Management control theory and its application to US medical practice: a critical review of contemporary literature and a call for research.

This paper examines the management control practices and systems that are used by US physicians through a critical review of the existing literature. US physicians play a pivotal role in the national economy both as a service provider and economic agent (White 1999). Healthcare spending currently amounts to 13 percent of US GDP. Collectively, payments to medical practitioners amounted to more than 20 percent of total healthcare spending. More importantly, allopathic and osteopathic physicians, acting as their patients' agents, serve as the leaders of the service delivery team. By state law and tradition, their decisions determine the onset, type and extent of care that their patients receive. Physicians prescribe medication, order diagnostic tests, initiate hospital admissions, and make referrals to other providers for a host of other health care services. As a result, physician behavior has a profound impact on both the overall cost of care as well as the quality of care.

Historically, US physicians practiced medicine as sole practitioners or members of small partnerships. Consequently, there were only weak organizational influences on physicians' practice behavior. Philanthropic hospitals typically offered them free access to the technological infrastructure of modern medicine without imposing an employment relation or any other type of formal contract. Noncompetitive fee for service payments from private insurance companies and government agencies provided physicians with a comfortable income without oversight of insurers or regulators. As a result their behavior was constrained only by personal ethical standards, peer oversight, public sector regulation, and the fear of occasional litigation (Robinson 1999). However, dramatic changes in health care markets and clinical technology in the last ten years have motivated many US physicians to form and join larger organizations (Burns 1999). By 1994 the proportion of physicians in solo practice fell to less than 30 percent (Kletke et al. 1996). Not only are physicians increasingly joining and forming larger group practices but many medical practices are being integrated into larger organizations as well. As physician behavior becomes increasingly influenced by organizational factors, academic researchers who are well grounded in the organizational behavior literature and field research methodologies will find many significant research opportunities in this field of inquiry. Specifically, this paper describes several areas for future research that are not yet fully explored in the contemporary literature.

1. MANAGEMENT CONTROL AND CONTINGENCY THEORY BASED CONTROL RESEARCH: A THEORETICAL FRAMEWORK

Large complex organizations are a central feature of the contemporary US economic landscape. These organizations marshal considerable financial resources to harness modern technology and coordinate the activities of thousands of individuals to deliver goods and services to a global market place. An organization's leadership will use their management control activities and systems to influence other members of the organization to implement their strategies and attain its goals (Anthony & Govindarajan 1998). As such, the management control function provides the link between the organization's enduring goals and long-term strategies and its day-to-day activities. For this reason an organization's management control function must be designed to operate in a systematic fashion if operating activities are to be efficient, effective, and contribute to its enduring objectives (Anthony & Young 1999).

A corporate culture, an integrated system of general and formal controls, and a system of compensation and incentives are the three primary components of an organization's management control system. Goffee & Jones (1996) describe corporate culture as a sense of community within an organization based upon shared values and mutual obligations. This sense of community, mutual obligations, and shared values gives meaning to the activities of an organization's members and provides them with the rules of behavior within their organization. The functioning of corporate culture is reinforced by an integrated set of general and formal controls (Fisher 1995). General controls are based upon the organization's behavioral norms and are most often reflected in various forms of written standard operating procedures. General controls are applied through interpersonal interaction in the workplace and the formal direction of subordinates in their activities. Green & Welsh (1988) describe formal means of control as a system in which standards of performance are determined, measuring systems gauge performance, comparisons are made between the standards and actual performance, and feedback provides information on the variance. Formal controls are supported by and operate through general controls. Consequently, the need for and efficacy of an organization's formal controls is often determined by the strength of its general controls. An organization's compensation and incentive system specifies the appropriate financial rewards for desired individual performance. As such, a compensation and incentive system is the tangible motivational linkage between an individual's work activities and their organizational role.

An effective set of formal controls includes three major mechanisms--operational plans, performance measurement systems, and feedback mechanisms.

i. Operational plans include the firm's annual master budget and the related supporting work plans. They provide the linkage between senior management's strategic plans and the day-to-day organizational activities of each employee. Operational plans are the standard for operating performance.

ii. Performance measurement systems compile and report the results of the collective work activities on a periodic basis. An effective performance measurement system presents both financial results and operating data on a responsibility basis.

iii. Feedback mechanisms report the variances between actual and planned performance. Feedback is generated from the comparison of planned and actual performance. This comparison is communicated to managers and others within the organization periodically through various interim reports.

The establishment of performance standards, the measurement of performance, and the generation and communication of feedback is the essence of the formal control process.

A comprehensive set of general controls includes six components--supervision; recruitment, orientation, and training practices; collegial interaction and peer review; professional training and socialization; organizational structure; and strategic alliances.

i. Supervision reflects the interpersonal interaction that is sanctioned by the official chain of command

ii. Recruitment, orientation, and training practices involve the selection of new organizational participants and the process of teaching them "the way that we do things around here."

iii. Collegial interaction and peer review reflects the informal organizational structure of the corporation.

iv. Professional training and socialization includes the requisite formal education for all organizational participants as well as the ethics and the other behavioral norms that are common to most professionals.

v. Organizational structure represents a system of behavioral norms, roles, authority, and responsibility. Typically, a firm's formal organizational chart is based upon a formal position description for each employee. This position description will outline the authority, responsibilities, and the reporting relationship of each position. Collectively, these position descriptions describe the structure of personal interactions within the firm.

vi. Strategic alliances represent the significant value chain relationships between the organization and its suppliers, distribution channels, and customers.

