Bypassing bureaucracy to answer important questions quickly

Sometimes there is a need to answer important clinical questions quickly. The Research Governance Framework (RGF) for Health and Social Care, which was introduced in England in 2005, aimed to improve the efficiency and quality of research in the NHS, but has introduced a new layer of bureaucracy which is preventing important questions being answered. 2007 saw the publication, in the JRSM, of two studies highlighting the unacceptable bureaucracy in the Research and Development system (JRSM 2007;100:101–4 and 234–8).^1,2 These studies estimated 44 days and 318 staff hours to gaining R&D approval for a low-risk multicentre study. This is a problem which could affect every single clinical researcher.

Other systems have successfully removed unnecessary bureaucracy, and a useful model is the Multicentre Research Ethics Committees (MREC). Variations in the functioning of ethics committees had erected barriers to research and a new system of multicentre ethics approval was introduced in the UK during the last decade to address some of the problems highlighted by researchers when undertaking studies involving more than one site.³ The use of standardized means of assessment by the new multi-centre research ethics committees means that their approval is transferable throughout the entire UK.⁴ Local R&D offices are now throwing up the same barriers that local research ethics committees once were and currently constitute the major impediment to multi-site research.^4,5

Sometimes failure to answer important questions quickly can have major implications for clinical practice. For example, we attempted to survey UK psychiatrists in response to an urgent need to describe their prescribing practice. In December 2003, the Committee on Safety of Medicines (CSM) issued guidelines suggesting that fluoxetine was the only safe antidepressant for use in child and adolescent depression. Clinicians were struggling with having to implement an immediate and radical change in their practice and we wanted to find out how they (and, by inference, their patients) were coping with this by implementing a nationwide questionnaire survey.

As our study was low risk, specific written approval by R&D offices other than our local one was not deemed necessary by MREC, but we were advised inform all relevant R&D offices. However, nearly one-third of all primary care trusts then requested further information. To our dismay, we found ourselves in the midst of a slowly emerging and eventually insurmountable bureaucratic response from the R&D departments, as demonstrated in Figure 1. There was an alarmingly wide variation in requirements, ranging from requests for simple documentation to the request that all members of the Scottish research team apply for honorary clinical contracts in a particular English Hospital Trust.

Figure 1.

Survival plot for R&D approval (Kaplan-Meier method)

Table of statistics.

Mean	Median	Interquartile range
65.6351	71	0
59.9410	*	*

The survival graph shows that after 70 days, we only had formal approval from 15% of PCTs. From the shape of the curve, it is clear that our study would never have gained approval from all the relevant parties, at least within our research project's (or researchers') lifetime. Reluctantly, we realized that our urgent and important question had been condemned to a bureaucratic trashcan.

We attempted to drive forward a national response to the problem by describing our abandoned study to key UK policy figures, but despite significant efforts it proved impossible to influence change in the system for the UK as a whole. Fortunately the Scottish R&D Consortium was more receptive, and as a direct result of our findings, there are now plans to pilot a Multi-Centre Research and Development Office (MRAD) that will create a standardized system in Scotland.

What could such an organization look like? It would be made up of R&D managers and co-ordinators from local offices. Members would be experienced in research and have a range of skills such as finance, statistics, ethics, risk assessment and clinical/pharmaceutical knowledge. The study's suitability for the NHS would be considered by MRAD, leaving only local issues to be approved by the individual offices. The researcher would benefit as all R&D issues and paperwork would be dealt with at the outset by one contact, clarified and then disseminated to all applicable sites. The individual R&D offices would still be entitled to reject a favourable MRAD review but only if a local issue is recognized that had not been appreciated by the committee. It would be helpful if such a system, as it developed, were overtly linked to the MREC system so that there would be a ‘one-door entry’ system for researchers and MRAD would follow the same strict time frame of approval as MREC. There is some evidence that MREC has managed to achieve a national system and has improved the time taken for ethics approval.³ Can R&D do the same?

Footnotes

DECLARATIONS —

Competing interests None declared

Funding None

Ethical approval Not applicable

Guarantor JEB

Contributorship All authors contributed equally