Friday, October 07, 2011

Lucy Reynolds and I offer our views on the provisions for competition in provision in the Health and Social Care Bill

“It is not from the benevolence of the butcher, the brewer or the baker that we expect our dinner but their regard to their own self-interest.”
Adam Smith, 1776[1]

“That any sane nation, having observed that you could provide for the supply of bread by giving bakers a pecuniary interest in baking for you, should go on to give a surgeon a pecuniary interest in cutting off your leg is enough to make one despair of political humanity.”
George Bernard Shaw 1913[2]

Without basis in any solid evidence, the NHS is about to undergo a market reform which may well be, for all practical purposes, irreversible. It is likely to cost the country much more than the existing system, because of extra bureaucracy, loss of economies of scale and full bargaining power with suppliers, maintenance of redundant capacity, and the need to cover return to capital and the costs of commercial borrowing. It will deliver less, not just because of the diversion of effort into the copious administration of commissioning and contracting, but because the embedded incentives of market-based healthcare will lead to physician-induced demand (unneeded tests and treatment undertaken to increase provider income, which may harm patients subjected to them). More effective regulation will be needed to control the problems to which this system is prone, again elevating costs. Overall, the reforms will reduce the quantity of front-line care which can be provided for each pound of the NHS budget spent.  This paper comments specifically on the theoretical weakness of the Cooper and Propper econometric studies which are being used to justify the reform.
In conclusion, the proposed market-based reform to the NHS will be an expensive step in the wrong direction.  It was rushed through the Commons and its content needs to be carefully scrutinised: in addition the policy consequences of adopting a market-based reform should be examined.


This market-based reform (see Figures 1 and 2 above for the financial changes involved) has been presented to the electorate as essential in the interests of cost-saving for the taxpayer. It has been justified on the basis of saving money and improving outcomes, neither of which hold up to more than cursory inspection.

The only “evidence” that competition can benefit healthcare performance is at best debatable
After a quarter of a century of assertions that competition benefits healthcare[3], lobbyists finally, in 2010-11, found three UK-based studies which purport to test and prove this theory.  They are known as the Cooper et al paper[4] and the Propper et al[5],[6] studies.
As these authors could not measure competition per se they used the density of service provision as a proxy, comparing it with routine health service data on heart attack mortality. These three papers stand against a background of many other studies which do not find a positive association (as noted in a review of the literature in one of them5) and in contrast to the theoretical case for why market-based competition in healthcare would be harmful, as set out by the Nobel Laureate Kenneth Arrow and explained in section 4 below. While the three papers contain elaborate mathematical models corrected for many relevant confounding factors including distance to hospital, they suffer from a basic scientific error, the confusion of association and causation: they document the former and impute the latter. A small but perceptible difference in the outcomes before and after the introduction of small-scale competition mechanisms in the NHS is present, but the authors do not offer any convincing mechanism for how reduction of heart attack mortality might result from higher hospital density (which, as noted, they treat as a proxy for greater market competition between providers). Instead they postulate a general “halo” effect of greater operational efficiency arising from bidding for some types of work in the same hospitals that treat heart attacks, via a more “competitive” culture. Alternative explanations for the outcomes generated from the models they have devised exist, such as under-correction for the time taken for ambulances to drive heart attack patients to the nearest hospital. They also ignore the possibility of “upcoding”, although this exaggeration of the gravity of illness in incoming patients is a well-known consequence of paying hospitals according to the quantity and type of cases they treat[7], a change which is part of the competitive internal market introduced in the NHS. If hospitals are judged on mortality and outcome data there is an incentive to record medical problems as more severe than they are, a practice that is hard to detect even by tracking back to case notes. For this overlooked explanation both motive and mechanism are evident, and so is much precedent7. Even if there is a true effect, these studies provide no evidence that it was competition that brought it about: it could as easily have been due to other changes taking place at the same time. None of these papers actually provide any evidence that the market competition reform proposed will in any way improve the NHS, and their citation as the best proof available tells its own story.
 Contrary to the government’s line, the new arrangements will be much more expensive than the Beveridge system that they replace[8]. The change will benefit only corporate market entrants, many of them foreign. The main reasons for the extra costs are:

1.  Fragmentation of purchasing among more entities
The single purchaser system that the NHS has moved away from in recent years can provide excellent value for money. This is demonstrated by the success of the PHARMAC single purchasing agency in New Zealand[9],[10], and the contrasting failure of cost control in many parts of the American health care market[11]. The arrangement we are moving to, with purchasing decentralised into hundreds of consortia, involves multiplication of procurement arrangements, reducing the bargaining power of each purchasing entity, and requiring creation of extra bureaucracy to enable this more costly arrangement.
2.  Competition between providers involves redundant capacity and loss of economies of scale
Government insistence on trying to run the system through “consumer choice” (a nonsense in this context, as explained in section 4 below) involves the need for redundant capacity; providers will need to charge more for the services they do provide in order to cover the costs of having capacity lying idle in order merely to allow creation of a competitive market.  In addition, market competition generally involves spending on marketing.  Obviously, unification of provision as in the Beveridge-system NHS allows services to be supplied more cheaply because of economies of scale, avoidance of redundant capacity and no need to spend money on marketing.

