www.bmj.com

BMJ -- Press Releases
Releases Saturday 31 May 2003
No 7400 Volume 326
 

 
 FIVE ARTICLES BELOW

 (1)  CONCERN THAT RESEARCH SPONSORED BY
 DRUG COMPANIES IS BIASED
 
 (2)  DRUG TREATMENT LIKELY TO BE BASED ON
 BIASED EVIDENCE
 
 (3)  WEEKLY CONTACT WITH DRUG REPS LINKED
 TO UNNECESSARY PRESCRIBING
 
 (4)  MEDICAL PROFESSION ATTEMPTS TO "CLEAN
 UP" RELATIONS WITH DRUG INDUSTRY
 
 (5)  RELATIONS BETWEEN THE DRUG INDUSTRY
 AND PATIENT GROUPS SHOULD BE OPEN
 ______________________________

 (1)  CONCERN THAT RESEARCH SPONSORED BY
 DRUG COMPANIES IS BIASED
 
 (Pharmaceutical industry sponsorship and research
 outcome and quality: systematic review)
 http://bmj.com/cgi/content/full/326/7400/1167
 
 Research funded by drug companies is more likely to
 produce results that favour the sponsor's product than
 research funded by other sources, claim researchers in
 this week's BMJ.
 
 They reviewed 30 studies that analysed research
 sponsored by a pharmaceutical company. Research
 funded by drug companies was less likely to be
 published than research funded by other sources. Studies
 sponsored by drug companies were also more likely to
 have outcomes favouring the sponsor than studies with
 other sponsors.
 
 However, none of the 13 studies that analysed methods
 reported that studies funded by industry were of poorer
 quality.
 
 These results apply across a wide range of diseases,
 drugs, and drug classes, over at least two decades and
 regardless of the type of research being assessed, say
 the authors. This suggests that there is some kind of
 systematic bias to the outcome of published research
 funded by the pharmaceutical industry.
 
 Possible explanations include the selection of an
 inappropriate comparator to the product being
 investigated and publication bias, they conclude.
 
 Contact:
 
 Joel Lexchin, Associate Professor, School of Health
 Policy and Management, York University, Toronto,
 Canada
 Email: joel.lexchin@utoronto.ca
 

 (2)  DRUG TREATMENT LIKELY TO BE BASED ON
 BIASED EVIDENCE
 
 (Evidence b(i)ased medicine - selective reporting from
 studies sponsored by pharmaceutical industry: review of
 studies in new drug applications)
 http://bmj.com/cgi/content/full/326/7400/1171
 
 Drug treatment is likely to be founded on biased
 evidence because drug companies tend to publish studies
 with more favourable results, suggest researchers in this
 week's BMJ.
 
 They identified 42 studies submitted to the Swedish drug
 regulatory authority to secure marketing approval for five
 antidepressant drugs. These studies were then compared
 with studies actually published between 1983 and 1999.
 
 They found evidence of three sources of bias: duplicate
 publication, selective publication, and selective reporting.
 For instance, 21 studies contributed to at least two
 publications each, and three studies contributed to five
 publications. Studies showing significant effects of a drug
 were published as stand alone publications more often
 than studies with non-significant results. The tendency to
 report the more favourable results only, in studies
 actually published, was a major cause for bias.
 
 These results should not be used to dispute the value of
 analysing the medical literature, say the authors.
 However, they are likely to be valid for other classes of
 drugs, so for anyone who relies on published studies
 alone to choose a specific drug, they should be a cause
 for concern.
 
 Without access to all studies (positive as well as
 negative, published as well as unpublished) any attempt
 to recommend a specific drug is likely to be based on
 biased evidence, they conclude.
 
 Contact:
 
 Hans Melander, Senior Biostatistician, Medical Products
 Agency, Uppsala, Sweden
 Email:  hans.melander@mpa.se

 
 (3)  WEEKLY CONTACT WITH DRUG REPS LINKED
 TO UNNECESSARY PRESCRIBING
 
 (Characteristics of general practitioners who frequently
 see drug industry representatives: national cross sectional
 study)
 http://bmj.com/cgi/content/full/326/7400/1178
 
 General practitioners who see drug industry
 representatives at least once a week are more likely to
 express views that will lead to unnecessary prescribing
 then those who report less frequent contact, finds a study
 in this week's BMJ.
 
 Over 1,000 general practitioners randomly selected from
 200 practices in England completed a questionnaire
 about their personal and practice characteristics, and
 their prescribing attitudes and behaviour.
 
 After analysing the responses, the research team found
 that frequent contact with a drug industry representative
 was significantly associated with a greater willingness to
 prescribe new drugs and to agree to patients' requests to
 prescribe a drug that is not clinically indicated.
 
