Do NICE and CHI have no interest in safety? Opinion of the book NICE, CHI and the NHS reforms. Enabling excellence or imposing control?

Seventeen eminent and experienced people have contributed to this most valuable review of NICE and CHI and their potential impact on clinical practice in the UK. There is essentially 100% agreement that the basic concept is a good one; we all want to have the highest possible quality of clinical practice and improvements in health care. This is all motherhood and apple-pie stuff which goes without question but the problem is how it is put into effect. The contributors are also in agreement and fear that central desire for control will outweigh the benefits. The most recent NICE action, which was leaked to the media as a 'preliminary opinion', concerned the use of beta-interferon for the treatment of multiple sclerosis (MS). The opinion seems to be that beta-interferon is very expensive, that, yes, it does help some sufferers but, no, it does not help others and because it costs more than the NHS can afford no one can have it. This seems to me to be a most unsatisfactory outcome. Surely what clinical excellence demands is the refinement of diagnostic capabilities so that those who will benefit may be distinguished from those who will not. In the meantime we do the best we can even if it does mean that the NHS has to pay for some patients who do not respond. This is the inevitable consequence of the belief that a 'free' and comprehensive health service can be provided out of general taxation. Beta-interferon for the treatment of MS is an example of the observable fact that medical science is advancing at a rate considerably in excess of possible increases in funding. Possibly the most important problem identified in this book is the absence of a relevant, high quality data source for the preparation of the numerous guidelines that NICE is expected to produce each year. In a fully grown science a starting point for a quantitative procedure is the establishment of a baseline and, having done that, the scientist's next step is to produce a standard curve for use in the measurement of further investigations. I have said previously that medicine is not a fully grown science (which is one of the problems) but that does not mean that basic scientific method can be abandoned. What is the baseline for the evaluation of clinical practice? The best would be records of the progression of a disease-state in untreated patients. That, for obvious reasons, is clearly not possible so a compromise is unavoidable. Unfortunately we do not even have that compromise baseline so how do we know what is better and what is worse? In simple, single disease states and within the limits of RCTs that is sometimes possible but in a population composed of many elderly people with multiple pathology it is greatly more difficult. If NICE is to produce authoritative guidelines then its first task is to define a (compromise) baseline. For the readers of this journal the absence of safety as one of the measures of clinical excellence must be a matter of concern. All clinical interventions may be casually related to adverse reactions which may, on occasion, be serious or even fatal. Perhaps excluding safety was a conscious decision by those who created NICE. At the time of market approval information on safety is almost always limited to events occurring more frequently than 1 in 1000 exposures which is far below the desirable level of precision. If NICE is to provide advice at the time of market authorization or shortly after then it will never be in a position to include an acceptable evaluation of safety. So why give a hostage to fortune by mentioning it in NICE's remit? Time alone will tell whether NICE and CHI achieve health improvement or whether they prove to be no more than a political gesture.