Design of an activity based costing system for a public hospital: a case study
This paper describes a case study of ABC system design in a new public foundation hospital in Spain focusing on design elements leading to successful conflict-free implementation, including managing resistance. The design stage has been largely ignored in the ABC literature (Arnaboldi and Lapsley, 2005). It expands on previous studies of institutional forces’ impacts on hospital management and is based in a non-English-speaking country responding to calls for research in other countries (Marcon and Panozzo, 1998).
Furthermore, it focuses on costing rather than control functions, the primary focus of previous studies. The next section discusses institutional sociology in hospital cost management contexts, followed by descriptions of hospital management and ABC within this framework. The following sections describe research methods, the Spanish healthcare system, details of the ABC design in the case-study hospital, and a concluding discussion presenting implications and contributions. 2 Institutional sociology
Organisational theories posit that organisational diversity results from adaptation to environmental pressures and organisations’ survival requires conformity to institutional pressures (Covaleski and Dirsmith, 1988; Oliver, 1991; Carmona and Macias, 2001). Hence, public hospitals should respond to institutional pressures to seek legitimacy from Design of an activity based costing system for a public hospital 3 governmental health and finance ministries and organisations directly supervising their operations.
Institutional theory emphasises that organisations’ behaviour does not always reflect rational analysis (Isaksson and Cornelius, 2004) but may attempt to conform to social and cultural norms (Scott, 1987, 1995). For extensive discussion and critique of institutional theory and sociology see Hopper and Major (2007). Institutional sociology emphasises cultural, normative and cognitive factors and suggests organisations must appear legitimate; behaviour may be directed more towards environmental acceptance than technical efficiency (Scott, 1992, 1995, 2001; Baxter and Chua, 2003).
These pressures produce isomorphisms, processes that force organisations to resemble others facing similar environmental conditions (DiMaggio and Powell, 1991). Hence, organisations ultimately conform to institutional pressures to achieve legitimacy, institutional support and stability. By contrast, market-based theories view organisations as economic transacting and rational decision making sites (Scott, 1995, 2001); Baxter and Chua (2003) suggest behaviour is directed towards internal technical efficiency due to market pressures. Isomorphism is either competitive (Hannan and Freeman, 1977, 1984) or institutional (DiMaggio and Powell, 1983).
NPM advocates presume that rational behaviour improves efficiency and effectiveness and can coexist with isomorphic legitimacy seeking. Furthermore, increasingly in private and public organisations, legitimacy depends on demonstrating efficiency and rationality (Lapsley, 2001). Isaksson and Cornelius (2004) conclude that institutional sociology approaches have provided new and different insights into contracting between venture capitalists and entrepreneurs, indicating that venture capitalists have more influence because of more contracting experience (also see Cornelius, 1997).
In this study, the new Spanish foundation hospital is an entrepreneurial organisation. Health ministry contractors are not entirely analogous to venture capitalists, but recent changes in contractual relationships between them and healthcare providers have introduced private-sector methods like ABC into the process similar to those required by venture capitalists (Ferguson and Lapsley, 1989; Ruef and Scott, 1998). Typically, healthcare providers have been in weaker positions than funders because of limited knowledge of their costs.
Exogenous and endogenous forces Researcher(s) Coercive isomorphism Mimetic isomorphism Normative isomorphism DiMaggio and Powell (1983) Political influence: formal and informal pressures Uncertainty encourages imitation Professionalisation of the work force: more concerned with status than efficiency Scott (1995) Regulative: coercive rules, laws and sanctions Cognitive: taken for granted, symbolic, mimetic aspects of social life Normative: social beliefs and norms
Covaleski and Dirsmith (1988) Formal and informal coercive pressure A more passive response Normative: legitimisation of professional autonomy Source: Irvine (1999) and Isaksson and Cornelius (2004) 4 S. D. Eriksen et al. Irvine (1999, discussed by Isaksson and Cornelius, 2004) outlined an approach to understanding institutional theory social traditions in accounting for religious organisations with changing environments. Her delineation of religious organisations and changing environments has similarities to those of public hospitals.
Table 1 presents exogenous and endogenous factors acting on and within organisations making them more homogeneous over time. DiMaggio and Powell (1983) classified these forces as coercive, mimetic and normative isomorphisms. In coercive isomorphism, external pressures such as the regulatory environment (Scott, 1995), or demands of taxpayers, legislators, patients and their families and the public on hospitals affect institutional choice. DiMaggio and Powell (1983) suggest coercive isomorphism impacts are greatest when organisations like public hospitals are subject to legislative initiatives, economic, or moral imperatives.
