Structural interests

The context for Alford's study was the health system in New York City. Rising costs, poor life expectancy, high rates of infant mortality, and poor access to services for the significant proportion of the population not covered by health insurance generated both a continual sense of crisis and a string of ‘commissions of investigation’ in the two decades following the second world war. Analysing the reports of these commissions, Alford concluded that: ‘the principal barriers to health care reform lie in the powers of strategically structured interests’ (Alford, Reference Alford1975: xiii). He defined these ‘structural interests’ in terms of the extent to which their ‘interests [are] served or not served by the way in which they “fit” into the basic logic and principles by which the institutions of society operate’ (ibid: 14). Alford identified three categories of ‘structural interests’. ‘Dominant interests’ are those ‘served by the structure of social, economic and political institutions as they exist at any given time. Precisely because of this, they do not have to organise and act to defend their interests’ (ibid: 14). Nevertheless, Alford argued that internal and external social pressures and resulting shifts in structures will give rise to ‘challenging interests’, seeking to establish themselves as the new legitimacy. Finally, he argued that ‘repressed interests’ are ‘the opposite of dominant ones . . . the nature of institutions guarantees that they will not be served unless extraordinary political energies are mobilised’ (ibid: 14, italics in original). Alford does not define ‘interests’, other than stating that they should not be seen as synonymous with pressure groups or interest groups. This suggests that he implicitly espoused an ‘ideal-regarding’ concept (Fishkin, Reference Fishkin1979) whereby the analyst (rather than the actor) identifies interests, while it is interest groups that act in pursuit of their perceived interests (Alford, Reference Alford1975: 13–14). This distinction explains how internal conflicts can arise within structural interests (p. 14): Alford states that ‘my assumption in making these key distinctions is that the similarities of structural location vis-à-vis each other of these major structural interests warrant an emphasis on the common aspects of each rather than on their internal differentiation’ (Alford, Reference Alford1975: 192, italics added).

In the specific context of New York in the decades prior to 1970, Alford identifies the ‘dominant interest’ as the ‘professional monopolisers’: that is, doctors, medical schools and public health agencies who have an interest in maintaining ‘control of the supply of physicians, the distribution and cost of services and the rules governing hospitals’ (Alford, Reference Alford1975: 14). As an example, Alford argues that the medical staff who dominated the running of New York hospitals at that time were so committed to maintaining lucrative private practices that they ran the hospital as their ‘workshop’, resisting the expansion of emergency room care or other services for uninsured customers (ibid: 196). Furthermore, he argues that the structural privilege enjoyed by dominant interests was such that: ‘even health reforms accommodate themselves to the existing interests and privileges of physicians. Medicare [the US federally funded service for older people], for example, explicitly allows hospitals to provide doctors with free equipment and space’ (ibid: 196). For Alford, the structural nature of dominant interests means that their dominance becomes ‘taken for granted as legitimate, as the only possible way in which these health services can be provided’ (ibid: 17, italics added).

Alford identifies pressures within the New York health system that tended to generate or empower a challenge to the status quo as: the increasing complexity of medical care, increasing specialisation among physicians, and the rising costs associated with the introduction of increasingly complex technologies. For managers of the system, these pressures implied a need for greater coordination between services and more efficient bureaucratic management. Alford labels this challenging interest as ‘corporate rationalisers’, and argues that this group has an interest in ‘breaking the professional monopoly of physicians over the production and distribution of health care’ (Alford, Reference Alford1975: 15). Finally, for Alford the ‘repressed interests’ at that time and place were the poor, the uninsured and those whose incomes pushed them just above the threshold for Medicaid (US state programmes for the very poorest people) assistance: ‘These are “repressed” or negative structural interests because no social institutions or political mechanisms in the society insure [sic] that these interests are served’ (ibid: 15).

Alford's analysis implies a resistance to change, rooted in the ‘dominance of the private sector and of the middle class’, going on to argue that ‘government policy is not fundamentally important, except insofar as the policy is that of not interfering with the private sector’ (Alford, Reference Alford1975: 254). He also argues that one explanation for the durability of the structural relationships that he describes lies in the fact that both dominant and challenging interest groups had a common interest in reaching a compromise, which acted to entrench the status quo (ibid: 262). For the professional monopolisers, corporatist reforms at the margins that brought more low-income groups into the pluralistic market for health care could only be a good thing, while for the corporate rationalisers any reform that allowed them to plan some portion of the service, if only at the margins, was a step in the right direction, enhancing the status of the managers involved, and providing kudos for the politicians responsible. Thus, the New York scene in the 1960s was, according to Alford, dominated by the development of small-scale schemes to provide care for disadvantaged groups, while the overall problem of a large uninsured population was not addressed.

