Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Physician Evolvement in Hospital Decision Making Stephen M. Shorten People in the future will need to learn organization the way their forefathers learned farming. Peter Drucker Medical decision making is an organizational process. Even at the level of individual clinical judgment, a myriad of prior "organiza- tional" decisions have been made that affect what appears to be an autonomous clinical judgment made by a trained professional. For example, a surgeon's choice of a given technique for a particular op- eration has been conditioned by prior decisions, such as the number and types of operating rooms available, types of equipment purchased, the quality and mix of surgical assistants and nursing staff, and the organization of the operating room schedule itself. The surgeon's de- cision may also be influenced by prior decisions made by the hospital's quality assurance committee. In brief, "micro" decisions involving individual clinical judgment and "macro" decisions involving larger organization-wide resource allocation and policy issues are highly in- terrelated. The nature of physician involvement in hospital decision making must be understood within this context. There are five major themes to this paper. The first is that the major hospital decision makers- trustees; administrators; voluntary staff physicians; hospital-compensated physicians; and, increasingly, nurses - will view the decision-making process primarily as a function of their actual degree of involvement in the organization, the degree of involvement that they fee} they should have, and the nature of the issue at stake. Physicians and nurses typically will be most concerned about decisions affecting patient care the ultimate goal. Adminis- 73
74 STEPHEN M. SHORTELL trators and trustees, although also concerned about patient care, will focus most of their energies on resource acquisition and management issues the instrumental goals for facilitating cost-e~ective patient care. The second theme of this paper is that the distinction between "clin- ical" and "administrative" decision making is becoming blurred. New technology, regulation, and competitive forces are giving rise to a number of decisions in which no single professional group has con- trolling interest and participation by all groups is required. ~~ ~ ~ . , ~ ~ . ~ ~ ~ ~ . . . ~ ~ . ~ . ~ ~ ~ the tulro theme IS that physician involvement in Hospital cleclslon making is affected by whether the hospital is voluntary or investor- owned, a freestanding hospital or a member of a multi-unit system.* For example, a hospital that becomes part of an investor-owned chain may find its physicians more actively involved in hospital governing board activities than previously. The fourth theme is that decision making may be moving away from the "dual authority" model of split administrative and clinical decisions to a more "shared authority" model based on increasing collaboration between administrators and physicians. This is partly because of the blurring of decisions noted above but is also due to a number of other factors that will be discussed. The fifth theme is organized around some evidence that suggests that greater physician involvement in hospital-wide decision making is associated with lower costs and higher-quality care. In this context the relationship between cost containment and quality of care also is examined. Where relevant, these themes are specifically considered for their implications regarding for-profit ownership of hospitals. This is par- ticularly true in regard to the types of decisions faced, dual authority and shared authority decision-making models, and specific forms of physician involvement in decision making. At the same time it is important to recognize that the differences in economic orientation between for-profit and nonprofit hospitals may be narrowing, with some interesting implications for hospital behavior. For brevity's sake, this paper will not describe the historical evo- lution of physician involvement in hospital decision making. The main concern here is with current developments and implications for the immediate future. The paper will not serve as an exhaustive review * It is recognized that there are also important differences between teaching and non- teaching hospitals, but the primary focus of this paper is on ownership and system versus nonsystem differences.
Physician Involvement in Hospital Decision Making 75 of the literature. Rather, it will highlight some of the more significant studies and major findings. This paper is divided into five major sections. First, the major kinds of decisions made by hospitals are described. Second, those individuals primarily involved in hospital decisions are noted, and two models of decision making are examined. Third, the nature of the involvement is highlighted. Fourth, evidence bearing on the relationship between physician/hospital decision making and the cost and quality of care is summarized. Finally, a number of future issues influencing phy- sician/hospital decision making are discussed. A Typology of Hospital Decision Making A simple typology of decision making is shown in Table I, which suggests that decision-making strategies used by hospitals depend on (1) the degree of agreement or certainty among the key parties as to their preferences for specific outcomes and (2) the degree of confidence or certainty in the cause-effect relationships involved i.e., whether the decision will actually produce the desired results. For example, in the first cell where all parties agree on preferences about outcomes and the certainty of cause-effect relationships is relatively straight- forward, decisions can be made on a fairly routine "computational" basis. Decisions involving the amount of standard supplies to keep in inventory in the hospital's central supply department serves as an example of a computational decision strategy. For the most part, phy- sicians do not get involved in such decisions, which are primarily made by hospital support department heads and increasingly are being com- puterized or otherwise automated. TABLE 1 A Typology of Physician/Hospital Decisions Certainty of Preferences about Outcomes Certain Uncertain Certainty Of Certain Cause-Effect Relationships Uncertain 2 Computational Compromise 4 Judgmental Inspirational SOURCE: Adapted from Thompson, J. D. Organizations in Action. New York: McGraw- Hill, 1967.
76 STEPHEN M. SHORTELL The second cell involves situations in which there is certainty about cause-e~ect relationships but in which the parties involved disagree about desired outcomes. These decisions are labeled "compromise" decisions. For example, a hospital may be faced with the decision of whether to purchase a CT scanner or expand the laboratory depart- ment's capabilities. In either case the cause-effect relationships are known (the decision will most likely result in improved patient care), but the parties disagree as to the areas of hospital operation (radiology or lab) in which they wish to see the improvement. It is important to note that general economic forces, external regulation, and competi- tive pressures are increasing the number of compromise decisions that hospitals must make. These are situations where the efficacies of decisions are known but where there are insufficient funds to imple- ment all of them. Compromise decisions are a major area of physician involvement in hospital decision making, as each specialty group strives to maintain or expand its scope of responsibility. Thus, for the most part, physicians become involved in these decisions in order to protect their interests. The third cell involves situations in which preferences about out- comes are known and agreed upon but where there is uncertainty about the cause-effect relationships. These situations constitute "judg- mental" decisions. For example, the decision to improve a hospital's financial position may be agreed upon by all parties, but uncertainty may exist about the best strategy or combination of strategies to ac- complish this. Physicians are becoming increasingly involved in judg- mental decisions but for a different reason than their involvement in compromise decisions. In compromise decisions they become involved primarily to protect their interests, but in judgmental decisions they become involved because their expertise as physicians is needed. For example, many administrators have relied heavily on physician advice in justifying major capital purchases or expansion projects to health systems agencies. The fourth cell describes situations in which uncertainty exists about both preferences for outcomes and cause-effect relationships. These decisions are labeled "inspirational." For example, a rural hospital with low occupancy may be pondering whether to develop an ambu- latory care program or affiliate with an urban medical center. In terms of cause and erect it is not clear that either option will increase admissions. Furthermore, with either option the parties involved may disagree about likely outcomes. To reduce the uncertainty surround- ing such decisions, hospitals are increasingly adopting methods of formal environmental assessment and long-run strategic planning.
