I am particularly grateful to Vanessa Burholt for an opportunity to speak under this heading. Gerontology is a lifelong view of health in all its dimensions, of health care as applied human biology. With examples from my own experience, I hope to show how we might squeeze this idea into a public agenda now filled almost exclusively by body repairs and replacements, commodities vulgarised and promoted to credulous consumers by even more credulous politicians, claiming to act on their behalf.

Gerontology aims both to explain life through research and to change it through social action; not just personal action, because no person can be fully understood without knowledge of his or her social context. Primary care can provide the continuing interface between medical science and the people it's supposed to serve, which is necessary for gerontological knowledge to be applied in practice. Practice without theory is blind, theory without practice is sterile. Without application, gerontology cannot usefully expand, so the health of your subject in part depends upon taking primary care seriously.

Effective social action always needs some research. First, because each locality and population is different; second, and less obviously, because knowledge applied to whole populations must either originate from biologically and culturally similar populations, or fail to provide effective solutions to the most important real problems. The most cost-effective way to achieve this is to bring service, teaching and research functions together in combined units.

All general practitioners in the UK National Health Service have since 1948 had at least two of the three most essential prerequisites for all field research on human populations. First, all NHS GPs have had defined, relatively stable, listed and regularly updated populations at risk for events significant for health, concentrated in areas for which a mass of circumstantial data is regularly recorded by census and other social agencies. Second, all of them have had, at least until recently, continuing initial responsibility for recording these events and coping with them. Rational and quantified approaches to research in populations begin from a numerator/denominator approach: a numerator of defined events or characteristics, and a denominator population at risk for these events or characteristics. GPs in the UK NHS - but not in the more or less shopping-around fee-earning markets of USA and Western Europe - have had this numerator/denominator opportunity for planned rational care ever since 1948.

Of course, only a small minority of GPs have ever used it, but the opportunity to make at least some sort of start was there, even if the costs (in all senses) of getting any further than this were and still remain almost prohibitive, and no social agency ever took responsibility for developing the full service, teaching and research combination in primary care even as a demonstration project, though this combination has been accepted in teaching hospitals for at least one hundred years.

To study any particular class of events in a population over a given period of time, one has also to study non-events in that population. To study any particular kind of ill-health one has to study not only people who have it, but also those who don't. The question why many people don't get ill is just as important as why a few of them do. This question is impossible to answer, and therefore almost unthinkable to ask, in hospitals and clinics which concentrate pathology and exclude health. As workers in primary care rediscover the world outside hospitals, they become capable of asking these more fundamental questions. All they then need is public support in doing so. The problem, of course, is not the public, but the governments claiming to represent them.

GPs acting as shopkeepers responding to presenting illness may not seem the best people to develop either research or rational proactive practice, but they still hold most of the power in the place where these can be done most cost-effectively. Until the 1990s GPs were moving steadily away from competitive shop keeping towards co-operative and proactive public service - far too slowly, but in a sustained and consistent direction - and I was one of them. Having the rare benefit of an elementary political education, and using the despotic power then possessed by GPs (and now being handed to commercial providers) I believed it was possible to develop a project of my own, which would demonstrate the greater effectiveness of practice combined with research and teaching.

Tolerable and effective medication for severe high arterial pressures began to be available in the late 1950s. As arterial pressures at any level rarely prompt symptoms before they have caused organ damage, it seemed obvious that any rational approach to its control should be proactive, reaching our entire adult population. Shop keeping would not be enough. It took my wife and I seven years of extremely hard work to get sufficiently ahead of demand for traditional demand-led care in a very sick population with very poor clinical traditions, before we could search for needs - and begin to address even secondary prevention in a systematic, proactive way, which we did from 1968 onwards.

I chose whole-community arterial pressure control as our first target, mainly because event rates were obviously high and effective treatment was available. But I also had a deeper agenda. If we could solve the problems of whole-community control of arterial pressure, we might also have a basic model applicable to every other problem with a quantifiable risk of organ damage or premature death, and soluble through continuing care. And so it proved. In 1987 we reviewed the results of 25 years of increasingly proactive search for opportunities to prevent an increasing range of common disorders by early diagnosis and treatment. We compared death rates under 65 for the five years 1981-86, in Glyncorrwg (which had developed a cumulative proactive programme since 1968) and in the socially similar community of Blaengwynfi (which had received only traditional demand-led care).(1) Age-standardised death rates under 65 were 28% lower in Glyncorrwg than in Blaengwynfi over the five-year period. Differences were mainly in deaths in the first year of life and cardio respiratory causes of death, the pattern expected when care becomes more effective.(2)

This evidence is limited by small numbers and its design as an at least partly natural experiment, but on the potential effect of proactive care when added to traditional reactive care; it is still virtually all we have. The 28% difference is certainly subject to a wide margin of potential error, but it seems unlikely that these large apparent differences were either not real or not caused by our proactive policy.

