Abstract
In explaining his economic reforms and the introduction of private ownership, Deng Xiao Ping famously said ‘It doesn't matter whether the cat is black or white as long as it catches mice’. Is there any reason to assume that this does not apply to the provision of healthcare?
Although the government's policy to bring in independent sector providers in hospital, community and primary care sectors is in line with policy in many other countries, it has led to some heated debate 1 and adverse reactions from those who see this as the first stage in breaking up the NHS. 2 This paper reviews the history of the debate and the evidence that is available about whether ownership matters.
Why are there changes?
Across Europe here has been a general trend to more autonomous providers through changing the ownership, governance and accountability arrangements of existing publicly owned providers or by encouraging new entrants to the market.
There are a number of reasons for this but generally the trend has often been the result of an increased interest in applying market mechanisms to public services and a belief, partly supported by economic and political arguments, that market sector management techniques and expertise can be useful in the public sector. Most notable of these was the concept of‘ new public management’, a conceptual framework that redefines the role of the state from provider of public services to one of‘ guarantor’, with an associated belief that governments should‘ steer’ and not‘ row’. 3
Managers and owners drawn from outside the public sector are believed to be more willing and able to confront difficult issues with clinicians, implement productivity improvement and respond to the incentives in the market based systems that have often been created at the same time as changes in ownership.
A more sophisticated approach suggests that performance will be better where the key decision-makers in an organization are also its owners. 4,5 One way to achieve this is by either privatization or creating non-profit organizations and developing social entrepreneurs - this is a particular feature of social care provision in England. Parker's study of privatization showed that in the majority of sectors in the UK this was associated with higher productivity and performance. 6 How far these arguments can be applied to healthcare is not clear.
A further reason for changing the ownership of healthcare providers is the evidence that innovation is more likely where there are opportunities for market entrants to bring in new ideas. 7 This has been used by the government as an argument for the development of Independent Sector Treatment where a significant increase in innovation and productivity is attributed to this. 8 This argument is not universally accepted.
The greater autonomy of health care organization can also be justified on political grounds. There has been a loss of faith in the ability of central government to successfully manage a large number of complex organizations in a fast moving environment. ‘New localism’ has raised an alternative model of public accountability that replaces large state-run bureaucracies under national political control with autonomous organizations accountable to local populations.
Lastly, the devolution of responsibility for health care provision to relatively autonomous organizations may suit the political interests of government. A process of ‘creative destruction’ may be needed to make space for new providers. 9 Any ownership model that distances governments from problems associated with poor management or the consequences of ever-rising health care costs may be welcomed. While autonomy for health care providers and competitive markets are often linked, one is not a necessary condition for the other. Furthermore the actual level of autonomy granted may well be heavily circumscribed: nor are the hospitals necessarily safe from capricious government intervention.
The role of govenments in provision
It is very striking that in most of the policy debate outside the UK there is no presumption that governments need to have a role in the direct provision of health care and a wide variation of ownership arrangements exists (Table 1).
There are, however, some areas where there is agreement that governments should retain a role if high quality health care services are to be maintained:
Standard setting Policing dysfunctional behaviour Competition regulation Emergency planning Protection of vulnerable specialist services Ensuring the provision of services important for public health (immunization
and vaccination, infectious disease control, sexual health etc).
In the large majority of countries, including the UK, primary care is provided by independent for-profit contractors. In many systems not-for-profit providers play a significant part in hospital care, and there is a smaller, but still significant, role for the for-profit sector. Where the hospitals are publicly owned this is often through arms' length arrangements through local government structures and it unusual for central government to take such a significant role in the direct oversight of provision as is the case in the UK. In a number of countries there has been a general shift to non-governmental models of provision including significant responsibility to local, county and regional government. However, there are important contextual differences as, in many continental European countries, local governments have greater powers than in the UK.
In most other European countries where bed numbers have fallen, it has generally been the public sector that has seen the most significant reduction.
England has more of a mixed pattern of provision than is sometimes assumed. The estimated involvement of the independent sector in secondary care is shown in Box 1; total spending 2003/4 was £59 billion (excluding capital).
The acknowledgement of the role of the independent and voluntary sector is not a new departure. Long-term care of the elderly was privatized in the 1980s 10 and there has been a long history of non-statutory providers in mental health.
Does ownership matter?
