The damage done A study by Andrew Wakefield, right, helped fuel media attention to the vaccine-autism story, until Brian Deer exposed his work as deeply flawed. (Left: Courtesy of Brian Deer; Right: Anthony Devlin / Associated Press)

In 1998, The Lancet, one of the most respected medical journals, published a study by lead author Andrew Wakefield, a British physician who claimed there might be a link between the vaccine for measles, mumps, and rubella (MMR) and autism, the developmental disorder that afflicts one out of every 88 children in the US. The paper coincided with growing concern among parents in the US and UK about a possible connection between the rising number of childhood vaccinations and the rising rate of autism among kids. Although the trends were only coincidental, Wakefield’s paper helped spark a debate about the supposed link that has played out in the media over the last 15 years.

Among scientists, however, there really was never much of a debate; only a small group of researchers ever even entertained the theory about autism. The coverage rarely emphasized this, if it noted it at all, and instead propagated misunderstanding about vaccines and autism and gave credence to what was largely a manufactured controversy. As Ben Goldacre, a British doctor and media critic, wrote in his 2008 bestseller, Bad Science: “[Y]ou will see news reporters, including the BBC, saying stupid things like ‘The research has since been debunked.’ Wrong. The research never justified the media’s ludicrous over-interpretation. If they had paid attention, the scare would never have even started.”

The consequences of this coverage go beyond squandering journalistic resources on a bogus story. There is evidence that fear of a link between vaccines and autism, stoked by press coverage, caused some parents to either delay vaccinations for their children or decline them altogether. To be sure, more than 90 percent of children in both the US and the UK receive the recommended shots according to schedule, but in 2012, measles infections were at an 18-year high in the UK, reflecting low and bypassed immunization in some areas. In the US, vaccine-preventable diseases reached an all-time low in 2011, but the roughly one in 10 children who get their shots over a different timeframe than the one recommended by the medical establishment, and the less than 1 percent who go entirely unvaccinated, are enough to endanger some communities. And American and British authorities have blamed recent outbreaks of measles and whooping cough on decisions to delay or decline vaccination.

Beginning in 2004, Brian Deer, a British investigative journalist, brought a measure of redemption to journalism’s performance on this story, publishing a series of articles about improprieties in Wakefield’s work that culminated with the British General Medical Council stripping Wakefield of his license to practice in 2010, and The Lancet retracting his paper. For most journalists, that should have effectively put an end to the autism story. But those who never bought the vaccine-autism link—in the press and elsewhere—have been waiting for the proverbial nail in the coffin on this story for years, and it never seems to come. In April, for instance, The Independent in London published an op-ed by Wakefield, in which he trotted out his argument about the mmr vaccine in the context of the current measles outbreak in Wales.

* * *

Contrary to popular belief, the autism scare didn’t begin immediately after publication of Wakefield’s 1998 paper. Initially, science and health journalists who, as Goldacre and others have noted, “were often fairly capable of balancing risks and evidence,” handled most of the coverage and kept the story in its proper context. But the scare began to gain momentum in 2001, driven in large part by Wakefield, but also by the refusal of then-Prime Minister Tony Blair and his wife to say whether or not they had vaccinated their son, Leo, which raised suspicions nationwide. (Years later, they acknowledged that Leo was, in fact, vaccinated on schedule.)

In the US, Wakefield’s paper didn’t garner much media attention at first. Concern about a link between vaccines and autism had quietly built among parents and some physicians throughout the 1990s, but it revolved around vaccines containing the preservative thimerosal, not around Wakefield’s specific concerns about the MMR vaccine. It wasn’t until a year later, when the Food and Drug Administration recommended removing thimerosal from childhood vaccines as a precautionary measure—stressing that it could find no positive link with autism—that the American press tucked into the debate. In 2000, Dan Burton, a former Republican Congressman from Indiana who believes that vaccines caused his grandson’s autism, held congressional hearings wherein he asked the Department of Health and Human Services to study the alleged link, and Wakefield made his way into The New York Times for the first time. The 820-word story, buried on page 20, emphasized the danger of sowing mistrust of vaccines and the fact that the mainstream medical community considered them safe. Then, six months later, Wakefield appeared on 60 Minutes, where he linked vaccines to what he called an “epidemic of autism.” In 2002, Burton held more hearings that led to more stories on the dangers of vaccines. Major reports from the Institute of Medicine, part of the National Academy of Sciences, in 2001 and 2004, rejected the link and drew a lot of coverage, but the level of concern among the public remained on the rise.