Alford's analysis is essentially situated, and the categories identified as ‘dominant’ and ‘challenging’ (professional monopolists and corporate rationalisers respectively) relate to New York City in the late 1960s, although, as noted above, he was pessimistic about prospects for substantial change. Perhaps partly because of this pessimism, the situated nature of his theory has often been missed or elided by those analysts who have used it as a framework to explore the nature of health systems at other times and in other places. Some make no reference to the time or place of Alford's study (for example, Ham, Reference Ham2004; Lewis and Considine, Reference Lewis and Considine1999), while others seem to take his original categories as read. Duckett's analysis of the Australian health system states that Alford ‘identifies the dominant interest in the health sector as being “professional monopolist”’ (Duckett, Reference Duckett1984: 959, our italics), suggesting that Alford's analysis applied to ‘the health sector’ as a whole. Similarly, Cho examines interactions between practitioners of traditional eastern forms of medicine and those advocating western-style pharmacology and states that: ‘[Dominant structural interests] are manifest in the form of “professional monopoly”’ (Cho, Reference Cho2000: 124), suggesting that such interests are permanently dominant. Some authors have stressed that the identity of structural interests is an empirical matter (Barnard and Harrison, Reference Barnard and Harrison1986; Harrison and Ahmad, Reference Harrison and Ahmad2000: 138). However, it is analysts who question the relevance of Alford's categories to UK general practice (North, Reference North1995; North and Peckham, Reference North and Peckham2001) that have provided the most critical uses of his work. In addition to Alford's pessimism about change, the durability of his categories as manifest in the health policy literature is perhaps due to its ‘uncluttered’ character (North, Reference North1995) and the strong face validity of ‘professional monopoly’ and ‘corporate rationalisation’, within an academic community for whom the analysis of professionalism and the conflict between professionals and managers within health systems have long been prominent topics for discussion and investigation.

Although several analysts have applied Alford's approach to the NHS, reaching different conclusions about whether professional monopolists or corporate rationalisers were dominant at particular times (for example, Harrison and Ahmad, Reference Harrison and Ahmad2000; Ham, Reference Ham2004; Baeza, Reference Baeza2005), none has questioned his categories. In this article we abandon this implicit assumption that professional monopolists and corporate rationalisers are immutable analytic categories, thereby liberating us to ask more general questions about structural interests. Drawing on an empirical study of so-called ‘practice-based commissioning’, we return to Alford's original question: what interests are systematically served by current structures? We suggest that new structures of resource allocation within the NHS systematically serve a structural interest that we term ‘corporate monopolists’, and that the contemporary challenge to this interest is from a group of doctors that we term ‘professional rationalisers’.

Contemporary structural context in the English NHS

Since the late 1970s, money was allocated to the NHS according to a formula which took into account factors such as population size, age-sex profile, service usage and mortality. However, this relatively crude allocation could not easily compensate hospitals for the fact that general medical practitioners (GPs) were free to refer patients to any hospital; it could take two years for hospitals to be paid for the treatment of patients from outside their area (McDonald and Harrison, Reference McDonald and Harrison2003). The system also included built-in incentives for hospitals to limit activity, since any increase in patients would have to be paid for out of the existing allocation. Between 1991 and 1997 the Conservative government attempted to address these anomalies by introducing the ‘internal market’ in the NHS (Flynn and Williams, Reference Flynn and Williams1997), whereby local Health Authorities (and, to a limited extent, GP fundholders; Glennerster et al., Reference Glennerster, Matsaganis, Owens, Hancock, Robinson and Le Grand1994) were expected to ‘purchase’ care for their local population from hospitals on an annual contract basis. The Labour Party opposed this policy while in opposition (Labour Party, 1994) and, on coming to power in 1997, abolished GP fundholding and modified the internal market, replacing annual contracting with longer-term ‘service agreements’. ‘Contracting’ was renamed ‘commissioning’, and Health Authorities were gradually replaced by Primary Care Trusts (PCTs), who took over responsibility for the commissioning process (Department of Health, 1997). The NHS Plan (Department of Health, 2000) emphasised the role of patient choice in improving services, and this, along with a desire to provide incentives for Trusts to improve efficiency, provided the impetus for the development of a payment system that was more responsive to the movement of patients about the system. Payment by Results (PbR) was introduced in 2002 (Department of Health, 2002) and implemented gradually; elective surgery was the first area included, then out-patient attendances, and A&E by 2008/2009. Community and mental health services are not yet covered.

