Cost and performance information for doctors: an international comparison

Abstract

In the last decade many OECD countries have undertaken substantial changes to their system of healthcare administration and funding. These changes have been characterised as being economically and managerially focused. However, doctors remain central to any system of healthcare and therefore to the implementation of these reforms. Therefore, if these changes have had any real impact on doctors then there should be a clear interest in and access to information on the cost and outcome of clinical activity. This paper attempts to evaluate whether this is the case. We have conducted a case study of hospital sites in the UK, Germany and Italy. We have found that although there was some interest in cost and activity information that clinical staff generally did not have access to it. Cost and activity information were only available to clinical staff at the most senior levels. Therefore, serious questions must be raised about the extent of penetration that these reforms have had at the clinical level.

Introduction

In this paper, we seek to explore the question of clinical access to and interest in cost, performance and feedback information. Ham (1997) described the reform of health care administration and funding as an ‘international fashion’. Within the UK these changes have occurred in the context of a wider programme of restructuring and marketisation. Many aspects of the public sector have been subject to change and commentators such as Hood (1995) have suggested that the old public service ethos has departed and that a new set of values, which he calls New Public Management (NPM), now characterise the service. It is well established that accounting and accounting visibility are central to these NPM values and to the operation of many aspects of contemporary public administration (Lapsley, 1999). A shift towards a NPM model is accepted in relation to the UK NHS (Ballantine et al., 1998, Kurunmäki et al., 2003). While accounting measures and controls are not a new innovation, they have been transformed from an outside threat to a central part of the function and operation of many aspects of the UK NHS (McNulty and Ferlie, 2002).

Ferguson and Lapsley (1989) conducted a postal survey of doctors in three Scottish hospitals to identify their information needs and attitudes towards cost and performance information. This paper seeks to explore whether now, over fourteen years later and after the implementation of a number of different NPM style reform initiatives, doctors have both a greater awareness and access to cost and performance information. Ferguson and Lapsley (1989) drew on two earlier pieces, Davis and Miles (1984) and Carter and Magee (1983), both of which were critical of the quality of the financial and performance information provided to clinical staff, however, there was some ambiguity over how useful financial information would be to clinical staff. Ferguson and Lapsley (1989) were clear that costs were important, particularly to more senior staff and for developments or purchase of equipment. However, there was scope to explore what kinds of information doctors have access to.

Europe is far from unitary in terms of health care arrangements (Freeman, 2000). Different forms of health funding, provision, patterns of professionalisation and professional control have emerged in different countries. It also seems evident that NPM is short-hand for a far more complex and divergent development (Ferlie et al., 1996, Hood, 1998) and that there is no pre-prepared, off the shelf reform package (Olson et al., 1998b). Pollitt and Bouckaert (2000) highlight the elusive nature of NPM in the international context, particularly when researchers focus on results rather than policy proposals. Marcon and Panozzo (1998, p. 205) are critical of simplistic anglo-centric analysis of these changes which suggest “an all-too-easy convergence in the way in which accounting intervenes in health care reform”. They call for “systematic and empirically grounded cross-national comparisons” (p. 206) that move beyond a focus on English speaking countries. This paper represents one response to that call and, as such, seeks to unpack the role of accounting in European health care reform, particularly the focus on costs and measurement of performance.

Central to the implementation of any reform policy are the staff who deliver the service. Lipsky (1980) called these people the street level bureaucrats. Within health care it seems relatively uncontroversial to suggest that medical staff are likely to have a substantial influence over the nature and progress of any reform initiative. Also, because of the significance of medical staff in resource allocation in health care they are likely to exercise significant influence over how a given policy is implemented and how the services are actually provided.

It is well established that hospitals are complex organisations and seem to contain two separate elements – a medical staff or demand division and an administration or supply division (see for example, Harris, 1977, Harrison and Pollitt, 1994) and that these two divisions are ‘loosely-coupled’ (Weick, 1976, Kitchener, 1999). Reform proposals that only address the administrative or supply division and make no real impact on the medical or demand division are likely to have limited efficacy. In short, if NPM is really an international trend then doctors in different countries should be concerned with questions of cost and performance measurement.

