Click for next page ( 134


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 133
E The Development of Clinical Guidelines for Primary Care Practice: A Systems Approach Steven A. Wartman, M.D., Ph.D. ABSTRACT The division of health care delivery into primary and specialty care is somewhat artificial because the actual delivery of medical care is not so nicely hierarchical and compartmentalized. In reality, primary care is part of a continuum of care that ranges over a broad spectrum of health issues. The diverse nature and content of primary care practice requires a unique perspective for guidelines development, one that addresses the underlying complexities that exist in actual health care delivery. A systems perspective is offered as a means to approach the structure that underlies this complex situation. Using this perspective, the proposed framework for primary care clinical guidelines development is based on the principal characteristics of the type of care delivered. The three essential descriptors of type of care for primary care services are: (1) the unselected nature of the patients, (2) continuous and longitudinal care over extended periods, and (3) integrative care that links other health care services to the patient. Three key health system elements are defined that, when combined with the descriptors of primary care, produce a model for the generation of guideline topics. Built in to this proposed model of guidelines Paper prepared by Steven A. Wartman, M.D., Ph.D., Chairman, Department of Medicine and Director of Medical Services, Mount Sinai Medical Center of Greater Miami, and Professor of Medicine, University of Miami. 133

OCR for page 133
134 SETTING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES development is the direct incorporation of outcomes studies in a continuous feedback loop so as to facilitate further guidelines improvement and refinement. The advantages of this model for guideline development in primary care is the linking of process and outcome in a systems approach to primary care. It is suggested that specific funding be devoted in addition to enlarging our understanding of medical practice so that attempts to influence this practice through clinical guidelines will have a greater chance of success. The development of clinical guidelines that address issues in primary care poses both theoretical arid practical challenges since the scope and content of primary care practice is neither easily defined nor well circumscribed. This becomes evident if one attempts to adopt a disease-specific approach to guidelines development for primary care. While clinical practice guidelines for kidney disease, for example, would include such disorders as nephrolithiasis, renal failure, and pyelonephritis, the diagnosis and management of kidney disease in primary care medicine does not generally permit such neat patterns of categorization. Patients often present to their primary care providers with vague or ill-defined complaints such as fatigue or back pain. These complaints require a diagnostic evaluation before disease-specific guidelines can be employed. Further, patients' decisions to seek care are often highly individual and complex, and include the full array of psychosocial factors that can interface with health and disease. The characteristics of the health care delivery system to which the patient belongs, including factors such as availability and accessibility, also can play a pivotal role. These important factors lie outside the scope of clinical practice guidelines based on disease entities. The nature and content of primary care practice simply defies a traditional, biomedical-based approach to the development of clinical practice guidelines. It is my intent to suggest an alternate perspective be applied to guidelines development for primary care based on the assumption that a conceptual paradigm different from that which is generally applied to the analysis of most health care issues is needed. BACKGROUND: FLAWS IN THE CURRENT CONCEPTUALIZATION OF PRIMARY CARE Primary care is usually defined as the continuous, comprehensive health care of patients regardless of the presence or absence of disease and which integrates other health resources when necessary (Alpert and Charney, 1973~. It stresses accessibility as well as comprehensiveness, and is coordinated, continuous, and accountable first-contact care (IOM, 1978~. Primary care is often viewed as a parallel partner of specialist care. In this view, both "systems" of care operate within well-defined boundaries, with patients shuffled back and forth as necessary (Figure E.1~. In a sense, the age-old "breadth" versus "depth" debate is reflected in the attempts to dichotomize care into the two realms of primary

OCR for page 133
APPENDIX E 135 and specialty care. As depicted in Figure E.2, the practice patterns of generalists and specialists diverge over time; the former are characterized by increasing breadth at the expense of depth, the latter by increasing depth at the expense of breadth. This fundamental divergence reflects not just practice patterns, but also different intellectual paradigms. Specialty care lends itself more readily to a reductionist approach while primary care favors a more integrative, generalizeable approach. Over time, the primary care physician has greatly extended his or her scope of practice to include a wide variety of clinical entities; the specialist physician has narrowed his or her scope of practice to include a great deal of information about fewer entities. But breadth and depth are not mutually exclusive from the point of view of the patient. For example, the primary care physician's scope of practice certainly includes some elements of kidney disease such as pyelonephritis and early renal failure. The nephrologist also treats these two entities. At what point does the patient "belong" to one or both realms of care? LL cr hi: a: cc FIGURE E.1 Parallel systems of care. 1 ~ En m :> m This problem of categorization becomes clear in reviewing the proposed, in progress, and completed lists of clinical practice guidelines funded by the Agency for Health Care Policy and Research (AHCPR). If the following question were posed, "Does this guideline fall within the purview of primary or specialty care?", at least thirteen of the twenty-one guideline topics listed could be considered as both within the purview of the primary care physician and the specialist Table E.1~. For example, headache is an exceedingly common problem in primary care practice; a guideline for the management of chronic

OCR for page 133
136 SEfTING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES ._ o ce a, / / / / G/ ~/ G - Generalist : , / / / / / / / S Depth over time S = Specialist FIGURE E.2 Practice patterns of generalists versus specialists over time. headache would be helpful to both primary care providers and neurologists. The same is true for many of the guideline topics listed in Table E.1. Thus, primary care is not a parallel system of the specialty system of care. The actual delivery of medical care is not often so hierarchical and compartmentalized. Even in the managed care model of practice, patients and their health concerns pass through both "systems" of care freely because the boundaries of care for individual health problems are not well established. Primary care is part of a continuum of care that ranges over a broad spectrum of health issues. A systems model perhaps best illustrates this point (Figure E.3~. This model reflects the considerable overlap that exists in most areas of medical practice. The boundaries of care, between the specialties themselves and between primary and specialty care, are not well defined. The systems model, especially the "chaotic" center, is more reflective of the real-world practice of medicine. The nephrologist, cardiologist, and general internist all have roles in the treatment of congestive heart failure. The general surgeon, dermatologist, and plastic surgeon overlap in the care of many skin lesions. The systems model of care reflects this complexity and avoids the artificially tidy approach suggested by the traditional compartmentalization of medical care into a series of specialty domains. As will be elucidated farther in this paper, the systems approach offers a new framework for the development of clinical guidelines for primary care.

OCR for page 133
APPENDIX E TABLE E.1 AHCPR's Clinical Practice Guidelines by Type of Care* . . . 137 Type of Care Guidelines Specialist Primary Both Acute pain management Urinary incontinence in adults Pressure Ulcers in Adults Cataracts in adults . Depression in primary care Sickle cell disease Early HIV infection Benign pro static hyperplasia Cancer pain Unstable angina Heart failure Acute low back problems in adults Otitis media with effusion in children Mammography Post stroke rehabilitation Recognition and initial assessment of Alzheimer's and related dementias Cardiac rehabilitation i/ Smoking prevention and cessation Anxiety and panic disorders Screening for colorectal cancer Headache pain ~ TOTAL 5 3 13 *As categorized by the author (SAW). The point should be clear: primary care is not a discrete field confined to the management of a particular set of health care issues. Applying the standard view of primary care as a parallel system to the specialty system of care in the development of primary care guidelines will not work. It will lead to a confusing selection of topics that artificially separates what is, in reality, a continuum of care from the patient's point of view. A different conceptualization of primary health care is needed in order to properly address the issue of clinical practice guidelines for primary care.

OCR for page 133
138 SETTING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES PRIMARY CARES ~ j ~ DOMAIN \~ ~ \ ~ / SPECIALTY CARE | ,~\~,-\~4~'< \ / DOMAINS ~ ~ ,' ~ Of - ~ \ / / an/ \ >a SPECIALlY CARE / >a ~ \ DOMAINS / / ~\ \ FIGURE E3 Systems model of care. PRIMARY CARE: A DIFFERENT VIEW In approaching the development of clinical guidelines for primary care, it is essential to define the perspective from which the guidelines are to be derived. This perspective can be sought from a version of the "unit of analysis" question. Is the unit of analysis the problem, the provider, the patient, or the type of care? I have already suggested why analysis of the "problem" might not offer the best approach. As pointed out previously, medical "problems" do not neatly fit into primary or specialist care indeed it is impossible to confine many problems to one sphere or the other. There is significant overlap and any approach to guideline setting for primary care by "problem" is likely to fall prey to conflicting constituencies and interpretations. From the viewpoints of training, certification, and reimbursement, the world of providers is certainly divided into primary care and speciality. Perhaps it makes more sense to focus on the provider. While it is certainly possible to do so, a closer view reveals the futility of this approach. First, there is considerable variety in the kinds of primary care providers, including many physician assistants, nurse practitioners, family physicians, osteopathic physicians, general internists, and general pediatricians. Each has very different scopes of practice

OCR for page 133
APPENDIX E 139 based on education, training, patient population, type of health care environment, and so forth. It is unclear that a single guideline for primary care practitioners could effectively embrace this large and diverse constituency. Second, there are many specialists who practice some amount of primary care; the exact numbers are unknown but are felt to be substantial (Aiken, Lewis, Craig, et al., 19794. Thus, an approach by type of provider in our current pluralistic system of health care delivery is likely to be confusing if not misleading. Could an approach be developed based on the patient? The idea initially is attractive. After all, the patient should be the focus of the health care system. The needs of patients, both individually and collectively, could be determined on an epidemiologic basis. Guidelines could then be constructed for the most pressing of these needs, based on the patient's social, environmental, economic, and other circumstances. While initially challenging, such an approach would simply be impractical because it would call for the manipulation of resources well outside the domain of current health care practice, involving such areas as housing, controls on violence, economic disenfranchisement, and the like. Lastly, the perspective to be used could be based on type of care. This approach focuses on the systems involved in health care delivery. It is compatible with the model depicted in Figure E.3. It facilitates the use of "systems thinking," which develops a framework for "seeing interrelationships rather than linear cause-effect chains," and is useful in situations where "cause and effect are subtle and where the effects over time are not obvious" (Serge, 1990~. Systems thinking is particularly applicable to health care because health is affected by a myriad of forces, including individual values, beliefs, expectations and genetic predispositions, all of which are immersed in a particular environment, social milieu, and health care network. In such a complex environment, it is not unusual for small events, such as a change in medication or weather, to have large arid unpredictable effects on the patient (Lipsitz and Goldberger, 1992~. Rather than be overwhelmed by all this complexity, systems thinking helps see the structure underlying complex situations (Serge, 1990~. Taking a systems perspective offers the opportunity to discern those issues that can have a great impact on health. PRIMARY CARE AS A SYSTEM OF CARE Type of care analysis lends itself to a systems perspective. It offers an approach that deals with the underlying complexities that exist in primary care practice. Instead of pursuing simplistic cause-and-effect relationships (as when using the "problem" as the basis of the analysis), type of care analysis emphasizes the circumstances surrounding the occurrence of events (e.g., symptoms, decision to seek care, how complaints are expressed) and the

OCR for page 133
140 SETTING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES interrelationships among such variables as the provider's work-up, role of stress, and family traits. It reflects the observations that, in attempting to look at health and well-being, (a) the whole is greater than the sum of its parts; (b) whatever affects the system as a whole affects each part; and (c) any change in one part affects other parts and the system as a whole (Glenn, 1984~. Given the overlap between specialty care and primary care as depicted in Figure E.3, it makes sense to distinguish primary care from other Morons of care by the type of care provided. Primary care is a system of care that involves three basic components: care of an unselected panel of patients, care that is continuous and longitudinal over time, and care that is integrative in scope. Patients are unselected in that those receiving primary care services have no thematic set of reasons for doing so. Their reasons are theoretically unbounded and range from periodic health maintenance exams, vague and ill-defined complaints, to chronic and acute disease. The primary care system must be prepared to deal with this wide variety of patients and health care issues. Continuous and longitudinal care is received by patients in an accessible primary care setting, is ongoing and comprehensive, and is provided by the same health care professional over considerable periods of time in both illness and health. Specialty care may or may not fulfill this criteria; primary care always must do so. Integrative care includes other health-related services (e.g., specialist consultations, rehabilitation, counseling) that are applied and coordinated to the patient's overall care in a socially responsible (i.e., accountable) manner. These three components compose the essential descriptors of the primary care system. The descriptors of primary care operate, of course, within the complex ecology ofthe patient and the larger hearth care system. The practice of primary care involves the interactions of these descriptors with certain elements in the patient's environment. These elements represent the key environmental dynamics that influence primary care health care delivery. From a systems perspective, these elements include: the context in which the care takes place; the interactions that bring the various components of the health care system together in the management ofthe patient; and the reaction ofthe health care system to changes in the patient's health needs (Glenn, 1984~. The context is the social, cultural, and medical milieu of the patient. It largely determines the basis for the patient's decision to seek care. From the point of view of the health care system, care seeking ignites the medical care apparatus. From a policy and practical perspective, it is important to understand end attempt toinfluence hearth seeking behavior. Theinteractions are to the combination of factors that permit the health-care system to interact with a given patient. These include the accessibility and availability of health care services, financing of care, established practice patterns, and so forth. The reaction of the health care system to changes in the patient's symptoms, needs, and disease states is an important element of health care delivery because a

OCR for page 133
APPENDLY E 141 patient's state of health is not constant, but constantly varying. Health needs change frequently over time in both predictable and unpredictable ways. New issues arise and change in their relevance and importance. Old issues wax and wane in their severity. The manner in which the health system reacts to these changes is the third key element in the ongoing primary care of the patient. This conceptualization of primary care as a system of care consisting of a set of descriptors and elements (l able E.2) facilitates a rational approach toward the development of clinical practice guidelines that are uniquely suited for primary care medical practice. TABLE E.2 Descriptors and Elements of the Primary Care Health System Descriptors (basic features of the delivery of primary care) Elements (key environmental dynamics that influence the delivery of primary care) Unselected Patients: Primary care Context: The social, cultural, and patients who seek care for an unlimited medical milieu of the patient. set of reasonse. Continuous and Longitudinal: The Interactions: The combination of ongoing, comprehensive care of a factors that permit the health system patient over time by the same provider. to interact with a given patient. Integrative: The coordination of all ~ The ~a ~r i health system resources in the care of a which the health system reacts to given patient. changes in the patient's symptoms, needs, and disease states. DEVELOPMENT OF GUIDELINES FOR PRIMARY CARE In developing guidelines for primary care, the intersections of the descriptors and elements of primary care practice suggest the type of strategic issues that can serve as a foundation for specific guidelines. These issues tend not to be disease specific, rather, they are "systems" oriented. They involve such questions as: When should patients seek primary care services? Who practices primary care, and if there is more than one type of practitioner, what are their appropriate scopes of practice? Is all primary care the same? What are the key differences between generalist and specialist care? What is the nature of the interface between generalist and specialist care?

OCR for page 133
142 SEITING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES These kinds of questions lend themselves to an enlargement of the disease orientation for guidelines development in primary care to include a systems orientation. An example of the utility of this approach may be taken from patients with chronic illnesses. In general, chronic illness cannot be cured; it must be managed. The patient's biologic abnormality is but one of many interacting components, including treatment effects, emotional state, ability to perform work, healing capability, and so forth; further, the pattern of the patient's illness changes over time as the interacting factors change (Holman, 1993~. The precise cause-and-effect relationships among these interacting factors are often not apparent. It n~akes sense therefore to focus on the broader issues, such as appropriate referrals, intervals between arid timing of follow-up visits, and the impact of social support. The kinds of issues for guidelines development appropriate to primary care are outlined in Table E.3 . In this table, the essential descriptors of primary care are cross-tabulated with the health system elements to produce nine cells that outline suggested guideline topics. Cell #1 (l~escr~tor: Unselected Patients; Element: Context of Care) The context of health care when applied to the essential primary care descriptor unselected suggests a number of issues related to the social, cultural, and medical milieu of the patient. Because primary care patients have such a wide variety of reasons for seeking or needing medical care, it is proposed that guidelines development be explored for patients' decisions to seek health care. Rather than viewing the patient as a passive "victim" or as a greedy "abuser," this approach places some of the responsibility on the patient, something which makes sense in a constrained health care system. These guidelines could be symptom oriented and health maintenance oriented. The former could address such issues as the level, intensity, or tinning of specific symptoms. Specific examples might include chronic fatigue or back or abdominal pain. From the health maintenance point of view there exists already a loose amalgam of these kinds of guidelines, including those from the American Cancer Society and the U.S. Preventive Services Task Force. The challenge is to convert these into meaningful, workable guidelines for patients. The development of patient guidelines offers a particularly unique opportunity to bring the patient into the primary care health system as a partner.

OCR for page 133
APPENDIX E 143 TABLE E.3 Potential Issues for Clinical Guidelines Development In Pnmary Care Using Me Systems Approach Health Systems Elements Primary Care Health System Reaction to Descriptors Context of Cares Interactions Change Unselected Patients Cell #1 Cell #4 Cell #7 When patients Scope of practice of When changes in should seek care various types of pri- symptoms in pa (e.g., chronic fa- mary care providers tients with diagno tigue, abdominal sed illnesses pain, back pain) warrant care (e.g., congestive heart failure, chronic obstructive pul monary disease) . Cell #2 Cell #5 Cell #8 Behavioral and life Communications Intervals for style modifications among providers appropriate follow (e.g., stress reduc- (e.g., content, up and routine lion, dietary modify- frequency, role of care (e.g., arthritis, cations electronic peripheral vascular communications disease Cell #3 Cell #6 Cell #9 Role of other Boundaries of Health care health-related generalist and decision making services (e.g., specialist care (e.g., (e.g., provider limited psycho- hypertension, patient communi therapy, rehabilita- diabetes cation, patient lion post CVA, role preferences, of social support in medical-legal chronic disease) ~issues Continuous/ Longitudinal Care Integrative Care

OCR for page 133
144 SEITING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES Cell #2 (Descriptor: Continuous and Longitudinal; Element: Context of Care) The descriptor continuous and longitudinal, when applied to the context of care, suggests the development of guidelines for specific behavioral and lifestyle modifications. Such guidelines could pertain to both patients and providers. Examples are: stress reduction and selected dietary modifications. Smoking cessation, a guideline already under development, fits into this category. Cell #3 (Descriptor: Integrative Care; Element: Context of Care) This descriptor refers to other health-related services that are applied ~ the care of the patient. Given the context of care, examples for guidelines development include: beef, limited psychotherapy for reactive depression, rehabilitation post cerebral vascular accident, and the role of social support in patients with specific chronic illnesses. Cell #4 (Descriptor: Unselected Patients; Ele'``ent: Health System Interactions) The second element involves the interactions of the various components of the health system that impact on the patient, such as the availability and accessibility of services and the financing of care. From the point of view of the unselected patient, a major guideline topic involves the scopes of practice of the various types of primary care providers. At least six kinds of such providers can be identified and include: family physicians, general internists, general pediatncians, nurse practitioners, physician assistants, and osteopathic physicians. Can their individual roles and scopes of practice be defined? How can they best be differentiated from each other? Can guidelines be developed to facilitate these providers working together in a coherent system of primary care? Cell #5 (Descriptor: Continuous and Longitudinal Care; Element: Health System Interactions) The descriptor continuous arid longitudinal suggests guidelines that address health system interactions among multiple providers involved in the care of the saline patient. What kinds of communication, in terns of content and frequency, would maximize efficient and timely care of the patient? How should such communication be unmanaged by the primary care provider? What is the role of

OCR for page 133
APPENDIX E 145 communication be managed by the primary care provider? What is the role of electronic information in this regard and should there be a standardized system of such communication among providers? Cell #6 (Descriptor: Integrative Care; Element: Health System Interactions) A central theme for the descriptor integrative care involves the interactions of the various providers involved in the care of the patient. What should be the boundaries of generalist and specialist care? Can guidelines be developed for appropriate referral? For example, when should a patient with hypertension or diabetes be referred from a primary care provider to a specialist? Cell #7 (Descriptor: Unselected Patients; Element: Reaction to Change) From the point of view of the unselected patient, guidelines could be developed that address when patients with specific health problems should seek medical attention because of a change in symptoms. This is somewhat similar to the patients' decision to seek care guideline mentioned above, but differs in that the patients are already diagnosed. Examples could include patients with congestive heart failure and chronic obstructive pulmonary disease. Cell #8 (Descriptor: Continuous and Longitudinal Care; Elel1'ent: Reaction to Change) The continuous and longitudinal descriptor together with the element reaction to change suggests topics for guidelines development that include the appropriate intervals for patients with particular problems to be seen in follow-up by their providers. How does this vary for specific chronic or recurring medical problems? Can realistic goals or standards be set for each visit? What are the appropriate intervals for routine care and can guidelines for such care be established? Cell #9 (Descriptor: Integrative Care; Element: Reaction to Change) The descriptor integrative care with this element raises issues concerning health care decision making. Can guidelines be responsive to patient preferences? Can they play a role in facilitating provider-patient communication,

OCR for page 133
146 SEITING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES particularly in explaining tile medical, epidemiologic, and prognostic information? Can they help providers explain risks and benefits to patients? Can guidelines serve a positive role in reducing medical-legal practices such as the perceived or actual need for malpractice actions? THE ROLE OF OUTCOMES IN GUIDELINE DEVELOPMENT As noted above, the three health system elements~context, interactions, and reaction to change) when combined with the principal descriptors of primary care (unselected patients, continuous and longitudinal care, and integrative care) form the basis for generating a series of generic topics for clinical guidelines development in primary care. These generic topics may be applied to specific health-related concerns as desired. For example, the treatment of hypertension could be selected as a topic for guidelines for specialist referral. Abdominal pain could be used as a symptom to define guidelines for when patients should seek care. The specific entities chosen for further study may be selected on epidemiologic, cost, academic, or theoretical bases. The proposed model for primary care guidelines development is, as yet, incomplete. The health care outcome resulting from the guideline must be viewed as an intrinsic part of the guidelines development process. The advantage gained is the linking of process and outcome in a systems approach to primary care. The measured outcome should be fed back in a continuous loop such that a modified guideline results. The general model suggested is depicted in Figure E.4. An example illustrates how the model in Figure E.4 works. When a primary care provider feels that a patient needs more specialized care, a referral is generally made. We know relatively little about the criteria for referral. When is a referral usually necessary? What kinds of communication should take place between the providers involved? How often should the patient to be seen by the specialist? How best can multiple referrals be coordinated? The issue of patient referral touches on the integration descriptor of primary care. From the point of view of the health care system, to interaction of the various caregivers form part of the health system framework. A topic for guidelines development ill this framework could be, "When should primary care providers refer a patient with hypertension to a specialist?" Following the selection of a topic, the types of primary care providers and specialists to be included in the guideline must be addressed. This is a complicated systems issue since the United States is unique in having a pluralistic system of primary care providers along with many specialist physicians who also deliver primary care type services. Since it is obvious that We competency, knowledge, patient base, arid legal basis for practice of all these

OCR for page 133
APPENDIX E Topic \ Primary Care | Descriptors ~ ) '1 ~ Health System Elements . <` \ OCR for page 133
148 SETTING PRIORITIES FOR CLINICALPRACTICE GUIDELINES The model suggested brings a dynamism into guidelines development by linking outcomes with the systems issues involved in primary health care delivery. A SPECIAL CONCERN REGARDING GUIDELINE DEVELOPMENT FOR CLINICAL PRACTICE Guidelines, despite their recent proliferation, have been viewed as largely voluntary and their use in clinical practice has not been well studied. What evidence does exist suggests that they often do not impact on clinical practice (Kosecoff, Kanouse, Rogers, et al., 1987; Lomas, Anderson, Domnick-Pierre, et al., 19891. A recent national survey of a random sample of members of the American College of Physicians is enlightening in this regard (Tunis, Hayward, Wilson, et al., 1994~. Only 18 percent of respondents reported that a change had occurred in their practice during the past year as a result of any guideline. Physicians' confidence in guidelines appears to be related to their familiarity with the organizations issuing the guideline, and physicians in fee-for-service environments were less positive about guidelines in general. The authors note that guidelines are perceived to have many different purposes and that "factors other than validity may influence which guidelines are followed." Creating guideline after guideline does not address their viability in clinical practice and may be an exercise in futility. This is particularly apt for primary care practice, given its diversity and broad scope. Short of enforcing guidelines through some form of regulation, more work is needed to develop strategies to enhance their implementation into clinical practice. The challenge here clearly goes beyond one of dissemination; it gets to the heart of why providers do what they do. The model presented in Figure E.4 suggests one approach: linking guidelines to medical outcomes. If some guidelines can be shown to lead to good (or better) outcomes, then the rationale for their use becomes more compelling and more easily translated into "standards of care." This further supports the recommendation that guidelines for primary care (or other areas) be programmatically linked to outcomes studies. In addition, specific research funds should be devoted to enlarging our understanding of medical practice so that attempts to influence this practice will have a greater chance of success. The funding for this research must be viewed as a legitimate part of the process of guidelines development. CONCLUSION The development of clinical guidelines for primary care practice would benefit from a systems approach. The key features that distinguish primary care

OCR for page 133
APPENDIX E 149 medical practice encompass a dynamic system that does not lend itself readily to a static (e.g., single-focused) analysis. As pointed out in Table E. 1, many of the guidelines that have been disseminated, or are currently under way or planned, may legitimately be viewed as falling within the boundaries of the primary care delivery system. A disease orientation for guidelines development will not work well for primary care practice, where clinical guidelines must reflect the deeper dynamics of health care delivery. A theoretical model for guidelines development in primary care is presented based on the interfacing of the essential descriptors of primary care practice with the health system elements that result in health care delivery for the patient. A series of generic topics can be derived from this model that may then be applied to specific health care issues related to the practice of primary care. These issues, which reflect the epidemiologic realities of primary care practice, may be readily found in the literature ~J.S. Department of Health and Human Services, 1990; Schappert, 19921. They include the common reasons for office visits by patients and the principal diagnoses given by primary care providers. A critical feature of this model is the direct incorporation of outcomes studies in a continuous feedback loop so as to facilitate further improvement and refinement of each guideline. This approach to primary care guidelines development reflects the "real world" of primary care practice, in which an imposing array of forces and contingencies ultimately result in a particular medical care outcome. In this model, guidelines arise not just from specific illnesses but from the broader characteristics that embrace the practice of primary care medicine. REFERENCES Aiken, L.H., Lewis, C,E,, Craig, J., Mendenhall, R.C., Blendon, R.J., and Rogers, D.E.. The Contribution of Specialists to the Delivery of Primary Care: A New Perspective. New England Journal of Medicine 300:1363- 1370, 1979. Alpert, J.J., and Charney, E. The Education of Physicians for Primary Care. Washington, DC: Department of Health Education and Welfare, 1973. Glenn,M.L. On Diagnosis. A Systemic Approach. New York: BrunnerlMazel, 1984. Holman, H.R. Qualitative Inquiry in Medical Research. Journal of Clinical Epidemiology 46:29-36, 1993. Institute of Medicine. A Manpower Policy for Primary Health Care: Report of a Study. Washington, DC: National Academy of Sciences, 1978. Kosecoff, J., Kanouse, D.E., Rogers, W.H., McCloskey, L., Winslow, C.M., and Brook, R.H. Effects of the National Institutes of Health Consensus

OCR for page 133
150 SETTING PRIORITIES FOR CLINICAL PRACTICE GUIDELINES Development Program on Physician Practice. Journal of the American Medical Association 258:2708-2713, 1987. Lipsitz, L.A., and Goldberger, A.L. Loss of"Complexity" and Aging. Potential Applications of Fractals and Chaos Theory to Senescence. Journal of the American Medical Association 267:1806-1809, 1992. Lomas, J., Anderson, G.M., Domnick-Pierre, K., Vayda, E., Enkin, M.W., and Hannah, W.J. Do Practice Guidelines Guide Practice? The Effect of A Consensus Statement On the Practice of Physicians. New England Journal of Medicine 321:1306-131 1, 1989. Schappert, S.M. National Ambulatory Medical Care Survey: 1991. Advance Data from Vital and Health Statistics. Number 230. Hyattsville, MD: National Center for Health Statistics, 1992. Senge, P.M. The Fifth Discipline. The Art and Practice of the Learning Organization. New York: Doubleday, 1990. Tunis, S.R., Hayward, R.S.A., Wilson, M.C., Rubin, H.R., Bass, E.B., Johnston, M., and Steinberg, E.P. Internists' Attitudes about Clinical Practice Guidelines. Annnals of Internal Medicine 120:956-963, 1994. U.S. Department of Health and Human Services. Healthy People 2000. National Health Promotion and Disease Prevention Objectives. Washington, DC: USDHHS, 1990.