Under PbR, each episode of care is allocated to a casemix category called a ‘Healthcare Resource Group’ (HRG), each of which attracts a standard payment under a national tariff. Hospitals that provide care more cheaply than this can retain the surplus, while hospitals with higher costs must cover the loss (Department of Health, 2002; Dixon, Reference Dixon2004). It is of note that hospitals themselves categorise cases, allocating episodes of care to HRGs; thus the system of PbR puts the hospital in the position of control. Cases may be ambiguous, allowing hospitals to choose to allocate the episode to the HRG attracting the most income, and studies have demonstrated the coding process to be approximate, at best (Bajaj et al., Reference Bajaj, Crabtree and Tucker2007). Furthermore, whereas prior to PbR hospitals and PCTs agreed a cap on the number of cases that would be reimbursed, under PbR everything is paid for on a per-case basis, with PCTs expected to pay for every intervention carried out on one of their patients, with resulting pressure on their budgets (Bevan and Robinson, Reference Bevan and Robinson2005). Klein puts it thus:

Contracts (‘service level agreements’) specify quality requirements, such as penalties for failing to operate on patients with fractured hips within a specified time, or for the number of patients contracting a hospital-acquired infection, but PCTs may not refuse to pay for patients that have been treated within the terms of the agreement, even if their budget is exhausted. It is clear, therefore, that this new system has potential implications for PCTs, as hospitals undertaking more work than they had done historically would be paid more; consequently the official documents setting out these changes also emphasised the importance of commissioning by PCTs to maintain a balance in the new environment (Department of Health, 2002: para. 3.2). In this context, Practice Based Commissioning (PBC) was introduced in 2005, in part to generate front-line medical engagement with the commissioning process:

While not prominently expressed in these terms in official documents, there is little doubt that policy-makers hoped that this involvement of GPs would improve the effectiveness of commissioning (Checkland et al., Reference Checkland, Coleman, Harrison and Hiroeh2008) and go some way towards countering the incentives for hospitals to treat more patients that commentators suggest are inherent under PbR (Marini et al., Reference Marini, Miraldo, Jacobs and Goddard2008). Under PBC, general practices are allocated an indicative commissioning budget, but it has become clear (Coleman et al., Reference Coleman, Harrison, Checkland and Hiroeh2007) that most PBC is undertaken by groups of practices (‘consortia’) pooling their budgets and working together. Guidance suggests that this budget should cover all PbR activity, prescribing costs, community services and mental health (Department of Health, 2006a), although it is perhaps significant that PCTs seem more likely to devolve budgets for activity that is already coded as discrete events (HRGs) than those which are not yet covered by PbR (Checkland et al., Reference Checkland, Coleman, Harrison and Hiroeh2008). Any savings made under PBC can be reinvested in additional services, providing an incentive to keep patients away from relatively expensive hospitals and to seek cheaper alternatives.

A further ingredient in this policy mix is the Foundation Trust (FT), semi-independent from the NHS and introduced in 2003 in the teeth of considerable opposition (Toynbee, Reference Toynbee2003; Wilmoth, Reference Wilmoth2004). Government rhetoric asserts that FTs will ‘incentivise innovation and entrepreneurialism’ (http://www.dh.gov.uk/en/Healthcare/Secondarycare/NHSfoundationtrust/DH_4062806, accessed March 2008). FTs are regulated outside the mainstream NHS hierarchy by an organisation known as ‘Monitor’, which also emphasises the need for FTs to behave in a business-like manner (http://www.monitor-nhsft.gov.uk/publications.php?id=998). FTs are able to employ staff according to locally negotiated terms and conditions, enter into partnerships with the private sector and are governed by a board that includes elected patient/public representatives (Department of Health, 2005a). Unlike ordinary hospitals, FT Board meetings are not open to the general public. Crucially, and also unlike ordinary hospitals, FTs are both able to build up surpluses and at risk of insolvency. It therefore seems likely that their behaviour towards commissioners will be generally more combative than has been seen in the NHS in the past (Marini et al., Reference Marini, Miraldo, Jacobs and Goddard2008).