The potential contribution of secondary prevention to longer and healthier life and healthier deaths is small compared with the potential contribution of primary prevention. Even within the bounds of present knowledge, the scope for primary prevention of most causes of premature and/or unhealthy death is grossly underestimated. This is obvious for breast cancer(3), cancers in childhood, (4) and even more so for gross poverty and medical neglect of at least one-third of the world population despite unprecedented growth in material production. Increasingly polarized distribution of economic power and consequently of wealth maintains its rule through a generally shared culture of fatalism, which wrongly assumes that we are ignorant of the causes of most premature or unhealthy deaths, and that the best way to help poor people is to make already rich people richer still, so that they may find profitable ways to employ them. These assumptions rest not on a balance of objective evidence, but on faith that so many successful professional politicians, business tycoons, press barons, editors, journalists, celebrities and their attendant post-modernist philosophers of all grades can't all be wrong - and the terrifying thought that if they are wrong, we shall have to develop for ourselves a new philosophy of just about everything - a new common sense.

However, though the potential contribution of proactive secondary prevention to longevity may be relatively small, the potential contribution of health workers to the political and social changes necessary for a general turn toward primary prevention could be decisive. Nothing has been more confusing and counter­productive for progress toward rational health care than the unnecessary and grossly unequal contest between advocates for clinical medicine and advocates for public health over the past hundred years or so. Taxes not spent on care do not go to public health, housing or education, but to more prisons and wars. Speaking rationally, public health is not a minor specialty subordinate to clinical medicine, but vice versa - until clinical medicine accepts its role as a subset of public health, it can have no intelligently planned direction. Instead of addressing needs, it will just keep on trying to meet wants.

The NHS in England has already passed the crossroads where we might still choose to develop rational policies addressing the lifelong agenda for health care demanded by a gerontological approach, within the existing administrative frame. Responsibility for primary care has become so fragmented that the gatekeeper function of GPs, through which the NHS became for its first 50 years the world's most cost-effective care system, is falling apart. The NHS in England is now converging on the absurd waste of competitive corporate care in USA. It has become an entanglement of profiteering with public service, opened to administrative gaming, insider trading, and corruption.(5) Continuity for citizens and co-operating professionals within registered populations, the first requisite for any lifelong approach either to research or service, has been sacrificed to create so-called consumer choices. These are in fact, and in intention, business opportunities for competing multinational corporations(6).

However, you are meeting in Wales, where, about half a century before its birth, the NHS was conceived by its true parents, the South Wales coal-mining, steel and tinplate making communities which created their own prepaid medical care schemes. These were funded from poundage - sixpence or so from every £ of wages, deducted at source - effectively, a local income tax. These schemes included whole communities; wives and children, but also teachers, milkmen, ministers and even vicars. Some even sought to provide integrated and comprehensive care including hospital specialties.(7) Miners throughout the UK created prepaid medical care schemes, but only in Wales were these provided from all according to their ability, to all according to their need. Though painfully slow to anger, there are hopeful signs that Wales as NHS birthplace may also become its active guardian. Having experienced health care as a human right, the NHS as an idea has become entrenched in the minds of the people. This makes it too powerful a potential vote-winner to be permanently discarded by political leaders, even in England. Somehow we shall eventually find the courage to renationalise the NHS throughout the UK, this time with its patients and their carers recognised as participants rather than supplicants, and its staff recognised as applied scientists rather than shopkeepers. The aim of longer healthier lives, and shorter healthier deaths, will become their core measure of shared output - and gerontology will reach maturity as public intellectual property, useful for longer, more intelligent and effective living.

1. Hart JT, Thomas C, Gibbons B, Edwards C, Hart M, Jones J, Jones M, Walton P. Twenty five years of audited screening in a socially deprived community. British Medical Journal 1991; 302:1509-13.

2. Kaul S. Twenty five years of case finding and audit. British Medical Journal 1991 ;3Q3:524-5.

3. UK Working Group on the Primary Prevention of Breast Cancer. Breast Cancer: an environmental disease. The case for Primary Prevention, www.nomorebreastcancer.org.uk and info@nomorebreastcancer.org.uk 2005.

4. Kaatsch P, Steliarava-Foucher E, Crocetti E, Magnani C, Spix C, Zambon P. Time trends of cancer incidence in European children (1978-1997): report from the Automated Childhood Cancer Information System project. European Journal of Cancer 2006;42:1961 -71.

5. Talbot-Smith A, Pollock AM. The New NHS: a guide. London: Routledge 2006.

6. Hart JT. The Political Economy of Health Care: a clinical perspective. Bristol: Policy Press, 2006.

7. Hart JT. Storming the Citadel: from romantic fiction to effective reality. In Health and Society in Twentieth-Century Wales. Cardiff: University of Wales Press, 2006:208-15.