A number of studies linked coverage by the British media in that early period to declining rates of vaccinations and outbreaks of rare diseases. But again, the effect was slower to take hold in the US. In 2008, a group of epidemiologists in Philadelphia compared annual mmr immunization rates from 1995 to 2004 to coverage that mentioned a link with autism. Their study, published in the journal Pediatrics, found that MMR vaccinations started to decline in the US years before news coverage took off in 2001, suggesting “a limited influence of mainstream media on mmr immunization in the United States.”

That influence soon began to grow, however. In 2005, an unvaccinated Indiana teenager returned from a church trip to a Romanian orphanage, where she’d unknowingly contracted measles. The next day, she attended a gathering of fellow congregants, many of whom were also unvaccinated, and triggered what at the time was the largest measles outbreak in the US in nine years. “Concern about adverse events, particularly related to media reports of a putative association between vaccinations and autism and of the dangers of thimerosal, appeared to play a major role in the decision of these families to decline vaccination,” according to a 2006 study published in The New England Journal of Medicine.

The year of the Indiana outbreak was a banner year for promoting the autism-vaccine link in the media. That summer, Rolling Stone and Salon published Robert Kennedy Jr.’s article alleging that the federal government covered up the danger of vaccines. A laundry list of corrections and clarifications followed, and in 2011, Salon retracted the article (Rolling Stone never did).

But it was the work of two veteran journalists, not Kennedy’s shameful piece, that really kept the story simmering. In February 2005, St. Martin’s Press published Evidence of Harm by journalist David Kirby, in which Kirby didn’t reach any specific conclusions about a link but presented a litany of parental suspicions that suggested one. And that winter, Dan Olmsted, a senior editor at United Press International, turned out a series called “Age of Autism,” for which he conducted an admittedly unscientific survey that found lower autism rates among ostensibly unvaccinated Amish communities (other studies found that vaccination rates are high in those communities). Few newspapers picked up Olmsted’s articles, but they got the attention of Representative Carolyn Maloney, a Democrat from New York. In March 2006, Maloney held a briefing at the National Press Club, where she cited Olmsted’s work as her motivation for drafting legislation that would compel the federal government to study autism rates in unvaccinated populations.

Maloney’s bill went nowhere, but Kirby and Olmsted went on to build their careers around the idea that a link exists in some children. Olmsted launched Age of Autism in November 2007, branding it the “Daily Web Newspaper of the Autism Epidemic”; it continues to be one of the most popular sites for those who doubt, or are concerned about, the safety of vaccines. And Kirby has written numerous columns on the subject for The Huffington Post. (HuffPost has long been a sympathetic home for the vaccine-autism crowd; it published a number of misleading pieces by celebrity-advocate Jenny McCarthy, for instance, whose son has autism. McCarthy’s fame allowed her to spread her theories far and wide in the media, including via influential TV programs like Oprah and Ellen.)

CJR, too, played a role in sustaining the vaccine story. In a 2005 piece, Daniel Schulman, who’s now an editor at Mother Jones, advised that it was “too soon for the press to shut the door on the debate” about vaccines and thimerosal.

Yet evidence in support of closing that door continued to pile up, and if history remembers no other journalist who fought back against the spurious claims about vaccines, it will remember Brian Deer. Between 2004 and 2011, the investigative reporter produced a series of reports for The Sunday Times of London, the UK’s Channel 4 Television, and the British Medical Journal (BMJ) that exposed how Wakefield had exhibited a pattern of gross medical misconduct in his work on the vaccine-autism question, including the unethical treatment of children and undisclosed conflicts of interest. After The Lancet retracted Wakefield’s 1998 paper and he was stripped of his medical license, the British Medical Journal published Deer’s coup de grace: a series revealing that Wakefield had actually doctored medical histories presented in his 1998 paper. In an accompanying editorial, the BMJ accused Wakefield of perpetrating an “elaborate fraud.”