Physician Involvement in Hospital Decision Making 77
78 STEPHEN M. SHORTELL issues are raised by such a decision. They include cost considerations, likely impact on turnover, absenteeism, job satisfaction, ability to recruit nurses, relationship with other departments, continuity of care, and quality of care. These issues are interrelated and difficult to sep- arate, even though each group will approach the question from its particular area of concern nurses from the perspective of job satis- faction and quality of patient care, physicians from their perspective of quality of care and how the change will affect nurse/physician re- lationships, and administrators from the perspective of costs and ad- equacy of staffing in addition to concerns about quality of care. Other examples could be used to illustrate the blurring of admin- istrative and clinical decision making. Some additional examples are provided in Table 2, categorized according to the computational, com- promise, judgmental, and inspirational frameworks. Although it provides some insight, the typology described above is oversimplified. At least two other sources of complexity appear to be important in understanding the nature of the hospital decision-mak- TABLE 2 Examples of Physician/Hospital Decisions 1. Computational Maintaining inventory levels. Hiring ancillary staff. Hiring additional nurses to increase coverage. 2. Compromise Suspending privileges of a popular physician. Admitting a new physician in a specialty that is already well supplied. Purchasing a CT scanner or a major piece of lab equipment. Developing a new compensation arrangement for the director of the laboratory. 3. Judgmental Expanding physician continuing education efforts to improve quality of care. Hiring a full-time director of medical education to improve quality of care. Changing to computerized billing system. Establishing a long-range planning department. 4. Inspirational Adding a new clinical service. Developing a hospital-sponsored group practice. Affiliating with a medical school. Merging with another hospital. NOTE: The examples are not necessarily mutally exclusive. They will obviously vary de- pending on people's perceptions of the cause-effect relationships and preferences for out- comes. They also depend on the stage of the decision-making process. For example, as more information becomes available, some inspirational examples may become compromise or judgmental decisions.
Physician Involvement in Hospital Decision Making 79
80 STEPHEN M. SHORTELL serve. They also tend to have high turnover in upper administrative ranks, and, therefore, many lack strong continuous managerial di- rection. In brief, there are likely to be more debates about the pref- erences for different kinds of outcomes in voluntary community hospitals than in investor-owned hospitals. As such, in voluntary hospitals the decision-making process may be somewhat more complex and inde- terminate than in investor-owned hospitals. In general, hospitals belonging to a multi-unit system seem likely to be more involved in computational and judgmental decisions than are freestanding individual hospitals. This is due in part to the influ- ence of a corporate headquarters office with greater managerial staff expertise, which can reduce the uncertainty of cause-effect relation- ships surrounding given decisions. Also, the presence of an overall corporate mission and value system can help orient individual hos- pitals toward achievement of more common objectives, resulting in less disagreement regarding desired outcomes. In contrast, individual hospitals, often lacking such expertise and direction, may become involved in more compromise and inspirational decisions. These sug- gested differences, however, also depend on other factors, including the maturity of the multi-unit system and its emphasis on innovation. For example, a multi-unit system in the early years of existence may face a greater number of compromise decisions as it attempts to gain agreement among member hospitals regarding overall directions. Fur- thermore, a system at the cutting edge is experimenting with new programs, services, and organizational arrangements and may thus face a high number of inspirational decisions. Decision-making strat- egies will also be influenced by the degree of centralization that exists between the corporate headquarters office and individual member hospitals. The suggested differences by ownership and system status are summarized in Table 3. TABLE 3 Most Prevalent Types of Decision-Making Strategies, by Type of Hospital Decision-Making Voluntary Investor-Owned Single Strategy Hospital Hospital Hospital Multi-Unit System Hospital Computational Compromise Judgmental Inspirational + + + + + + + +
Physician Involvement in Hospital Decision Making Convergence versus Divergence of Interests 81 Determining where physician and hospital interests overlap and where they diverge is difficult because the relationship is subject to complex and rapidly changing forces. In general, hospital and physician in- terests coincide most often in areas involving expansion of hospital programs and services that are complementary rather than substi- tutable with physician services. Examples include increasing the number of beds; acquiring sophisticated technology, such as nuclear magnetic resonance scanners; and adding selected support services, such as occupational therapy, physical therapy, and social work, which are uneconomical for most physicians to incorporate into their private practices. Interests also coincide when physicians and hospitals can assist each other in responding to external regulation or changes in payment. A noteworthy example is the development of quality assur- ance committees in response to the establishment of Professional Stan- dards Review Organizations (PSROs). Conversely, hospital/physician interests diverge when physicians perceive the hospital to be in direct competition or when the hospital believes physicians are acting counter to the Tong-run objective of the hospital. Thus, as previously noted, hospitals' efforts to expand their ambulatory care activities may meet medical staff opposition because of fear of direct competition for patients and hospital beds.2 Opposition may also be based on philosophical objections to the "corporate practice of medicine." Regulations or changes in payment also can create con- flict rather than representing the "common enemy" against which hospitals and physicians can unite. For example, limiting hospital revenues by reimbursing on a case-mix basis may create conflict be- tween a hospital's economic interests and the physicians' economic and professional interests. It is important to note that there is frequently more disagreement among physicians than between physicians and hospitals. Physicians are not a unitary group and seldom act in concert on a given issue. Differences exist by specialty, years in practice, and geographical lo- cation, in addition to individual differences in personality and phi- losophy. For example, surgeons and other specialists are typically strong supporters of hospital ambulatory care programs because they usually benefit directly from increased referrals. Primary care phy- sicians, in contrast, are likely to be the most vocal critics because of perceived competition. Even here, differences exist depending on the patients to be served. For example, if the primary purpose of an ex- panded ambulatory care program is to serve more Medicaid patients,