In this paper, a new explanatory model of an organization's management control package that draws its motivation from a contingency theory perspective is presented. Contingency theory has become one of the dominant conceptual frameworks for control research over the past two decades (e.g. Otley 1980, Dent 1990, and Fisher 1995). Contingency theory rests on the notion that the best way to organize is based on your circumstances and the work that you want to accomplish. This perspective states that efficient and effective organizational structures will vary with contextual factors such as environment and technology. However, while an organization's control structure is contingent on its context, its effectiveness and efficiency is measured by the organization's performance relative to its goals and objectives (Waterhouse & Tiessen 1978). For this reason, the ultimate goal of contingent control research should be to develop and test a comprehensive model that includes multiple control systems, multiple contingent variables, and multiple outcome variables (Fisher 1995).

Fisher (1995) describes a generalized management control model that relies on a contingency theory perspective. The model includes multiple contingent variables, a management control package, and multiple outcome variables. Contingent variables that may be either endogenous or exogenous to the organization provide the primary motivation for its management control package. Exogenous contingent variables include organizational environment and technology. Meyers & Scott (1983) defined the organization's environment as the markets where it buys factor inputs and sells its services as well as the legal and regulatory setting which constrain its operations. Technology describes the hardware, software, and basic know-how through which an organization transforms labor and material resources into the goods and services that it delivers to its customers. While the initial choice of which markets to participate in and what technology to employ are the results of managers' strategic choices, these variables are typically not subject to short-run influence by an individual organization and are best viewed as exogenous to the organization. Endogenous contingent variables, which reflect the strategic choices of the organization's leadership, include its mission and business strategies. The firm's leadership uses its management control package to provide its organizational participants with the planning and coordinating infrastructure to direct its operational processes. These operational processes combine selected technology, environmental resources from selected product and factor markets, and the labor of the organization's personnel to generate goods and services for its targeted marketplace. Organizational outcomes reflect the firm's operational effectiveness relative to its mission, its efficiency in employing scarce resources, and the quality of its products in the hand of its customers. These outcomes are measured by the organization's performance measurement system. Reported performance is compared with planned outcomes to generate feedback. Feedback is communicated by the management control package to organization's personnel and leadership. Monetary rewards are distributed to employees through the organization's compensation and incentives system based upon this feedback. Finally, this feedback generates information that the organization's leadership use for future operational planning, strategy formulation, and mission articulation.

2. DOCTORS, COMPLEX ORGANIZATIONS, AND MANAGEMENT CONTROL

Based upon this framework, a review of the literature suggests four primary themes about the developing management control practices among US physicians:

i. Rapidly advancing medical technology, increasingly competitive health care markets, and a growing role for managed care organizations are three key contingent variables that have dramatically changed US physicians' work and environment (White 1999, and Robinson 1999).

ii. The combined influence of these contingent variables has motivated many US physicians to experiment with a wide variety of complex organizational forms over the past decade (Magel 1999, Evans 1998, and Unland 1995).

iii. The financial performance of these experiments has been very mixed--ranging from high profitability to insolvency (Bodenheimer 2000, MGMA 1999, and Peters 1999).

iv. The relationships between the contingent variables of rapidly advancing technology and rapidly developing markets, the management control systems variables, and medical group organizational outcomes are complex. Research into these relationships remains at an early stage of development (MGMA 1999, and Larkin 1998a).

Advancing Technologies and Changing Market Conditions faced by US Physicians

Rapidly improving clinical technology has been the most prominent exogenous contingent factor that is driving the entire US health care system since the middle of the twentieth century (Robinson 1999). New pharmaceuticals, less invasive imaging equipment, and radical new clinical procedures give physicians a dramatically enhanced ability to prevent, diagnose, and treat illness. Moreover, this explosion of clinical knowledge has lead to the expansion of the medical specialties. However, in order to utilize these new technologies physicians need access to complex and expensive equipment and highly specialized expertise. Thus, the availability of more complex and powerful technology has transformed modern health care from a solitary enterprise to a team enterprise. The delivery of modern clinical technology to patients requires the coordinated actions of many diversely skilled people in hospitals, pharmaceutical firms, equipment manufacturers and a host of other healthcare operating entities. As clinical, diagnostic, and therapeutic alternatives have multiplied, the physician's role has developed into that of the selector of the appropriate technology within this team enterprise. As a consequence the physician's task has moved from that of a direct service provider role to that of the leader of the service delivery team (White 1999).

This rapidly improving technology has made it possible for physicians, hospitals, and other members of the service delivery team to engage in ever more complex clinical interventions into the health status of their patients. Paradoxically, while these clinical interventions have led to improved health outcomes for millions of patients, the developing health care technology has created a drastic escalation of health care costs. This technology driven cost escalation was further fed through a growing demand for these services. The growing awareness among lay people of the potential benefits of new technologies has caused an increase in the demand for exotic and expensive medical technology. Consequently, the cost of health care services has also increased (Robinson 1999).

