entirely of the phrase “Get me off your f—ing mailing list” repeated over and over again, it was published by a “predatory open-access journal,” as the science blog i09 called it.
Prior to the chocolate hoax, John Bohannon posed as a researcher in 2013 and submitted a bogus work to 304 journals that pride themselves on the peer-review process.
“An unusual move; but it was an unusual paper, concocted wholesale and stuffed with clangers in study design, analysis and interpretation of results. Receiving this dog’s dinner from a fictitious researcher at a made-up university, 157 of the journals accepted it for publication,” the Economist reported that year.
Last year, the Journal of Vibration and Control retracted at least 60 papers after it was revealed that a researcher in Taiwan and others “had exploited peer review so that certain papers were sure to get a positive review for placement in the journal,” Hank Campbell, founder of the science-based website Science 2.0, wrote in a 2014 Wall Street Journal op-ed.
An author of one of the now-pulled works even wrote his own “glowing reviews” (all of it under a false name).
Until 2010, the National Academy of Sciences allowed researchers to select who would review their submissions.
This loose policy allowed University of California, Berkeley, Prof. Tyrone Hayes to choose friend and colleague Prof. David Wake to review his work on his studies published in 2002 and 2010, Campbell told the Examiner.
Hayes claimed to prove that a pesticide called Atrazine was responsible for causing sex changes in frogs, which offered disturbing implications for humans.
Problematically for the integrity of the studies, Wake functionally hand-walked Hayes’ work around the peer-review process, Campbell said.
“There’s no data. Hayes’ work has never been replicated,” Campbell told the Examiner. “All there are are a couple of screenshots. But it was published in the National Academy of Sciences, so of course it’s soon picked up by The New York Times, The New Yorker and so on. The EPA is even told it must conduct an investigation because this product is supposedly harmful.”
The EPA launched multiple reviews of Hayes’ work. Hayes refused to provide the EPA panels with a form of his data they could access. The EPA was unable in any review to replicate his findings.
Nevertheless, despite all the questions surrounding his work, Hayes’ research not only boasts of the “peer review” stamp of approval, but he is also presented by media as something of a heroic crusader: One man fighting a conspiracy of corporations who are determined to bury the truth.
The New Yorker, in a glowing profile of the scientist, reported that “after Tyrone Hayes said that a chemical was harmful, its maker pursued him.” The New Yorker profile makes no mention of allegations that Hayes’ research was hand-walked through the peer-review process.
Following Campbell and others claiming that Hayes had avoided scrutiny, the National Academy of Sciences adjusted its protocols to prevent any further potential conflicts. Data are now required for all submissions.
Hayes, who did not respond to the Examiner‘s request for comment, is just one example of what Horton calls the “endemicity of bad research behavior” afflicting science. In “their quest for telling a compelling story, scientists too often sculpt data to fit their preferred theory of the world,” he wrote in the Lancet.
As Folta noted, “Once something’s published, people will cling to it and say, ‘Well, it’s published in a peer-reviewed journal.’ This thwarts the entire scientific process.”
Yet, despite the troubling state of its checks and balances, science is regularly invoked as an absolute authority, and anyone who challenges the “research” is branded as backward.
President Obama, for example, derisively calls members of Congress who resist the White House’s climate change agenda, “flat-earthers,” despite questionable scientific claims and murky data. Opposing or even questioning carbon taxes and other regulatory actions that impede the energy industry is deliberately portrayed as being “anti-science.”
Who wants to be the lawmaker known as being a science “denier?” The answer, it would appear, is few.
Attacking legislation for being supposedly “anti-science” is common in Congress, meaning it’s a likely effective approach. Perhaps this is why the tactic is applied even to proposals that appear to be based on sound research.
Lawmakers who question the data showing that unborn infants feel pain at 20 weeks of pregnancy, for example, often resist legislation limiting a woman’s “right to choose” by attacking the supporting research for being “anti-science.”
For these members of Congress, it’s not enough that legislation banning abortions during the third trimester is characterized as “anti-woman,” but it must also be “anti-science.”