 Those who reported frequent contact were also more
 likely to be dissatisfied with consultations ending in
 advice only, and were more receptive to drug
 advertisements and promotional literature from drug
 companies.
 
 General practitioners who saw drug representatives most
 often tended to be singlehanded practitioners and
 working in deprived areas.
 
 More research on the nature of this relationship would
 help primary care trusts to adopt policies encouraging
 more cost effective prescribing, conclude the authors.
 
 Contact:
 
 Chris Watkins, General Practitioner, Backwell and
 Nailsea Medical Group, Bristol, UK
 Email: chris.watkins@Bristol.ac.uk

 
 (4)  MEDICAL PROFESSION ATTEMPTS TO "CLEAN
 UP" RELATIONS WITH DRUG INDUSTRY
 
 (Who pays for the pizza? Redefining the relationships
 between doctors and drug companies. 1: Entanglement
 2: Disentanglement)
 http://bmj.com/cgi/content/full/326/7400/1189
 
 As entanglement between doctors and drug grows, a
 two-part article by journalist Ray Moynihan in this
 week's BMJ explores the brewing conflicts at one of the
 world's leading medical institutions over how to redefine
 relations with big pharmaceutical companies.
 
 Across the United States, drug companies sponsor close
 to 300,000 events for doctors every year as part of their
 promotional efforts.
 
 Against this background, the University of California San
 Francisco (UCSF) is considering plans to end free
 lunches sponsored by drug companies and to remove
 drug representatives. Elsewhere, medical reform groups
 and student associations are also calling for
 disentanglement from drug companies and independent
 education and sources of information.
 
 "Relationships with industry are critical, but they need
 redefining," says Professor Haile Debas, Dean of
 Medicine at UCSF. "We have to make sure our
 relationships are more appropriate."
 
 "In some ways we are all addicts to big pharma's
 money," argues Chief Executive Officer of UCSF
 Medical Center, Mark Laret, "but we are going to have
 to wean ourselves off a dependency that is generally
 inappropriate. This relationship is one of those things we
 need to clean up. The sooner the better."
 
 Responding to the view that drug company funded
 education is necessary, Drummond Rennie, Professor of
 Medicine at UCSF says, "Nonsense. That argument
 presupposes that some of the most well off in our society
 can't afford to pay for their lunches, their education, or
 their conference."
 
 Many individual doctors, and their professional
 associations, are facing difficult choices about whether
 they remain part of the industry's extended promotional
 machinery or seek real distance in their relationships, to
 give prescribing, teaching, and advice that is truly
 independent, writes Moynihan. Growing moves toward
 genuine separation may well make previously acceptable
 conflicts of interest untenable, he concludes.
 
 Contact:
 
 Ray Moynihan, Journalist, Washington DC, USA
 Email: raymond.moynihan@verizon.net
 
 
 (5)  RELATIONS BETWEEN THE DRUG INDUSTRY
 AND PATIENT GROUPS SHOULD BE OPEN
 
 (Relationships between the pharmaceutical industry and
 patients' organisations)
 http://bmj.com/cgi/content/full/326/7400/1208
 
 Relationships between the pharmaceutical industry and
 patients' organisations must be at arm's length and
 transparent, and not affect the agenda and priorities of
 these groups, argues a senior doctor in this week's BMJ.
 
 Advertising drugs to patients in the United Kingdom is
 not allowed, so drug companies try to inform patients
 through partnerships with self help groups and patients'
 organisations, writes Dr Andrew Herxheimer of the UK
 Cochrane Centre.
 
 The Long Term Medical Conditions Alliance offers
 guidelines to voluntary health organisations on relations
 with the pharmaceutical industry. However, many
 organisations volunteer minimal information about their
 structure and funding, and some are reluctant to discuss
 the subject.
 
 If industry directly or indirectly funds a large part of the
 budget, say over 20%, the organisation comes to
 depend on it and this, if nothing else, will influence
 policies, he says.
 
 Most patients' organisations are poor and have little
 independent funding. Grants and joint projects with
 pharmaceutical companies can help them grow and be
 more influential, but can also distort and misrepresent
 their agendas. Relationships must therefore be
 acknowledged and open, without public relations
 flummery, stresses the author.
 
 Some modest public funding for patients' organisations
 should also be considered, and regulatory agencies have
 to distinguish between independent and extensively
 funded patient groups, he concludes.
 
 Contact:
 
 Andrew Herxheimer, Emeritus Fellow, UK Cochrane
 Centre, London, UK
 Email:  andrew_herxheimer@compuserve.com