Mimetic isomorphism is the tendency to mimic successful organisations and is more likely to occur when there is a high degree of environmental uncertainty (DiMaggio and Powell, 1983). In new hospitals, uncertainty can motivate administrations to mimic established hospitals. Equally important, it can motivate corollary participants like hospital physicians to mimic physicians in other hospitals. DiMaggio and Powell (1983), however, claim that institutions confident in their identity, internal structures, goals and abilities to achieve aims have little incentive to mimic others.
The final framework factor of Irvine (1999), Scott (2001) and Isaksson and Cornelius (2004) is normative isomorphism. Individuals trained in the same discipline and working in similar institutional settings share a common understanding of ‘normal’ behaviour and what is acceptable (DiMaggio and Powell, 1983). Because of normative isomorphism, institutions like hospitals that draw on a standard pool of ‘workers’ like physicians find their ability to improvise new approaches is compromised because the ‘workers’ often follow professional norms that are not always consistent with organisations’ norms.
Institutional sociology has been studied extensively, but not in system design contexts (Scott, 2001; Dacin et al. , 2002). Studies of managing resistance to change in system design processes are particularly lacking. Hopper and Major (2007) and Major and Hopper (2005) for example, use institutional sociology to discuss ABC implementation in a for-profit telecommunications organisation, but not design issues. Moreover, because the firm was well established, they could not address such issues in newly created organisations like the hospital in this study.
Difficulty in obtaining common strategies is supported by some studies (McGuire et al. , 1997; West and West, 1997). They identify problems in the design and implementation stages of ABC, including the identification of activities, cost pools and drivers. Suggestions for addressing problems include focusing the initial design on specific subunits of the organisation rather than the entire organisation (Beecham et al. , 2001; Orlewska, 2002). This paper expands studies of institutional forces’ impacts on management adaptation in hospitals by addressing managing resistance during ABC design.
This approach is consistent with Oliver (1991) suggesting successful adoption of NPM in Spanish hospitals should actively involve many strategic adoption choices rather than passively responding to the environment and complying with demands; successful NPM employment should recognise that hospitals’ responses to conformity pressures depend on why these pressures are being exerted, who is exerting them, what these pressures are, how they are being exerted and where they occur (Oliver, 1991). Design of an activity based costing system for a public hospital.
Hospitals and ABC in institutional settings 3. 1 Management processes of hospitals Hospitals are amongst public institutions most affected by NPM because of reduced public funding and emphasis on performance measurement (Arnaboldi and Lapsley, 2004, 2005) and one response has been the adoption of contemporary management systems including ABC [see Jones and Dugdale (2002) and Armstrong (2002) for history and critiques]. Extensive external pressure from NPM advocates, coercive isomorphism, has motivated hospitals to implement ABC.
Cost control is a major issue for hospitals changing from a fee-for-service model, in which providers could pass on cost increases, to managed care in which providers are paid fixed predetermined fees (Ferguson and Lapsley, 1989; Ruef and Scott, 1998). Traditionally, weak cost measurement systems have caused hospitals to contract with payers without having realistic knowledge of their costs (Long et al. , 1983; Fowkes, 1985; Ryan et al. , 1996; O’Connell and Feely, 1997; Reichert et al. , 2001). Changing relationships between hospitals and payers were major elements of introducing market discipline.
Success, even survival of hospitals, especially under capitated systems with rates paid per procedure or per patient, depends on appropriate resource utilisation and controlling costs per service unit (Bergthold, 1990); both are focus of ABC. In changed healthcare environments, hospitals have advantages when contracting with payers when there are prospective and retrospective reviews of high volume procedures that assess resource utilisation and ABC facilitates such reviews (Hussey and Holford, 1993). Moreover, adopting ABC provides not only economic benefits but also signals hospitals’ intentions to improve efficiency.
Such institutionalisation represents coercive isomorphism. 3. 2 Physicians and hospital cost management Scepticism, political self-interest and control constrain organisations’ willingness to conform, whereas capacity, conflict and awareness bound their ability to conform (Oliver, 1991). Responses to conformity pressures depend on why pressure is exerted, by whom, what pressures are exerted, how and in what environmental context. In hospitals, pressures to adopt ABC come from funding agencies seeking economic efficiency.
Physicians however, often perceive this objective conflicting with their clinical healthcare objectives. The non-profit nature of hospitals and lack of formal bureaucratic controls compound physicians’ scepticism (Mechanic, 1976). Major and Hopper (2005) reported similar scepticism and resistance to ABC by production engineers in the telecommunications company. Production engineers are analogous to physicians because each group has institutional norms and views themselves as dominant professionals crucial for organisation success.