Dominant interests

Alford argued that ‘dominant’ interests are those favoured by current structures. Under PbR, hospitals, especially FTs, have an incentive to maximise their activity, provided they can perform that activity at a cost that does not exceed the tariff (Marini et al., Reference Marini, Miraldo, Jacobs and Goddard2008). Thus increasing efficiency and overall activity, along with ensuring that all activities are coded and therefore claimed for, are obvious objectives (Kurunmäki and Miller, Reference Kurunmaki and Miller2008). However, since the rules of PbR make hospitals responsible for the categorisation of episodes of care into HRGs, it is also in hospitals’ interests to look for easier ways to increase revenues, such as coding cases so as to secure maximum possible payment for each activity. Thus, whereas under the system that existed prior to PbR (whereby hospitals were paid for providing a specified volume of cases) a consultant asking a colleague for an opinion would be unlikely to register this as a referral, under PbR such activity will be logged and coded, attracting payment as a ‘new outpatient’ event. PbR also encourages Trusts to ensure that telephone requests from GPs seeking advice are also coded and charged for. Thus activity that was previously invisible is visible under the new system, placing further pressure on PCT budgets. A priori, therefore, the system of PbR puts hospitals in a dominant position and encourages them to act to maximise their income in ways that will disadvantage commissioners of care. We found that, working within the rules, hospitals in all our sites had increased their income from commissioners without necessarily increasing activity.

In meetings and interviews, respondents emphasised the perceived distortions occurring as a result of PbR, with FTs regarded as particularly aggressive:

In all our sites, respondents complained in interviews and in meetings that hospitals had engaged in ‘gaming’ in order to maximise their income:

Indeed, the incentives inherent in PbR were felt to have actively prevented hospitals and GPs from working together to reduce admissions and overall costs to the NHS:

The perceived dominance of hospitals under the PbR regime was reinforced by the fact that ‘patient choice’ was generally not acting as the Department of Health assumed in the areas that we studied. Guidance relating to PbR suggests that patients will choose from a variety of hospitals, putting pressure on them to offer a better service (Department of Health, 2002, 2005b). In practice, however, commissioners believed (and in at least two cases had survey evidence) that patients generally wished to be treated in their nearest hospital:

This local loyalty entrenches the structural advantage enjoyed by hospitals under PbR, ensuring that commissioners police themselves and their activity so as to avoid action that might destabilise the hospital. Commissioners were anxious not to put their local Trust under pressure by taking work away:

Finally, hospitals were able to protect their monopoly by forbidding discussions about services between the consultants whom they employed and GPs involved in PBC:

Challenging interests

PBC is not explicitly or officially identified as a counterweight to PbR, but its structures seem likely to have this effect. Thus PBC consortia that make savings on their budgets (by, for example, reducing referrals and/or prescribing) are guaranteed at least 70 per cent of this saving to reinvest in new services for their patients. Furthermore, services equivalent to outpatient services that are provided outside the hospital setting are not subject to the PbR tariff, providing an incentive for consortia to establish community-based alternatives to hospital services that will be charged at less than tariff prices, saving money that can be reinvested.

Our respondents argued that PBC was the tool required to combat the power of Trusts under PbR:

We found evidence in all our sites that those involved with PBC saw it as their role to challenge hospitals’ coding of patients:

Referrals between consultants were a particular concern, as these had increased considerably since the introduction of PbR. Commissioners sought to ensure that consultants and other hospital clinicians did not cross-refer without approval from the referring GP:

Some PCT staff reported that GPs were prepared to be more challenging than PCTs had been prepared to be in the past:

Some respondents reported that they were intending to use contracts to specify in great detail how patients should be managed:

One of our sites had gone a long way down this route, and claimed to have made considerable savings from enforcing agreements. Conditions such as arthritis of the hip, deep vein thrombosis and cholecystitis were all the subjects of ‘patient pathway’ development. Some respondents saw ‘evidence based medicine’ as a means of challenging hospitals:

