Analysis

Difficulty accessing data from randomised trials of drugs for heart failure: a call for action

BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h5002 (Published 21 October 2015) Cite this as: BMJ 2015;351:h5002
  1. Robert Fleetcroft, clinical lecturer in general practice1,
  2. John Ford, NIHR doctoral research fellow1,
  3. Nicholas D Gollop, MRC doctoral research fellow in cardiology2,
  4. Pieter MacKeith, academic clinical fellow in primary care1,
  5. Kosala Perera, general practitioner3,
  6. Ashia A Shafi, specialist trainee, obstetrics and gynaecology4,
  7. Jhenaan Sorefan, medical student1,
  8. Caroline Thurlow, academic clinical fellow in primary care1,
  9. Rachel Wakelin, academic clinical fellow in primary care1,
  10. Martin R Cowie, professor of cardiology5,
  11. Nicholas Steel, reader in primary care1
  1. 1Norwich Medical School, Norfolk NR4 7TJ, UK
  2. 2Norfolk and Norwich University Hospital, Norfolk, UK
  3. 3JS Medical Practice, London N15 4JR, UK
  4. 4North Central and East London Deanery, London, UK
  5. 5National Heart and Lung Institute, Imperial College London, UK
  1. Correspondence to: R Fleetcroft r.fleetcroft{at}uea.ac.uk
  • Accepted 11 August 2015

When Robert Fleetcroft and colleagues attempted a systematic review of the treatment of heart failure, they were unable to answer their research question because of poor reporting and non-disclosure of data. And yet the necessary studies had been done. Here they discuss their experience and possible solutions

Access to individual patient data from published research has the potential to improve patient outcomes by enabling researchers to answer new questions with existing data, to validate findings, and to combine the power from individual studies, for example in meta-analyses.1 It may also deter selective reporting and analysis, and research fraud. Difficulty accessing data from clinical trials means that only part of the evidence is available, which may lead to erroneous clinical decisions.

A survey of corresponding authors of clinical trials found that, although in principle they agreed with sharing their data, they also had concerns about appropriate data use, investigator or funder interests, and protection of trial participants.2 Sometimes data are not available because the trial was not published. For example, the Carcinoembryonic Antigen Second Look Trial in colorectal cancer was not published until over 20 years after completion and reported no benefit for the common use of carcinoembryonic antigen to detect and resect disseminated cancer.3 Barriers to accessing data have led to delays in determining the effectiveness or adverse effects of commonly used drugs such as oseltamivir, rofecoxib, and rosiglitazone.4 5 6

To make research data more widely available several journals and funders have published statements supporting data sharing for all published articles.1 7 8 9 10 The Association of Medical Research Charities (AMRC) in the United Kingdom, the European Medicines Agency, and the Institute of Medicine in the United States all recommend data sharing between researchers.11 12 13

Here, we describe our experiences trying …

View Full Text

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial

Subscribe