Click for next page ( 34


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 33
Definitions of Key Terms "When I use a word," Humps Dumpy said, in rather a scornful tone, "it means just what I choose it to mean- neither more nor less." "~e question is," said Alice, "whether you can make words mean so many different things." "The question is," said Humpty Dumpy, "which is to be master that's all." Lewis Carroll, Through the looking Glass If AHCPR and the Forum are to proceed confidently with their mis- sion, they need clear, broadly acceptable definitions of four key terms, which were used in the legislation establishing the agency: (1) practice guidelines, (2) medical review criteria, (3) standards of quality, and (4) performance measures. Neither the final legislation nor preceding House or Senate bills offered definitions of these concepts. The four key terms employed in HERA 89 have been defined and used in quite disparate ways. Words like guidelines and standards may mean one thing to clinicians, another to purchasers, and yet another to attorneys. (The study directors for this project discovered early in their work that they were using these terms differently.) Moreover, the same person may use the same term differently in different contexts. Able 2-1 suggests how context influences usage and terminology. Further complicating the semantic and conceptual situation are other frequently used terms that are not mentioned in the legislation terms 33

OCR for page 33
34 TABLE 2-1 Influence of Context on Terminology CONICAL PRACTICE GUIDE~NES Context Influence Users may be practitioners, payers, or consumers. Common practitioner terms include Practice guidelines. and ~algorithms.. Consumer terms include Patient information (such as Postoperative instructions. For insurers, teens such as Medical necessity. and Standard practice. are used in discussions of health plan coverage, and lawyers refer to Community standards. in malpractice cases. Purposes Purposes may include practice management, practitioner or patient education] quality assessment, and payment determination. Purposes and users overlap to some extent, but even a single user such as a clinician may talk at one time about Protocols. or ~indications. for an intervention and at another time about Clinical indicators. or Occurrence screens. to flag potential problems for further review in quality assurance programs. Timing Prospective, concurrent, or retrospective uses may require veer different formulations. Practice management guidelines are intended to guide care prospectively, and Recertification criteria are used to review proposed care on a case-by~ase basis. Continued stay review has employed Length of stay norms.W Retrospective review may use ~screens. and analyses of practice patterns. Consensus A number of terminological distinctions have been based on the strength of scientific evidence or expert consensus about what is effective or ineffective medical care. One formulation uses ~standard,. Guideline, and adoptions to indicate decreasing levels of information about the likely outcomes of an intervention and decreasing levels of consensus about preferences for different outcomes. Other frameworks use gradations such as class I, II, and III to indicate conditions for which the application is justified, for which there is divergent opinion, and for which it is unjustified. Still other classifications use such teens as Approbate, "equivocal,. and ~inappropnate.~ Complexity At one extreme are joules of thumbs and ~dicta,. which are often stated very tersely. At the other extreme lie detailed algorithms, decision trees, and Criteria maps such as clinical indicators, pracizce parameters, norms, and practice policies (Meyer, 1989~. Some people may treat certain of these terms as synony- mous; others may make strong conceptual or practical distinctions. Given this abundance of terminology and meaning, it is helpful to recall what Donabedian (1981:409) said when discussing norms, criteria, and stan- dards: '~we have used these words in so many different ways that we no longer clearly understand each other when we say them. But we. . .do not have the liberty of abandoning them.... Our more reasonable course...

OCR for page 33
DEFINITIONS OF KEY TERMS 35 is to see whether we can clarify the existing nomenclature, barnacled and misshapen though it may be with the encrustation of careless past usage." Guided by the spirit of this observation, the committee's objectives are limited: to provide definitions that are clear, realistic, and practical for use with expert panels, contractors, and others. These definitions are not expected to serve all users for all purposes; however, they should permit clear communication between the Forum and the many organizations and interests with which it must work THE COMMITTEE'S APPROACH The committee began its definitional exercise by consulting the litera- ture on practice guidelines, quality of care, and related topics. This body of work contains a number of thoughtful efforts by practitioners, researchers, policy analysts, and others to define and explain key terms. In addition, the committee used standard English and American dictionaries, which provided common uses of the four terms. Congressional and agency staff were also consulted. In proposing definitions, the committee wanted to make its rationale and purposes explicit. ~ that end, it identified six desirable, although not fully compatible, characteristics of a definition. 1. Each definition should be parsimonious; that is, it should include the minimum necessary distinguishing characteristics of the concept in question. It should exclude elements that are not essential to make the definition clear and useful (for example, priorities for action, contingencies, or desired characteristics for "good" instances of the concept). 2. Each definition should be consistent with customary social and professional usage, insofar as reasonable given other objectives. 3. Each definition should be consisterll with legislative language, inso- far as reasonable. 4. Each definition should be practically and symbolically acceptable to important interests, insofar as reasonable. 5. Each definition should not be easily misunderstood or misused, insofar as reasonable. 6. A word or phrase should not be defined in teens of itself (for example, a practice guideline should not be defined using words such as "guiding medical practiced; the definition should not be tautolog~cal. With these criteria in mind, the committee reviewed definitions in common and professional use, developed draft definitions of terms, and then revised the definitions based on committee discussion. The following sections of the chapter describe the results of this process for each of the four key terms.

