HTA - 10/80/01: Neuraminidase inhibitors for influenza: a systematic review and meta-analysis of regulatory and mortality data
|Project title||Neuraminidase inhibitors for influenza: a systematic review and meta-analysis of regulatory and mortality data|
|Research type||Evidence Synthesis|
|Start date||February 2011|
|Link to journal publication|
|Chief Investigator||Professor Carl Heneghan|
|Co-investigators||Professor Chris Del Mar (Bond University), Mr Peter Doshi (Massachusetts Institute of Technology), Dr Rokuro Hama (Japan Institute of Pharmacovigilance), Dr Thomas Jefferson (Cochrane Acute Respiratory Infections (ARI) Group), Dr Matthew Thompson (University of Oxford), Dr Mark Jones (The University of Queensland)|
|Contractor||University of Oxford|
|Plain English summary||Pandemic influenza has been one of the highest public health priorities for governments across the world, including the UK, over the past few years. A central element of the strategic approach to preventing spread and treatment of influenza has been the use of neuraminidase inhibitors (NIs) such as oseltamivir (Tamiflu) and zanamivir (Relenza). Despite this, updated Cochrane Reviews published at the end of 2009 by this international research group, were unable to ascertain from the published research whether the drugs were effective in reducing the complications of influenza, and also the frequency and severity of treatment harms in adults or children. Following feedback to the previous version of the Cochrane Review of NIs in adults, the review authors realised that they would need to obtain from the drug manufacturers more details relating to unpublished trials of one of the neuraminidase inhibitors (oseltamivir - Tamiflu). The review authors published their revised findings in December 2009 in the BMJ and Cochrane Library. As a result of consequent publicity, the two drug manufacturers have undertaken to provide the trial data. The proposed project therefore attempts to evaluate both neuraminidase inhibitors for the effects on treating and preventing influenza symptoms and complications of influenza, and the treatment harms in both previously healthy adults and children, using the best available evidence.|
|Scientific summary||Neuraminidase inhibitors (NIs) have become global public health drugs, stockpiled against possible influenza outbreaks. Despite their popularity, the discovery of sizeable reporting bias, uncertainty over the number of trials run, ghost authorship and widespread discrepancies between published and unpublished versions of the same trials, have led to a reappraisal of the published evidence and calls for public release of hitherto confidential data. The project aims to review unpublished data on effectiveness and harms of NIs for influenza in all age groups updating and amalgamating two established Cochrane reviews on adults and children. The review will be in two stages. Stage one will identify, reconstruct and assess all NI trials conducted to January 2010 using regulatory and pharmaceutical reports. Only reliable* trial data will be included in the second phase of the review (an analysis following standard Cochrane methods). Because of the complexity of our task we have devised four additional types of tools for data management and information: a directory mapping citations to clinical trials in our data sources; its expanded version with annotations; a narrative of our review, documenting the evolution of methods and detailed summary of the regulatory documents and a reference resource holding details of all trials. In addition we have created a tool for assessing all known types of reporting bias in trials and sub-trials programmes. *Reliability is made up of 5 aspects: 1.Completeness - there is sufficient material to assess the other reliability criteria. 2. Internal consistency - all parts of the same CSR/unpublished pharma report are consistent and coherent. 3. External consistency - all versions of the various CSR/unpublished pharma reports are consistent and coherent. 4. Clarity - all parts and versions allow understanding of the design and conduct of the trial. 5. Robustness - exposure of reconstructions to regulatory comments do not lead the majority of voters (at least 4 of 7 in the team) to change the study conclusions or their trustworthiness.|
Protocol (PDF File - 253.7 KB)
Join mailing list
To receive funding alerts and other programme news, please join our mailing list.
NIHR on TwitterTweets by @OfficialNIHR
Make a research suggestion
Complete our online web form to make a research suggestion to our programmes.
Sign in to MIS
Login to the NETSCC Management Information System (MIS).