82 STEPHEN M. SHORTELL private practice physicians may be supportive because of their desire to limit the number of Medicaid patients in their practice. The diversity among physicians is important to recognize in con- sidering decisions involving almost any new program, service, tech- nology, or reorganization. In brief, each physician and specialty group will be concerned if the decision is likely to benefit other groups or interests more than their own. As Harris3 notes, in the extreme, this results in . . . each clinical service of the medical staff . . . striving to maintain and expand the magnitude of its own defensive position.... Each service gets its own intensive care unit. Each intensive care unit gets its own laboratory. The idea behind all of these arrangements is to insure the exclusive availability of a set of inputs to a small group of demanders. In that way no one is going to get bumped. Although this often creates problems for hospital administrators and trustees, it also is to their advantage in that it facilitates "divide and conquer" strategies and adores administrators some flexibility in playing off the interests of one group of physicians against another. How these relationships are influenced by competition, regulation, and related factors is described in a subsequent section. The Decision Makers The most important point to understand about decision making in hospitals is that there is no single decision maker. Rather, decision making is a complex and often diffuse process involving multiple co- alitions of key people, including physicians; administrators; trustees; and, increasingly, nurses. These coalitions exert different degrees of influence depending primarily on the topic. Typically, physicians exert the most influence over clinical matters, such as determining staff privileges, establishing practice protocols, reviewing quality of care, and determining patient admission and discharge. Executive-leve] administrators exert the most influence over hospital policy and plan- ning activities particularly as they relate to the organization's exter- nal environment. Middle-level executives and department heads typically exert the most influence over matters related to daily staff- ing, budgeting, and procurement of supplies. The influence of trustees is primarily felt in the areas of Tong-run strategic planning and ar- ticulating the overall mission and direction of the hospital. Nurses are striving to become more involved in all of these areas. From this general description, it is possible to highlight two general "models" of decision making in hospitals: the dual authority mode} and the shared authority model.
Physician Involvement in Hospital Decision Making 83 The Dual Authority Mode! The dual authority mode} is best developed by Pauly and Redisch4 and Harris5 and was first described by Smith.6 In the Pauly/Redisch version the hospital is seen as a physicians' cooperative in which physicians' decisions largely determine the nature of hospital opera- tions. The administration largely exists to provide the equipment, supplies, and facilities for physician use. Although two distinct lines of authority (administrative and clinical) are recognized, administra- tors seldom oppose physicians because the hospital's success and the administrator's own job security are closely tied to satisfying the de- mands of the physician staff. In the Harris version the administrative and medical split is con- ceptualized as two different "firms." The medical staff constitutes a "demand division" and the administration a "supply division." Each division has its own managers, decision-making strategies, operating rules, and policies. Third-party payers recognize this separation in the form of separate payment policies for ambulatory care versus inpatient care. In brief, although hospitals and physicians are in fact involved in a joint production process, they are largely organized as separate entities; therein lies much of the difficulty in hospital deci- sion making as it pertains to the allocation of scarce resources. Until recently, the "expert" power of the physician as legitimated by the state has dominated the decision-making process over the "legitimate" power (i.e., formal position authority) of the administration. Further- more, physicians control both their own and the hospital's inputs. As Harris notes: Doctors are in a position to deem all sorts of demands as necessary for their patients. This is not the same thing as saying doctors order useless tests to satisfy some ulterior motives. Additional demands for inputs above the hypothetical scientific minimum are going to be regarded by doctors as improvements in qual- ity.7 The issues suggested by the dual authority mode] of decision making are more complex for voluntary hospitals than for investor-owned hospitals. If one assumes that the goals of investor-owned hospitals are somewhat more homogeneous and targeted than are the goals of voluntary hospitals, the interests of physicians and the hospital may be more closely aligned. In contrast, voluntary hospitals may pursue a variety of community objectives, not all of which may contribute to financial viability and which may in fact detract from or even compete with physician interests. But as cost containment pressures continue, voluntary and investor-owned hospitals are becoming more alike in
84 STEPHEN M. SHORTELL their orientation to financial viability. In single hospital communities, this may lead to further hospital competition with the medical staff. In multiple hospital communities where physicians have alternatives for admitting patients, hospitals are more likely to pursue initiatives that will complement rather than compete with staff interests. The Shared Authority Model The shared authority mode] is the product of recent developments. Briefly stated, it involves more conjoint or shared decision-making power between administrators and physicians and increased integra- tion of clinical and administrative information. This mode] has emerged as a result of legal, economic, and societal forces. From a legal per- spective, the Darling decisions in 1965 established the hospital's ul- timate legal responsibility for the quality of care. This responsibility can be delegated to the medical staff, but the final accountability resides with the hospital and its governing board. Although subse- quent cases have modified and refined this landmark ruling, it has resulted in a fundamental change in the behavior of hospital admin- istrators and trustees toward physicians in regard to establishing institutional accountability for physician behavior. It has provided administrators and trustees with a degree of legal clout. The economic forces are twofold. First is the general concern about the inflationary economy, which has made it more costly for many organizations to function. Second, and more specifically, has been the concern over the continued above-average increases in the cost of medical care and hospital services in particular. This has led to a number of regulatory cost containment initiatives, including health systems agencies (HSAs); the PSROs; and, in a number of states, hospital rate review commissions. In addition, some states have ex- perimented with hospital reimbursement based on comparable diag- nostic case mix. It is beyond the purview of this paper to address the efficacy of these approaches to cost containment, but there is no ques- tion that hospitals have been operating in a environment of increas- ingly constrained resources, particularly over the past five years. From the perspective of hospital decision making, the most important con- sequence has been that administrators have gained power and influ- ence in their negotiations with physicians to contain costs. Ad- ministrators may not agree with the regulations, but they can use them as an "external scapegoat" for promoting more efficient decision making by physicians in the use of hospital resources. In brief, hos- pitals and hospital administrators have been provided with greater economic clout.
