Too good to be true? The case of influenza antivirals

shutterstock_159623486There truly is a day for everything, and according to the trusty Internet, this Tuesday, January 13 is Skeptics Day. What better way to get out of holiday mode and back into the routine of January than to turn a critical eye on the evidence, right? Even if you’re not quite ready to give up all of those feelings of goodwill from December, it does give us a good opportunity to look at a high profile case that has shown that some healthy skepticism has its place: that of influenza antivirals such as Tamiflu.

First, the evidence. Neuraminidase inhibitors (NIs), including oseltamivir (Tamiflu) and zanamivir (Relenza), are drugs that are used to treat and prevent seasonal and pandemic influenza. They are widely used internationally, are stockpiled at the recommendation of public health agencies, and oseltamivir has been classified as an essential medicine by the World Health Organization. A previous Cochrane review, published in 2006, describing the potential benefits and harms of NIs supported their use. The 2009 flu pandemic, however, raised a lot of questions about the broad population benefits of Tamiflu and other NIs, suggesting the review was due for an update.1 Questions had been raised about the validity of relying solely on the published literature, so the new review was updated using information from:

  • clinical study reports from the drugs’ manufacturers (GlaxoSmithKline [Relenza]; Gilead Sciences and F. Hoffman-La Roche [Tamiflu]), and;
  • regulatory documents from the US Food and Drug Administration (FDA), European Medicines Agency (EMA), and the Japanese regulators.2

This represents a very (very!) thorough look at the evidence, and marks the first time that a systematic review has relied on these sources of data for its analysis.

This review included 46 randomized, placebo-controlled trials that evaluated the benefits and harms associated with oseltamivir (20 studies; 9,623 participants) or zanamivir (26 studies; 14,628 participants) in all age groups. Four of the oseltamivir and two of the zanamivir trials were conducted with children. For kids, the main findings from this review were:

Time to first symptom alleviation: Oseltamivir reduced this time by a mean difference of 29 hours (95% confidence interval (CI) 12 to 47 hours) in otherwise healthy children, but did not have a significant effect in asthmatic children. Zanamivir did not have a significant effect.

Hospitalization rates: Oseltamivir did not have a significant effect; data for zanamivir were unreported.

Risk of developing serious influenza complications or those leading to study withdrawal: Oseltamivir did not have a significant effect; the zanamivir trials had insufficient events to compare.

Risk of pneumonia: Neither drug had a significant effect.

Risk of bronchitis, sinusitis, and otitis media: Neither drug had a significant effect.

Harms of treatment: Oseltamivir induced vomiting with a risk difference of 5.34% (95% CI 1.75 to 10.29) and number needed to treat to harm of 19 (95% CI 10 to 57). Zanamivir was well tolerated.

Prophylaxis: No data available for either drug.

The authors’ conclusions were that while NIs reduce the time to first symptom alleviation and may be beneficial in some populations, for example in immunocompromised patients, the trade-off between benefits and harms, the lack of evidence for a reduction in complications, and the self-limiting nature and low severity of influenza mean that the evidence does not support broad use and use policies for these drugs.

Now for the skepticism. The reason that this became such a high profile case comes down to transparency. This isn’t by any means limited to Tamiflu, but it happened to be the drug that got caught in the crosshairs.

As mentioned above, there was a previous systematic review on the effects of NIs, and it supported their use. Then, a reader of this review pointed out that the conclusions were based on a lot of unpublished data, so it couldn’t be known whether the results were valid.


Source: Wikimedia Commons

At this point, billions of dollars had already been spent worldwide, stockpiling a drug whose benefits were to become uncertain. The authors agreed that they needed further detail and this sparked a five year battle between them and the drug’s manufacturer (Roche) to obtain the information necessary to make evidence-supported claims. While the review authors attempted to get complete, unabridged clinical study reports (to avoid pitfalls like publication bias, which they’d already experienced in relying only on published trials), they were met with stonewalling, excuses, and data that didn’t match up with what had been published.

In 2013, after much persistence and media attention, Roche complied with the authors’ requests and ultimately made all of their data on Tamiflu available. When the updated Cochrane review was published in 2014, the results no longer supported widespread use of NIs, raising questions about the implications for policy worldwide. This process shone the spotlight on the issues of data accessibility and transparency in clinical trials, and the reversal of the review conclusions made it clear that those factors are pre-requisites for evidence-informed decision-making.

What next? Attitudes towards transparency in research are slowly changing, with more people embracing the idea that informed treatment decisions require full disclosure, both about what works and what doesn’t, to make sure that patients receive treatments that are beneficial, and don’t receive those that cause harm. Initiatives such as prospective trial registration and the AllTrials campaign, which aims to ensure complete reporting of methods and results, are attempting to address some of these gaps. While the implications for policy and practice are working themselves out, though, maybe now is a good time to embrace your inner skeptic.


  1. Jefferson TO, Demicheli V, Di Pietrantonj C, Jones M, Rivetti D. Neuraminidase inhibitors for preventing and treating influenza in healthy adults. Cochrane Database of Systematic Reviews 2006, Issue 3
  2. Jefferson T, Jones MA, Doshi P, Del Mar CB, Hama R, Thompson MJ, Spencer EA, Onakpoya IJ, Mahtani KR, Nunan D, Howick J, Heneghan CJ. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children. Cochrane Database of Systematic Reviews 2014, Issue 4; Cochrane Summary


-For a full history of the Tamiflu saga, check out Ben Goldacre’s summary:

-The BMJ has been highly involved in highlighting these issues and pushing for their resolution. Here is their most recent editorial on the subject: and a feature on the WHO’s current recommendations:

-A bit of a backlash from other influenza researchers, profiled in Nature News:

-The AllTrials Campaign:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s