Tonight channel 4 news and the BMJ have released a major story on the effects of Tamiflu. This is a must read as this is the second major systematic review to appear in the BMJ investigating the evidence underpinning "neuraminidase inhibitors" (the class of drugs including Tamiflu) for preventing and treating influenza in healthy adults.

Jefferson and his colleagues should be commended for systematically searching for the truth, and asking, what are the exact benefits and harms of Tamiflu and similar drugs? They are being used liberally in the current H1N1 pandemic and it is important that these types of studies – systematic reviews - are undertaken and independently funded by national bodies to ensure impartiality.

These results are not surprising, in that the benefits of neuraminidase inhibitors are modest, which most clinicians probably already knew. However, amongst policy makers and the wider public this message does not seem to be getting through. Significantly, this paper differs from previous published studies in that the investigators attempted to systematically investigate the potential harms of the drugs. What is disturbing is that only one study explicitly examined drug safety. This is an important omission given the widespread use of Tamiflu.

Moreover, we now know that the evidence-base for a reduction of serious complications such as pneumonia is uncertain. Simply this means the benefits of the drug may be outweighed by the complications, i.e. Tamiflu could cause more cases of pneumonia than it prevents, from what we now know.

What happened next in this study proves pretty shocking. To get key access to the data the BMJ and channel 4 news undertook their own investigation. It is unclear which drafts the Roche investigators saw,and we don't know who authored the papers. One major author named on the abstract of one of the most influential studies told the BMJ he had not even been involved in the study. Yet, we continue to be bombarded with news stories of apparent benefits of antivirals; whilst now knowing that major data are being withheld and manipulated to propagate the unwarranted use of these drugs.

The BMJ commissioned Freemantle to pass judgement on the observational studies. Freemantle rightly states that randomised trials rarely include people with "co-morbidities" (other health problems), as adverse events in these groups may be significantly worse. Therefore, this undermines attempts to establish the safety of the new drug. To be able to generalize results to the wider population what we require is observational data of the safety profile in routine practice. That is real life patients who are excluded from the trials for the reasons outlined above.

There is a pressing need to understand the benefits and potential adverse effects of these drugs as the current evidence base supporting their use is now so uncertain. In individuals with mild symptoms their use should be discontinued with immediate effect. Using antivirals liberally runs the risk of generating resistance. We have known this for many years. More worryingly in those at high risk the benefits remain uncertain and their use has to be considered by clinicians on a case-by-case basis. Experts and those that generate policy should look at these documents carefully and stick to the evidence.

Overall the benefits of these drugs are fairly limited. Tamiflu provides only a small benefit by shortening the duration of illness in children with influenza and has little effect on asthma exacerbations or the use of antibiotics. Its effects on the incidence of serious complications and on the current A/H1N1 influenza strain in both adults and children is unknown.