Sometimes, we selectively ignore information. We seek data that confirms our beliefs and biases, and brush off evidence that conflicts with our thinking.
Sometimes, though, it’s not OK to ignore evidence. Especially when it comes to public health. So Science-ish was astonished when the Public Health Agency of Canada announced plans to take “exceptional action” and release a supply of the influenza drug Tamiflu from the emergency national stockpile to “ensure Tamiflu remains available to those Canadians who need it.”
If Tamiflu, also known by its drug name oseltamivir, had meaty evidence behind it, the news would a boon. We’d encourage Canadians to access an effective medicine during flu season. If the drug did what the public health agency suggests— reduce the chances of getting the influenza and minimize the harms associated with the flu in vulnerable elderly folks and kids—the actions would be comforting.
But that’s not the case. The release demonstrated that the guardians of Canadian public health, and sometimes those who report on it, selectively ignore key information.
Allow Science-ish to explain.
When Tamiflu was approved for the treatment of influenza in the late-1990s it was considered an effective drug for shortening the duration of the influenza, minimizing person-to-person spread, and reducing complications such as pneumonia and even hospitalizations. This belief was based on early studies, all funded and carried out by the drug-maker Roche. The antiviral became one of the World Health Organization’s “essential medicines” and governments around the world began to stockpile it to the tune of $6.9 billion worldwide (2009 value).
However, when systematic reviews of Tamiflu were done—where researchers gather and synthesize all the best-available research evidence—they discovered big, unacceptable holes in the data. In particular, in 2009, academic reviewers found that just two of the 10 studies on Tamiflu had ever been published in academic journals. Incidentally, those two showed the drug was barely better than placebo.
To get a better understanding of Tamiflu’s efficacy, they asked Roche to release its full data so it could be vetted independently. That was in 2009. Roche still hasn’t complied, despite its promise to do so, releasing only a fraction of data.
As a British Medical Journal editorial put it, plainly, “This means that taxpayers in the United Kingdom and around the world have spent billions of dollars stockpiling a drug for which no one except the manufacturer has seen the complete evidence base.”
For this reason, an international campaign to get the Tamiflu data—spearheaded by some of the world’s leading researchers, including the editorial staff of the BMJ—has gained momentum. Roche has claimed that legal constraints got in the way of opening the data, but researchers aren’t buying it, even calling on governments to sue Roche and boycott its products.
What the evidence shows
In a statement for Science-ish, a spokesperson for Roche Canada said the company stands behind “the robustness and integrity of our data” and that Roche “plans to set up a multi-party advisory board comprising experts from academia and private institutions, including the Cochrane Collaboration, to review the totality of Tamiflu data and agree on a statistical analysis plan.” It may turn out the drug is more effective than current evidence suggests.
But the medical community can only work with the best, publicly available data. And it tells a pretty dismal story. A 2012 systematic review—the gold standard of evidence-based medicine—looked at Tamiflu for preventing and treating influenza in children and found evidence of “modest benefit,” though noted it was based on small and poor-quality studies. It also found that Tamiflu in kids was associated with a significant increased risk of vomiting. So the researchers called into question the efficacy of the drug and asked for more and better data.
Another 2012 systematic review looked at Tamiflu for healthy adults and children and found the evidence so biased they could not draw conclusions. They also discovered that Roche never published the biggest trial ever done on Tamiflu.
Science-ish forwarded the Public Health Agency of Canada news release to Dr. Tom Jefferson, the researcher who led that last systematic review and who has been calling on Roche to open their data. “It’s like a time warp,” he said over the phone from Rome. “I just couldn’t believe my eyes. Do they know about the international campaign?”
He took issue with the fact the Canadian public health agency suggested Tamiflu be used in children and the elderly, and that it can reduce the chances of getting the flu following contact with an infected individual. “In the case of the elderly, there was one trial done which was never published and it shows Tamiflu doesn’t work, there was no effect,” he said. As for the statement about reducing chances of transmission, “That’s false,” Jefferson said. “Only one trial has been done on [the spread of influenza in] close contacts, and the trial design was faulty, so no such evidence exists.”
The only evidence-based benefit Tamiflu offers is that it shortens flu symptoms by about a day—but that’s if you manage to see your doctor, get a prescription and take the drug within 48 hours of symptoms appearing. Rather unremarkable.
Plus, on outcomes that should matter to the governments that stockpiled the drug—like whether Tamiflu cuts the risk of flu complications such as pneumonia and hospitalizations—Jefferson and the Cochrane team concluded the evidence was too weak to know. Worse still, the evidence made it hard to form a full picture of Tamiflu’s adverse effects. This conservative verdict is similiar to the U.S. Food and Drug Administration stance on Tamiflu.
So, asked Jefferson, “What other drugs are around that are less toxic, and cost less than Tamiflu, and do the same? The answer is of course something like an anti-inflammatory, Aspirin or whatever.”
In light of these questions about the pharmaceutical, Science-ish asked the Public Health Agency of Canada and Health Canada to explain what research they had access to in making decisions and recommendations about Tamiflu and whether they are aware of the gaps in the evidence base.
They seemed to take no issue with the paucity of good evidence for the drug. A Health Canada spokesperson said in an email that recommended uses for the drug in Canadians “are based on the evaluation of clinical trials data submitted by the sponsor to support the indications.”
Science-ish asked to see what information PHAC’s claims were based on, and they pointed to the product monograph for Tamiflu—which was prepared by Roche, includes all Roche-authored research, and does not reflect any evidence produced after 2004, including those important systematic reviews.
Dr. Ben Goldacre wrote about the Tamiflu evidence base in his new book Bad Pharma. “Currently, Roche is withholding vitally important information about half of the clinical trials of Tamiflu,” he told Science-ish. “We know about half have never been published. So we cannot tell you as doctors whether Tamiflu works because Roche is withholding information.”
The flu medication is a symbol of a broken system, he said. Regulators have allowed industry to withhold data, to control what information about medical interventions is made public. Goldacre said, “No effective legislation has ever fixed this problem. This is an international scandal, and it means doctors are effectively prescribing in the dark.”
So if what’s happening with Tamiflu is happening all the time with other treatments, why all the fuss and focus on Tamiflu? “The researchers were so dogged, they refused to be sloughed off when the company said we refuse to give you the data,” explained Goldacre.
Then the question becomes: Why aren’t our regulators and public health agencies doing the same?
For more details about the 2013 flu season, tune in tomorrow. Science-ish is a joint project of Maclean’s, the Medical Post and the McMaster Health Forum. Julia Belluz is the senior editor at the Medical Post. Got a tip? Seen something that’s Science-ish? Message her at firstname.lastname@example.org or on Twitter @juliaoftoronto