However, while patients' demands for expensive medical technology served to fuel health care inflation, taxpayers and consumers rebelled against the higher prices and taxes that this inflation engendered (White 1999). This rebellion, in turn, encouraged cost containment pressures from historically passive public and private sector payers. There has been a general trend toward deregulation and a greater reliance on market competition among the nation's health care markets. Public sector payers such as the US government and private payers such as large employers and insurance companies are exercising their market power relative to health care providers. Many of these payers turned to managed care organizations as intermediaries to assist them in controlling their benefits costs. Managed care organizations combine various mechanism that limit the utilization of expensive technology with market-based pricing which shifts financial risk to service providers. The proliferation of managed care organizations such as HMOs and PPOs in the nation's health care markets have increased price-based competition among physicians, hospitals, and other health care providers (Robinson 1999).

Experimentation with More Complex Organizational Forms

The increasingly competitive healthcare markets have rapidly created incentives for the transition of the US medical practice industry from a cottage industry to increasing levels of organizational complexity (Evans 1998). Physicians have been increasingly abandoning sole practitioner and small partnership organizational forms in favor of employment relations in more complex, larger administrative structures over the past decade. Kletke et al. (1996) concluded that between 1983 and 1994, the proportion of patient care physicians practicing as employees rose 24.2 percent to 42.3 percent. During the same period, the proportion of physician care physicians practicing in self-employed group practices fell from 35.3 percent to 28.4 percent while the proportion of patient care physicians practicing as solo physicians fell from 40.5 percent to 29.3 percent. The economic reasons for consolidation in medical care delivery are very straightforward. In a competitive marketplace, access to patients and patient care revenues, developing technologies, and capital often depend on market position. Market position is built on reputation and the ability to deliver valued services at a financially' viable cost structure. Gains in market share translate to more clout in contract negotiation with managed care organizations. Moreover, the formation of larger group practices through merger, acquisition, or contractual arrangements create opportunities for reduction of operating costs by the sharing of administrative talent, the elimination of redundant processes and facilities, and taking advantages of various economies of scale (Magel 1999).

Burns (1997) reported that there are two predominant models for the organizational design of large medical practice groups--the equity model and the independent practice association (IPA) model. Equity model medical groups are profit-motivated corporations that purchase the tangible assets of individual physicians and partnerships. These physicians provide medical services for the corporation as independent contractors or salaried employees. Outside investors and member physicians typically share equal representation on a joint policy board that is responsible for strategic planning and corporate governance. On the other hand physicians remain in their own independent practice in IPA model medical groups. There are no common assets created by the association of group practices. Instead, many IPA model group practices form a management service association to handle the shared administrative activities of the associated group practices. These activities include physician recruitment, administrative support, management services, practice administration, and management information services. The management service association may also function as the group's marketing arm, negotiating contracts with HMOs, insurance companies, and other payers.

Medical groups and IPAs began to increasingly affiliate with physician practice management companies (PPMC) to achieve their objectives of practice consolidation and horizontal integration (Burns 1997). PPMCs may either purchase the tangible assets of the physicians' practice (equity model) or contract with the IPA to provide management and marketing services. In either case, the PPMC acts as an agent for physicians by providing them with managerial expertise and negotiating collectively with managed care organizations. While PPMCs were generally privately financed by venture capitalists until the late eighties, the bull market of the early and mid nineties saw several very successful initial public offerings and secondary offerings that attracted billions of dollars to the physician practice management industry. MedPartners, FPA Medical Management, and UniMed used much of this capital to achieve industry leadership through an aggressive growth strategy based upon rapid expansion and acquisition (Robinson 1998).

Robinson (1999) reported that freestanding hospitals and hospital based integrated service delivery systems were also major investors in physician practices. These investments were often an integral part of their vertical integration strategy. In particular, individual hospitals and regional hospital chains would purchase the dominant primary care practices in its target market. These hospital-based systems would obtain the inpatient service referrals that were generated by these acquired practices. These integrated systems sought to maximize their related service revenues in this way. Physician-Hospital Organizations (PHOs) were the results of these investments. Physicians were typically salaried employees of the individual hospital, integrated delivery system, or PHO.

Mixed Financial Results for Large Medical Groups

The existing literature is rich in descriptive case studies that portray the diversity of financial outcomes that have been achieved by many group practices and integrated delivery systems (Burns et al. 2000, MGMA 1999, Peters 1999). For a variety of reasons, California saw the strongest culmination of these trends toward practice consolidation and the rise of large equity model medical groups and IPAs during the mid-nineties. However, despite the great promise and rosy financial projections of the mid-nineties, the California Medical Association reported that nearly 90 percent of its members were on the brink of insolvency (California Medical Association 1999). While Bodenheimer (2000) believed that this observation was overly pessimistic, he stated that many physician groups are losing money and cannot continue to lose money much longer. Furthermore, he ascribed the financial difficulties of many California medical groups and IPAs to the following four factors:

i. Responding to pressure from California's two largest organizations of employers--the California Public Employee Retirement system and the Pacific Business Group, capitation payments from HMOs to physician groups fell by 20 to 25 percent during the mid nineties.

ii. In 1998, the financial collapse of two large physician practice management companies--MedPartners and FPA Medical Management--left many medical groups and IPAs with over $100 million in defaulted obligations and billions of near worthless equity.

iii. Increasing hospital utilization and the growing costs of prescription medication and specialty care increased medical groups' operating expenses significantly.

iv. Many groups failed to implement sound management and fiscal practices.

