NICE: a panacea for the NHS?

Harold Wallace Ross, the great editor of the New Yorker, had a continuing fantasy that the next person he hired would bring “grace and measure” out of chaos.¹ They never did. But his is a common fantasy, entertained at some time by most employers, secretaries of state, and prime ministers. The National Institute of Clinical Excellence (NICE),* which begins its assault on Olympus next week, makes me think of Ross. I have heard NICE mentioned as the solution of most of the NHS's problems: rationing, poor practice, the failure of good practice to spread, postcode prescribing, the mindless adoption of technology, the absence of a sensible mechanism to introduce new drugs, and variations in outcome. How much can we realistically expect?

Like any other institution NICE will evolve, but it begins with three main functions: appraising new technologies, including drugs, before they are introduced into the NHS; issuing and kitemarking guidelines; and encouraging national audit. Most of its initial energy will be put into the first function, and this is the beginning of explicit, national rationing. It is also the appearance of the “fourth hurdle” in that to become widely used in the NHS new drugs will have to prove themselves not only to be pure, efficacious, and safe but also better in some way than what is currently available. The mess over the introduction of sildenafil (Viagra) into the NHS—the delay and the botched criteria on who would get it²—shows that a better mechanism is needed. And the government's focus groups will have told it that the public doesn’t like at all the fact that you can get new and expensive treatments if you live on one side of a street in one health authority area but not if you live on the other side, in another health authority.

So the case is strong for NICE appraising new treatments and technologies, and it seems set to do it well. The discussion document produced by the NHS Executive in January on how NICE appraisal will work promises horizon scanning for new technologies; transparent, rapid, evidence based appraisal that considers effectiveness and cost; input from patients and companies; and a clear outcome.³ There will be essentially three possible outcomes: use routinely in the NHS; use only in the context of trials; or don't use. Routine use may be recommended for everybody or for particular specialists. It will, however, be for ministers to decide exactly what the NHS should do, and here the system begins to creak.

There seems little point in NICE adopting a transparent process if its recommendations then disappear into a black box at the Department of Health only to emerge in garbled form six months later. Ministers do have the great advantage of accountability, but is the accountability of being a member of a tightly controlled party that gets elected every five years adequate for 1999? And what about the accountability of NICE? Who appointed its first chairman Sir Michael Rawlins, pleasant fellow that he is, and what process was used? There is an inevitable sense that although the government has learnt the rhetoric of transparency, accountability, and evidence based appraisal it would rather avoid living with it day to day.

And although NICE is to make a beginning with rationing (avoiding the words at all costs for fear of startling the horses) it won’t achieve much by simply considering what’s new. Intellectually sound rationing would mean weighing what’s new against what’s already there, and it would necessitate finding a way of choosing among resources spent on new drugs and on the number of nurses at night in geriatric wards or on facilities in the community for people with learning disabilities. Rationing in Britain works mostly by dilution rather than denial: it's politically so much easier, particularly if you dilute services for the most marginal.

NICE will be concerned with what’s already there through its work on guidelines, and Sir Michael has a vision that “doctors will go to work with the British National Formulary in one pocket and a copy of NICE guidelines in the other.”⁴ Sadly, this vision may reflect Sir Michael's naivety about guidelines. Firstly, guidelines that covered every eventuality would be carried in a wheelbarrow not a pocket. Secondly, guidelines are difficult and expensive to produce, and the most tricky part is making the jump from evidence to recommended actions. Those making that jump resort not only to wisdom but also to prejudice and self interest. Thirdly, guidelines on their own change nothing.⁵

Here we arrive at what may be the biggest failing of NICE. Centralist direction is a poor way of solving the NHS's biggest problem, the fact that good practice may flourish in one clinic and fail to spread even to the clinic next door let alone the rest of the NHS. Meanwhile, poor practice gaily continues. Those who try to run the NHS are understandably frustrated by these failures and naturally turn to organisations like NICE and its less often mentioned brother CHI (Commission for Health Improvement, or “nasty” as it’s widely known) to put things right. But their controlling instincts are probably wrong. “Over the long run,” writes Peter Senge, an academic at the Massachusetts Institute of Technology and one of the originators of the idea of the learning organisation, “superior performance depends on superior learning.”⁶ And control limits learning. “Control limits space. Learning needs space,” said Arie de Geus, probably the originator of the learning organisation.⁷ “It is simply no longer possible for anyone to ‘figure it all out at the top’ ”⁶ and “little significant change can occur if it is driven from the top.”⁸ Ironically, both Senge and de Geus were speaking at a symposium organised to identify how to sustain the NHS for the next 50 years.

In conclusion, NICE should help with rationalising the introduction of new technologies into the NHS, and the less politicised and more transparent its process the better. It might develop into an effective means of rationing all health care, but it is likely to struggle with solving the important problem of variable performance throughout the NHS. No one institution could produce so much.