86 STEPHEN M. SHORTELL and physicians. It is no longer in the physician's economic interest to stand aloof from the process, and the hospital stands to gain by bring- ing resources more under the control of the organization although at the "price" of greater physician involvement in hospital-wide de- · . . clslon ma sing. Strain Among Decision Makers and Between the Two Models There are inherent strains between the needs of organizations and the needs of the professionals associated with them. These strains particularly affect hospital/physician relationships and the two models of decision making described above. Some of the more important strains are summarized in Table 4 and briefly noted below. First, organizations have a high need for predictability in order to achieve their goals. In contrast, professionals have a high need for freedom to operate in the face of uncertainty. This is most widely recognized in the "exceptional cases" syndrome whereby health care professionals, physicians in particular, can assert that a given case is an "emergency" and thereby set aside the usual rules and regulations. The organization also has a high need for goal commitment, par- ticularly in regard to survival and effectiveness. Professionals, on the other hand, have a high need for professional goal commitment, which TABLE 4 Inherent Strains Between Organizational and Professional Needs ORGANIZATIONAL NEEDS 1. Predictability. 2. Commitment to organizational goals- maintenance of the organization. 3. Coordination/integration across tasks, services, and departments. 4. Control and feedback to ensure public accountability. 5. Diffuse division of labor (specialization) to accomplish organizational-relevant tasks. PROFESSIONALS' NEEDS 1. Freedom to operate in the face of uncer- tainty; the "exceptional case" syndrome. 2. Commitment to the goals of one's profession and peers; more narrowly fo- cused than the organization's goals. 3. Freedom to function within specialized interest areas; "loose" coordination that does not interfere with one's profes- sional work. 4. Emphasis on individual accountability to patients and professional peers. 5. Fairly rigid specialization to accomplish individual-specific tasks. SOURCE: Shortell, S. M. "Theory Z.: Implications and Relevance for Health Care Manage- ment." Health Care Management Review 7 (Fall 1982), p. 11.
Physician Involvement in Hospital Decision Making 87 is less widely focused than organizational goals and tends to be cen- tered more on individual patient treatment. Organizations also have a high need for coordination and integra- tion across tasks, services, and departments. In contrast, professionals have a high need for freedom to function within specialized interests. As Weisbord notes: "In medicine, professionals believe in their bones that procedures and organizational needs for . . . survival will be in- imical to theirs." Organizations have a high need for control and feedback, particu- larly concerning their public accountability. In contrast, professionals have a high need for individual accountability to patients and to professional peers. Finally, organizations have a relatively high need for specialization to accomplish tasks. Professionals also have a high need for special- ization but not necessarily in a manner compatible with the needs of the organization. Overall, the organization's needs are largely macro in nature, re- flecting the overall goals of the organization and the relationship of the organization to its larger environment. At the same time the organization's needs are primarily local in the sense that the com- mitment is to the organization, with professionals viewed as a vehicle for achievement of the organization's goals. In contrast, health care professionals are largely concerned with micro issues centered on in- dividual patient care but with a cosmopolitan orientation character- ized by a commitment to professional growth in the development of one's speciality. In brief, the organization is seen as the vehicle for the achievement of professional goals. These are some of the fundamental differences that must be taken into account and managed whether one adopts a dual authority or shared authority decision-making model. In general, the above dif- ferences tend to reinforce the dual authority mode] and make it more difficult to bring about a shared model. Types of Physician Decision-Making Involvement Decision-making involvement takes two primary forms: formal and informal. The principal modes of formal physician involvement in hospital decision making are participation in meetings of the board of trustees and in the committee structure of the board, the admin- istration, and the medical stab itself and in hospital/physician com- pensation arrangements whereby the physician is economically tied to the hospital's welfare.
88 STEPHEN M. SHORTELL
Physician Involvement in Hospital Decision Making 89 investor-owned hospitals and among voluntary hospitals than be- tween the two types. The Hospital Corporation of America (HCA), for example, operates on a strongly decentralized basis in which local hospital governing boards maintain some degree of discretion and influence, although accountability remains centralized. In contrast, Humana, Inc., operates on a highly centralized basis with more op- erating decisions made at the corporate office level and less autonomy provided to individual hospitals. The relative influence of hospital governing boards in investor-owned hospitals, freestanding voluntary hospitals, and voluntary multi-unit systems is an important issue for further investigation. For example, it is commonly believed that the influence of individual hospital boards is diminished in multi-unit systems, whether investor-owned or voluntary. But no systematic in- formation is available to indicate the extent to which this is true or in what specific areas or types of decision making such differences may exist. Overall, hospitals have approximately 4 medical staff committees per 100 beds. Two key committees are the joint conference commit- tee, made up of trustee, hospital administration, and medical staff leaders, and the medical staff executive committee itself. Although all accreditated hospitals are required to have these two committees, their actual influence and practice varies widely. Other common com- mittees are medical audit, utilization review, credentials, and the infection committee. The average number of committee members is 6, and the committees meet an average of Il times per year. Approx- imately one-third ofthe medical staff committees have a nonphysician, usually an administrator or a nurse, with voting representation. For 58 percent of the committees the members are appointed either by the medical staff president, the hospital administrator, or by both acting jointly, rather than being elected by the staff. In regard to physician compensation arrangements, nationally 25 percent of active staff physicians have some type of hospital financial arrangement, either part time or full time.~9 Twenty-eight percent of department chiefs are on contract. Of all active staff with a contract, 23 percent are salaried. Arrangements whereby physicians are com- pensated by hospitals are more often found in teaching hospitals than in nonteaching hospitals.20 They also are more common in for-profit hospitals than in voluntary hospitals.2i Systematic longitudinal data are not available, but a general read- ing of the literature and conversations with hospital administrators and medical staffs suggest that physician involvement in governing board activities, participation in committees, and hospital-based com- pensation arrangements is growing. For example, over the past five