Across the nation many PHOs and their integrated delivery systems have also suffered serious financial reverses. On July 21, 1998 Allegheny Health, Education, and Research Foundation, Pennsylvania's largest integrated delivery system, became the nation's largest nonprofit health bankruptcy. Burns et al. (2000) report that the implementation of the Foundation's vertical integration strategy through the purchase acquisition of numerous primary care physician groups and specialty practices was a significant explanatory factor in its failure. In another instance, Boston Regional Medical Center filed for Chapter 11 bankruptcy protection and closed its doors in February 1999. During the nineties Boston Regional had borrowed funds to finance the acquisition of 26 primary care physicians. Weak financial controls, inexperienced practice management, and high physician salaries coupled with flagging physician productivity generated practice losses for the hospital that exceeded $100,000 per physician (Peters 1999). PHOs and integrated delivery systems have generally been unable to align physician incentives with system incentives and effectively manage practice costs in order to achieve the desired synergies and economies of scale and scope from practice acquisitions.

Despite these problems in California and elsewhere, some medical groups have been very successful financially. The Medical Group Management Association (MGMA) (1998) conducted a survey of its membership's 1997 financial performance. Out of 353 multispecialty groups that participated in the survey, the MGMA highlighted the performance of 16 groups. Based upon self-reported data these 16 groups reported median physician compensation in the 75th percentile while their operating costs per non-surgical procedure were below the median. While survey results suggest that group size, regional difference in prevailing labor costs, and managed care market penetration help explain medical group financial performance, group management practices were also significant explanatory factors. Larkin (1998a) reports that four factors appear to be critical:

i. an incentive based compensation formula,

ii. an organizational culture that emphasized productivity,

iii. an ability to effectively monitor individual physician practice patterns, and

iv. an equitable method of allocating and controlling overhead costs.

The 1999 MGMA survey (MGMA 1999) supplemented quantitative survey data with a qualitatively based multiple case study on the relationship between superior medical group financial performance and various aspects of medical group management control system. This study found that financially successful medical groups shared the following five management control variables:

i. A physician compensation formula that is firmly aligned with strategic, operational, and financial goals.

ii. An effective governance structure that is staffed by strong physician and administrative leadership which allows the group's physicians to function together rather than a collection of individual practitioners.

iii. Careful selection, orientation, and training processes of new physicians.

iv. Considerable investment in state of the art budgeting and accounting systems.

v. Patient satisfaction is the fundamental shared value of organizational culture.

In summary, the existing literature posits a variety of explanatory factors that differentiate financially successful and unsuccessful medical practice groups. Contingent environmental variables such as the extent and strength of managed care market penetration are portrayed as factors that provide considerable explanatory power. The structure and competitiveness of the provider market in a given local market may also prove to be a critical factor in explaining a medical group's financial performance. Prudent fiscal management and effective management control practices are common characteristics shown by financially successful medical groups and IPAs. Considerable work, however, remains to be done to specify and adequately describe the management control variables that are associated with financially successful medical groups. In particular, there are many major research opportunities for field research into the relationship between specific control practices and superior financial results among US medical groups.

Elements of Effective Management Control among US Medical Groups

Organizational Culture and Supporting General Controls

The literature suggests that the implementation of organizational strategy through effective management control systems and practices within successful physician practice groups requires the creation, implementation, and maintenance of a strong organizational culture. Physicians' clinical decisions and their resulting practice patterns are the primary drivers of health care costs. Traditional budgetary and administrative controls are not effective mechanisms for managing physicians (Sandrick 1996). Clinical diagnostic and therapeutic processes are too complex and uncertain to enable the programming necessary to support formal controls. Instead, professional control is based upon social and self-control processes. For this reason, physician practice behavior is best controlled through professional socialization and collegial oversight. These informal control mechanisms provide the most effective means of influencing clinical behavior (Abemethy & Stoewinder 1995). However, competing successfully for managed care contracts and gaining market share in competitive regional health care markets usually requires instituting externally motivated cost and utilization controls (Larkin 1998b). Jacobs (1998) found that physicians are very resistant to administrative pressure as a means of influencing their practice patterns. On the contrary, his findings suggest that physicians are much more accepting of externally motivated controls when combined with a peer review and education process.

Many medical groups seek to establish and maintain a strong organizational culture through carefully conceived and implemented recruiting, training, and shared governance programs. Larkin (1998a) reports that many successful medical groups pay careful attention to potential recruits' attitudes toward the established work ethic, clinical philosophy, and group decision-making processes. He also notes that one medical group aggressively managed the orientation and training process of its new physicians by assigning each an experienced physician-proctor. Proctors, who present formal quarterly evaluations to the group's board of directors, are responsible for helping new doctors to meet group performance expectations as well as getting oriented to the group's administrative and governance practices. These governance practices typically include an active board of directors and standing committee structure. These committees, which are staffed by both experienced and new physicians, are responsible for reviewing major capital expenditure requests and monitoring individual physician practice patterns.

The role of organizational structure in the creation and maintenance of a strong organizational culture appears to be significant. Robinson (1999) suggests that IPA model groups and PHOs find it difficult to provide the personalized feedback to physician that is necessary to create an organizational culture based on a single accepted set of clinical norms. IPA model groups are geographically dispersed and each affiliated member practice retains considerable autonomy. These factors lead to the development of diverse clinical practice patterns within most IPAs. PHOs typically find a considerable clash of culture between their physician and hospital administrative staffs. Often meaningful physician involvement in the development of clinical operations and treatment protocols is lacking as a result.