Lee and Mahenthiran (1994) observed different groups’ conflicting interests when investigating structural elements influencing healthcare system implementations, arguing that successful ABC implementations require common strategies, which is difficult in hospitals given different service delivery goals. Many organisations are confronted with inconsistent institutional expectations and internal objectives. Institutional expectations of funders and hospitals administrators may be based on NPM whereas those of physicians are autonomy over decision making. 6 S. D. Eriksen et al.
In hospitals, perceived conflicts of interest between physicians and management result from different socialisation and ensuing values. In Spain, hyper-bureaucracy compounds conflicts because physicians are civil servants employed by the state not hospitals. The medical profession is the physicians’ dominant socialisation agent (Lurie, 1981; Derber and Schwartz, 1991) with physicians oriented towards patient care (Alexander et al. , 1986) and the medical profession as the primary control mechanism (Mintzberg, 1979). Furthermore, physicians are ‘dominant professionals’ who control core clinical processes (Friedson, 1975).
Hospital management, however, is oriented towards efficient and effective use of economic resources for all patients consistent with overall needs (Alexander et al. , 1986). Core hospital processes dependent on physicians’ expertise and significant autonomy, not subject to bureaucratic controls (Mechanic, 1976), compound conflicting orientations (Barley and Tolbert, 1991; Derber and Schwartz, 1991; Zucker, 1991). Consequently physicians have substantial authority – their decisions commit 70 to 80% of hospital resources.Two trends have intensified perceived conflict: firstly, physicians are increasingly integrated into the management structure (Abernethy and Stoelwinder, 1990). Secondly, funding via prospective payment schemes (PPSs) which pay predetermined amounts for standard treatments called diagnostic related groups (DRGs) transfers economic risks from payers to hospitals.
In response to PPSs, hospitals seek to develop sophisticated budgeting and costing systems (Comerford and Abernethy, 1999) like ABC to capture resource consumption and profitability of product lines (Chua and Degeling, 1991; Preston, 1992). 3. 3 Cost accounting and ABC in hospitals In traditional hospital cost systems, direct costs are related to revenue-producing units, and indirect fixed costs are not. Indirect costs are allocated to direct cost units using a step-down approach. Ultimately, all costs are allocated to stand-alone medical services, and then to DRGs and patients using arbitrary relative costs or relative values.
Two problems with traditional approaches in Spain are that allocations are based on the US published rates with questionable validity in Spain, and per-unit intermediate costs are determined by actual volume not capacity measures. Studies in English-speaking countries have employed institutional sociology to examine hospitals’ responses to pressures to adopt sophisticated management planning and control tools. Some (e. g. , Abernethy and Chua, 1996) addressed fundamental criticisms of institutional sociology, namely neglect of power and interests.
Others assumed that practices adopted to secure legitimacy are only symbolic and decoupled from operative internal systems (Mouritsen, 1994; Carruthers, 1995; Chua, 1995). Also, institutional sociology only provides insight into processes rather than achieved states (DiMaggio, 1988). Considerable public sector reform introduced since the 1980s has met minor resistance (Hood, 1991; Olsen and Peters, 1996), but public-sector implementation of ABC has been difficult (Jones and Dugdale, 2002), especially in hospitals (Cobb et al. , 1993; Arnaboldi and Lapsley, 2004, 2005).
Attention has focused on technical considerations for successful implementation (Hussey and Holford, 1993; Lapsley and Design of an activity based costing system for a public hospital 7 Moyes, 1994; Rotch, 1995; Urrutia, 2001), with less on managing resistance. Studies of ABC applications in hospitals (Hussey and Holford, 1993; Lapsley and Moyes, 1994; Rotch, 1995; Urrutia, 2001; Arnaboldi and Lapsley, 2004), focusing on technical considerations note a perceived conflict of interest between physicians and management and potential conflicts when physicians collaborate.
Physicians who place medical priorities above administrative ones and attempt to mitigate control systems are perceived as the primary cause of resistance; they often consider costing information as obstacles to the best possible treatment and perceive costing systems as another part of complicated hospital bureaucracy (Abernethy and Stoelwinder, 1990). Moreover, ABC systems are expensive to install, maintain and update, further complicating the issue (Cobb et al. , 1993).