go STEPHEN M. SHORTELL years a number of hospitals have added cost containment committees, medical equipment purchase committees, and strategic long-range planning committees, all with physician participation. Thus, there appears to be a growing trend toward the shared authority model of decision making described earlier or at least deliberate attempts to blur the clear demarcation suggested by the dual authority model. Some of the effects of these changes on the cost and quality of patient care are examined below. Hospital/Physician Decision Making and the Cost and Quality of Care The issue of hospital/physician decision making is important primar- ily as it affects the delivery of patient care services. The relevant question is whether certain patterns of decision making are associated with improvements in the cost-effectiveness of the care delivered. Present research does not provide a clear-cut answer in terms of cause and effect, but the majority of the existing evidence suggests consistent associations between greater physician involvement in hospital de- cision making and lower costs. Existing research also suggests con- sistent associations between greater physician participation and higher quality of care. There is little evidence that costs can be contained only at the expense of lowering the quality of care. If anything, the evidence suggests that efforts to contain costs can be associated with improvements in the quality of care. It is important to note that the research on the relationship between physician involvement in hospital decision making and the cost and quality of care has almost all been conducted in voluntary hospitals. Thus, little is known about this relationship in for-profit hospitals. This is another area for future research. Evidence Regarding Costs A number of studies have examined the relationship between various aspects of physician involvement in hospital decision making and cost of care.22 These studies generally indicate that the more aware phy- sicians are of the organization's performance and the greater the num- ber of scheduled meetings between such key clinical and patient care departments as radiology, laboratory, and nursing service, the lower the costs will be in specific medical support departments. Some evi- dence also suggests that for-profit hospitals have a higher ratio of nurses and physicians to support personnel, which in turn is more
Physician Involvement in Hospital Decision Making strongly associated with occupancy rates in for-profit hospitals than in not-for-prof~t hospitals.23 This may be due to the stronger economic orientation of the for-profit hospital, although in the current climate of economic constraint, not-for-profit hospitals also have a high need for surplus revenues. Thus, differences in economic orientation of for- profit versus not-for-profit hospitals may be narrowing. The percentage of hospital-based physicians on contract also has been found to be positively associated with lower costs per admission, and physician presence on the executive committee of the governing board also is associated with lower costs per admission.24 91 Evidence Regarding Quality A number of investigators have examined how the relationship be- tween physicians and hospitals may affect the quality of patient care.25 In general, these studies suggest that greater physician participation in hospital decision making is positively associated with higher qual- ity of care, as measured by such indicators as severity-adjusted death rates and postsurgical complication rates. There is also evidence that the greater the hospital administrators' ability to influence decisions within their domain, the higher the quality of care.26 Others have found positive relationships between quality of care and more highly structured medical staffs, as measured by appointment procedures, number of control committees, and percentage of physicians on con- tract.27 Morlock et al. also found evidence of a strong relationship between hospital trustee involvement in hospital decision making and the quality of care.28 In their study, hospitals with influential trustees were much more likely to have medical stab committees that met frequently and were more likely to produce frequent internal moni- toring reports on quality of care statistics. Evidence Regarding Possible Trade-Offs Between Cost arid Quality A major issue in physician/hospital decision making is the extent to which control of costs or improved efficiency can be achieved only at the expense of the quality of care. Most of the studies to date, however, suggest that efforts at containing costs are positively associated with quality. For~example, a study of Chicago-area hospitals found that the more efficient hospitals, as measured by lower costs and lower man- hours per standardized unit of output, also provided higher-quality care, as evaluated by outside experts and as indicated by accreditation and severity-adjusted death rates.29 A study of hospitals in Massa-
92 STEPHEN M. SHORTELL chusetts revealed that higher cost per case was associated with higher medical/surgical death rates, even when differences in case mix were taken into account.30 Other studies have generally found similar re- sults.3i However, Flood et al. found that hospitals that provide a greater number of certain specific medical services that increase cost also had better than expected patient care outcomes.32 In this study the rela- tionship between overall cost and measures of quality of care was not examined. It is important to note that the above results are preliminary and suggestive at best, and they must be viewed with caution. Nonetheless, existing evidence offers little support for the argument or expectation that efficiency or cost containment goals are inherently incompatible with effectiveness or quality of care. It may be that greater physician involvement in hospital-wide administrative decision making facili- tates cost containment decisions that protect or even enhance the quality of care provided. For example, changes to improve the turn- around time for laboratory tests not only improve hospital efficiency but may also improve quality of care by expediting the physician's diagnostic and treatment plans for the patient. Clearly, this is a major area for future research and public policy development. The effects of physician involvement in hospital-wide decision making on the overall use of hospital services is another important area for further inves- tigation. Future Issues It should be evident from the above discussion that physician involve- ment in hospital decision making is in flux. As indicated, this is pri- marily due to changes in the external environment of health care delivery, which is causing physicians and hospitals to view themselves and each other in a different light. As a clue to the future it is useful to consider the changing context of both clinical decision making and institutional decision making. The possible demise of the traditional voluntary medical staff organization can be foreseen. It then becomes possible to consider the factors that either promote or constrain the movement toward more shared, collaborative decision-making models. The Changing Context of Clinical Decision Making Pellegrino has commented that: The process of making clinical decisions is the balance wheel of hospital operation. It is central to all the patient-oriented functions of the hospital, and it has remote effects on all major elements of hospital organization the patient, the health
Physician Involvement in Hospital Decision Making 93 care professional, administrators, trustees, and the community. It is also the process least accessible to organizational control, the most in need of freedom, and yet the most potent of hospital processes for good and evil. The clinical decision is the most zealously guarded of the physician's prerogatives and at the same time the most in need of some kind of surveillance for individual and public good. It is, moreover, the most difficult process to evaluate in a definitive way.33 Five factors are redefining the context of clinical decision making: (~) the realization that resources are scarce- a "logic of scarcity," (2) the continued impact of new technology, (3) changes in the mix of diseases being seen, (4) the increased institutionalization of all aspects of medical care, and (5) the ejects of the consumer movement. The concern over the cost of health care has resulted in a logic of scarcity that is beginning to permeate medical practice. There exists a subtle and still-developing change from the norm of"doing every- thing possible for the patient at all costs" to one of"doing only those things that might reasonably yield positive outcomes" and choosing the most cost-effective ways of doing those procedures. In the extreme this is resulting in the use of cost-benefit assessments in making decisions to treat some patients and not others. This is a profound and very important change. Never before has such a logic been a part of the "micro-level" of the health care system, the level of individual clinical decision making. Continued advances in technology require continual rethinking of diagnostic and treatment protocols and clinical decision-making rules. This increases the rate of change and uncertainty, which in turn leads to greater specialization of function and greater competition among specialties. One example is the recent dispute among pathologists, radiologists, and internists over developments in nuclear medicine. Specifically, pathologists claim they have the facilities, space, and personnel to handle large-scale procedures; radiologists maintain they have the techniques; and internists, of course, note that they have the patients. A partial solution appears to have been worked out in the development of a conjoint board that is sponsored by all three spe- cialties and that allows access to certification in nuclear medicine from each of them. There have also been appreciable changes in the mix of diseases being seen specifically in the chronic, complex conditions associated with aging. One implication of this change is that teams of different kinds of specialists and providers are needed to provide effective care. This further complicates the clinical decision-making process and raises a number of issues involving who should be the team leader and who should assume various roles and responsibilities. As previously noted, medical care is increasingly an organizational
