Professional and Organizational Differences in Physician Satisfaction
John C. Lammers and Joshua B. Barbour
The health care system in the United States has changed dramatically in the last 30 years. Health organizations are under pressure from patients for access to high quality, affordable care. Employers, insurance companies, and federal agencies want lower costs, accountability, and safety. Hospitals and medical groups, as well as insurance companies and government agencies, are adopting management methods that emphasize effi ciency, predictability, cal- culability, and control (Ritzer, 2004). Despite the efforts to more tightly man- age health care in the United States, however, medical expenses continue to rise (Strunk, Ginsburg, & Gabel, 2002), and patients and payers are not alone in their expressions of dissatisfaction. Both nurses and physicians also have expressed concern about working conditions and quality of care in managed care arrangements (Harvard School of Public Health, 2000). The concerns of so many involved in the health care system at multiple levels (patients as well as providers, organizations as well as government agencies) present an oppor- tunity to health services researchers in the form of research of keen interest to a broad audience. The changes in health care also present a mandate to address our fi eld’s lack of theoretical models that explain developments at both personal and organizational levels in health services.
The widely-used term managed care (see Real & Street, this volume) is the catch-all that describes developments in health care over the last 30 years.
Managed care commonly refers to health services arrangements that employ prepayment for services, require pre-authorization from offi cials for services, and involve contracts among patients, providers, and administrators (Lam- mers, Barbour, & Duggan, 2003). The broadest defi nition includes
Health insurance plans intended to reduce unnecessary health care costs through a variety of mechanisms, including: economic incentives for physicians and patients to select less costly forms of care; programs for reviewing the medical necessity of specifi c services; increased benefi ciary cost sharing; controls on inpatient admissions and lengths of stay; the establishment of cost-sharing incentives for outpatient surgery; selective contracting with health care providers; and the intensive management of
high-cost health care cases. The programs may be provided in a variety of settings, such as health maintenance organizations and preferred provider organizations. (National Library of Medicine, 2008)
These types of arrangements now dominate health care in the United States (Barbour & Lammers, 2007; Lammers et al., 2003; Lammers & Duggan, 2002).
The administrative context or “peri-consultation interaction” (see Real &
Street, this volume) of patient-provider communication now has changed.
Administrative policies for care delivery and payment form important new organizational structures (McPhee, 1985) in which communication between physicians and patients takes place. These structures also contextualize the relationships between physicians and managers, and among managers of the health care systems, which tends to bureaucratize health care (Lammers &
Geist, 1997).
Managed care itself is not a unifi ed entity, however, as a number of writers have observed (Hacker & Marmor, 1999; Real & Street, this volume). The term loosely covers many arrangements from highly corporate approaches to cost controls, to loose confederations of cooperating providers. From its least organized manifestation (independent practitioner associations) to its most comprehensive arrangements (staff model health maintenance organizations), the meaning of managed care involves the establishment of written guidelines for the provision of medical services (Mechanic, 2000, p. 103). The formal rules and guidelines of managed care have been the subject of considerable dis- sent in the medical community (Bloche, 1999; Feldman, Novack, & Gracely, 1998; Gonsoulin, 1997; Mechanic, 2000; Potter, 1999; Rodwin, 1998). For example, Sullivan (1999) worried that managed care and its fi nancial impera- tives threaten the professionalism of medicine.
For the most part, these changes in the context of the physician-patient relationship have not been linked to physicians’ work by theoretically guided systematic research. Researchers concerned with the medical care setting of physician-patient communication, for example, have understood its context in terms of the privilege and authority of the physician and the privacy of the physician-patient encounter (Ray & Donohew, 1990, p. 29). Real and Street (this volume) pointed out that the specifi c formal organizational context, however, is rarely distinguished. Moreover, specifi c types of providers—such as primary care providers or specialists—are rarely distinguished in research on physicians’ work (although they are distinguished in health policy research;
see Robinson, 2001).