The relationship between the organizational culture of a medical group, its supporting general controls, and organizational outcomes present an interesting area of future inquiry. The existing literature is based exclusively on a small number of case studies. Given the problems of managing the activities of professionals, the role of less formal control mechanisms would appear to be critical. However, research into this field is fraught with many challenges. There are few sources of quantitative empirical data. Some quantitative data is available from the Inter-university Consortium for Political and Social Research's Health and Medical Care Archive (www.icpsr.umich.edu/hmca). The Health and Medical Archive include physicians' surveys of nationally representative samples of US physicians. The physician surveys are is intended to track changes in health service delivery, health system characteristics, and physicians' perceived ability to provide quality care. Moreover, the development of appropriate constructs would seem to be the most daunting. However, the rewards to our understanding of US medical practice would be considerable if these challenges could be hurdled.

Compensation and Incentives in Physician Group Practices

Financial incentives and base compensation are the two principal elements of the physician's total compensation. While organizational culture is the primary mechanism for implementing effective management control, economic compensation is an essential mechanism for influencing physicians' practice behavior (Hillman 1987). Financial incentives for physicians can be grouped within four broad categories (Magnus, 1999): (a) withholdings, (b) bonuses, (c) profit sharing, (d) penalties. The use of financial incentives varies according to whether physicians receive their base compensation by fee for service, capitation, or salary (Kwon 1996). Moreover, financial incentives payments to physicians are also subject to statutory and regulatory overview. For example, many states have outlawed withholding as a legitimate means of physician compensation. While the existing literature is suggestive, the exact nature of the relationship between environmental factors, organizational culture, compensation system design, and organizational performance in medical groups has not yet been adequately examined. Consequently, an empirical examination of this multifaceted relationship using a case study or survey research methodology would make a significant contribution to the literature.

Formal Control Systems--Sophisticated Cost Accounting and Service Utilization Planning and Reporting Systems

Physicians need both sophisticated cost accounting systems and service utilization reporting systems (Marino 1996) to monitor their progress toward their strategic objectives and control their operating activities. These systems generate the empirical data that is the basis of professional control in a managed care influenced environment. Mays & Gordon (1996) note that a cost accounting system that provides reliable, detailed information about actual and prospective service delivery costs forms the core of an integrated performance measurement system. Activity based costing (ABC) systems are being used increasingly by physician groups to better model the relationship between resource consumption, clinical activities, and patient care (West & West 1997, Preston 1998, Siegel et al. 1999, Zeller et al. 1999). The development of clinically accepted treatment protocols and critical pathways within medical groups (Cooper 1996) when combined with an effective case management function (Grandinetti 1998) provide mechanisms for reporting and controlling the utilization of expensive resources, controlling costs, and improving service quality.

Integrated performance measurement systems have been used increasingly by health care organizations. Stewart & Bestor (2000) and Chow et al. (1998) document the formulation and use of these performance measurement systems in hospital-based delivery systems. Curtright et al. (2000) describes the development and implementation of a balanced scorecard reporting package for Mayo Clinic of Minnesota's outpatient operations group. Mayo's leadership developed a listing of critical performance indicators by benchmarking itself against other leading academic medical centers. The conceptual design of the outpatient operations group's performance management and measurement system was developed from this group of performance measures and organized into the performance categories that were suggested by prior research (Kaplan & Norton 1992). Mayo Clinic implemented its conceptual design for the outpatient operations group's performance management and measurement system on to the organization's intranet to facilitate the broadest possible dissemination and utility of system output. To date, the effects of these systems upon medical group performance has not yet been studied. Future case studies could examine the implementation and use of balanced scorecards for other physician practice groups. These inquiries would extend the literature in a fruitful new direction.

Strategic Alliances with Hospitals and Managed Care Organizations

The current practice environment demands close coordination of physician activities with the activities of hospitals and managed care organizations. The creation, implementation, and operation of the control systems that are essential to the planning and coordination of this complex range of activities were often implemented by the vertical integration of related providers and managed care entities through legal ownership (Unland 1995). Many health care organizations have acquired physician group practices to create primary care networks, increase the utilization of their specialty and tertiary care facilities, and broaden their managed care market penetration (Robinson 1999). Ortiz (1997) reports that few are realizing a positive return on their investment after acquisition. In particular, these organizations have struggled unsuccessfully with the problems of developing appropriate incentives for salaried physicians to achieve goal congruence between themselves and the organization. While the poor operating performance and financial failure of many ownership-based integrated service delivery systems (e.g. Burns et al. 2000, and Peters 1999) suggests that these problems many be insurmountable, future research is necessary to ascertain the optimal scale of a centrally controlled conglomerate of physicians;, hospitals, and managed care organizations.

Alternatively, many integrated delivery systems have been formed through contractual rather than ownership mechanisms. Shortell & Zajac (1990) observed that turbulent healthcare markets and rapidly advancing technologies have encouraged a variety of multi-institutional arrangements between medical groups, hospitals, and payers. In their view these arrangements were motivated by the need to rapidly acquire new organizational competencies through the exchange of information. This alliance model of an integrated service delivery system gives service providers the benefits of increased economies of scope, more operational flexibility, and greater market power without the need for large capital investments. These systems typically rely on shared governance structures and integrated utilization management systems to effectively plan and control their joint service delivery activities (Terry 1998). The development, implementation, and operation of interorganizational control systems are an essential managerial infrastructure for alliance model integrated delivery systems. Stewart & Covaleski (2001) found that a Wisconsin-based alliance model integrated delivery system relied on various control systems that spanned organizational boundaries to manage the system's information flows and coordinated its joint clinical activities. However, future research is necessary to better specify the components of an effective interorganizational management control system and determine its relationship to organizational performance.