94 STEPHEN M. SHORTELL process, subject to organizational forms of social control. The Darling decision, which held hospitals and their governing boards ultimately responsible for quality of care, helped give rise to PSROs and related institutionalized forms of review. Finally, there is continued interest by the public in having more control over their own lives, and, as previously noted, this has affected the health care professions. The public has a desire to know more and to be given more choices, including the choice not to seek or comply with medical advice. Manifestations are emerging both in collective bodies such as health planning agencies and at the level of the indi- vidual provider-patient relationship. As such, they have affected clin- ical decision making, if only as a sensitizing factor that further complicates the decision-making process. The effect of these five factors has been to transform the context in which clinical decisions are made. In brief, such decisions are no longer within the exclusive domain of the medical profession; the boundaries have become more permeable, allowing participation by other provid- ers, health care organizations, regulatory groups, consumers, and oth- ers. The issue is whether the continued prevalence of dual authority decision-making structures or the continuing emergence of shared decision-making authority structures provides a better forum for deal- ing with the increased complexity and diffuseness of clinical decision making. The Changing Context of Institutional Decision Making Not only are hospitals under increased public scrutiny because of the continuing rise in costs, but it also seems likely that hospitals will remain under such scrutiny permanently. This is not only because of the continued concern regarding the cost-effectiveness of patient care but also because hospitals, individually and collectively, have taken on more characteristics of industrial enterprises central to the Amer- ican economy. Many individual hospitals are joining multi-unit sys- tems to gain greater economic and political clout. Approximately 26 percent of all hospitals belong to a multi-unit system now, and esti- mates suggest that close to 80 percent may belong to such systems by i990.34 Even among individual hospitals there has been growth in professional managerial staff specialists, marketing specialists, long- range planning departments, and health services research units. Regulation of capital and operating expenses plus an inflationary economy have forced hospitals to compete more with each other for patients, physicians, and nurses. In many areas of the country, vol-
Physician Involvement in Hospital Decision Making 95 untary hospitals are competing directly with investor-owned hospi- tals, and teaching hospitals are competing with nonteaching hospitals. The result is that voluntary and investor-owned hospitals are becom- ing more alike, ironically as a result of trying to differentiate their services in an attempt to find new markets for growth. Thus, some voluntary hospitals are entering into management contract relation- ships with other voluntary hospitals and are forming systems that are similar to those of investor-owned hospitals, and some investor-owned hospitals are beginning to offer outreach and satellite services similar to those offered by voluntary hospitals. Teaching hospitals are becom- ing more like their community hospital counterparts in offering more general primary care services and community outreach services, and community hospitals are striving to expand their markets by adding the more sophisticated technology found in teaching hospitals. American hospitals are no longer a cottage industry; they are part of an industry that is becoming more highly concentrated, more com- petitive, and more heavily interdependent with other organizations. It is also an industry that is extremely vulnerable to economic, reg- ulatory, and technological changes. As such, decision making, partic- ularly at the upper policymaking levels of the organization, has become a very complex and difficult process. The number of inspirational decisions relative to computational decisions has increased. There is an increased need to turn more of these inspirational decisions into judgmental or compromise decisions. There also is a greater need for clinical participation in the admin- istrative decision-making process and consideration of more admin- istrative and economic matters in the clinical decision-making process. The following question may be raised: Is the current relationship of physicians to hospitals, in the form of the voluntary medical staff, able to meet the challenge of the new decision-making environment? In brief, is the voluntary medical staff organization structure rapidly becoming an anachronism? The Demise of the Voluntary Medical Staff Fundamental changes in the structure of medical staff organization may be taking place already. A growing number of physicians are affiliating with hospitals as a cost-effective way of starting practices, a growing number of speciality-trained physicians are contracting with one or more hospitals to deliver secondary and tertiary care services, and a growing number of hospital medical staffs are entering into HMO arrangements of various forms.35 As the predicted physician
96 STEPHEN M. SHORTELL surplus materializes over the next decade, competition among phy- sicians will grow, and many will look to the above kinds of arrange- ments to gain competitive advantages. But what effect wait these trends have on physician/hospital decision-making relationships? Although it is safe to say that the dual authority mode} will continue to prevail in most settings, it is likely that shifts toward more shared models will become more prevalent, depending on a number of factors, high- lighted below, that may facilitate or constrain such a movement. Factors Promoting or Impeding Shared Decision-Making Models Expectations of more shared decision making between physicians and hospitals can be based on several arguments. The first is that the physician surplus will make physicians more dependent on hospitals for privileges and services to build and maintain their practices; thus, their economic well-being will become more closely identified with that of the hospital. This will provide a stimulus for more joint phy- sician/hospital involvement in decision making. Second, as regulation (at any governmental level) continues, physicians and hospitals may perceive increased incentives to unite against the "common enemy." Consistent with the "capture" theory of regulation (whereby the in- dustry itself desires the regulation so as to protect its own interest), physicians and hospitals will work together to make sure their mutual interests are protected. Hospital reimbursement based on case mix also may require more collaborative decision making as such reim- bursement requires administrative and cost data to be integrated with clinical data. Third, as physicians become more closely aligned to hospitals, they may demand greater participation in hospital-wide decision making than they currently have through traditional medical staff organi- zation channels. In brief, they may seek to have greater influence with an organization that is gaining greater importance in their professional lives. Finally, shared decision-making models may be facilitated by more sophisticated and enlightened physicians and professionally trained hospital administrators. More physicians are being exposed to the importance of cost-effective medical care and associated cost-e~ec- tiveness and cost-benefit methodologies. Some, such as graduates of the Robert Wood Johnson Foundation's Clinical Scholars Program, have received broad exposure to health services and health policy issues. Thus, there may be emerging a new cadre of medical leadership with a broader understanding of the hospital both as an economic and