3.  Shift to individual billing to facilitate competition between providers multiplies bureaucracy
To facilitate competition between providers, the future arrangement in the NHS will be that “the money follows the patient”, in the words of the White Paper. This results in a shift from efficient pooling of cost and risk across the nation toward the use of individual healthcare budgets. The change will require both providers and commissioning groups to maintain administrative capacity to set up and monitor individual billing.  Again this requires a multiplication of unnecessary bureaucracy which provides no benefit to patients, while increasing costs to the taxpayer.

4.  The structural problem of supplier-induced demand results from competition in healthcare
Henceforth all hospitals will be required to generate their income from selling services to private patients or to Clinical Commissioning Groups (CCGs: composed of GPs and others). This is to occur through what is known in general as fee-for-service and in the NHS as “payment by results”, although in fact it is activity not results which is rewarded. Except for a few facilities which offer essential services which are otherwise unavailable, they will need to cover their costs from such sales or face insolvency and closure. The government has explicitly rejected calls from the Chief Executive of the NHS to maintain the mechanism for a failing hospital to be taken back into public ownership[12]. Hospitals will be permitted to retain any profits they make, for reinvestment or payment to their shareholders or partners.  This arrangement ensures that hospital staff are keen to provide services, and is represented as encouraging them to maximise efforts to provide patient-friendly service of high medical quality so as to generate repeat custom and patient recommendations.  However, it also incentivises clinicians to recommend and provide services that are unnecessary or of marginal medical benefit, and hospital managers to overprice services where this can be arranged. In this era of light-touch regulation[13], both of these undesirable options would be much easier and cheaper to arrange than genuine improvements in service provision, and would more reliably boost the bottom line. One corporation that is now offering commissioning services to our CCGs[14] has previously found it necessary to compensate the US government following allegations of its fraud against the Medicare scheme[15].
The setting of a fixed tariff for each procedure is intended to block over-pricing, but experience in countries which use this system shows that the profit-making provider’s response to this form of control is to miscategorise patients’ conditions, so that they are registered as suffering a more severe medical condition than is actually the case, “upcoding” or“DRG-creep”[16].  A patient miscategorised in this way could be treated more aggressively than is appropriate to their condition, so might for instance undergo unneeded surgery or suffer side-effects from treatment with stronger medication than appropriate. The patient may thus be harmed in order to generate money for the provider. 
The planned new system will be vulnerable to this sort of malpractice because the rationalisation for this market based reform (welfare economics and its offshoot public choice theory[17]) relies on a consumer versus supplier model which does not fit healthcare.  In this model, providers wishing to sell more services at higher prices to generate more profits oppose customers trying to acquire the services they need in adequate quantities at lower prices, and the two come to a mutually agreeable arrangement. This model relies on a number of assumptions including one which is rarely met in the case of medical services, as explained in 1963 by the economics Nobel Laureate Kenneth Arrow[18], the possession by both buyers and sellers of complete information about the nature and quality of the services to be supplied.  Such symmetry of information allows a balance of power between buyer and seller to facilitate a fair negotiation process between these two principals in the transaction.  However, this is evidently not the situation where medical services are concerned. Few patients are medically qualified, and they visit doctors in order to ask what is wrong with them and how it may be treated. That is, the patient relies on the clinician to provide advice as their agent.  When the clinician is also allowed or required to act as a seller of services, the two roles of agent for the patient and principal in the transaction can come into direct conflict, resulting in an increase in demand (through an upward shift of the demand curve in economist’s jargon) induced by the physician’s inclination to increase income by maximising services provided[19].
Where the survival of the service and the owners’ requirements for financial returns put pressure on clinicians to increase the money they bring in by selling more services to patients, the stage is set for systematic exploitation of patients’ ignorance through unneeded billable tests and investigations, diagnosis of minor problems as severe, unnecessary treatment, and over-aggressive treatment[20].  The ability of a patient-consumer to assess the quality of medical services received is for many types of treatment limited to such peripheral issues as waiting time, comfort of waiting rooms and wards, and friendliness of staff, while the clinician-seller is able to exploit their inability to detect profit-led over-treatment, possibly at the cost of subjecting the patient to the side-effects and risks of unnecessary procedures.  In the future NHS, the proposed use of “fee for service” payment arrangements for clinical services adds a dimension of moral hazard, incentivising this form of abuse. Where exploitation does not ensue, it is because the professional ethics of the medical profession constrain it.  But unnecessarily creating such a temptation is surely unwise.  
The more market-driven the health care system the more it is prone to this problem:
Comparing two societies of similar economic and cultural status, in 1993 doctors in the US performed hysterectomies two and a half times as often as doctors in Sweden, and caesarean sections twice as often. And doctors in the US performed 4.4 times more coronary bypasses than doctors in Canada.
Another revealing comparison can be made between certain professions and everyone else. The American Medical Association conducted a study of medical intervention in cases of stage II prostate hypertrophy. They asked urologists what they would do if it were their own case. Just 40.5% of urologists asked said they would opt to have a transurethral prostate resection, yet for the population as a whole the rate of transurethral resections actually carried out in cases of stage II prostate hypertrophy was 80%.[21]
The Economist calculated that the market-driven, corporate-dominated US health care system in 2009 generated between $250 and $325 billion of charges for unnecessary care; this comprised 10-12% of US healthcare spending that year[22].

