LEFT: University of Miami pediatrician Judith L. Schaechter gives an HPV vaccination to a 13-year-old girl in her office. (Joe Raedle/Getty Images)
RIGHT: A newspaper box in front of the Toronto Star building in 2008. (Mark Blinch/Reuters)

Corrected

The Toronto Star’s front-page feature on the “dark side” of a widely used HPV vaccine had all the makings of a blockbuster: a grim, gripping headline, vivid accounts from teenagers who died or were debilitated, a wrenching image of a woman holding a framed photo of her dead daughter.

But it lacked a crucial component of any scientific investigation: good data.

“It’s too bad there isn’t a vaccination to prevent journalistic misstep,” wrote the paper’s public editor, Kathy English, who called the story “alarmist.” “I suspect we’d all line up for that shot about now.”

All reporters face dueling pressures when covering medicine and other science issues. On the one hand, they want to craft a compelling narrative. On the other, science values statistics, not stories and anecdotes. The Star’s vaccine article, which the paper retracted Friday after it was thoroughly panned by doctors and other experts, epitomized the failure to strike that balance. As the paper’s publisher put it in a note, the Star’s story “led to confusion between anecdotes and evidence.”

The Star, Canada’s highest-circulation daily paper, won a reputation for its investigative reporting after its high-profile exposé of Toronto Mayor Rob Ford. The HPV vaccine article was meant to be one installment in a series of investigative pieces on drug safety.

The story was built around the accounts of five young women who suffered muscle pain, swelling, heart attacks and other health problems in the weeks after receiving shots of Gardasil, a vaccine that protects against the sexually transmitted human papillomavirus. Published Feb. 5 under the headline “A wonder drug’s dark side,” it said that doctors and health officials are pushing the vaccine on patients without fully informing them of potential side effects. The Star also reported that it had found more than 50 self-reports of “serious” adverse reactions to the drug in a regulatory database maintained by Health Canada and thousands of suspected cases in a similar American database known as VAERS (an acronym for Vaccine Adverse Event Reporting System).

The article acknowledged that none of the cases it discussed were conclusively proved to have been caused by the Gardasil injections and obliquely alluded to “comprehensive clinical trials and other data that show the vaccine’s well-studied safety and efficacy.” But those caveats were overwhelmed by descriptions of a 13-year-old athlete sapped of her strength by debilitating joint pain, a 14-year-old hospitalized for over a month after suffering a heart attack and, horrifyingly, another 14-year-old found drowned in her bathtub two weeks after receiving the vaccine.

Vox science writer Julia Belluz called it “everything wrong with vaccine reporting in one dangerous package.”

“These tales of suffering and death are awful. Stomach turning. But they are just that: stories,” she wrote.


Left: A screen grab shows a cached page of the Star’s HPV story. Right: After the retraction.

Doctors almost immediately responded to the Toronto Star piece with links to scientific studies of the HPV vaccine, none of which were cited in the article. Among them were a Journal of American Medical Association study showing that Gardasil is no more likely to cause adverse reactions than any other vaccine, a Journal of Internal Medicine study that found no connection between the Gardasil vaccine and autoimmune disorders and a Pediatric Infectious Disease Journal study reporting that 0.05 percent of Gardasil clinical trial participants experienced a serious reaction to the treatment (none of them fatal).

The JAMA study also warned against reliance on self-reporting mechanisms like the Health Canada and VAERS databases, from which the Star got its only data about vaccine-related incidents.

“Not all reported events are systematically validated, and many may have only coincidentally followed vaccination,” the study said, adding that underreporting, inconsistency in the quality and completeness of reported data, stimulated reporting due to extensive news coverage and reporting biases could also skew these numbers.

Meanwhile, people on Twitter accused the paper of “fear mongering” and compared the report to claims from vaccination skeptics that the measles vaccine causes autism.

A week after the story came out, it was updated to reflect doctors’ criticism of the piece. The online headline was changed to the less-inflammatory “Families seek more transparency on HPV vaccine and a subhead was added stating “there is no scientific medical evidence of any ‘dark side’ of this vaccine.”

Last Friday, the story was removed altogether.

“All vaccines, including Gardasil, have side-effects. The better known they are, the more safely the vaccine can be deployed,” publisher John Cruickshank wrote in a note on the paper’s Web site. “… We remain committed to this line of reporting. However, we have concluded that in this case our story treatment led to confusion between anecdotes and evidence.”

The lingering outrage over the story among medical experts points to frustration with science journalism that allows narrative to eclipse hard data. An opinion piece submitted to the Star by an immunologist and a medical law expert and signed by 63 other medical specialists said that the article’s “litany of horror stories and its innuendo give the incorrect impression that the vaccine caused the harm.” Even the Star’s public editor acknowledged that the paper’s reporting wrongfully emphasized anecdotal evidence.

“In giving disproportionate and dramatic play to the heartbreaking stories of young women who suspect their illnesses are linked to having received the vaccine … the proven scientific evidence of the vaccine’s safety was not made clear enough to readers,'” she wrote in a Feb. 13 column.

It’s a criticism that has been lobbed at science journalism before. Some of science’s most compelling writers — Macolm Gladwell and Oliver Sacks, for example — have been accused of privileging individual stories over mass amounts of data.

The trouble is that individual stories are what make reporting on science interesting. Gladwell said as much himself in a 2013 interview with Brian Lehrer.

“I am a story-teller, and I look to academic research … for ways of augmenting story-telling,” he said. “The reason I don’t do things their way is because their way has a cost: it makes their writing inaccessible. If you are someone who has as their goal … to reach a lay audience … you can’t do it their way.”

It’s worth noting that the Star didn’t just fumble in its science reporting. The paper’s initial response to the backlash was to dismiss its critics. Here’s a tweet from editor-in-chief Michael Cooke:

Vox reporter Belluz says Cooke used similarly crude language in an e-mail responding to her questions about the article.

“Stop gargling our bathwater and take the energy to run yourself your own, fresh tub,” he wrote, according to her piece.

Correction: The initial version of this story incorrectly said that the opinion piece in the Toronto Star was submitted by two immunologists. One author was an immunologist, the other was a medical law expert who has done work related to the HPV vaccine.