OCR for page 33
36 CLINICAL PRACTICE GUIDELINES PRACTICE GUIDELINES The committee began by defining the term guidelines because it appears to be the term most commonly and comprehensively used in professional and policy discussions. It is employed in this general way in the legislation establishing AHCPR. COMMON USAGE: THE DICTIONARY The Random House Dicizonary of the English Language (1987) dates the American origin of guideline to 177~1785, presumably in its literal usage as a "rope or cord that serves to guide one's steps especially over rocly terrain, through underground passages, etc." Its more metaphorical use- "any guide or indication of a future course of action" is a recent addition. The verb "to guide" is given several meanings including "to supply with advice or counsel." This dictionary notes that the term "implies continuous presence or agency in showing or indicating a course" as distinct from pilot, steer, escort, direct, or lead. The Compact Edmon of the Oxford English Dictionary (OEDj (1971) does not define guideline directly but includes it in a set of examples of attributes of technical appliances and machinery parts (e.g., guideline for a saw). It defines the verb "to guide" in much the same way as above: lo direct the course of (a physical action, for instance) and to lead in a course of action or the direction of events. PROFESSIONAL AND TECHNICAL USAGE American Medical Association (AMA) (J. Kelly, director of the Office of Quality Assurance, letter dated April 26, 1990): "Guideline: Recom- mendation for patient management which identifies a particular manage- ment strategy or a range of management strategies. Practice parameters: Strategies for patient management developed to assist physicians in clinical decision-making. " The Foam (`S. King, Forum director, personal communication, January 1990~: "Aguideline is a description of the process of care which will permit health to improve, and which has the potential of improving the quality of medical decision-making" (provisional definition). U.S. General Accounting Office (G. Silberman, assistant director, Pro- gram Evaluation and Methodology Division, letter dated October 16, 1989, to medical specialty societies): "[G]uidance by whatever name that aids practicing physicians and others in the medical community (and consumers, if included) in day-to-day decisions by describing the degree of appropri- ateness and the relative effectiveness of alternate approaches to detecting,

OCR for page 33
DEFINITIONS OF KEY TOWS 37 diagnosing, and/or managing selected health conditions" (working defini- tion). IOM (1990:304~: "Appropriateness guidelines describe accepted indi- cations for using particular medical interventions and technologies, ranging from surgical procedures to diagnostic studies." Joint Commission on Accred~tanon of Healthcare Organizations (1989a): A guideline gives "indications or contraindications for appropriate patient care.?' Physician Payment Review Commission: A) "Practice guidelines are standardized specifications for care, either for the use of a particular service [e.g., preventive screening] or procedure or for the management of a specific clinical problem" (1988a:134. (2) "[Practice guidelines refer to] formally developed guidelines based on the clinical research literature and the collective judgments of experts" (1988b:223~. (3) ". . Essentially clinical recommendations for patient care. They provide guidance to physicians and others who must make decisions. . .'' (1988b:223~. U.S. Preventive Services Task Force 1989:xx~rvii): "Recommendations appearing in this report are intended as guidelines, providing clinicians with information on the proven effectiveness of preventive services in published clinical research. Recommendations for or against performing these maneuvers should not be interpreted as standards of care but rather as statements regarding the quaky of the supporting scientific evidence." Mark Chassin (~1988~: "They "standards of care or practice guidelines] are statements describing specific diagnostic or therapeutic maneuvers that should or should not be performed in certain specific clinical circum- stances." David Eddy (forthcoming): "Pathway guidelines (protocols and algo- rithms) are intended to direct a practitioner along a preferred management path. Boundary guidelines (limits or criteria) are intended to define the limits of proper practice." In distinguishing different types of practice poli- cies (standards, guidelines, and options), Eddy states: "A practice policy is considered a guideline if the outcomes of the intervention are well enough understood to permit meaningful decisions about itS proper use, and if it iS preferred (or not preferred) by an appreciable but not unanimous majority of people." (Note in the section below that Eddy requires more stringent agreement for a standard.) Lucien Leape (1990:43~: Practice guidelines are "standardized specifi- cations for care developed by a formal process that incorporates the best scientific evidence of effectiveness with expert opinion."

