opinion & debate Beware the problems of centralisation. Commentary on . . . Mental health research system in England {

It is difficult to disagree with Chilvers & Clark that, overall, recent work to bring a more systematic approach to the organisation of mental health research in England has been a good thing. It is also necessary if mental health is to compete for research funding with other branches of healthcare

It is difficult to disagree with Chilvers & Clark that, overall, recent work to bring a more systematic approach to the organisation of mental health research in England has been a good thing. It is also necessary if mental health is to compete for research funding with other branches of healthcare. However, recent changes in the research system have not all been positive and there is a danger that the process of centralisation, which is inherent to the model they describe, will have unintended adverse consequences.
Chilvers & Clark acknowledge that 'the UK has a strong reputation for outputs from mental health research' compared with other industrialised countries. This reputation was gained under the 'laissez-faire approach to R&D' that Chilvers & Clark argue has prevailed for the past 40 years. This suggests that there were some good things about the old system; care must be taken that the baby is not thrown out with the bath water.
The problems, real and potential, are discussed below.

The stifling of research by bureaucracy
This has been most apparent in three areas: (a) the overcomplex process for obtaining ethical approval for multicentre studies (Alberti, 2000;Wald, 2004;Department of Health, 2005); (b) cumbersome, diverse and slow procedures for obtaining agreement from local research governance committees for research to be conducted in National Health Service trusts; and (c) the detailed and varied reporting requirements of the system for allocating support for science funding. Far from these changes achieving the Department of Health's stated intention to 'minimise bureaucratic process and facilitate high quality research' (Department of Health, 2004), it has made multicentre health services research very difficult to conduct, discouraged local small-scale research and led to a high proportion of the support for science funding, allocated to non-academic trusts, being spent on managing the bureaucracy of accounting for that funding.
Although changes in research ethics and governance procedures were partly in response to the need to implement the EU Clinical Trials Directive (European Parliament and Council, 2001), the way this has been handled in England is the product of the centralisation of the 'stewardship' of research. For example, the current Standard Operating Procedures for Research Ethics Committees, issued by the Central Office for Research Ethics Committees, runs to 219 pages (National Patient Safety Agency, 2005).

The narrowing of decision-making about research priorities
The changes described by Chilvers & Clark will inevitably result in existing and new research funding being channelled more into priorities set by the Department of Health and academics. Although this is understandable, it can only be at the expense of those priorities identified by people working at the coalface of mental healthcare.

The suppression of innovation
Chilvers & Clark state that 'we need to find ways of ensuring that the innovative research of the future is supported'. However, a more centralised system for setting priorities and allocating funding will not naturally achieve this. Presumably, the new system will not include frequent calls for 'responsive funding' bids for grants to support studies on topics chosen by an individual research team.

The creation of cartels
The greatly increased cost of obtaining research ethics and governance approval for multicentre studies works in favour of research institutions with substantial administrative infrastructure. Also, the mental health research network and its hubs will favour the academic institutions involved in their management. Although this is good in that it creates collaborations that can compete on a more equal footing with other medical specialties, it is important that it does not lead to a reduction in competition for research grants; particularly for the larger scale studies that might consume a higher proportion of future funding.

Conclusion
The future health of mental health research in England depends on ensuring that the potential benefits of rationalisation and centralisation are realised and the pitfalls avoided.

Declaration of interest
The author is a recipient of research grants, including grants from NHS R&D.

G E O R G E S Z M U K L E R
The ground is in great shape, but can we field a kitted-out team? Commentary on . . . Mental health research system in England { Chilvers & Clark make it clear that research and development (R&D) in the National Health Service (NHS) is now a managed process. There is a coherent strategy and much thought has been given to the use of resources. I welcome this development and the 'Health Research System' that forms its framework. Mental health R&D now has its own voice and leadership within this system, and the benefits are already evident. Especially noteworthy are initiatives such as the Mental Health Topic Working Group and the Strategic Review of Mental Health R&D and their recommendations, and the consolidation of the UK Mental Health Research Network (UK MHRN; originally established via the National Institute for Mental Health in England; NIMHE). The UK MHRN provides a much needed national infrastructure for studies on a scale which is scientifically necessary, and has in its brief existence already adopted over 20 clinical trials.
The creation in 2004 of the UK Clinical Research Collaboration is another notable step. It brings together in partnership with the NHS, the key stakeholders -the main funding bodies, academic medicine, patients, industry, and even the National Institute for Clinical Excellence -to reinforce the position of the UK as a world leader in clinical research.
Chilvers & Clark recognise that all of this is 'only a beginning' but it is one of which they can be proud.
There are, however, two issues, both alluded to in the paper, which are troubling and on which I would like to expand.

Funding for mental health research
The government has announced extra investment in R&D, including mental health, and this has seen an expansion of the UK MHRN. The Medical Research Council (MRC) has funded 57 projects (out of 642 applications) under its »9 million 'Brain Sciences Initiative', supplemented by »1 million from the Department of Health to pump prime further research in mental health. Around half of the MRC-funded projects are directly related to mental disorders. The establishment of a Mental Health Funders Forum, assuming a commitment from them to cooperate and succeed, should lead to a more coordinated approach to research funding. These are all positive developments.
However, will the funding of mental health research ever reach a level commensurate with the 12% of the global burden of disease (around 20% in Europe) imposed by mental disorders (World Health Organization, 2001)? Perhaps it approximates the global figure for the NHS R&D spend (total of »540 million for 2003/4) -the Clarke Report in 2002 stated it was around 11%. On my calculations the total NHS R&D spend is just 0.8% of the total NHS budget, far below the 1.5% to which the first Director of R&D aspired. Furthermore, most of this funding is tied up in services, with only a small proportion available for new projects. A service innovation can only be funded at the expense of an existing service. Sadly, there is no longer an element of 'responsive' (or