Summary--Elements of Effective Management Control among US Medical Groups

In summary, this review of the existing literature suggests that the various components of management control in physician practice groups are well documented. Specifically, organizational structure, corporate culture, business strategy, integrated performance measurement systems, and interorganizational planning and control systems are well-known as management control system components in the management accounting literature (Shields & Young 1989). What appears to be lacking, however, is a systematic mapping of these management control system components on to the contingent factors and organizational outcomes of physician practice groups. The absence of this holistic perspective represents a significant research opportunity.

3. DIRECTIONS FOR FUTURE RESEARCH

While the existing literature is suggestive about the linkage between the exogenous contingencies of rapidly advancing clinical technology and developing healthcare markets, endogenous contingencies of business strategy and organizational mission, the management control systems characteristics of large US medical groups, and superior organizational outcomes, there remains a wealth of opportunities for future original research in this field. Fisher (1995) classified the contemporary contingent control literature into four categories of analysis complexity. At the first level of analysis, one contingent factor is correlated with one control mechanism. This approach examines whether the presence of a contingent factor is related to a control mechanism. The second level of contingent control analysis examines the joint effect of a contingent factor and a control mechanism on some aspect of organizational performance. At the third level of analysis the joint linkage between multiple control mechanism, a contingent factor, and some measure of organizational performance is examined. By examining multiple control mechanisms simultaneously, such analyses reflect the notion of that there might be substitution and complementary aspects of control system design. The fourth level of analysis acknowledges the need of managers to simultaneously respond to multiple contingent factors with multiple control mechanisms in order to achieve multiple faceted organizational performances. Consequently, level four analysis examines multiple contingent factors, multiple control mechanisms, and multiple measures of organizational performance. The strength of level four analyses is their external validity. Table 1 summarizes the contingent factors, control mechanisms, and organizational outcome measures that are associated with group practice in the US. Future contingent control studies at each of the four levels of analysis complexity could be conducted with such combinations of contingent factors, control mechanisms, and measures of organizational outcome as excited the intellectual curiosity of individual researchers and could be supported with available appropriate empirical data. Some quantitative data is available from the Inter-university Consortium for Political and Social Research's Health and Medical Care Archive (www.icpsr.umich.edu/hmca). The physician surveys are intended to track changes in health service delivery, health system characteristics, and physicians' perceived ability to provide quality care.

Since data on the control mechanisms that are utilized by medical groups and integrated delivery systems and business strategies are not generally publicly disclosed and may be deemed proprietary by management, access to this data represents the greatest challenge to the conduct of contingent control research studies. To obtain this access researchers must develop and maintain solid working relationships with medical group executives, hospital administrators, and consultants. These relationships can be cultivated through the researcher's active membership in key professional associations (e.g. Healthcare Financial Management Association and Medical Group Management Association) and the conduct of continuing professional education for practitioners. Those researchers who are based in a medical school setting cart also use the databases of the school's alumni relations departments as well. However initiated, good working relationships with the practitioner community can only be maintained where academic researchers are sensitive to issues of confidentiality and the time pressures that are faced by busy executives who running complex businesses in a demanding environment. On the other hand, these same busy administrators and physicians can find significant benefits from their associations with academic researchers. This association can give practitioners a fresh perspective on their operations from an informed observer as well as access to the field's state of the art practices from the researchers' reading and analytical activities.

The use of field research methodologies is essential to the conduct of contingent control research. Specifically, researchers will need to rely upon survey and case study methodologies to gather data from practitioners. Field surveys based upon well-designed survey instruments can enable researchers to develop large-scale databases needed to permit cross-sectional analyses. Where there are well-developed databases on environmental contingencies, survey research can be a useful methodology for Level 1 contingent control analysis. A case study research methodology (Yin 1994) is most appropriate for the conduct of levels 2, 3, and 4 analyses. Bruns & Kaplan (1987) suggest that the researcher must employ various data collection methodologies in order to obtain adequate empirical observations of the many and diverse contextual factors that are critical to the understanding of an existent management control phenomenon in its real world setting. The multiple data collection methods (Otley & Berry 1994) include:

* the extensive review of internally generated reports and other documents;

* the use of semi-structured interviews with managers;

* the unobtrusive observation of management meetings;

* the collection of information on the external context in which the company operates.

