'Tribes' need a dose of data?

Every four or five years the government, regardless of whether it is Conservative or Labour, "redisorganises" the structure of the National Health Service. This is a phenomenon familiar to all in public and private organisations and is a product of "innovative management" and/or a defensive inducement of organisational chaos to fog the vision of observers and convince them that beneficial change is on the way.

The prime minister, before the Iraq madness, had focused his energies on public-sector reform. The NHS has been given unparalleled growth in its resources for the next three years in particular. Tony Blair required in return that all decision-makers would "act smarter". This is quite a challenge for an organisation that is extremely labour intensive and much constrained by cartels - public, such as the British Medical Association, and private, such as the international pharmaceutical industry.

This book analyses the attempts of managers in one organisation - Leicester Royal Infirmary - to re-engineer processes of work and improve levels of activity (which influence waiting time) and outcomes both in terms of the processes of care (for example, staff being nice to you) and health outcomes (that the care given increases your length and quality of life).

Their study is qualitative, offering systematic descriptions of how process redesign was attempted in Leicester in the mid-1990s and how its success was limited.

Hospitals are unique organisations. Their activities are dominated by doctors whose diagnostic and therapeutic choices determine the commitment of society's scarce resources. This "tribe" defends its clinical freedom tenaciously - hospital consultants in England in 2002 rejected a generous new contract because they feared they would be managed, and their salaries part determined, by non-clinical managers. The authors point out that doctors believe that they can manage their organisations in the same way as they manage patients.

However, a fundamental problem with healthcare worldwide is that the profession poorly manages medical practice. Furthermore, those who pay for it, whether taxpayers or insurance premium payers, often have little evidence that they get good value for money. The implicit contract internationally in medicine is that those who pay trust the doctors to deliver timely and good-quality healthcare.

The research literature demonstrates that all may not be well in medical practice. Medical practice variations have been analysed for three decades by researchers such as Jack Wennberg in the US and Klim McPherson in the UK. While to vary is human, the quantity of variation in medical practice is remarkable. For example, the French consume four times more antibiotics per capita than the Dutch, and have correspondingly higher levels of antibiotic resistance.

This variation is accompanied by evidence of inappropriate care and medical errors. Some estimate that as many as 70 per cent of women who have their wombs removed due to heavy bleeding could be managed medically with pharmaceuticals. It is also estimated that one in ten hospital admissions is associated with an "adverse event" in the UK (the Australians estimate the error rate may be 16 per cent). These can be major, for example - the removal of a healthy kidney. The Americans estimate that such errors kill more of their citizens than breast cancer, motor-vehicle accidents or HIV-Aids. Medication errors kill twice as many Americans a year as died in "September 11".

Poor data collection and poor management of available data perpetuate dangerous inefficiency in all healthcare systems. For hundreds of years writers have demanded greater accountability but have been largely ignored.

Florence Nightingale demanded systematic and routine data collection in hospitals and advocated the measurement of success in terms of whether patients were "dead, relieved or unrelieved" after care. No healthcare system has such measurement mechanisms in the 21st century, and so consumers cannot make well-informed choices.

Against this background Peter Homa (then chief executive of Leicester Royal Infirmary and now chief executive of the government's major NHS inspection agency, the Commission for Health Improvement) and his colleagues sought to "re-engineer" processes in the Leicester Royal Infirmary by "recreating and redefining the value chain".

Central to the re-engineering was the creation of specialist management directorates in which clinical leaders were given an increasing responsibility for their budgets and the delivery of activity targets. Each of these nine directorates was to work to "the continuous improvement model" associated with US "guru" Donald Berwick. The impact of these reforms was very uneven - considerable in gastro-enterology and small in neurology and emergency medicine. An important explanation of this variation appears to be the "engagement" of the clinicians in the change process, the collaboration of nursing staff (who generally consume more than 35 per cent of hospital expenditure) and management capacity to support clinical and nursing change.

Margaret Thatcher's NHS reforms, further developed by Blair, have focused on the development of market-like organisations and improved management.

Terry McNulty and Ewan Ferlie describe insightfully how managing within these markets is complex and slow. Theirs is a fascinating and frustrating book. The fascination comes from the "stories" of the change process. The frustration comes from the rejection of quantitative methods and the failure of the designers of this re-engineering process to achieve systematic engagement of, and change in, clinical practice.

Such change will be possible only when routine data are integrated into management processes alongside the evidence base about "what works" in medical care. Most managers do not know what their consultants do, let alone whether they do it well or fatally as in the case of paediatric cardiac surgery in Bristol.

Management is about the control of decision-making in relation to corporate objectives. This nice description of the expensive Leicester re-engineering process demonstrates how difficult it is to lead and manage when there is local resistance to change amid the continuing political "redisorganisations" of the NHS by ill-informed politicians.

Blair's "redesign" of the NHS is encountering the same problems as those met in Leicester. He may fail, as those in Leicester did, because of a reluctance to use data to inform and incentivise clinicians. Getting doctors to "act smarter" is never easy, particularly when they are in short supply and exercising their monopoly power. Hundreds more Leicesters are now being funded by Blair. One hopes they will deliver demonstrably better care to the population.

Alan Maynard is professor of health economics, University of York.