Hussey and Holford (1993) found that need for significant changes in attitudes of hospital administrators who rejected ABC believing healthcare is fundamentally different from other sectors where it had been implemented. Lapsley and Moyes (1994) found that physicians directing clinical departments did not believe that sophisticated cost systems were necessary and were suspicious and unsupportive. These authors stressed the need to educate clinical directors on benefits of ABC. This need for education was supported by Urrutia’s study (2001) of the Spanish healthcare sector.
Data collection to implement ABC was identified as another conflict area because it depends entirely on owners of the data (Hussey and Holford, 1993). Physicians are the primary data owners in hospitals and implementation teams must negotiate with physicians rather than making information demands. ABC design necessitates a horizontal process view of hospitals’ operating processes, contrasting with typical vertical view of hospitals based upon specialised functional divisions or departments (Hussey and Holford, 1993).
Due to technical complexities of hospitals, ABC designers must rely on physicians to capture intricacies of hospital processes (Lapsley and Moyes, 1994; Arnaboldi and Lapsley, 2004). DRGs represent operating processes with protocols for related activities, tasks and operations that can be understood only via the descriptions and specifications. No common language exists, however, for these protocols. ABC requires standardised protocols but design is frustrated by lack of common language.
Innes and Mitchell (1995) indicate that process knowledge requirements removes ABC ownership from accounting functions, especially in hospitals. Furthermore, when there are cost reduction policies physicians may believe that ABC-generated information will be used for purposes other than the original stated objectives, such as analysing under-utilised productive capacity (Lapsley and Moyes, 1994), resulting in friction between physicians and management. To address this potential resistance, Lammert and Ehrsam (1987) advocate education and training of clinical staff.
All observations suggest that physicians have power positions within hospitals and are expected to resist accounting systems that control or curtail their behaviour (Abernethy and Stoelwinder, 1995). Studies of information systems needed to capture medical resources utilised in DRGs (Hussey and Holford, 1993; Rotch, 1995; Urrutia, 2001) conclude that physicians are critical for design and implementation of ABC because of their proprietary knowledge. Without physicians’ full collaboration and participation, even the first step in the ABC design may not be achieved. 8 4 S. D.
Eriksen et al. Research method The research method involved an intensive case study of ABC design in Fundacion Hospital Alcorcon (Foundation Hospital Alcorcon) in a Madrid suburb conducted between June 1998 and May 1999, which coincided with the first phase, design and implementation; the second phase involved the actual usage of the system. The study occurred when the Spanish healthcare sector was undergoing significant NPM reforms, including creating new foundation hospitals like the Fundacion Hospital Alcorcon, utilising ABC from the outset, as models for other hospitals.
In addition to data collected during interviews and site visits, we examined large numbers of public documents. We also made four visits to three other Madrid area hospitals to identify common and contrasting costing practices. Recognising the value of longitudinal case studies (Gallmeier, 1991), additional visits were made to the case hospital six years later in 2006 for interviews with three key persons involved in ABC design and implementation. The study began by analysing public documents including annual reports, governmental studies and internet literature, which provided important background information.
The main data for analysis were obtained through 15 hospital visits: three with the chief financial officer (CFO) and the controller totalling ten hours and 12 remaining visits totalling approximately 36 hours. Notes were taken in all interviews. The first interview with the CFO and controller discussed and diagnosed hospital cost accounting systems. We received ABC design documentation and the operational plan which included organisational charts and portfolios of medical services.
The second interview analysed top management’s reasons for implementing ABC and discussed potential problems identified in the literature discussed above (Hussey and Holford, 1993; Lapsley and Moyes, 1994; Rotch, 1995). The third interview discussed external problems including relations with the taxing authority, the hospital proprietor Instituto Nacional de la Salud (INSALUD, National Institute of Health) and the consulting firm hired to assist ABC design. Each of the remaining 12 visits to the hospital verified statements made by top management through observation and discussions with medical personnel.
The follow-up visit six years later included three one-hour interviews with now the ex-CFO, the controller and the medical director about the usefulness of ABC. While the second phase on ABC implementation and use is outside the scope of this study, these interviews clarified and expanded the original objectives of ABC. All interviews and virtually all documents were in Spanish. The authors, all fluent in Spanish and one a native speaker, translated when necessary. Validity and reliability are as important for case studies as for any empirical research (McKinnon, 1988; Atkinson and Shaffir, 1998; Baxter and Chua, 1998).