Central to the assumption of economic rationality contained in much of the health reform literature, particularly that reflecting NPM, is the assumption of the availability of information (Marcon and Panozzo, 1998, Robinson, 1999). Information relating to cost, performance, standards and targets is assumed to exist and to be provided to the right people (McNulty and Ferlie, 2002). Also central to the operation of many of the private sector management technologies described by Hood, 1991, Hood, 1995 is the function of an effective organisational information system, providing financial and performance information to key decision-makers. Historically doctors are seen to be the key decision-makers however, more recently both managers and nurses are playing an increasingly important role (Ong et al., 1997, Griffiths and Hughes, 2003). However, it is clear that within the structural arrangements for healthcare provision that doctors exercise considerable power and autonomy (Harrison and Pollitt, 1994, p. 35). Therefore their access to, attitude towards and use of financial and performance information is a critical indicator of whether NPM is anything more than rhetoric in the health care setting. In addition, if it proves possible to understand this area better and to improve the quality of information provided to doctors then gains in both quality and efficiency could result. Therefore, this paper provides an international comparative analysis of doctors’ access to and attitudes towards cost and performance information.

The UK, Italy and Germany were chosen as the locations for the study as they represented contrasting models for health care funding and management. A study comparing the UK, Germany and Italy also represents a serious attempt to respond to Marcon and Panozzo (1998) call to go beyond English speaking countries and to critically evaluate what is described as NPM in contextually informed empirical research. As much of the research on doctors, information and cost is UK or US based, there is clearly a need for this work.

The UK, Germany and Italy also have very different professional histories; the medical profession is constructed differently, has developed differently and has different levels of political power and influence. Each of these factors contributes to interesting and potentially valuable contrasts between the different countries and potential factors in doctors’ attitude towards and access to cost and performance information.

Germany has been seen as the archetypical health insurance system combining both involvement with and separation from the state. The state has been strongly involved in a supervisory and regulatory role, ownership is spread throughout different state, charitable and private entities and funding is almost completely separate, vested in the over 750 statutory health insurance associations (Gesetzliche Krankenversicherung). Some of these associations are regional while others are associated with particular professions or firms although alternative private health insurance companies are available to those who are self-employed or who are high-earners who choose to opt-out of the normal system. The needs of the poor and the unemployed who do not have insurance are covered by local and regional authorities from general taxation.

Many of the German hospitals are owned and operated by local or regional authorities (öffentlich) however an almost equal number are operated by private-non-profit bodies (freigemeinnützig), often associated with religious or charitable organisations and a few are owned and operated by for-profit businesses (privat). Until 1996 German GPs and specialists who practice outside of the hospitals (and therefore historically have dealt with ambulatory care) were reimbursed on a fee-for-service basis, while ultimately being constrained by a global budget. However, in 1996 this changed to a capitation based approach. The hospital system was reimbursed on a bed-day basis however that is shifting towards a diagnosis based reimbursement model similar to the US style DRG system within a capped budget arrangement.

German doctors are amongst the best paid in Europe (in relation to the average employee salary) (Freeman, 2000, p. 91). However, considerable pressure has been placed on doctors to control their numbers and to contain costs. In 1989, reforms obliged doctors to consider the cost-effectiveness of referrals to hospitals and in 1993 provided for individual indicative drug budgets. The introduction of DRG style cost reimbursement represents a further incursion of economic and the state in the medical jurisdiction.

German hospitals have generally maintained the three-part collegial management structure. Management is shared by the Medical Director (Ärztlicher Direktor) the Director of Nursing (Pflegedirektor) and the Administration Manager (Verwaltungsdirektor). On the medical side there are three different levels of medical staff under the medical director. The Clinical Directors (Chefarzt) are responsible for a clinic or speciality (Abteilung) within the hospital. Senior doctors (Oberärzte) answer to each of the Chefarzt, but have considerable autonomy in their medical practice. Junior doctors (Assistenzärzte) generally operate independently but may consult their Oberärzte if they have questions on matters of medical treatment. Doctors who are still in training are known as Ärzte im Praktikum.