5.  The proposed means of controlling the extra cost of supplier-induced demand involves a second layer of profit-taking and incentivises denial of needed care
It is well-known by health economists that fee-for-service medical provision causes inflation of medical spending[23].  This is the reason why the US medical system provides such poor value, consuming 18% of GDP (compared to our 8%) and yet leaving a high proportion of the population without access to decent medical care (according to the Commonwealth Fund report comparing seven developed world health systems[24]). Because medical providers in the USA are too politically powerful for the government to be able to introduce the universal coverage the population[25] and the President[26] would prefer, other means have had to be found to try to contain soaring medical costs. A mechanism, “managed care”, was devised to oppose the incentive for providers to over-treat with another which uses the profit incentive to motivate limitation of the amount of care provided.  This is the arrangement which the Health and Social Care Act would move us to, with commissioning at GP level used to restrain overcharging by hospitals through gate-keeping of referrals.
In this model, managed care is undertaken by health maintenance organisations (HMOs) which invite members of the public to pay an annual insurance premium in order to access treatment for any new illness that befalls them (in our new system this premium will initially be fully replaced by an allowance from the NHS budget, which will for now also cover treatment for pre-existing conditions).  The HMOs employ doctors who diagnose and recommend referral to hospitals as appropriate (as our GPs do).  The HMOs keep any difference between the total of the premiums they receive and the care they pay for (as our CCGs will).  This creates a financial incentive for the HMO (or CCG) to reduce total referral cost.  Proponents emphasise that this deters referrals undertaken more to keep the patient happy than for any medical benefit, and that it creates a financial incentive for referring doctors to control the amounts charged by referral services. Both are good goals if provision is under fee-for-service arrangements, which encourages excessive charges, as explained above.  Medically qualified HMO employees are more knowledgeable customers than are patients, and can better control supplier-induced demand, also clearly beneficial.  Unfortunately for patient welfare, the crude use of the profit motive also incentivises refusal to refer patients for services justified on the basis of medical need. 
Criteria for assessing efficiency in medical service provision should involve the correct and timely treatment of medical conditions which are amenable to treatment. Instead, managed care applied to profit-making provision simultaneously gives rise to both under-treatment (exclusion from medically needed care because of denial of referrals[27]) and over-treatment (medical procedures undertaken for the benefit of providers not patients[28]).
 These mismatches occur because the choices triggered by the profit motive map poorly on to the priorities set by the meeting of medical need. It is known that financial incentives raise doctors' activity levels[29] but this does not translate to better health outcomes unless doctors paid this way are underprovided in a community, in which case more activity might improve outcomes. If they are oversupplied and competing for business, then providers’ need to make a living incentivises overtreatment28,[30]. Since no treatment is devoid of adverse effects, poorer outcomes ensue28.  If the market competition that the government insists is central to the reform[31],[32] is introduced throughout the system, we will be in the second situation, where competition between providers for enough patients to cover their fixed costs (rent, salaries, etc) will create motive and opportunity for exploitation of patients.  As noted by a concerned QC, it is also likely to create stresses on staffing caused by cost-cutting which could result in many more medical errors[33].
 In contrast, the Beveridge system does not encourage overtreatment, overcharging or denial of needed care, because doctors are under no financial pressure to over-treat patients. It scores highly on value for money8,[34], quality of health outcomes[35],[36], and patient satisfaction[37], and produces a system where clinician efforts are expended on addressing medical need not on generating income. The proposed reform involving two extra layers of profit and a great deal of unnecessary administration being extracted from the NHS budget for hospital treatment appears to be in the interests of neither patients nor taxpayers; it may however produce a bonanza for foreign corporations[38] and plenty of new work for accountants and lawyers[39],[40].