Others expressed a more general desire to use the incentives and levers available to them to challenge hospitals:

However, respondents also reported limits to the impact of their challenges. Even where cases had been demonstrably miscoded by a hospital, there was reluctance to exert financial pressure:

Furthermore, those PBC consortia that did not commission mainly from a single hospital had little influence:

However, towards the end of the fieldwork period, there were signs of the development of joint ventures between hospitals and PBC consortia, although such movements toward joint working were rudimentary, and seen most clearly in the site with least competition between providers:

Conclusions: theory and implications

We have argued that our study of the early stages of Practice Based Commissioning in England has confirmed Alford's contention that health service structures (in this case the development of Payment by Results) systematically privilege certain interests above others. In this specific context, the dominant structural interest comprised by the hospitals can be labelled as ‘corporate monopoly’. Hospitals are monopolists under this structure because their ability to define their workload and income, combined with the reluctance of patients to travel and of commissioners to destabilise local services, gives them a local monopoly in a context where the market has come to be seen as the natural way to organise health services. Contrary to Alford's picture of managers across an area working together as ‘corporate rationalisers’, rationally planning to service the health needs of a population, hospitals are incentivised by PbR to compete with and draw business from their local competitors. It is important to note that this monopoly is corporate in that it is the hospital as an institution that possesses this status, and that hospital medical staff share in this interest, with their ability to provide clinical care and pursue their clinical interests (and, more remotely, perhaps, their jobs) dependent upon the hospital's economic fortune. This does not mean that there are no intra-hospital conflicts; as noted above, hospital doctors may be expected to act in ways that run counter to their own immediate interests: for example, where consultants were forbidden to have any contact with PBC consortia. In our analysis, these are interest group conflicts, occurring within a structural interest.

The challenging structural interest constituted by PBC can be labelled as ‘professional rationalisers’. This challenging interest is professional, in that it was being clearly led by GPs. Nurses and other practice employees rarely played major roles within PBC groups, and GPs and PCT employees commented in interviews that the strength of PBC lay in the clear lead role played by GPs, who have both the clinical knowledge and the appetite to challenge the hospitals. We found evidence of PBC consortia challenging hospitals’ coding of cases and being prepared to challenge consultants on clinical grounds, using their professional knowledge to argue about clinical care. PBC consortia constitute a rationalising interest in the sense that they are behaving in ways that are essentially formal and bureaucratic, justifying their demands in terms of guidelines such as those produced by NICE and its Scottish counterpart, and on National Service Frameworks, and seeking to govern their relationship with hospitals through formal rules, such as ‘patient pathways’. Like the corporate monopoly, the professional rationalising interest is likely to exhibit internal cleavages; for instance, it is clear that initiators of PBC in specific localities had their own agendas (Checkland et al., Reference Checkland, Coleman, Harrison and Hiroeh2008).

In disrupting the association made by Alford and subsequent analysts between corporate organisations as rationalisers and the medical profession as monopolists, we have sought to identify new structural interests in the context of the English NHS. What are the implications? First, Alford argues that, over time, dominant and challenging structural interests will move towards establishing a modus vivendi. In our study this was manifest as embryo joint venture schemes. From a structural interest point of view, this movement towards joint working makes sense, in a context where competition is rudimentary, particularly at the level of individual patient preferences. The corporate monopolists within hospitals would be able to neutralise the challenge from the commissioners by holding them close and tying them in to joint ventures. The professional rationalisers would gain influence over the services provided by the hospital, embedding their rationalising agenda and thereby enhancing their status. The development of patient pathways requires agreement between referring and providing clinicians, as a pathway that hospital clinicians refuse to follow will be ineffective. It is therefore in the interests of professional rationalisers to work closely with a small number of hospitals to develop an agreed ‘rational’ pathway rather than putting services out to tender to a wider network. Moreover, this observation connects with current proposals for the development of what have been labeled ‘Polyclinics’ in the UK (Department of Health, 2007). In a report of an ‘inquiry’ into the current state of the NHS (http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_085825 accessed August 2008), Lord Darzi, a surgeon, has suggested that new GP clinics should be established in health centres that also provide some diagnostic and therapeutic services currently provided by hospitals. The government has responded by requiring every PCT to commission one such centre. Our analysis suggests that corporate monopolists will resist such developments (since they threaten to remove business from hospitals) while professional rationalisers within PBC consortia will be wary of structures that might disrupt carefully developed care pathways and services. The implication is that actors within the two structural interests are faced with strong incentives to cooperate to seek the development of such clinics as joint ventures, and indeed Alford argued that the durability of structural relationships derived partly from the willingness of structural interests to reach compromises.

