The Cochrane Collaboration, an independent non-profit organization that reviews health information, released a report last week on the antiviral influenza treatment drugs Tamiflu and Relenza. The report examined both published studies and previously unreported trial data. The review confirmed that the drugs shorten the duration of flu-like symptoms. However, the drugs are less effective than previously thought and do not reduce transmission or complications associated with the flu. The organization then went on to criticize governments that have stockpiled Tamiflu in order to prepare for a pandemic. While the analysis sheds light on the importance of accessible data, Cochrane has made some excessive claims based on it.

The United States has spent $1.5 billion stockpiling Tamiflu, and the United Kingdom has spent $710 million. The initial decision by the United States to stockpile was made in 2006, based on studies showing a significant reduction in symptom time. The Tamiflu website states that the drug reduces symptoms up to 1.3 days, or 30 percent, faster than a control group, but the Cochrane review estimated it only reduces symptoms by half a day in adults. The authors went on to compare Tamiflu’s effect to those that Tylenol might have. This comparison has raised concerns that the money spent on Tamiflu was wasted and should have been used on health initiatives more efficiently.

Critics of the study have come from both the pharmaceutical industry and the medical field. U.K. Medical Director Daniel Thurley said the statistical analysis used by Cochrane led researchers to “systematically underestimate the benefits” of the antiviral drugs. Others have pointed out that the reduction of symptoms seen is substantial enough to warrant their use, and that the authors clear position against stockpiling may have influenced the work. But the heaviest criticism involves the design of the review, which only looked at randomized, placebo-controlled trials. These criteria exclude observational studies of patients who have already been admitted to the hospital with the flu. One of the showed showed a significant reduction in adult mortality.

There is no doubt that this report is a useful and informative review, but it is not sufficient to support Cochrane’s call for governments to suspend stockpiling. Until there are better treatments and the flu is far less common, governments need to have a plan in place for a pandemic flu that includes treatment. Even if Tamiflu cannot reduce the transmission of the virus, it definitively reduces both the time people are affected by symptoms and the mortality of the infection. In a pandemic year, this could save millions of lives. Review studies like the ones Cochrane performs are critical to the progress of health policy but must be inclusive of all the data before jumping to conclusions that could have large-scale implications.

A version of this article appeared in the Tuesday, April 15 print edition. Tess Woosley is a staff columnist. Scientific Society is published every Tuesday. Email her at [email protected].