'My local supermarket does not call me an inappropriate shopper - ever'

??When Iuse a word,? Humpty Dumpty said, in a rather scornful tone, ?it means just what I choose it to mean, neither more nor less.??

Alice in Wonderlandwas not the first book to explore the wonderful ambiguity of words, but it remains one of the funniest. Meanings change: sometimes deliberately, sometimes without us even noticing. ?The question is,? said Alice, ?whether you can make words mean so many different things.?

And the answer, in the health sector, is ?Why certainly, ma?am.? Perhaps it is something learned from doctors who occasionally have to say particularly unpalatable things to their patients, but many health managers become exceptionally skilled in the use of euphemism.

Take, for instance, the phrase ?managing expectations?, used to describe a necessary and wise approach to communicating clinical prognosis ? as in, ?this may hurt?, or ?with that fractured skull you?d be better off giving up goalkeeping for a while?. It has now slithered into management parlance as code for ?you can?t have everything?.

Or at least that?s what I think it means. Actually, as Mr Dumpty says, it means just whatever each user chooses it to mean. Rather like ?demand management?.

What does ?demand management? actually mean? Typically it is used, not unlike ?performance management?, as a thinly veiled threat, something we?ll unfortunately ?have to do? (to someone else). The health economy is in deficit, we can?t afford all this extra activity: we?ll have to manage demand.

But the phrase actually seems to encompass at least three separate meanings. The first is a rational, or at any rate transparent, approach to deciding what the NHS will and will not fund, and then withdrawing funding from anything that falls below a cut-off line.

In an idealised world this is what commissioners would do, basing their local cut-off line on rigorous health needs analysis rather than the pleas of providers.

It is a logical enough approach, but there are serious flaws. My opinion on what should not be funded will probably differ from yours. You may not think supplies of incontinence pads are an appropriate use of public funds; I happen to think they are essential. Some say the NHS should not pay for psychotherapy; others insist it should. And there is no right answer.

What?s more, no sooner have we decided something will not be funded than the clamour for it grows, and along with it the pressure on politicians. In fact, were we to wish to stimulate demand for a treatment, denying access to it would seem a pretty effective approach.

This principle has allowed the pharmaceutical company behind Herceptin to build the drug?s high public profile with negligible investment in advertising; and one or two primary care trusts have learned a hard lesson in political realities along the way.

So why not try a second approach: rationing without transparency? Why not create a process for reviewing hospital referrals so that unnecessary (or, in NHS-speak, ?inappropriate?) referrals can be returned to sender?

This is in essence the approach, based on the clinical assessment and treatment services (CATS) model pioneered in Lancashire, that an increasing number of PCTs in England are now adopting. Instead of an explicit cut-off line, it relies upon peer review and typically hinges upon the opinion of a panel of GPs.

Does it work? Well, it depends what you mean by ?work?. At least in the short term, the CATS approach can reduce hospital referrals, either through explicit challenge, or the fear of challenge, or simple bureaucratic delay.

The whole-system impact is, however, harder to gauge. Will GPs continue to play the gatekeeper role if it brings them into direct conflict with their peers? And how will acute hospitals, at whose expense most of the savings are being made, react? They are currently being encouraged to market their services more professionally, presumably to stimulate demand.

And, crucially, how will patients react? In a consumer age, who readily accepts refusal? My local supermarket does not call me an inappropriate shopper ? ever. It does not suggest that, with my meagre shopping list, I would be better off at the corner shop. No, it accepts my money happily, sometimes even with a smile.

So how will you, intelligent reader, react when you are the patient and your GP says, with a sad smile, that they can?t get you a hospital referral for your condition? Shrug your shoulders and accept your lower priority?

What CATS certainly is not is demand management. It is referral management. Demand may be suppressed, but has a habit of bubbling up elsewhere. We have never yet, for instance, been able to reduce demand for accident and emergency services.

Genuine demand management, the third approach, actually reduces the call for NHS treatment and care, and the crucial area is the effective management of chronic illness. Not just the small group of regular users of NHS hospital services, and not just the big killer conditions, but millions of people who successfully manage illnesses ranging from hypertension to arthritis to depression from year to year.

From this viewpoint, the evaluation of the NHS?s Evercare trials, published in the British Medical Journal last month, makes fascinating reading. Has Evercare failed? Well, it depends what you mean by ?failed?. The evaluation raises many good questions, but in the background lurks perhaps the biggest question: why measure the success of chronic-disease management by counting reduced hospital admissions. Yes, there is a relationship, but it sort of misses the point. Would we measure the effectiveness of surgery by counting GP attendances?

Successful demand management isn?t really about thresholds of care, gatekeepers or case management. It?s actually about the effective self-management of long-term conditions: the effective de-mystification of numerous aspects of care. We do, indeed, need to manage expectations. Upwards.

Noel Plumridge is a former NHS finance director.