Physician Involvement in Hospital Decision Making 97 a social institution, which overrides the notion of the hospital as sim- ply the "doctor's workshop." As noted by the Hospital Association of Pennsylvania: The hospital-medical staff relationship is currently the weakest link in the hospital corporate management structure. It is this weakness, together with the rising cost issue, which will force a new relationship between physicians and hospitals in the very near future. Joint decision-making involving medical staffs will need to be developed to gain their participation in an acceptance of change in institutional procedures.36 On the other hand, several factors could impede the development of shared decision-making models. First, increased physician compe- tition, resulting from the developing surplus of physicians, could result in more physicians offering services in direct competition with hos- pitals. Emerging examples include emergency care, sports medicine, and health promotion. Under increasing competition, primary care physicians in particular may seek to develop special services. Whether they choose to compete directly with hospitals will depend on a number of local market factors and customs, including the power of local hos- pitals, the demographic composition of the community, and the or- ganization of the medical practice community itself. For example, it would be difficult for a new solo practitioner to compete with a hos- pital, but it would be easier if new physicians could join well-estab- lished group practices and develop new programs and services from that base. A second factor that may cause physicians to keep an arm's-length relationship with hospitals is the physician's desire to escape the reg- ulation and reimbursement controls imposed on hospitals. If physi- cians see little opportunity to change the regulatory or payment climate by working with hospitals, some will move to distance themselves from its consequences by becoming as autonomous as possible. This will have essentially the same effect as noted above in regard to com- petition, i.e., the provision of more services in the physician's offices. For such services as radiology and pathology this has already resulted in the purchase of more sophisticated equipment for physician's offices (e.g., computed tomographic scanners), as opposed to locating them in the hospital. A third deterrent to the development of more shared decision-mak- ing models may be the unwillingness of hospital administrators to open up the decision-making process to physicians. This is likely to be a significant issue in many areas and is understandable given the historical evolution of administrator-physician relationships in U.S.
98 STEPHEN M. SHORTELL hospitals. Essentially, administrators have used informal and per- suasive skills (in addition to the legitimate authority derived from their positions) to gain influence over medical staffs. In particular, they have used their role as intermediary between the medical staff and the board of trustees to control communication and information flow and thus to keep some control over the medical staff's influence on the board. The idea of involving physicians more systematically in hospital-wide policymaking presents a major challenge for adminis- trators and physicians alike. Summary This paper has attempted to capture some of the complexity and dy- namics of changing physician/hospital decision-making relationships. A typology and a number of examples of physician/hospital decision making were developed to provide a framework for considering current developments. Some differences were suggested in decision-making strategies by hospital ownership and whether the hospital belonged to a multi-unit system. Two major models of physician/hospital re- lationships were described the dual authority mode] and the shared authority model. The implications of each of these along with the forces influencing their continued development were examined. Evi- dence regarding the association of more shared decision-making mod- els and the cost and quality of care was summarized. A number of issues pertaining to the changing context of clinical and institutional decision making were presented, suggesting that some fundamental changes may take place in the structure of hospital medical staffs. These points have a number of possible implications for for-profit hospitals. First, they are likely to continue to be somewhat more selective than voluntary hospitals in their choice of services to offer the community. Specifically, they will tend to offer services that en- hance the return on the overall portfolio or mix of services provided. Because of the greater involvement of physicians in hospital gover- nance, for-profit hospitals may be more reluctant to compete directly with their medical staffs and more likely to offer services that are complementary to rather than substitutable for physician services in the community. Second, for-profit hospitals, particularly those owned by investor- owned chains, may be better able than voluntary hospitals to deal with "compromise" or "judgmental" decisions. This is because they have a more clearly defined and homogenous group of constituents (stockholders) and generally more overall centralized direction from
Physician Involvement in Hospital Decision Making 99 the corporate headquarters office. As a result, preferences regarding desired outcomes may be more clear. Investor-owned hospitals may, therefore, be more able to make rapid adjustment to external changes (e.g., changes in third-party reimbursement or changes in competi- tion) than most voluntary hospitals can. Third, because the interests of physicians and the hospital may be more closely aligned in for-profit hospitals, the dual authority mode} of decision making is less problematic. Perhaps the lesser degree of physician involvement in daily committee work that characterizes for- profit hospitals reflects a higher degree of agreement on a more ho- mogenous and targeted set of goals and greater physician involvement in the governance process. In contrast, voluntary hospitals deal with the issues created by the dual authority mode] through a rather elab- orate system of committees attempting to achieve increased physician participation and involvement. Although both types of hospitals may be shifting toward a more shared authority model, investor-owned hospitals may be able to make the adjustment more quickly and easily because of the greater degree of agreement on overall goals and the history of physician involvement in decision making at the governance level of the organization. But it is also important to note that the above differences and their implications may be attenuated by some growing similarities between for-profit and voluntary hospitals in their economic orientations. Un- der pressures for cost containment, plus increased competition, vol- untary hospitals have had to give more concerted thought both~to their short-run operational needs and to longer-run capital formation requirements. A number of voluntary hospitals have corporately re- organized, in many instances creating for-profit subsidiaries to expand the hospital's sources of revenue. Some of the above differences may also be attenuated by the continued growth of multi-unit systems among not-for-profit hospitals. Through their corporate office exper- tise and structure, such systems may be able to offer the same kinds of advantages as the investor-owned systems. In brief, although im- portant differences still exist between the mission, philosophy, struc- ture, and decision making of for-profit and not-for-profit hospitals, forces are currently in motion that over time may diminish some of these differences. References and Notes 1. Thompson, J. D. Organizations in Action. New York: McGraw-Hill, 1967. 2. Shortell, S. M., T. Wickizer, J. Wheeler and Associates. Hospital-Physician Joint Ventures: Results and Lessons from a National Demonstration. Ann Arbor, Michigan: Health Administration Press, 1984.