OCR for page 33
38 CONICAL PRACTICE GUIDELINES THE COMMITTEE'S DEFINITION: PRACTICE GUIDELINE Practice guidelines are systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances. Practice guidelines assist patients and practitioners in malting decisions about health care. 1b that end, they describe clearly appropriate care, clearly inappropriate care, or care about which the scientific evidence and consensus are equivocal. Guidelines focus on specific clinical circumstances, which may sometimes include clinically relevant organizational factors, community characteristics, social variables, and similar influences on health care derive. I-hey should be developed in a formal, systematic way that is fully documented, as discussed in Chapter 3 of this report. As defined here, practice guidelines include such varied means of as- sisting clinical decisionmaking as pathway guidelines or practice algorithms, boundary guidelines or appropriateness criteria, and practice parameters. (Appendix B contains examples of different styles of guidelines.) The choice of approach in developing guidelines will depend on their purpose, the intended users and sites of care, and the clarity and quality of the scientific evidence on which they are based. Judged in terms of the principles the committee set forth for good definitions, this definition is reasonably succinct, generally consistent with common, professional, and legislative usage, and not mutological. The element of advice or counsel in the committee definition differentiates guideline from such terms as suggestion (which is weaker than advice or counsel), information (which is less goal oriented), and cntenon (which, as discussed below, adds the element of evaluation). The definition goes beyond the dictionary by adding the requirement that the advice or counsel be "systematically developed." This element is mentioned by many health care researchers and medical organizations either explicitly or implicitly through such phrases as "formal development" and "well understood" and references to the scientific literature. Based on its discussions in meetings and in other quarters, the com- mittee believes its definition of practice guideline is, for the most part, practically and symbolically acceptable lo varied groups. When combined with the other definitions offered here and the specification of desirable attributes of guidelines, the definition should be as little subject to misuse and misunderstanding as is possible, given the semantic diversity described 1 Clinically relevant organizational factom include the equipment, facilities, personnel, and skill levels available within a particular health care institution or local community. Clinically relevant social factom include the patient's home and family circumstances.

OCR for page 33
DEFINITIONS OF ~ TEAS 39 earlier. A few areas of potential controversy can be anticipated, however, and they are discussed below. Guidelines and the Strength of Evidence In and of itself, the committee's definition of practice guidelines does not explicitly require that guidelines describe the strength of the scientific evidence or consensus associated with a set of guidelines. Rather than being part of the definition, the committee views this proper as an attribute that good guidelines must have (see Chapter 3~. Every set of guidelines developed or adopted by the Forum should be accompanied by clear statements about their strength, and the evidence for such statements should be cited. When the evidence is extremely strong and professional judgment is virtually unanimous, the guideline may be treated as a standard of practice permitting few if any exceptions. When the evidence is equivocal, the guideline may only identify currently acceptable practice options.2 This use of the phrases standard of practice and practice option is consistent with Eddy's usage as described earlier. However, the committee's use of the term guideline is equivalent to Eddy's umbrella use of the phrase practice policy. The committee's primary reason for using guideline as the general label was that the term is used this way in the legislation establishing AHCPR and by other sources such as the Physician Payment Review Commission. Admittedly, the committee risks creating some confusion by using the phrase standard of practice, given the OBRA 89 reference to "standards of quality," but it judges that the risk is acceptable because the term is useful and, indeed, hard to avoid. Relation of Guidelines to Review Criteria and Other Evaluation Tools In the committee's reading, OBRA 89 does not use the terms standards of quaky, performance measures, and medical review criteria as synonyms for practice guidelines. Instead, it links these three terms to evaluating practice rather than to assisting practitioners and patients.3 Although the Forum is required to arrange for the development of these tools, other sponsors of guidelines may do no more than offer general observations on how the guidelines should or should not be used in evaluating practitioner decisions and outcomes. For various practical or technical reasons, some elements 2 Some health care organizations or health benefit plans may rule out certain of these options given their objectives, resource limits, or other constraints (Havighurst, forthcoming). Such deci- sions, however, are distinct from scientifically and professionally based judgments about practices that are acceptable options for clinicians. 3 See sections 912(a)~1) and (amp) of Title IX of the Public Health Service Act as amended by PA 101-239.