Nonetheless, managed care alters both professional roles as well as the orga- nizational contexts of physicians’ work (Barbour & Lammers, 2007; Lammers
& Geist, 1997). As physicians are drawn into ever more rational social struc- tures (Scott, 2003), principles and theories of organizations are likely to apply to their work. McPhee (1985) labeled organizational structure as “explicit, authoritative, metacommunication” (p. 162). This view of structure is espe-
Exploring the Institutional Context of Physicians’ Work 93 cially congruent with the circumstances of managed care, because now, in contrast to their traditional autonomous roles as professional healers, physi- cians work by fi xed routines (Weber, 1946, 1947), subject to organizational rules (Euske & Roberts, 1987). Thus, the case of physicians’ roles in managed care provides an opportunity to study specifi c roles and the structural context of health services.
Role confi gurations, such as physicians’ specialization (a socially micro-level phenomenon), and structures, such as practice settings (a socially macro-level phenomenon), can be understood together through the application of the institutional theory of organizational communication (ITOC) (Lammers &
Barbour, 2006). An institutional approach considers “established and endur- ing patterns of beliefs and practices that apply at both the microlevel within organizations and at the macrolevel across organizations” (p. 262). ITOC thus provides a vehicle for examining how medical professionals’ institutionalized beliefs and practices in particular organizational settings may result in greater or lesser satisfaction and effectiveness. Moreover, a hallmark of institutions, and a guiding aspect of ITOC, is the role of formal, written rules, contracts, and regulations that cut across organizational boundaries. From the regula- tions that govern Medicare reimbursement to the contracts the bind providers to patients in health plans, managed care represents an institutional structure of infl uence on health organizations and the providers who work in them.
This study identifi es institutional contexts and role confi gurations in the provision of medical care. To study institutional contexts, we surveyed phy- sicians in a community dominated by three organizations: a specialty clinic established as a treatment center for medical problems, a community clinic established with a public interest philosophy, and a loose federation of solo physicians known as an independent practitioner association (IPA). Physi- cians’ role confi gurations in this study include their perceptions of satisfaction, autonomy, and ability to make clinical decisions, their attitudes toward medi- cal practice, and their reports of communication with patients and managed care organizations. This characterization of role confi guration is defi ned by individuals’ perceptions of their roles. In other words, our analysis focuses on the relationship between individual physicians’ perceptions and their institu- tional situations.
To this end, we fi rst review physicians’ administrative contexts and their professional medical roles as primary care or specialists providers. We suggest that both the administrative context and professional roles can be understood using an institutional perspective, which we also outline below in more detail.
Next, we provide background on the community and the three organizations we studied, and we hypothesize the likelihood of satisfaction or dissatisfac- tion for physicians in the practices. Survey data of physician satisfaction is then presented and compared for specialist and generalist physicians practic- ing in each of the three organizations. Results suggest that the histories of the organizations, as well as specialty and practice arrangements, affect physicians’
attitudes toward medical practice both in terms of satisfaction and clinical autonomy. Implications for health services research bridging organizational structures with physicians’ work are discussed.
Assumptions about Professions and Organizations:
Toward an Institutional Perspective
Although professional powers may transcend administrative or organizational arrangements (Freidson, 1986), many medical care providers today view man- aged care as a challenge to their professional discretion (Feldman, Novack, &
Gracely, 1998; Real & Street, this volume; Rodwin, 1998). Perhaps the auton- omy of physicians could be taken for granted before 1980, when the majority of practitioners saw their patients under fee-for-service reimbursement arrange- ments, and the number of physicians in group practice was smaller than it is today (Scott & Lammers, 1985). Many medical groups today depend on prepaid contracts or membership in preferred provider organizations to sustain revenues (Bodenheimer & Grumbach, 2002, p. 197); however, the adminis- trative arrangements that accompany group practices intervene in physicians’
conduct (Gross & Budrys, 1986). Physicians themselves have called into ques- tion managed care rules as they relate to their decision-making powers as well as their ethical responsibilities (Feldman, Novack, & Gracely, 1998; Friedman
& Savage, 1998; Kralewski et al., 1998; Minogue, 2000). In this section, we review the role confi gurations and the organizational context of physicians as understood by past health services research. We make the case that the advent of managed care signifi cantly affects these confi gurations and contexts, and we suggest that an institutional perspective is an appropriate lens for the study of these changes.
There are at least two directions in which professional medical roles have developed (Thomas, 1983). The literature on medical professional socialization has for many years highlighted the tension between the humanistic and scien- tifi c requirements of the role (Becker, Geer, Hughes, & Strauss, 1961; Conrad, 1988; Miller, 1993). This tension has been characterized in terms of caring versus curing (Conrad, 1988); holistic versus biomedical approaches (Longino, 1997); and the healer versus the scientist (Laine & Davidoff, 1996). A number of writers have laid these developments at the feet of the medical school train- ing experience (Becker et al., 1961; Conrad, 1988; Miller, 1993).