The Italian national health system (Servizio Sanitario Nazionale SSN) is a relatively recent innovation, tracing its origin to the ‘Institution of the National Health Care Service Act’ of 1978. It combines taxation and national insurance funding and has a relatively high level of delegation to the regional and the local level. Regional government is responsible for the provision of health through the 228 regional health authorities (Aziende Sanitarie Locali, ASLs – known as USLs – Unità Sanitarie Locali prior to 1992). The 1992 reforms made the regions responsible for shortfalls in funds. The new title Azienda (business/firm) was intended to indicate a shift in orientation and practices in these organisations towards a business orientation. In practice, the trusts are referred to as Aziende USL. Since the 1992 reform some of the larger hospitals have been made organisationally autonomous and have been devolved budgets and have become ‘independent hospital agencies’ (Aziende Ospedaliere). Since 1995 there have been experiments with contracting and a form of quasi-markets. Many Aziende USLs now purchase services on a diagnostic related group (DRG) basis from hospitals. However, in reality the local area trusts and the individual hospital institutions are not separate. The independent hospitals (normally large university associated teaching hospitals) are controlled indirectly through the funding arrangements.

Italy currently has one of the highest number of practising doctors per head of population in Europe, more than three times that found in the UK. Doctors working in public hospitals are paid a salary but they are forced to choose whether to operate in the private or the public sector as split-appointments and private patients are no longer allowed for SSN doctors. Previously, there were many and complex levels of medical seniority within hospitals. However, in the last couple of years this has been significantly reduced and now there is only two: the Primario who is the head of the unit (unità operativa) and the other doctors (aiuti and assistenti). Answering to the Primario is the unit head of nursing (Capo Sala), who is responsible for all of the nurses, nurse aids and ancillary staff in the unit. The Primario answers to the hospital director, the Dirigente Medico Ospedaliero (normally a doctor), who is responsible for all of the hospitals in the local area. It is evident that the existing management structure of the hospital is deeply embedded, and indeed indistinguishable from the management of the local area. The embedded relationship between the local region and the hospital is even more evident when one takes a closer look at the management structure within the hospitals in a locality. The one Primario may be responsible for the laboratory services of all of the hospitals in the area, with practical (although not structural) responsibility being delegated to an assistant at the other site. Therefore, it is not possible to really conceive of these as separate hospitals, rather more one hospital with multiple locations as they share common systems of management and accountability.

While for many years, the UK was seen as the advocate of market reforms, more recent changes have seen a shift away from this position and a move towards ideas of co-operation, planning and co-ordination. The formation of the UK NHS occurred in July 1948 (Klein, 1995). Central to this structure were the power of the hospital clinician and the promise of free access to the public. Health services were funded from a general taxation base, were administrated by the Department of Health and multiple regional public agencies. Services were primarily provided by the publicly owned hospitals. In 1989/1990 the UK government initiated a number of changes, which created what was known as the ‘internal market’. In 1997 it was announced that competitive market was being rejected and that the new model was one of ‘integrated care’. These reforms also required UK clinicians to have access to necessary information and to be aware of the costs and cost implications of their actions:

It will be important for right information to be made available to clinicians and for high professional standards to be set and maintained (DHSS, 1997, p. 52).

It is important that managers and clinicians alike have a proper understanding of the costs of local services so that they can make appropriate local decisions on the best use of resources (DHSS, 1997, p. 73).

Within the UK there is a low number of doctors per head of population. All doctors are required to be registered with the General Medical Council, which licences individual practitioners, approves medical schools and investigates accusations of medical misconduct.

Within the UK most hospitals are operated as part of a NHS trust. Trusts are managed by a chief executive, a finance director, a medical director and a head of nursing. Within the trust the chief executive has the freedom to organise the managerial structure. Consultants are the most senior grade of hospital doctor and may engage in private practice in addition to their hospital work. One of the consultants will normally be appointed with management and budgetary responsibilities as a clinical director. Staff who have not completed their specialist training are known as registrars and junior staff who are still in their basic specialist training and have not yet qualified for membership in one of the specialist colleges are called senior house officers.

The UK NHS contrasts to the German system because of the direct tax funding and the extensive state-ownership and management of the hospitals. However, in these areas there are similarities between the UK and the Italian structures. In regard to the internal management structures of the hospitals, the UK is different to both Germany and Italy as in the UK the trust is run by a chief executive rather than a doctor. The UK also has fewer doctors per head of population than the other two countries and the lowest health expenditure as a percentage of GDP of the three countries.