Much of the evidence from the USA reviewed by Devereux et al shows little difference between for-profit and not-for-profit hospitals in terms of costs per case although there were some difference in the total bill paid with for-profit providers billing larger amounts for the same level of care as not-for-profits. 11 A similar review of research by Rosenau & Linder found 56 studies since 1980 that assessed cost differences. 2,12 The not-for-profit providers performed better on costs and efficiency in all but 23% of cases. It should be remembered that a US not for profit organization operates in a highly competitive environment and has to use the full range of market orientated management approaches employed by their for profit competitors. Duckett and Jackson produce a similar finding for hospitals in Australia despite the fact that non-profit hospitals tended to pay higher wages. 3,13
Duckett's review of the evidence suggests that there may be higher costs for purchasers where markets are dominated by for profit providers due to a higher propensity among such providers to cost shift, drive up demand for services (‘supplier induced demand’) among other reasons. 1 However, as with much of the evidence in this area the results are equivocal and may be more related to the characteristics of the local health market than the model of ownership.
There is little empirical evidence to support the contention that for-profit providers will increase levels of innovation and technological adoption. In fact, one study suggested an advantage for non-for-profit hospitals as their front line clinicians had more influence on top management. 14
Some impact of the type of ownership on quality of care is suggested by the research evidence. A meta-study of research covering 38 million US patient episodes suggests that non-profit ownership is associated with longer patient survival. 15 Rosenau& Linder's review found 69 studies (largely from the USA) that looked at least one or more quality measures, 41 (59%) found that non-profit providers were superior and 20 (29%) found no difference. They suggest that over time the growing power of market mechanisms will tend to reduce the differences between alternative types of provider.
Ownership and social mission
Providers reimbursed on a per-case basis are only likely to deliver services that are not directly reimbursed if these produce other benefits or are in line with their ‘social mission’. In the USA for-profit providers (share-holder or privately owned) are less likely to pursue social goals. Not-for-profit providers often have better links to their community and are better able to mobilize voluntary effort and philanthropy. 12
Ownership and whole system transformation
The experience of whole system transformation, with the whole or partial replacement of publicly managed health care systems with autonomous or quasi-autonomous providers acting within market environments, has been mixed.
The experience of outsourcing the building and operation of hospitals in Australia 1,16 seems to have been unhappy - La Trobe hospital was bought back by the government of the State of Victoria. The Alzira hospital in Spain required a major injection of cash and a number of Australian hospitals had to have payments increased to maintain quality. Experience of transferring responsibility for management to private sector operators in Portugal, Germany and Sweden appears to have been more positive.
There is limited evidence that the creation of autonomous hospitals in central and eastern Europe has had a major impact on performance and some anxiety that their freedom could be used to pursue objectives that are not in the interests of their owners (usually local municipalities). 17 Similarly, the introduction in the 1990s of provider markets and the transformation of publicly owned hospitals into for-profit ‘crown companies’ in New Zealand was abandoned after only a few years when expected efficiency gains did not materialize. 18
More positively, there is some tentative evidence that the greater freedoms afforded from governmental control as part of the NHS quasi market of the 1990s served to increase some measures of health care efficiency, 19 although there is some suggestion that the impact of the competition that created this was negative. 20 The benefits of greater autonomy for healthcare providers may extend beyond those of efficiency. The Healthcare Commission's review of NHS foundation trusts provides some evidence of service improvement but is limited due to it being undertaken early in their gestation. 21 Research by the King's Fund suggests that foundation trusts do represent a positive change in the level of community engagement. 22 This suggests that the degree of freedom enjoyed by providers, rather than simply their formal ownership status, may be important in shaping performance.
Conclusion
The NHS has never been exclusively state owned and operated and views about changing the basis of ownership have to be based on practical concerns and evidence rather than simple ideology.
All the models have advantages and problems and there seems to be little theoretical or empirical basis for claiming superiority for one model of ownership over others; the question may be one about the values and objectives society wants providers to hold and how innovative they are. 23 Even here, no one model has a monopoly.
It is noticeable that where there has been a move away from state owned and operated services in Europe not-for-profit models have been preferred. This preference may be cultural but it may also be that organizations already trying to align the interests of patients, the public, clinicians and payers find that shareholders or owners (including government) introduce an even greater layer of complexity. It may be this that explains the tentative evidence in favour of not-for profit models.
Changing the ownership of providers is only one part of a mix of complex policies and the impact of ownership can be mediated by other environmental factors - the behaviour of commissioners, regulators and government and the financial and other incentives that are applied.
If it is the case that the incumbent managers are risk averse, fail to innovate and are unresponsive to patients, changing the management or ownership of the provider may help. But without addressing the wider environment this is unlikely to solve some of the underlying problems that the NHS faces.
In summary, autonomous not for profit providers in environments that force them to be innovative seem to have an edge. For profit providers, come a close and arguable second.
Footnotes
DECLARATIONS