This study used standard reliability and validity conventions (Cunningham, 1992; Cunningham and Harris, 2005), including multiple interviews at different levels, inviting respondents to read and comment on written drafts of results and establishing good rapport with interviewees. Conducting interviews in Spanish, the native language of interviewees and the ability to read documents in Spanish adds to the validity and reliability. Design of an activity based costing system for a public hospital 5 9 NPM in the Spanish healthcare sector
The Spanish State established the right to public health service in 1978 and INSALUD, a health ministry agency, was created to administer healthcare, a role it exercised until 2002. Later legislation transferred responsibilities and resources to autonomous regional governments called Comunidades beginning in 2002 with INSALUD retaining supporting and consulting roles. In 1996, a new conservative government installed new management in INSALUD mandated to identify problems and solutions (Nunez Feijoo, 2000).
It found high inefficiency levels, de-capitalisation and deterioration of hospital equipment due to lack of investment. The centralised bureaucracy lacked management tools, overemphasised controlling rather than managing costs, and lacked valid measurement tools and adequate information systems (Nunez Feijoo, 1998). Expected transfers of hospitals to regional governments precluded long-term perspectives resulting in a ‘residual’ or ‘minimalist’ management approach. Hospitals controlled few resources because INSALUD employed personnel and owned the infrastructure.
Healthcare administration deteriorated further and reduced motivation of physicians due to lack of initiatives and unclear objectives (Nunez Feijoo, 1999). Consequently, INSALUD produced a general strategic plan and implemented structural and cultural reforms that not only improved healthcare management but also established behaviour patterns promoting flexibility (Gonzalez Gonzalez, 2002). Changes reflected needs for higher quality healthcare administration to facilitate the transfer to local governments (Uribe Ladron de Cegama, 2002).
His leadership qualities and proven ABC experience were primary factors in his appointment as hospital CFO. He was familiar with ABC implementation requirements as well as potential hospital ABC implementation problems identified in the literature (Hussey and Holford, 1993; Lapsley and Moyes, 1994; Rotch, 1995). Due to his prior successful experience, he championed implementing ABC in Fundacion Hospital Alcorcon becoming a critical ‘enabler’ (Arnaboldi and Lapsley, 2003, 2004). At this point, we received the ABC system design documentation.
Twelve subsequent visits verified statements made by management via observation and collaboration with medical personnel. No differences of opinion between management and medical personnel were detected. Design of an activity based costing system for a public hospital 11 When considering minimum requirements for successful ABC design, the second phase one interview with the CFO and controller addressed three potential problem areas: the implementation team, the ABC design and its operating environment.
Consequently, the implementation team management should be multi-disciplinary and understand hospital operating processes. Management also considered classification of costs and patients and cost accumulation processes (which included the proper selection of cost drivers) was necessary. Management especially realised the need for collaboration with physicians. A subsequent hospital visit involved three interviews with physicians from three medical areas to ascertain potential resistance with none detected. Reports on information system output were collected at this time.
The third phase one interview with the CFO and controller discussed their presumptions regarding positive collaboration with physicians, our interviews with the sample of physicians, the conclusion that their presumptions were well founded, and that there did not appear to be problems of potential physician resistance. Having effectively discarded internal sources of problems, the discussion then focused on external issues with such parties such as taxing authorities, INSALUD, and the consulting firm assisting in ABC system design.
There were for example, design-related difficulties resulting from the consulting firm’s incomplete understanding of the clinical processes that produced technical problems mapping activities and documenting critical path protocols. These issues were eventually resolved through collaboration of clinical personnel. Six years later, three phase two interviews with the then ex-CFO, controller and medical director verified original conclusions and assessed success. The ex-CFO, when asked about lack of physician resistance, said: “…
The foundation was created with excellence in mind; professionals were sought who were tired of traditional bureaucratic systems and who were motivated by the idea of making public healthcare a different place. Otherwise, if doctors had complained and blocked the design of ABC at the first sign of change, it would have created huge contradictions. The doctors who came to work at the hospital had to resign their positions; they took this risk because they believed in an innovative project for change. There was nothing in the atmosphere that obstructed management’s initiatives (translation by authors). ” When asked how excellence is created he answered:
Through economic and professional incentives: because we ceased to be a public body, management decided to design a salary package that enables medical excellence and organisational efforts to be combined. Furthermore, doctors and even administrative staff were encouraged to write in specialised journals in attempts to transform the hospital into an example of good management. In fact, when the top 20 ranking of the most important hospitals came out, the foundation put itself forward to demonstrate its commitment to excellence.
We were visited by other hospitals and INSALUD proposed that we take part in the good practice guides… (translation by authors). ” When asked why he was so interested in implementing ABC model, he answered: “…Because I believed in order to direct an organisation in the best possible way, be it private or public, it is necessary to have the best possible information on costs and I don’t understand how someone could possess information systems without having a good costing system.