* * *

Between 1998 and 2006, 60 percent of vaccine-autism articles in British newspapers, and 49 percent in American papers, were “balanced,” in the sense that they either mentioned both pro-link and anti-link perspectives, or neither perspective, according to a 2008 study by Christopher Clarke at Cornell University. The remainder—40 percent in the British press and 51 percent in the American press—mentioned only one perspective or the other, but British journalists were more likely to focus on pro-link claims and the Americans were more likely to focus on anti-link claims.

While it’s somewhat reassuring that almost half the US stories (41 percent) tried, to varying degrees, to rebut the vaccine-autism connection, the study raises the problem of “objectivity” in stories for which a preponderance of evidence is on one side of a “debate.” In such cases, “balanced” coverage can be irresponsible, because it suggests a controversy where none really exists. (Think climate change, and how such he-said-she-said coverage helped sustain the illusion of a genuine debate within the science community.) A follow-up study by Clarke and Graham Dixon, published in November 2012, makes this point. The two scholars assigned 320 undergrads to read either a “balanced” article or one that was one-sided for or against a link between vaccines and autism. Those students who read the “balanced” articles were far more likely to believe that a link existed than those who read articles that said no link exits.

In that context, Susan Dominus’s 2011 profile of Andrew Wakefield in The New York Times Magazine is problematic. Dominus trailed Wakefield around Texas, where he now lives, as he continued to proselytize to one crowd after another. And while her story was highly critical of Wakefield, the decision to publish it at all was controversial among science journalists. Some worried that people would undoubtedly read it as martyr story; others argued that journalists should simply stop paying attention to Wakefield.

Reporters don’t need Wakefield, however, to keep this story alive. Also in 2011, Robert MacNeil, a former of host of PBS NewsHour, came out of retirement to produce a six-part series for the program, called “Autism Now.” In part one, MacNeil interviewed his daughter, Alison, whose son has autism, and let her make unfounded claims about vaccines. MacNeil, who narrated the series, told viewers there was no scientific evidence to support those claims, but it was a throwaway line that allowed MacNeil to claim “balance” while sowing serious misunderstanding about vaccines.

Thankfully, the Web is now full of watchdogs who are looking out for such shenanigans. One is Seth Mnookin, author of The Panic Virus, who wrote a blog post calling the PBS series “an embarrassing coda” to MacNeil’s career.

* * *

Today, people who worry that childhood inoculations trigger autism prefer to be described as “vaccine-hesitant,” rather than “anti-vaccine,” and think the CDC’s immunization schedule “overwhelms” kids’ immune systems. This rhetorical shift is illustrates how those who claim a link exists keep moving the goalposts. For the last three years, the idea that the shots are “too much, too soon” has the been the argument of last resort in the face of mounting evidence that vaccines have nothing to do with autism. Accordingly, federal authorities have stepped up efforts to reassure people that the number, frequency, timing, order, and age at which vaccines are given is safe.

In January and March, the Institute of Medicine and the CDC both released evaluations of the current vaccination schedule—which includes as many as 24 immunizations by a child’s second birthday—and reiterated that the shots are unrelated to autoimmune diseases, asthma, hypersensitivity, seizures, or learning and developmental disorders. While it’s true that children today get more shots than they once did, it’s not the number of shots that the body notices, but rather the amount of antigens—the substances that produce an immune response—they contain. These days, thanks to the development of more efficient vaccines, a child is exposed to a maximum of 315 antigens by the time he turns two, compared with “several thousand” in the late 1990s.

The US media greeted the reports with a collective yawn. In some sense, the media’s apathy is welcome, as there was never any proof that the vaccination schedule was unsafe to begin with. But it would be unfortunate if part of the autism story’s legacy is that reporters and editors are wary of tackling any story about vaccine safety. Because there are rare, but genuine, safety issues with vaccines that the public needs to know about. In a series of articles for Reuters in January and February, reporter Kate Kelland described how a Finnish researcher endured months of ridicule and accusations from colleagues while trying to establish a link between a flu vaccine called Pandemrix and an outbreak of narcolepsy among children in Europe. Eventually, other studies confirmed the link, Kelland reported, but she added a cautionary note: “After the false alarm sounded by British doctor Andrew Wakefield, some scientists say they are more hesitant to credit reports of potential side effects from vaccines.” That chilling effect might extend to journalists as well; Kelland was one of only a few reporters in the US or the UK to cover the Pandemrix story.

Curtis Brainard is the editor of The Observatory, CJR's online critique of science and environment reporting. Follow him on Twitter @cbrainard.