UK and US papers dominate the literature on doctors’ attitude towards cost information, however Mishra and Satpathy (2001) illustrate that the results hold true for other countries. The research shows that doctors have a poor idea of the costs of the resources they use. Fowkes (1985) found that 77% of doctors have no idea about the real costs of the drugs that they used, the hospital costs or the therapeutic costs incurred. O’Connell and Feely (1997) found that the majority of doctors were unable to accurately estimate the cost of the medicines they were using and Ryan et al. (1996) found that only one third of the doctors studied were able to accurately estimate the costs of drugs. In the absence of accurate information doctors operate on the basis of inaccurate estimates. Long et al. (1983) found that the physicians generally underestimating prices, particularly when they were ordering higher priced tests and/or a large number of tests. Ryan et al. (1996) and Reichert et al. (2001) also found that doctors underestimate the expensive costs and overestimate the cheap ones.

There is evidence that the provision of accurate information alters clinician behaviour. Frazier et al. (1991) were clear that cost information made a significant different to prescribing and costs. However, while Cohen et al. (1982) found that providing physicians with information about the costs of tests led to some change in test usage, there was also contrary behaviour as some physicians made very little change. Cohen et al. (1982) concluded that simple cost feedback systems were not sufficient and that it was essential to prepare physicians to use the data, particularly in a team/clinical review context. This is consistent with Jacobs (1998) who suggested that systems of cost visibility can be welcomed by clinicians and make a significant difference when they are produced by other clinicians and were seen as part of a system of medical quality control. Hoey et al. (1982) also had a mixed result and found that while price information did have some influence on the medical tests ordered a large proportion of the tests were ordered because of clinically absolute reasons and these physicians were insensitive to price.

There have been attempts in many countries to involve clinical staff in management and administrative roles, which indicate that, in some circumstances, clinical staff are both capable and willing to engage with cost and performance information (Ham, 1997). UK initiatives such as the recent ‘primary led health care’, budgetholding and clinical directorates are all evidence of this objective. Within Germany there has been an interest both in the management responsibilities of clinicians and the relationship between clinicians and management (Kuck, 1999, Anheier, 1997, Hellmann, 1999). Within Italy (Zanetti et al., 1996) and most other European countries (Ham, 1997) there are also similar interests.

While the interest in the management responsibilities of clinicians is shared, the national focus on information provision has been emphasised the most in the UK. The UK Royal Commission on the NHS (1979) and Körner (DHSS, 1982–1984) emphasised the need for national level comparative statistics and national performance indicators. The reports of Griffiths (1983) and the Resource Management Initiative focused on the roles of budgets and budget delegation to clinical staff. The directive of 1986 (DHSS, 1986, appendix, section II) identified that earlier initiatives had failed because the information needs of clinical staff were ignored and that doctors and nurses should be centrally involved in specifying the new system (Lapsley, 2001). DHSS (1986) specified that one objective of the Resource Management initiative was “to give a better service to patients by helping clinicians and other managers to make better informed judgements about how the resources they control can be used to the minimum effect” (p. 2). Packwood et al. (1991) indicated that although the initiative succeeded in improving cost awareness among clinical staff, the provision of data relevant to patient care and of value to clinicians in providing care had been “provided patchily to sub-units and [to] individuals” (p. 159). It is questionable whether the objective was actually achieved. Simpson (1994) was clear that the requirement of accurate, timely and appropriately presented information on clinical activities was often not met. A UK Audit Commission (1995) study of information management and systems in acute hospitals also found that most of the information systems were for administrative or financial purposes. Relatively little attention was devoted to clinically relevant systems and the clinical systems that existed were generally isolated from the administrative and finance systems. The UK Audit Commission (1995) argued that data must be collected in a form that is clinically useful.

Lapsley (1996) addressed the question of doctors and information, although the primary focus of the paper was on hospital doctors’ decision-making. He found that the there was an absence of information to support management and development decisions and what was there tended to be informal and ad hoc. Lapsley (1996, p. 4) argued that there was a need to develop better information for doctors.

Lack of rigor in prediction was felt to be very difficult to overcome without the development of information systems which would highlight best practice through the linking of input, medical procedures used and the outcome. An example was cited where attempts were made to identify the success criteria in a certain surgical procedure.

Lapsley (1996) challenged the view that doctors have no interest in management and cost information. This view is also challenged by the medical literature where themes such as cost-benefit analysis are increasingly discussed (for example see Kangis and van der Greer, 1996 and Phillips and Holtgrave, 1997).

In summary, although many reports, some government initiatives and a number of academic researchers have argued for the importance of clinically relevant information there is little evidence that this has been provided in any coherent way. Therefore, this paper seeks to specifically address this lacuna and address not only the issue of clinical attitudes towards costs and performance information but also the simple question of what kinds of information doctors currently have access to.