One could argue, on the basis of this literature, that the longer a physi- cian in training is exposed to the training environment, the more likely that she or he will adopt a scientifi c outlook. The scientifi c professional outlook is cultivated in the recruitment and selection process for medical schools. Candi- dates are selected on the basis of scientifi c acumen and an orientation toward systematic knowledge. Faculty members employed in careers of research rather than treatment reinforce the professional scientifi c outlook throughout gradu- ate and postgraduate education (Bloom, 1989; Jefferys & Elston, 1989). The
Exploring the Institutional Context of Physicians’ Work 95 scientifi c outlook is further reinforced by the drive toward specialization in medicine that began during the middle part of the 20th century. Specialists became even more closely tied to academic medical centers to remain well informed and to have access to sophisticated technologies, facilities, and col- leagues (Simmonds, Robbins, Brinker, Rice, & Kerstein, 1990).
In contrast, the American Academy of Family Physicians (2001) defi ned the primary care physician as “a generalist physician who provides defi nitive care to the undifferentiated patient at the point of fi rst contact and takes con- tinuing responsibility for providing the patient’s care” (p. 2, emphasis added).
The generalist outlook is an older, more traditional confi guration of the health professional’s role. Starr (1982) discussed the confl ict early in this century between specialists, who sought control over particular types of procedures and practices, and their generalist forebears. In the United States, this eventu- ally resulted in higher barriers to entry into medical practice by nonphysicians (such as technologists and midwives), but “fl uid boundaries within the profes- sion” (p. 325). Nonetheless, as in other fi elds, because of the cultural value placed on technical knowledge, physicians have found it diffi cult to remain generalists (Vanselow, 1998).
These two strands of professional roles in medicine are not exclusive. It would be inappropriate to say that some physicians are healers whereas others are scientists, but the emphasis on science in medical training is well-docu- mented, and the differences in distribution, prestige, and earnings of generalists (including internists, family practitioners, and pediatricians) versus specialists (such as dermatologists or surgeons) is also well-documented (Bureau of Labor Statistics, 2000; Donabedian, 1986). Perhaps most important for our purposes, the undifferentiated nature of generalists’ work compared to the specialist would lead us to expect different communication patterns, behaviors, and atti- tudes. For example, as gatekeepers in a managed care regime, generalists make initial diagnoses and referrals to specialists, whose narrow range of practice and expertise allows them to focus on treatment options. Indeed, Smetana et al. (2007) found a bias toward effi cacy in research evaluating specialists, who typically focused on a single condition, in contrast to generalists, who concerned themselves with a wider spectrum of patient diseases and ailments.
Stille, Primack, and Savageau (2003) found that the practice styles and com- munication habits of pediatric generalists and pediatric subspecialists made coordination of care diffi cult.
Managed care may be expected to affect these role confi gurations in various ways. As a program driven by cost-conscious insurance companies, employers, and governments, managed care does not reimburse physicians for training, research, or education. It favors the healing role in this respect. But managed care also involves practice guidelines, review panels, credentialing, schedul- ing, and other mechanisms that create effi ciencies in the provision of care.
We should, therefore, be able to observe managed care arrangements frustrat- ing physicians in their ability to serve patients. In general, we suspect that
managed care is likely to frustrate patient care efforts of physicians by rais- ing questions about their autonomy over care decisions and by raising ethical issues with respect to the physicians’ obligations to patients.
The organizational context of the provision of care often is regarded as undifferentiated, even when understanding the context might have been theoretically informative (Sharf, 1993). In a departure from studies of con- sultations in examining rooms, Wissow et al. (1998) studied physician-patient communication in emergency departments, but did not compare these data to nonemergency consultations. Suchman, Roter, Green, and Lipkin (1993) studied physicians’ satisfaction in patient interviews, including a wide range of patient characteristics and circumstances in their analysis, but they included no organizational variables. Key variables in the study, respondents’ percep- tions of the patient physician relationship, the data collection process, the appropriateness of the use of time, and the absence of excessive demands on the part of the patient all were likely linked to (noncontrolled) organizational issues. Research in technologically mediated communication also suggests that organizational and institutional forces must be taken into account in order to understand their application to the physician-patient relationship (Baur, 2000;
Lehoux, Sicotte, Denis, Berg, & Lacroix, 2002).