6.  Privatisation will increase the costs of regulating adequately; failure to do so will result in abuse of patients
 It should be noted that once the reform has loaded our health system with incentives to over-treat patients, deny referrals, falsify or withhold information on patient outcomes and to maximise corporate profits, the cost of regulating adequately will rise.  Failure to do so will save money on “red tape” but at the cost of harming vulnerable people and squandering taxpayers’ money.  We can see how this works in practice from the poor record of the Care Quality Commission, for example at Winterbourne View[41] , mainly due to an underfunded regulatory system which operated largely on the basis of self-certification and not inspection.  It should be noted that recruitment advertising for the CQC over the past year has not required any medical qualification for any post[42].

 While it is clear that ways exist to improve the Beveridge system NHS, it is equally plain that replacing it with a market-driven health system is a move in the wrong direction which is likely to harm patients and waste scarce resources. 
 Market competition in healthcare does not produce desirable results, a conclusion evident from inspection of medical systems which rely upon it and also from economic theory. Despite the highest spending in the world both as a proportion of GDP and as a per capita figure, health outcomes achieved in the USA are mediocre: for instance, in 2010, the US was 44th/193 in the world ranking for the probability of dying by age five34. In the past decade,  the American health care system has made minimal progress in reducing deaths from causes amenable to medical care (the UK Department of Health’s chosen high level indicator of health system performance), falling increasingly far behind other industrialised nations; in the UK, progress on this measure has been among the fastest[43].
 This “reform” serves only corporate interests. They will gain new profit opportunities, as recently explained by Earl Howe to an audience of private sector investors[44], and by past NHS Director Mark Britnell to a group of US private equity companies[45] as follows:
 “In future, the NHS will be a state insurance provider not a state deliverer. In future ‘any willing provider’ from the private sector will be able to sell goods and services to the system. The NHS will be shown no mercy and the best time to take advantage of this will be in the next couple of years.

GPs will have to aggregate purchasing power and there will be a big opportunity for those companies that can facilitate this process.

The monolithic arm of state control will be relaxed which will provide a huge opportunity for efficient private sector suppliers.”

We believe the Bill as presently worded is largely unworkable, while attempts to implement it are likely to increase costs and reduce the level and quality of care that is provided. It is laden with incentives for opportunistic behaviour. Furthermore, as noted by the House of Lords Constitution Committee,[46] it is potentially unconstitutional. The Secretary of State has condemned his critics for relying on anecdote rather than evidence. In this paper we have presented the evidence that, we argue, he must respond to if he is to make a convincing case that his Bill should be passed.


[1]    Smith A. An Inquiry into the Nature and Causes of the Wealth of Nations. 1776
[2]    Shaw GB. Preface on doctors. The doctor's dilemma: a tragedy. Baltimore: Penguin, 1913: 9

