Rapid Responses to:

EDITORIALS: J A Muir Gray Evidence based policy making BMJ 2004; 329: 988-989 [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Cause and effect are often less clear and too late

Tim Wilson   (2 November 2004)

Policy making in the NHS

Daphne I Austin   (15 November 2004)

Cause and effect are often less clear and too late 2 November 2004
Tim Wilson, GP Mill Stream Surgery, Benson, Oxon. OX10 6RL Send response to journal: Re: Cause and effect are often less clear and too late	Surely one of the major barriers to better evidence based policy is that the evidence base is weak and too late. In clinical practice, evidence in randomised controlled trials often uses selected groups of patients, excluding those with inconvenient co- morbidities who would spoil the trial design. And yet these patients are the very patients who would benefit from the evidence base. Therefore, when the GP tries to explain the risks and benefits of the options for managing atrial fibrillation to a patient with depression, the decision has to be made on some trial, some knowledge of pharmocokinetics and lots of guess work. So it is with policy; how can we tell if a policy will or will not work either in a different time frame, different environment, or different context if the trials do not exist? All we can be certain of is that the policy will have some effect and be reasonably sure of the sense of direction. Policy making is, as Dr Gray outlines, within a social context. And that changes rapidly, both in terms of broader social issues, but also politically. The evidence base (research) has difficulty keeping up with these changes, so by the time the evaluation, report, or study has been published, things have moved on. This implies the need for a formative rather than summative reserach methodology. The one hope is when the policy comes round again. Practice based commissioning, although different in many important ways from GP fund holding, and occuring in a different system, can learn from previous research on primary crae led commissioning. Maybe we need to look to constantly rediscover lost policies in the hope they will have a sound evidence base behind them. Competing interests: None declared
Policy making in the NHS 15 November 2004
Daphne I Austin, Consultant in Public Health West Midlands Specialised Services Agency Send response to journal: Re: Policy making in the NHS	Editor As a public health practitioner, I find it hard to argue against the principles of evidence based medicine (EBM). However a worrying trend can be observed - EBM is becoming a substitute for good policy making. I agree with Muir Gray’s thesis that evidence based policy making has to take into account both the evidence and the needs and values of the population. I would add a few other things of which affordability and practicability are two. Policy making should also always be done within the context of priority setting. Take for example NICE’s clinical guidelines relating to in-vitro fertilisation (IVF). This is a classic example of uni-dimensional policy making (and NICE guidelines are de facto NHS policy in the eyes of the public, patients, politicians and many clinicians). NICE’s focus was purely on IVF. However any policy maker would also need to take into account the evidence base concerning the consequences of IVF. The result would be a requirement to balance the needs of adults with that of seeking the best outcomes for the children potentially born as a result of IVF. Add to this the ethical dilemma which comes from the fact that most disabled IVF children (arising from complications of a multiple pregnancy) find themselves in this position as a direct consequence of a medical intervention – something over which choice can be exerted. Disability, in this instance, is iatrogenic. Finally, implementing NICE recommendations would impact on obstetric and neonatal services. Many of us commissioning newborn services would question whether the neonatal care capacity needed to implement a ‘two egg, three cycle’ IVF policy can readily be met. The current recommendations are therefore not practical. Finally, because one cannot escape affordability, policy makers, having reconciled the above tensions, must decide the priority to be afforded to their preferred policy option. The above illustrates the complexity of even an apparently simple policy decision. It also demonstrates that a purely evidence based approach will not necessarily give us a sound answer. By all means let us continue to develop EBM but this should be balanced with developing and strengthening good, affordable and ‘joined up’ policy making within the NHS. Competing interests: None declared