It seems false, therefore, to assume uniformity across all administrative arrangements in which physician-patient relationships occur. Public or pri- vate auspices; the volume of resources; the reimbursement methods; the rural or urban location of the facility; whether the encounter occurs in a hospital, clinic, or private offi ce; and the size, history, climate, and culture of the orga- nization may all contribute to differences in the communicative encounter (Kralewski et al., 1998). Moreover, many of these factors have been changing rapidly in recent years as managed care arrangements have developed.
In the administration of managed care, patients are seen as members of populations, and administrative arrangements are based in part on the wealth of those populations (Lammers et al., 2003). For example, physicians’ reve- nues per member per month for commercial HMOs in 2000 averaged $167.32, whereas Medicare revenues per member per month averaged $53.18 (Managed Care Online, 2001), and the disparity was even greater for hospitals ($198.51 and $40.23, respectively). It is diffi cult to imagine that the disparity in reve- nues between these two populations would not infl uence the medical encoun- ter. Indeed, wealthier, more intensively managed health plans are more likely to use pay-for performance schemes to reward physicians (Rosenthal, Landon, Normand, Frank, & Epstein, 2006); yet Medicare patients report higher satis- faction with traditional plans versus managed care plans (Landon, Zaslavsky, Bernard, Cioffi , & Cleary, 2004).
Financial pressures on medical care administrators, and, in turn, on physi- cians, are expected to continue to grow (Van de Ven, Engleman, & Rogers, 2001). The percent of employees of medium and large establishments who use traditional indemnity insurance has been shrinking since 1980, whereas the
Exploring the Institutional Context of Physicians’ Work 97 percent of such employees who use various forms of managed care has been growing (Ellis, 2001). As the funding environment becomes more constrained, administrative constraints also may be expected to grow.
Managed Care in an Institutional Perspective
These changes in professional roles and administrative context can be under- stood using an institutional perspective (Lammers & Barbour, 2006; Scott, Meyer, & Associates, 1994). The institutional perspective emphasizes the historical context and the extra-organizational environment of organizations (Lammers et al., 2003). Although particular organizations may be thought of as institutions, the concept more usefully applies to “clusters of conventions”
(Barley, 2008, p. 496) that form the foundation upon which organizing occurs.
Thus, the use of intensive management practices in medical care, the profes- sional autonomy of physicians, the tendency for physicians to work as general- ists or specialists, and the establishment of specialized or primary care medical clinics, together represent the institutions of medical care in the United States today.
The institutional roots of managed care may be found in the oldest prepaid medical practices like Kaiser Permanente, but the movement really gained momentum in the mid-1960s. Since the advent of Medicare and Medicaid in 1964, the vast majority of physicians’ offi ces, medical groups, and hospitals have elected to comply with the regulations that authorize reimbursement for services through these programs. A few sustained standards but elected not to participate. Later, as private managed care efforts mounted throughout the 1980s, some hospitals and clinics resisted the changes, retaining niche ser- vices or patients. As the regulatory environment became more complex, with the advent of prepayment using the Diagnosis Related Groups (DRG) system (Geist & Hardesty, 1992) and eventually the Resource-Based Relative Value Scale (Harris-Shapiro, 1998), more experienced organizations and providers developed an advantage over inexperienced organizations and providers who had earlier resisted the trend toward fi xed prepayment and pre-authorization, and other contractual guidelines. Therefore we are led to expect that providers who work in organizations with longer exposure to the regulatory environment (that is, the institutions) of managed care would be less inclined to complain about its pressures.
Barbour and Lammers (2007) explored physicians’ satisfaction and dis- satisfaction with managed care using an institutional perspective. Like oth- ers, Barbour and Lammers found that exposure to managed care—in terms of number of contracts and frequency of communication with managed care representatives—as a dissatisfi er for physicians in general. But they also found that physicians’ institutional beliefs about autonomy and their commitment to the profession acted as moderators on their feelings of dissatisfaction associated with managed care (p. 225). Paradoxically, however, Barbour and Lammers