Second, although we did not seek to collect direct evidence about patients and public, it is clear that neither patient choice nor patient and public participation is facilitated by the structures that we identified. Patient choice is lauded by the UK government as the key to improving quality in the NHS (Milburn, Reference Milburn2003; Department of Health, 2006b: 6, 2006c: 18), but (leaving aside the question of whether public appetite for choice of provider really exists) our ‘challenging interests’ showed little inclination to pursue a diversity of providers under their commissioning plans, other than as a rhetorical threat to use to bring the local hospital into line. Our data also suggest that patient and public involvement is unlikely to be anything more than rudimentary within the current structures. From a rationalising perspective, the ideal service is one that delivers identical care to all patients identified as suffering from a defined condition, based upon ‘the evidence’ of what is the ‘best’ care for that condition. The most efficient mechanism for implementing this is a bureaucratic pathway or algorithm that states explicitly what care should be provided in what circumstances, and we found evidence of the development of these types of bureaucratic pathways in our study sites. From this perspective, patient or public participation has little logic; indeed, we found that patients and the public were not involved in strategic decisions about commissioning in our sites, even though official PBC guidance mandates that this should occur (Department of Health, 2006c: 6). Although these findings might be seen as regrettable, they may well reflect the extreme difficulty of meaningful participation (Martin, Reference Martin2008) and, more importantly for our analysis, cannot be taken to imply that patients and public are ‘repressed’ structural interests in a sense that Alford would recognise. As noted above, ‘interests’ for Alford imply an external view of what a particular group might be said to ‘need’. In the New York system that Alford studied, the uninsured ‘community population’ can be said to be failing to have their needs met in a very real sense, a level of repression that, as North (Reference North1995) has observed, is not paralleled in an NHS that provides services, mostly without charge, to all UK residents.

Finally, our argument that it is no longer possible to simply read off Alford's categories of structural interests into the current NHS situation raises questions about how the apparent restructuring, with hospital-based corporate monopolists as dominant interests and professional rationalisers as challenging interests, has occurred. Alford himself is somewhat unspecific about how new challenging interests come into existence, referring to ‘changing technology and division of labour in health care production and distribution, and the shifting rewards to social groups and classes’ (1975: 15), also arguing (as noted above) that such change has little or nothing to do with government. It would be surprising if this were to apply in a state-funded system such as the NHS and, unlike in Alford's New York, government health policies have had an effect, and indeed have been a necessary condition for the restructuring of interests that we have observed. This is not to say that the present structure of interests has been rationalistically created; rather, government has taken a number of decisions that have together brought about the restructuring. Most obviously, evidence-based medicine (increasingly promoted and bureaucratised by government since the mid-1990s: Harrison and Smith, Reference Harrison and Smith2003), the PbR system (introduced in 2002) and PBC (introduced in earnest in 2006) are government policies that have provided some of the necessary conditions. However, it seems likely that social changes resulting from earlier government reorganisations of the NHS have also contributed. First, the readiness of GPs to become professional rationalisers involved in the planning and commissioning of services under PBC is perhaps partly a legacy of the system of GP ‘fundholding’ (Glennerster et al., Reference Glennerster, Matsaganis, Owens, Hancock, Robinson and Le Grand1994) that obtained from 1991 until abolished by the incoming Labour government in 1997. Second, the ability of hospitals to function as corporate monopolies – that is, with an alignment of interests between managers and doctors – is likely to be a legacy of earlier NHS reorganisations. In particular, the roles of hospital chief executives became seen as legitimate after the Griffiths Report of 1983, even though they initially had little influence (Harrison et al., Reference Harrison, Hunter, Marnoch and Pollitt1992), and the so-called NHS ‘internal market’ of 1991–1997 served to remind doctors that their careers were bound up with the prosperity of their hospital, even though the market rarely led to real competition between hospitals (Checkland, Reference Checkland, Flynn and Williams1997).