4. CONCLUSIONS

This paper presents a descriptive model of management control. According to this model the motivation of individual effort toward the fulfillment of an organizational mission through an established strategy requires the establishment of a strong corporate culture that is supported by an integrated system of formal and general controls as well as an effective compensation and incentive system. Contingency theory perspective was utilized to create an explanatory model of an organization's management control package. This model provides the analytical framework for a review of the existing literature about the management control systems and the group practice of medicine in the United States. These findings suggest that a continuing explosion of clinical technology and progressive developments in the nation's healthcare markets are the driving influences upon US medical practice. US physicians are increasingly practicing their craft in the context of large corporate organizational structures in response to these driving influences. The financial performance of these medical groups has been very mixed--ranging from high profitability to insolvency. While environmental factors are significant explanatory variables for medical group financial performance, empirical observations and theory suggests that a functional management control system also may be predictive of superior financial performance. Effective management control practices within successful physician practice groups ultimately rely on a strong corporate culture that emphasizes cost-effective high quality patient care and a physician compensation system that is aligned with group objectives. A corporate strategy is typically formulated to communicate a medical group's mission and align it to the demands of its environment. This organizational strategy motivates the operating plans that form the foundation of the control process. Many medical groups rely on sophisticated cost accounting and service utilization reporting systems to effectively support their planning and control processes. These systems often extend beyond organizational boundaries in order to facilitate the coordination between doctors, hospitals, and managed care companies that is demanded by health consumers. While the existing literature is suggestive, many opportunities remain for future research. In particular, this paper briefly outlines some of the analytical and methodological issues that researchers need to consider to conduct future studies about US group medical practice using a contingency theory perspective. These studies would extend the management accounting and organizational behavior literature as well as expand our understanding of the nation's rapidly changing health care service delivery system.

TABLE 1

MANAGEMENT CONTROL SYSTEMS AND US PHYSICIANS
CONTINGENT FACTORS, CONTROL MECHANISMS, AND ORGANIZATIONAL OUTCOMES

Environment
  * Provider market structure
  * Managed care market
      penetration
  * New legislation

Technology
  * Medical specialty
  * New clinical procedures
  * New computer hardware
      and software

Mission
  * Profit vs. nonprofit
  * Market driven vs.
      community needs
  * Articulation process

Strategy
  * Formulation process
  * Planning process

Corporate Culture
  * Mission congruence
  * Cohesiveness
  * Consistency
  * Strength

General Controls
  * Formal org structure
  * Personnel practices
  * Peer review practices
  * Alliances with other
      providers & payers

Formal Controls
  * Budget formulation
      processes
  * Cost accounting systems
  * Utilization planning &
      reporting systems
  * Feedback communication
      systems

Compensation System
  * Base compensation
  * Incentive compensation
  * Mix of base and incentive
  * Performance criteria
  * Non-monetary rewards

Financial Performance
  * Cash Flow/Profitability
  * Net Worth
  * Liquidity/Solvency

Product Quality
  * Patient health outcomes
  * Patient satisfaction

Mission Effectiveness
  * Patient volume
  * Service volume

Resource Efficiency
  * Cost per patient
  * Cost per procedure

REFERENCES

Abernethy M. & J. Stoelwinder (1995) "The Role of Professional Control in the Management of Complex Organizations," Accounting, Organizations, and Society Vol. 20:1-17

Anthony, R. & V. Govindarajan (1998) Management Control Systems, Irwin / McGraw-Hill

Anthony, R. & D. Young, (1999) Management Control in Nonprofit Organizations, Irwin / McGraw-Hill

Bodenheimer, T. (2000) "California's Beleaguered Physician Groups--Will They Survive?" The New England Journal of Medicine, Vol. 342, No. 14.

Bruns, W. & R. Kaplan, (1987) "Field Studies in Management Accounting," Accounting and Management: Field Study Perspective. Harvard Business Press.

Burns L. (1997) "Physician Practice Management Companies," Health Care Management Review, Vol. 22:32-46.

Burns L. (1999) "Physician Responses to the Marketplace: Group Practices and Hospital Alliances." Leonard Davis Institute of Health Economics Issue Brief, Vol. 5, No. 8.

Burns L. J. Cacciamani, J. Clement, & W. Aquino, (2000), "The Fall of the House of AHERF: The Allegheny Bankruptcy," Health Affairs, Vol. 19, No. 1, (January/February).

California Medical Association. (1999) "The coming medical group failure epidemic," September 4, 1999, San Francisco.

Chow, C, D. Ganulin, O. Tekinka, K. Haddad, & J. Williamson, (1998) "The balanced scorecard: a potent tool for energizing and focusing healthcare organization management," Journal of Healthcare Management, Vol. 43: 263-80.

Conrad, K. C. Nagle R. Wunar (1996) "Cost Accounting Helps Ensure Group Practice Profitability," Healthcare Financial Management, November 1996.

Cooper, R. (1996) "Look Out, Management Accountants," Management Accounting, May 1996.

Dent, J. (1990) "Strategy, organization, and control: Some possibilities for accounting research." Accounting, Organization, and Society. Vol. 15. pp. 3-25.

Evans, J. (1,998) "Cost Management and Management Control in Health Care Organizations: Research Opportunities," Behavioral Research in Accounting, Vol. 10, Supplement.

Fisher, J. (1995), `Contingency-Based Research on Management Control Systems: Categorization by Level of Complexity', Journal of Accounting Literature, Vol. 14, pp. 24-53.

Goffee, R. & G. Jones (1996) "What Holds the Modern Company Together?" Harvard Business Review. November-December 1996. pp. 133-48.

Grandinetti, D. (1998) "How groups are profiting from case management." Medical Economics, August 10, 1998.

Green, S. & M. Welsh (1988) "Cybernetics and dependence: Refraining the control concept." Academy of Management Review. Vol. 13. pp. 287-301.

Hillman, A. (1987) "Financial Incentives for Physicians in HMOs: Is There a Conflict of Interest." New England Journal of Medicine. Vol. 317. pp. 1743-8.

