Research: what's the point?

Much emphasis is placed on doing research, but why bother? Polly Brown explains

On first reading, "Research: what's the point?" seems a perverse question; without research we would never have come up with most of the medical interventions we take for granted in practice today, and we would never improve on them. Can anyone seriously question the importance of clinical and academic research in the development of medical care? A cynical point of view may be that the point of research is to add weight to the CV of the researcher, to help the marketing strategy of drug companies, or to keep medical publishers in business. To inform, research needs to be well done, unbiased, considering the right questions, and applicable to practice. This is not always the case.

ALIX/PHANIE/REX FEATURESL

In vitro fertilisation: frozen sperm and human embryos

Allegations of fraud or bias often bring research to the public eye-for example, the case of Anjan Banerjee who in 1990 admitted to submitting an abstract to the British Society of Gastroenterology, the results of which were all based on his own urine samples.¹ The 2002 report of the Committee of Publication Ethics highlighted eight key areas relating to misconduct in research, including publication of poorly done research, authorship misconduct, falsification, fabrication, and plagiarism.² The BBC has published two articles on fraud in research in the past five years, one quoting a case in which a scientist who claimed to have transplanted black skin on to a white mouse had just coloured the mouse with a felt pen.³ Revelations such as these bring the reputation of research into question.

The relationship between the pharmaceutical industry and research is intrinsically problematic. Research is big business for pharmaceutical companies; the estimated cost of creating a new drug in the US in 2001 was $800m (£440m; €620m).⁴ Research funded by drug companies is often thought to contain bias, particularly with respect to publication bias-the reluctance to publish work with negative findings.

What about the question of whether research reflects health needs? Globally, the international research foundation reported that less than 10% of the world's budget is spent researching conditions which account for 90% of global disease.⁵ This lack of correlation is illustrated by analysing Clinical Evidence, a resource that searches for evidence to answer clinical questions likely to be asked by general practitioners. Of 1851 interventions nearly half are of "unknown effectiveness."⁶ The definition of unknown effectiveness is based on "interventions for which there are currently insufficient data or data of inadequate quality."^{6 7} Examples in which evidence is surprisingly sparse include treatment with carbamazepine for bipolar depression, drug treatments for postnatal depression, and even such routine activities as removal of earwax by syringing.

Even when research is reliable and asks the right clinical question, do doctors have the time to keep themselves, or their patients, updated? As doctors are increasingly busy and bombarded with more and more information, expecting them to critically appraise new papers and assess which ones they should be paying attention to is asking a lot. Even good quality research may not necessarily be applicable to their patient; doctors must consider age, comorbidity, prevalence of adverse events, and of course patients' preferences.

Despite these problems, research is an essential part of medical development. Clinical governance and the increased expectations of patients mean that simply following your nose or relying on advice handed down by your teacher or on the last discussion with a colleague is increasingly inadequate. Monitoring bodies set benchmarks to encourage doctors to back up practice with evidence. The focus should therefore be not on the point of research but rather on how we can improve it.

Electronic publishing enables medical journals to improve the reporting of research. The Cochrane Collaboration and the Lancet are leading the way by publishing study protocols, accelerating the path to publication, reducing publication bias, and helping fund applications.⁸ Current Controlled Trials is an online publisher which provides services for randomised controlled trials (www.controlled-trials.com). Organisations or individuals can register their trials via the international standard randomised controlled trial number (ISRCTN) scheme or host their trial register on the metaregister of controlled trials. The metaregister is an international compilation of registers of completed and ongoing trials hosting all randomised controlled trials sponsored by the UK Medical Research Council and the US Department of Veterans Affairs Cooperative Studies Program. By sharing trials, registers reach a much wider audience and add to the evidence base available to researchers, clinicians, and patients.

The trend in evidence based medicine is also increasing pressure on the pharmaceutical industry to be transparent; a recent review found that the research methods of trials sponsored by drug companies is at least as good as that of non-industry funded research and in some cases better.⁹

Legitimate limitations of research, however, should be recognised. Research cannot answer a^ll clinical questions, and it does not even need to. For example, no one would wait for a randomised controlled trial before giving oxygen to a patient with a severe asthma attack. Even for questions that are amenable to research there can be a considerable time lag in answering them-much of industry funded research starts with a hypothesis that can take on average 15 years to test.⁴ This means that a clinical question asked today might not be answered by research until 2019. Also, drop off is a problem in clinical trials-that is, trials are initiated but not completed. In the United States, only a fifth of clinical trials embarked upon are actually completed.⁴

Without research, most of the medical interventions we take for granted in practice today would never have been developed. In the past 10 years research has developed the human genome project, which shows massive potential in molecular medicine for improving diagnosis, rational drug design and gene therapy.¹⁰ Developments in vaccines for HIV are entering third phase trials, which could potentially halt the progress of the worldwide HIV epidemic.¹¹

Good quality research should answer relevant clinical questions, cover new ground rather than old, take advantage of tools to aid protocol design, and be accessible to everyone. Only then can it begin to keep up with and fulfil the demands of our rapidly expanding health agenda.

1. Ferriman A. BMJ 2000;321:1429.
2. Godlee F. COPE sets out an agenda for research. In: Committee of Publication Ethics. The COPE report 2002. London: COPE, 2002. www.publicationethics.org.uk/cope2002/pdf2002/21330_pp13_16.pdf (accessed 8 Jan 2004).
3. BBC. "Lives at risk" from research fraud. London: BBC, 1998. http://news.bbc.co.uk/1/hi/health/106186.stm (accessed 8 Jan 2004).
4. DeMasi J. Risks in new drug development: approval success rates for investigational drugs. Clin Pharmacol Ther 2001;69:297-307.
5. Roach JO’N. Research does not reflect global disease burden. studentBMJ 2000;8:183.
6. Clinical evidence. Issue 10. London: BMJ Publishing Group, 2003.
7. Enkin M, Keirse MJNC, Neilson J, Crowther C, Duley L, Hodnett E, et al. A guide to effective care in pregnancy and childbirth. Oxford: Oxford University Press, 1998.
8. Chalmers I, Altman DG. How can medical journals help prevent poor medical research? Some opportunities presented by electronic publishing. Lancet 1999;353:490-3.
9. Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ 2003;326:1167-70.
10. Human Genome Project Information. www.ornl.gov/sci/techresources/Human_Genome/home.shtml
11. Centers for Disease Control and Prevention. Questions and answers on the Thailand phase III vaccine study and CDC’s collaboration. Atlanta: CDC, 1999. www.cdc.gov/hiv/pubs/facts/vaccineqa.htm (accessed 8 Jan 2004).