OCR for page 33
40 CLINICAL PRACTICE GUIDELINES of a set of guidelines may have no corresponding review criteria or other practice evaluation tools at all. For example, one guideline of the U.S. Preventive Services Task Force (1989:108) states that "screening for congenital hypothyroidism is recom- mended for all neonates during the first week of life." This statement and a subsequent one defining specific tests to be used could be directly translated into a criterion for reviewing care for a specific neonatal case or for reviewing the pattern of neonatal care provided by a practitioner or an institution. In contrast, the further statement by the Disk Force that `'it may be pnu den t to perform regular physical examinations of the thyroid in persons with a history of upper-body irradiation" (emphasis added) would be difficult to translate into a criterion for assessing either case-by-case or aggregate performance. Of course, even though certain statements about neonates could be used for medical review performance, determining that they should be so used requires a decision that the benefits of such a review in improving health or other desired outcomes warrant the administrative steps and costs that a review would entail. Ib cite a different example with respect to screening for hearing loss, the Preventive Services Disk Force (1989:198), recommends that "screening of workers for noise-induced hearing loss should be performed in the con- text of existing worksite programs and occupational medicine guidelines." Such a statement may have some value for patients or clinicians without itself generating a review criterion in the context considered in this report. Definition of Appropriate Care Because the concept of appropnate care or appropriateness is crucial to the committee's definition of practice guidelines and is itself the subject of some differences and confusion in usage, the committee considered it necessary to explicitly define this concept. Brook and his colleagues at the RAND Corporation define appropriate care as follows: when "the expected health benefit [exceeds] the expected negative consequences. . .by a sufficiently wide margin that the procedure [is] worth doing" (see, e.g., Park et al., 1986:6~. Conversely, care is inappropriate when the expected harms exceed the expected health benefits.4 Care may range from clearly appropriate to clearly inappropriate. Care that is not described by scientific evidence and expert judgment as either clearly one or the other may be termed equivocal. A practice might be 4Health benefits include increased life expectancy, beuer functional status, and reduced mor- bidity, pain, and anxiety. Negative health outcomes are the opposites of these qualities. Both short-term and long-term positive and negative outcomes should be examined to determine ap- propmateness.

OCR for page 33
DEFINITIONS OF^KEY TERMS 41 labeled equivocal for empirical reasons; that is, the benefits and harms are uncertain. It might also be described as equivocal because of ambivalence, for instance, when the benefits are known to exceed harms but to such a trivial or small degree that raters are reluctant to call the practice appropriate. In this latter case, value judgments may have much to do with whether a practice is considered "worth" providing (or receiving), and these judgments may rest on considerations of cost and other matters. Guidelines and Costs The committee's definition of appropriate care does not require that guidelines be based on judgments about the cost-effectiveness of particular clinical practices; neither does it preclude it As discussed in Chapter 3, the committee concludes that, insofar as feasible, developers of guidelines should consider costs and should include information with the guidelines that allows others to make their own cost-benefit or cost-effectiveness judg- ments. The committee's decision not to incorporate an explicit reference to costs in the definition of practice guidelines or appropriate care reflects a value judgment that was not shared by all committee members. The majority, however, believed that the emphasis on clinical decisionmaking should be paramount. In addition, some committee members strongly disagreed with the committee's decision not to refer explicitly to third-party payers and oth- ers in the definition of practice guidelines. As a prachca1 matter, OBRA 89 requires that the needs of the Medicare program and peer review or- ganizations be considered by AHCPR in selecting topics for guidelines development, encouraging the dissemination and use of guidelines, and evaluating their impact. In addition, the agency must arrange for the de- velopment of medical review criteria and other practice evaluation tools. These factors appear sufficient to ensure that the needs of payers, con- sumer groups, and similar parties will be addressed during the guidelines development stage. EVALUATION INSTRUMENTS As defined above, practice guidelines are meant to assist patients and practitioners in making health care decisions. Medical review criteria, standards of quality, and performance measures, which the committee groups together as practice evaluation instruments, are designed to assist health care organizations, payers, and others (including practitioners and payers themselves) in evaluating those decisions and health outcomes. Sometimes such evaluations will focus on individual instances of care (for example, to determine whether a hysterectomy is appropriate for a patient