Inter-university Consortium for Political and Social Research. Health and Medical Care Archive. Available at http://www.icpsr.umich.edu/HMCA/archive.html. Accessed on July 30, 2001.

Jacobs, K. (1998) "Costing health care: a study of the introduction of cost and budget reports in a GP association." Management Accounting Research. Vol. 9. pp. 55-70.

Kaplan, R. & D. Norton (1992) "The Balanced Scorecard--Measures that Drive Performance." Harvard Business Review. (Jan/Feb 1992) pp. 71-9.

Kletke, P., D. Emmons & K. Gillis (1996) "Current Trends in Physicians' Practice Arrangements--From Owners to Employees." Journal of American Medical Association. Vol. 276. No. 7. pp. 555-60.

Kwon, S. (1996) "Structure of Financial Incentive Systems for Providers in Managed Care Plans." Medical Care Research and Review. Vol. 53. pp. 149-61.

Larkin, H. (1998a) "High Income, low costs: How do they do it?," Medical Economics. September 7, 1998.

Larkin, H. (1998b) "How successful practices plan their success." Medical Economics. October 5, 1998.

Magel, J. (1999) "Consolidation in the health care sector." Journal of Health Care Finance. Spring 1999. pp 22-8.

Magnus, S. (1999) "Physicians' Incentives in Five Dimensions: A Conceptual for HMO Managers." Health Care Management Review. Vol. 24. pp. 57-72.

Marino, G. (1996) "Medical Group Practice: Enhancing The Basis For Managed Care Planning," The CPA Journal. January 1996: 74-5.

Mays, J. & G. Gordon (1996) "Developing a Cost Accounting for a Physician Group Practice," Healthcare Financial Management October 1996.

McKinnon, S. & W. Bruns (1993) "Information and Managers: A Field Study," Journal of Management Accounting Research, Vol. 5. Fall 1993. pp. 84-108.

Medical Group Management Association (MGMA) (1998) "Performance and Practices of Successful Medical Groups: 1998 Report Based on 1997 Data."

Medical Group Management Association (MGMA) (1999) "Performance and Practices of Successful Medical Groups: 1999 Report Based on 1998 Data."

Meyers, J. & W. Scott (1983) Organizational Environments: Rituals and Rationality (Sage Publications).

Ortiz J. (1997) "Ensuring the profitability of acquired physician practices." Healthcare Financial Management. January 1997:71-2.

Otley, D. (1980) "The contingency theory of management accounting: achievements and prognosis." Accounting, Organization, and Society. Vol. 5. pp. 413-28.

Otley D. & A. Berry (1994) "Case Study Research in Management Accounting and Control," Management Accounting Research. Vol. 4:45-65.

Peters J. (1999) "Why medical groups fail: There's truth in the numbers," Health Care Strategic Management. August 1999.

Preston S. (1998) "How to figure the cost of every little (and big) thing you do," Medical Economics. May 1998.

Robinson J. (1998) Consolidation of medical groups into physician practice management organizations." Journal of the American Medical Association. January 14, 1998. Vol. 279. No. 2: 144-9.

Robinson, J. (1999) The Corporate Practice of Medicine: Competition and Innovation in Health Care, University of California Press.

Sandrick K. (1996) "How To Succeed With Doctors By Really Trying," Hospitals & Health Networks, February 5, 1996. pp. 23-8.

Shields M. & M. Young (1989) "A Behavioral Model for Implementing Cost Management Systems." Journal of Cost Management. Winter 1989. pp. 17-27.

Shortell S. & E. Zajac (1990) "Health Care Organizations and the Development of the Strategic Management Perspective," Innovation in Health Care Delivery: Insights for Organization Theory, Jossey-Bass Publishers: 144-80.

Siegel G., G. Flourin, A. Ghosh, D. Ghosh, N. Hill, G. Kaciuba, A. Lau, & T. Zeller, (1999) Applying Activity-Based Costing in Healthcare, Institute of Management Accountants.

Stewart, L. & W. Bestor, (2000) "Applying a Balanced Scorecard to Health Care Organizations." The Journal of Corporate Accounting & Finance, March/April 2000:75-81.

Stewart, L. & M. Covaleski (2001) "Managing the uses of horizontal accounting information in a strategic alliance," An unpublished working paper.

Terry K. (1998) "Now hospitals want to be your partner, not your boss." Medical Economics, August 24, 1998.

Waterhouse, J. & P. Tiessen, (1978) "A Contingency Framework for Management Accounting Systems Research." Accounting, Organization, and Society. Vol. 3. pp. 65-76.

West, D. & T. West, (1997) "Applying ABC to Health Care," Management Accounting. February 1997:22-33.

White W. (1999) "Physicians and the changing health care marketplace." The Quarterly Review of Economics and Finance, Vol. 39:439-44.

Yin R. (1994) Case Study Research: Design and Methods. Beverly Hills: Sage Publication.

Zeller T., G. Siegel, G. Kaciuba & A. Lau (1999) "Using Activity-Based Costing to Track Resource Use in Group Practices." Healthcare Financial Management. September 1999. pp. 46-50.

Louis J. Stewart
New York University (USA)

Address for correspondence: Louis J. Stewart, The Robert F. Wagner School of Public Service, New York University, Room 600 Tisch Hall, 40 W. 4th Street, New York, NY 10012, USA, [e-mail].