100 STEPHEN M. SHORTELL 3. Harris, J. E. "The Internal Organization of Hospitals: Some Economic Implications." Bell Journal of Economics 8 ( 1978), p. 479. 4. Pauly, M. V., and M. Redisch. "The Not-for-Profit Hospital as Physician's Coopera- tive." The American Economic Review 63 (March 1973), pp. 87-100. 5. Harris, op. cit., pp. 467-482. 6. Smith, H. L. "Two Lines of Authority Are One Too Many." Modern Hospital 84 (March 1955), pp. 59-64. 7. Harris, op. cit., p. 477. 8. Scott, W. R. "Managing Professional Work: Three Models of Control for Health Or- ganizations." Health Services Research 17 (Fall 1982), pp. 213-240. 9. Darling v. Charleston Community Memorial Hospital, 211 N.E. Second 253 (1965). 10. Parsons, T. The Social System. Glencoe, Ill.: Free Press, 1956. 11. Carlson, R. J. The End of Medicine. New York: John Wiley & Sons, 1975. 12. Weisbord, M. R. "Why Organization Development Hasn't Worked (So Far) in Aca- demic Medical Centers." Health Care Management Review 1 (Spring 1976), p. 18. 13. Allison, R. F., and J. W. Dalston. "Governance of University-Owned Teaching Hos- pitals." Inquiry 19 (Spring 1982), pp. 3-17. 14. Shortell, S. M., and T. Getzen. "Measuring Hospital Medical Staff Organization and Structures." Health Services Research 14 (Summer 1979), pp. 97-110. 15. Sloan, F. A. "The Internal Organization of Hospitals: A Descriptive Study." Health Services Research 15 (Fall 1980), pp. 294-230. 16. Ibid. 17. Shortell, S. M., and C. Evashwick. "The Structural Configuration of U.S. Hospital Medical Staffs." Medical Care 19 (April 1981), pp. 419-430. 18. Sloan, op. cit. 19. Bidece, L., and D. Danais. Physician Characteristics and Distribution in the United States1981. Chicago: Division of Survey and Data Resources, American Medical Asso- ciation, 1982. 20. Sloan, op. cit. 21. Shortell and Evashwick, op. cit. 22. Neuhauser, D. The Relationship Between Administrative Practices and Hospital Per- formance. Chicago: Center for Health Administration Studies, University of Chicago, Re- search Series #28, 1971; Shortell, S. M., S. W. Becker, and D. Neuhauser. "The Effects of Management Practices on Hospital Efficiency and Quality of Care." Organizational Re- search in Hospitals: An Inquiry Monograph, S. M. Shortell and M. Brown, eds. Chicago: Blue Cross Association, 1976, pp. 90-106; Rushing, W. "Differences in Profit and Nonprofit Organizations: A Study of Effectiveness and Efficiency in General Short-Stay Hospitals." Administrative Science Quarterly 19 (December 1974), pp.473-484; Pauly, M. "Medical Staff Characteristics and Hospital Costs." Journal of Human Resources, Supplement 13 (1978), p.77; Sloan, F., and E. Becker. "International Organization of Hospitals and Hospital Costs." Inquiry 18 (Fall 1981), pp. 224-239. 23. Rushing, op. cit. 24. Pauly, op. cit. 25. Neuhauser, op. cit.; Shortell, Becker, and Neuhauser, op. cit.; Flood, A., and W. R. Scott. "Professional Power and Professional Effectiveness: The Power of the Surgical Staff and the Quality of Surgical Care in Hospitals." Journal of Health and Social Behavior 19 (September 1978), p. 240; Roemer, M., and J. Friedman. Doctors in Hospitals: Medical Staff Organization and Hospital Performance. Baltimore: Johns Hopkins University Press, 1971; Shortell, S. M., and J. P. LoGerfo. "Hospital Medical Staff Organization and Quality of Care: Results for Myocardial Infarction and Appendectomy." Medical Care 19 (October 1981), pp. 1041-1056; Rhee, S. "Relative Importance of Physician's Personal and Situational Characteristics for the Quality of Patient Care." Journal of Health and Social Behavior 18
Physician Involvement in Hospital Decision Making 101 (March 1977), pp. 10-15; Morlock, L., C. Nathanson, D. N. Schumacher, and S. D. Horn. "Decision-Making Patterns and Hospital Performance: Relationships Between Case Mix Adjusted Mortality Rates and the Influence of Trustee, Administrators, and Medical Staff in Seventeen Acute Care General Hospitals." Medical Care, in press; Holland, P. P., A. Konick, W. Buffum et al. "Institutional Structure and Resident Outcome." Journal of Health and Social Behavior 22 (December 1981), pp. 433-444. 26. Flood and Scott, op. cit. 27. Roemer and Friedman, op. cit. 28. Morlock et al., op. cit. 29. Neuhauser, op. cit. 30. Shortell, Becker, and Nenhauser, op. cit. 31. Longest, B. B. "An Empirical Analysis of the Quality-Cost Relationship." Hospital and Health Services Administration 23 (Fall 1978), pp. 20-35; Schulz, R. I., J. R. Greenley, and R. W. Peterson. "Management, Cost and Quality in Hospitals." Medical Care, in press. 32. Flood, A. B., W. Ewy, and W. R. Scott et al. "The Relationship Between Intensity and Duration of Medical Services and Outcomes for Hospitalized Patients." Medical Care 17 (1979), pp. 1088-1102. 33. Pellegrino, E. D. "The Changing Matrix of Clinical Decision-Making in the Hospital." In Organizational Research in Health Institutions, B. S. Georgopoulos, ed. Ann Arbor: Uni- versity of Michigan Press Institute for Survey Research, 1972, p. 301. 34. Brown, M. "Multi-Hospital Systems: Trends, Issues, and Prospects." In Multi-Hospital Systems: Policy Issues for the Future, G. Bisbee, Jr., ed. Chicago: The Hospital Research Educational Trust, 1981, pp. 3-21. 35. Morrisey, M. "A Description of Hospital-HMO Affiliations." Paper presented at the American Public Health Association Meetings, Montreal, Quebec, Canada, November 1982. 36. Pennsylvania Hospital Association. Environmental Assessment. Hospital Industry and Hospital Association of Pennsylvania, Camp Hill, Pennsylvania, February 1981, p. 35.