[3]   Reynolds L. Competition and the 1987-2011 NHS reform. BMJ 2011

[4]   Cooper Z, Gibbons S, Jones S, McGuire A. Does hospital competition save lives? Evidence from the English patient choice reforms. 2010.
[5]   Gaynor M, Propper C, Serro RM. Death by market power: reform, competition and patient outcomes in the NHS. 2010.
[6]   Bloom N, Propper C, Seiler S, van Reenen J. The impact of competition on management quality: evidence from public hospitals. Centre for Economic Performance discussion paper No 983, 2010.
[7]   Pitches D, Burls A, Fry-Smith A. Snakes, ladders, and spin: How to make a silk purse from a sow's ear—a comprehensive review of strategies to optimise data for corrupt managers and incompetent clinicians. BMJ 327 : 1436 doi: 10.1136/bmj.327.7429.1436 (Published 18 December 2003)
[8]    Pritchard C, Wallace MS. Comparing the USA, UK and 17 Western countries' efficiency and effectiveness in reducing mortality. J R Soc Med Sh Rep 2011;2:60; doi: 10.1258/shorts.2011.011076
[9]   New Zealand Government
[10] Cumming J, Mays N, Daubé J. How New Zealand has contained expenditure on drugs. BMJ 2010;340:c2441.
[11] 2011 Milliman Medical Index: healthcare costs for American families double in less than nine years. May 2011.
[12] Watt N. NHS chief challenges Andrew Lansley's foundation hospitals plan. URL:
[13] Editor’s Blog. Health Policy Insight 14 September 2011.
[15] ,
[16] Simborg DW. DRG creep: a new hospital-acquired disease. N Engl J Med. 1981 Jun 25;304(26):1602-4.
[17] LeGrand J. Motivation, Agency, and Public Policy: Of Knights and Knaves, Pawns and Queens. OUP 2003
[18] Arrow, Kenneth J. (1963). "Uncertainty and the Welfare Economics of Medical Care". American Economic Review, 53 (5)., pp. 941–973
[19] Shafrin J. Cutting Medicare reimbursement increases the quantity of medical care. Healthcare Economist 5 May 2008.
[20] Wennberg JE, Barnes BA, Zubkoff M. Professional uncertainty and the problem of supplier-induced demand. Soc Sci Med 1982;16(7):811-24. PMID 7100999.
[21] Deppe H-U. The nature of health care: commodification versus solidarity. In Panitch L. and Leys C, Morbid symptoms: Health under capitalism. (Eds.). 2009 New York: Monthly Review Press.
[22] The Economist. Waste Measurements: US healthcare spending. 17 June 2011. care-spending, plus 2009 OECD spending figures and 2010 US Census data
[23]   Tussing AD, Wojtowycz MA. Physician-induced demand by Irish GPs. Soc Sci Med. 1986;23(9):851-60.
[24] Davis, K.; Schoen, C. and Stremikis, K. (2010). Mirror mirror on the wall: How the performance of the U.S. health care system compares internationally 2010 update. The Commonwealth Fund.
[25] Steinmo S, Watts J. It’s the institutions, Stupid! Why comprehensive national health insurance always fails in America. Journal of Health Politics, Policy and Law Summer 1995;20(2):329-372
[26]         Sirota D. Obama for Single-Payer Before He Was Against It: How Political Elites Have Made Single Payer "Politically Impossible" Alternet 11May 2009.
[27] Potter W. Health Insurance Corporate Profits Spiked Despite Dire Predictions of Health Care Reform Wreckage. 25 April 2011.
[28] Gawande A. The cost conundrum. Annals of Medicine. The New Yorker.
[29] Gaynor M, Pauly MV. 1990, "Compensation and Productivity in Partnerships: Evidence From Medical Group Practice," Journal of Political Economy, 98, 544-573.
[30]   Bennett D, Lauderdale TS, Hung CL. Competing Doctors, Antibiotic Use, and Antibiotic Resistance in Taiwan. September 16, 2008.
[31] Lansley A. Extract from: 'The Future of Health and Public Service Regulation' Speech 09 July 2005.
[32] Johnstone R. Lansley stands by his NHS competition plans. CIPFA Public Finance. 18 May 2011.
[33] Whitting J. The hidden cost of NHS reform. New Statesman 23 June 2011
[34] World Health Organization Global Health Observatory. Accessed 26 September 2011
[35] Appleby J. Does poor health justify NHS reform? BMJ 27 January 2011; 342:d566 doi: 10.1136/bmj.d566
[36] Desai M, Nolte E, Karanikolos M, Khoshaba B, McKee M. Measuring health system performance: amenable mortality and the achievements in the UK. J Roy Soc Med (in press)
[37] Helm T. Andrew Lansley accused of burying poll showing record satisfaction with NHS: Ministers are said to be withholding survey results that undermine health secretary's case for urgent radical reforms Guardian 19 March
[38] Player S, Leys C. The people behind ‘Liberating the NHS’ reform. Part 3. Hospital Dr 22nd November 2010
[39] Munro R. Emergency Service. Accountancy Magazine. June 2011.
[40] Reynolds L, McKee M. GP commissioning and the NHS reforms: what lies behind the hard sell? Journal of the Royal Society of Medicine, in press
[41] The Press Association. CQC publishes critical report of Castlebeck abuse failings. Nursing Times19 July, 2011.
[42]         Bousfield A. The Care Quality Commission and Turkeys. Medical Harm. September 15, 2011.
[43] Nolte E, McKee M. Variations in amenable mortality trends in 16 highincome nations. Health Policy 2011; doi:10.1016/j.healthpol.2011.08.002
[44] Pulse Online. ‘Big opportunities’ for private sector in health bill, says minister PULSE 7 September 2011
[45] Apax Partners conference, Opportunities Post Global Healthcare Reforms, October 2010
[46] Ramesh R. NHS shakeup poses threat to political control, Lords committee claims URL: