BMJ Education and debate

Bookmarkers beware:
Bookmarks to pages other than the home page may not work after we change servers in April

Editor's Choice | This Week in BMJ | Press releases

BMJ No 7126 Volume 316

Education and debate Saturday 17 January 1998

Is NHS purchasing serious? An American perspective
Donald W Light
In 1991 Britain's government forced through one of the most sweeping transformations ever of a health care system, from an administered governmental service to an internal market of contracts between purchasers and providers.(1) Confronted by hospitals running out of cash, the prime minister asserted that the NHS suffered not from underfunding but from inefficiencies that competition would eliminate.(2) The model of managed competition used seemed ill suited at the time(3,4) and has been shown to be deeply flawed.(5) A recent review of the past six years concludes that the reforms scored some gains but raised costs, created new inefficiencies and inequities, left most old inefficiencies intact, and caused fragmentation.(6)
The new Labour government aims to heal these wounds, reduce inequalities, and install a cooperative model of commissioning for health gain.(7) But it also risks not addressing why the NHS is still a fragmented, wasteful system. How can this government (not to mention doctors and their patients) avoid being pushed to the wall by clinical funding crises each winter, as Margaret Thatcher was in 1989?
Help may come from an analysis of what the most enlightened commissioning groups in America have learnt in their struggles to contain costs and increase value for money. Although America has the most costly, wasteful, and inequitable health care system in the world, one can extract from its struggles to control costs through purchasing, without a treasury to set a firm budget, powerful lessons for managing costs that the British government has not yet learnt.

Summary points

Despite almost seven years of an internal market, commissioning in the NHS is still ineffectual. In its revision of the NHS the Labour government is in danger of ignoring the lessons learnt in America - in a much more wasteful system - of the best ways to purchase clinically effective health care and gain the best value for money

Just as England (though not Scotland) is abandoning unified health authorities, the main lesson from America is that purchasers need to be big and strong

They need also to purchase for all services (to prevent cost shifting between budgets) and to start by building accurate comparative data systems that identify the most cost effective providers

From ineffective to effective purchasing

In America health care is paid for by employers and the government (through Medicare and Medicaid). In 1970 both payers felt that escalating costs would bankrupt them, and congress took the lead in efforts to rationalise services and contain costs through working with doctors and hospital executives.(8) For example, it established national systems to review capital projects, plan health care services, and review hospital practices. In each case, the payer's intentions were neutralised or subverted by the providers.(8)
Buyers' revolt
In the 1980s both the largest employers and congress became more aggressive,(9) commissioning experts to review the clinical services they were paying for. Prospective review (gatekeeping) consisted of asking, "Does this patient need to go to hospital? Does this procedure need to be done?" Concurrent review had nurse specialists questioning each morning how much longer inpatients had to stay in hospital. Retrospective review analysed months of computerised records to see how different hospitals and specialty teams compared in the costs of treating comparable cases.
By the end of the 1980s these efforts had developed into comparative profiles of hospitals and specialty teams that are now a common part of performance management in health maintenance organisations. The results were variously used to educate, cajole, or drop more costly hospitals or teams with similar clinical results. They spawned various forms of selective contracting with more cost effective providers and what is now called managed care. The key to managed care is a single budget so that costs cannot be shifted to someone else's budget and savings can be reallocated elsewhere, including incentive pay and profits.
The wiliness of providers
Behind the buyers' revolt was disillusionment with doctors and the hospital industry. Not only had they undermined a decade of cooperative efforts to modify costs, but research showed large unexplained variations of use - as is true in Britain as well.(10-12) These variations and ineffective procedures challenged medicine's longstanding claim to professional status based on scientific expertise. In response, buyers and their managerial agents took over collecting systematic data, monitoring, and auditing because they realised that medical societies had failed to assess the relative effectiveness of alternative interventions and ensured that the most effective were used. Congress created and funded a new institution to do this work, the Agency for Health Care Policy and Research.(13) Although outcomes research is conducted by doctor-researchers, they are under contract to answer buyers' questions about value for money. Meanwhile, congress also sponsored research to create new tools for restructuring contracts and incentives, like diagnostic related groups for hospital costs and relative value scales for specialists' costs.
Nevertheless, costs kept rising, though some payers in some areas succeeded in containing their costs. The more sophisticated employers came to realise that health care is much more complicated to purchase than anything else (even jet airplanes or mainframe computers), and the sellers have all the advantages. They could (and did) frustrate purchasers' efforts to increase value for money by providing less care, or different care; selecting healthier subpopulations; or shunting costs to other payers or budgets. Even Alain Enthoven, the father of managed competition, was so impressed with providers' abilities to circumvent efforts to make them compete on true value for money that he concluded that health care markets would have to be actively managed by a talented watchdog team.(15)
Building purchasing power
Purchasers responded in three ways. They developed increasingly detailed ways to measure what they were buying and its cost effectiveness. They moved towards contracts covering all services so that providers could not shift costs. And they increasingly combined forces, because even corporations as large as Chrysler realised that they lacked enough purchasing power within specific medical markets to control costs well. Today American employers are struggling to attain what the NHS already has - unified purchasing health authorities for a given population with a mandate to develop integrated strategies to meet the health needs of the population.(7)(16)
Many "business health coalitions" have formed,(17) but only a few have really matured. These purchasing cooperatives almost always start by collecting data on performance and cost, identifying gaps and weaknesses in the data, developing integrated data systems, and requiring any provider who wants a contract to collect and share the data they need to assess what they are getting for their money.(18) The most advanced purchasing groups have also identified or developed programmes of prevention, patient education, and self management which they require their providers to implement and monitor.(19) They are interested in developing primary care, but as a frontline service, not as a purchaser. They use their market power principally to contract for secondary and tertiary care, where they face powerful hospitals and subspecialty groups. They are keen to implement any clinical protocols, disease management, case management, or outcomes research (the American term for evidence based medicine) that promise to maintain or increase quality at lower costs.(20)
Strong purchasing - nine lessons

The above story introduces British readers to the driving force behind health maintenance organisations. These are risk taking, capitation based, purchaser-provider partnerships, one level below the basic purchasers. At the contractee level, the purchaser-provider split does not work, but without a strong purchaser in charge, purchaser-provider partnerships will be dominated by specialists, hospitals, and high technology. In the British case, the basic purchaser is the Treasury and parliament - potentially very strong - but their priorities are confused by running the NHS and not wanting to challenge hospitals, consultants, and other providers.

Nine lessons from American purchasers
Purchasers need to be large and strong
They need to contract for all services
Good comparative data on price, product, quality, and service are essential
Devolution and flexibility cannot be primary goals
Nor can patient choice or responsiveness
Purchasers need to be clever, well trained, and with good technical support
Redesigning organisations takes time and money
Primary care needs to be commissioned and monitored too
Professionals are accountable, not autonomous

Large purchasers
The first, and basic, lesson of this story is that if a nation is going to use a purchasing approach to health care, the purchasers need to be large and strong, not small, local, and weak like general practitioner fundholders, locality commissioning groups, or even the proposed primary care groups. Purchasers need to be large enough to rethink how high risk groups of patients are treated and managed, create cross sectoral teams, handle risk, subcontract skilfully, and spread transaction costs over a reasonably large base.(18)
Ironically, the British have large purchasers already, but the English (though not the Scots) are giving up on them. Health authorities already have a legal mandate to purchase the health services for a defined population. Yet their potential is undercut and fragmented in many ways, and now they are going to be reinvented on a smaller scale as primary care groups. Devolved decision making works for clinical services, but devolved purchasing creates greater inequalities and worse coordination.(21) The shortest road to effective purchasing that will reduce the many sources of waste inside the NHS, so that more patients can be treated, is to bolster the authorities that are already in place.
Integrated care
Secondly, purchasers need to contract for the integrated care for an entire course of illness or for whole populations in order to capture savings and reallocate them within integrated clinical teams. Contractees need to share in the savings or help pay for excess costs. In the NHS shifting of costs and services is rampant, and most other sources of waste identified in 1990 are still in place.(22,23) Budget splits, budget barriers, prohibitions against carry forwards and cross-subsidisations all prevent integrated care from developing.(24)
Unifying the family health authorities and district health authorities was an important step in eliminating a split budget around which there was considerable cost shunting. But fundholding then created a new kind of split budget by delegating the easy 20% of purchasing to independent contractors and leaving the tough 80% to hobbled health authorities. The general practitioners' contract constitutes another budget split that separates the very providers who do most of the health care and who determine through referral many of the other costs from the main NHS budget. Unified health authorities have great potential to use evidence based medicine, guidelines, and protocols to squeeze more care out of the current budget, if only they were given the power and resources to do the job.
Comparative data
Thirdly, effective purchasing cannot happen without good comparative data on price, product, quality, and service. Yet the government, as purchaser, has never insisted on such comparative data from providers and until recently has not seemed to give it high priority,(25) though it does feature in the white paper.(8) Total purchasing pilots have found progress stymied by this rudimentary problem.(26) General practitioners have incomplete data on consultants and judge them inconsistently on quality and value.(27)
Effective American purchasing groups start by deciding what outcomes they want and then develop good data systems for all providers in their market area. One of the most effective American purchasing cooperatives, the Pacific Business Group on Health, has worked with researchers on developing increasingly accurate ways to identify which providers deliver greater quality and efficiency, as distinct from "gains" through case mix, service differences, cost shifting, and other easy ways by which one group can appear to be more efficient than another.(20) Getting all contractees to measure and report even costs the same way has been a long struggle that only the best purchasing groups have achieved.(28) Yet without clear specifications about what one wants to purchase, with what outcomes, the present government will end up as before with new costs and little gain.(6)
Devolution should not be a primary goal
Fourthly, these lessons imply that devolution and flexibility cannot be primary goals because both will lead to lower value for money unless firmly framed by clear terms of performance. If so framed, then devolution and flexibility invite creative solutions.
Patient responsiveness should not be a primary goal
Fifthly, these lessons also imply that strong purchasing is not about patient choice or responsiveness in themselves. Both encourage demand beyond need and raise costs. Even in highly competitive American markets, the primary client is the purchaser, not the patient, and, although managed care plans work hard at offering choice and being responsive to patients, they do so within the limits set by the purchaser.
Good managers
Sixthly, purchasing teams need to be very smart, well trained, and with good technical support. Most British managers I have met who have worked with American managers of health maintenance organisations report that they are streets ahead of British managers in these three attributes. The NHS does not need fewer managers but more top flight ones, especially on the purchasing side. Instead, there seem to be thousands of poorly trained managers who bungle things and then move on before they have to face the consequences. Examples abound of managers who close ranks to ostracise critics and bury legitimate criticism. Small wonder that both political parties score popularity points for campaigns to reduce NHS management, but it's the wrong lesson from an American point of view. Managers need to be better and accountable. NHS Executive discussion documents emphasise lower transaction costs, but American purchasers have concluded that substantial savings require higher transaction costs, because forms of waste are so deeply entrenched.
A related problem is that many health authority executives have old, strong ties to hospitals and consultants, so they are unlikely to challenge the embedded inefficiencies of these groups.(29,30) The fact that health authority executives are paid less than their hospital counterparts and have their performance measured by what even they regard as perverse incentives weakens purchasing still more. And the extent to which poorly performing health authority teams are not held to account(31) suggests that the government is not yet serious about cost effective purchasing.
Invest in changing clinical organisation
Seventhly, purchasers must recognise that redesigning the organisation and clinical pathways of care for sick patients takes time and costs money. Purchasing for value without development funds for clinical and organisational innovation will not work.
Commission primary care
Eighthly, primary care needs to be commissioned and monitored. The entire delivery system depends on these frontline teams providing uniformly good care and not referring unnecessarily to costly specialty and hospital services. The large unexplained variations in quality, expenditures, and referrals are a major issue for strong purchasers in America. In Britain, however, these variations have been amplified and locked into fundholding budgets that vary by 250% per patient, independent of variation. Such demand led differences necessarily discriminate against the needs of lower socioeconomic groups.(32) To be "primary care led" mistakes a part for the whole, a means for the end - which is to maximise people's ability to lead healthy, productive lives. For example, as a largescale purchaser, the Pacific Business Group on Health has chosen health education and prevention programmes and now requires any primary care group which wants a contract not only to implement them but to collect and report data on changes in risk behaviour and risk profile.(19)(33)
Accountability
Finally, all this means that professionals are accountable, not autonomous. Perhaps the main reason why American payers have moved towards forms of strong purchasing is to take on organised professional groups which resist doing work on a basis of clinically and scientifically grounded criteria. Major savings have come from challenging the customary practices of autonomous physicians.(34) American purchasing groups have learnt that one cannot minimise costs generated by unnecessary services, ineffective services, unwarranted variations; by overqualified providers; or by overhospitalisation without challenging the professional autonomy that protects these forms of embedded inefficiencies.
The absence of strong purchasing

These lessons indicate how far away the British government is from serious purchasing. Parliament and the Treasury should be, as the Sioux might say, the Great Purchaser in the Sky. But since, after six years, they have not even addressed the fundamentals of developing clear, common measures of outcomes and have pursued strategies that fragment services and budgets, one is led to conclude that there is no serious purchaser in the NHS. This impression is reinforced by Day and Klein's study of how the NHS Executive evolved, in which NHS managers assume ever greater power.(35) Are the NHS Executive divisions structured to facilitate cost effective, integrated services, or to impede them? The executive is structurally impeded from giving ministers good advice on how to maximise value for money. Purchasing hardly appears anywhere in the story. Yet the executive should be a team of experts at contracting and performance management that find new ways every day to treat more sick patients per million pounds.
graph
California has succeeded in reducing bed days well below the national average. Enthoven and Singer argue that this is due to its powerful purchasers(34)

A key problem is that the secretary of state is legally responsible for everything that happens in the NHS. That makes him the chief executive and manager of provision and thus completely hog tied as a purchaser. The new white paper indicates that the government wants to strengthen and integrate purchasing (now increasingly called commissioning) in several ways. But at the top the English parliament and ministers need to clarify their role as the main purchasers, aided by a commissioning executive. And, on the ground, the government needs to rectify the ways in which its plans for local commissioning contradict and undermine its national goal.
This essay is based on a forthcoming report on the white papers and cost effective health care to be published by the Office of Health Economics.

University of Medicine and Dentistry of New Jersey,
Stratford,
NJ 08084,
USA
Donald W Light, professor of comparative health systems

References
1 Klein R. Big bang health care reform - does it work? the case of Britain's 1991 National Health Service reforms. Milbank Q 1996;73: 299-338.
2 Butler J. Origins and early development. In: Robinson R, LeGrand J, eds. Evaluating the NHS reforms. London: King's Fund Institute, 1994:13-23.
3 Light D W. Learning from their mistakes? Health Service J 1990; 4 Oct:1470-2.
4 Light D W. Bending the rules. Health Service J 1990; 11Oct:1513-5.
5 Light D W. Homo economicus: escaping the traps of managed competition. Eur J Public Health 1995;5:145-54.
6 Light D W. From managed competition to managed cooperation: theory and lessons from the British experience. Milbank Q 1997;75:297-341.
7 The new NHS: modern, dependable. London: Stationery Office, 1997.
8 Starr P. The social transformation of American medicine. New York: Basic, 1982.
9 Light D W. The restructuring of the American health care system. In: Litman T J, Robins L S, eds. Health politics and policy. Albany, NY: Delmar, 1997:46-63.
10 Paul-Shaheen P, Clark J D, Williams D. Small area analysis: a review and analysis of the North American literature. J Health Polit Policy Law 1987;12:741-809.
11 Sanders D, Coulter A, McPherson K. Variations in hospital admission rates: a review of the literature. London: King Edward's Hospital Fund for London, 1989.
12 Wilkin D, Dorman C. General practitioner referrals to hospital: a review of research and its implications for policy and practice. Manchester: Centre for Primary Care Research, 1990.
13 Raskin I E, Maklan C W. Medical treatment effectiveness research. Evaluation and the Health Professions 1991;4:161-8.
15 Enthoven A. Theory and practice of managed competition in health care finance. Amsterdam: North-Holland, 1988.
16 Scottish Office. Designed to care: renewing the National Health Service in Scotland. Edinburgh: Scottish Office, 1997.
17 Bergthold L. Purchasing power in health: business, the state, and health care politics. New Brunswick: Rutgers University Press, 1990.
18 Robinson J C, Casalino L P. Vertical integration and organizational networks in health care. Health Affairs 1996;151:7-22.
19 Schauffler H H, Rodriguez T. Exercising purchasing power for preventive care. Health Affairs 1996:15:73-85.
20 Robinson J C. Health care purchasing and market changes in California. Health Affairs 1995;14:117-130.
21 Glendinning C, Schunk M. Cure or care: reforms in long-term care for older people in the UK and Germany. Manchester: National Primary Care Research and Development Centre, 1997.
22 Light D W. Labelling waste as inefficiency. Health Service J 1990;18 Oct:1552-3.
23 Light D W. Biting hard on the research bit. Health Service J 1990;25 Oct:1604-5.
24 Dawson D. Regulating competition in the NHS. York: Centre for Health Economics, University of York, 1995.
25 Cross M. Making connections. Health Service J 1997; 5 Jun:11.
26 Mahon A, Stoddart H, Leese B, Baxter K. How do total purchasing pilots inform themselves for purchasing? King's Fund, London, 1998, forthcoming.
27 Yates J. Developing quality indicators from routine data in order to improve effective purchasing. Birmingham: Health Services Management Centre, 1997.
28 Meyer J A, Silow-Carroll S, Tillman I A, Rybowski L S. Employer coalition initiatives in health care purchasing. Vol 2. Washington, DC: Economic and Social Research Institute, 1996.
29 Roberts C J, Crosby D L, Dunn R, Evans K, Gundy P, Hopkins R, et al. Rationing is a desperate measure. Health Serv J 1995; 12 Jan:15.
30 Light D W. The real ethics of rationing. BMJ 1997;315:112-5.
31 Dixon J, Klein R. It's all in the balance. Health Service J 1997; 5 Jun:26-7.
32 Bevan G. Taking equity seriously: a dilemma for government from allocating resources to primary care groups. BMJ 1998;316:39-43.
33 Coulter A. Partnerships with patients: the pros and cons of shared clinical decision-making. J Health Serv Res Policy 1997;2:112-20.
34 Enthoven A C, Singer S J. Managed competition and California's health care economy. Health Affairs 1996;15:39-57.
35 Day P, Klein R. Steering but not rowing? London: Policy Press, 1997.

Home | Current issue | Past issues | Classified ads | Career Focus | Feedback
Collections | About this site | About the BMJ | BMA | Medline