OCR for page 33
42 CLINICAL PRACTICE GUIDEl rNES with specific clinical characteristics); at other times they may focus on patterns of care (for example, to compare rates of hysterectomies for groups of patients with similar clinical characteristics). Because the committee was not asked to provide principles for trans- lating guidelines into evaluation instruments, it did not give in-depth con- sideration to the issues the Forum may face in moving from a given set of practice guidelines to the corresponding evaluation instruments. The committee did not, for example, discuss whether different principles and translation strategies might be needed for guidelines that focus heavily on nursing care compared tO those that mainly emphasuze care provided by physicians. The recent IOM report on quality assurance in the Medicare program (1990) discusses some of the attributes that good review criteria should have. Given the expectations that guidelines will be used to improve the quality and effectiveness of health care, the task of clearly translating guidelines into evaluation tools is a critical one that needs to be considered during the process of guidelines development rather than at its end. At this early stage in the Forum's work, however, it may not be feasible for each set of guidelines to be accompanied by all three types of evaluation instruments.5 As the Forum and its expert panels tackle specific decisions about evaluation instruments, they may well suggest some adjustments in the definitions offered below. This seems a possibility in particular for the definition of standards of quality and the provisional definition of performance measures, both of which gave the committee substantial difficulty in their formulation. MEDICAL REVIEW CRITERIA COMMON USAGE: THE DICTIONARY According to the Random House Diciionary, a criterion is "a standard of judgment or criticism; a rule or principle for evaluating or testing something." The OED offers three definitions: an organ, faculty, or instrument of judging; a test, principle, rule, canon, or standard, by which anything is judged or estimated; and a distinguishing mark or characteristic, -attaching to a thing, by which it can be judged or estimated. 5 For the initial set of conditions for which guidelines are due by January 1, 1991, OBRA 89 is somewhat inconsistent about what is required of the Forum. Section 912(d) of the law (part of the amendments to the Public Health Service Act) calls for the development of guidelines, standards, performance measures, and review criteria, whereas section 1142(a)~3~(A) requires only guidelines.

OCR for page 33
DEFINITIONS OF KEY TERMS PROFESSIONAL AND TECHNICAL USAGE 43 AAL4 (J. Kelly, director of the Office of Quality Assurance, letter dated April Z6, 1990~: Review criteria are "bases for evaluating quality and appropriateness of medical care." IOM (1974:28~: " 'Criteria' are identifiable elements of care used to judge the appropriateness of that care...." Professional Standards Review Organzzanon fPSROJ Manual (cited in Donabedian, 1981~: Medical care criteria are predetermined elements against which aspects of the quality of medical services may be compared. Avedis Donabedian (1981:411~: Although he apparently wishes to avoid use of the term cntenon, Donabedian notes that it could usefully be employed to mean "an element or attribute that is to be used in evaluation, and that often comes accompanied by an explicit or implicit norm. . .or, even, by a standard." Elsewhere (p. 410), he says criteria are "discrete, clearly definable, and precisely measurable phenomena that belong within the categories of process or of outcome, and that, in some specifiable way, are relevant to the definition of quality." Kathleen Lohr and Robert Brook (1981:761-762~: "Quality-of-care cri- teria are commonly characterized in several ways. One is whether they are explicit (i.e., stated in detail in advance) or implicit (i.e., based on judgments of experts without direct reference to a priori standards).... Criteria may be thought of as minimal, average, or ideal.... [M]inimal criteria place a base under which the care-giver should not fall, lest the individual be judged as practicing severely substandard care. Those who fall below this level might well be classified as outliers. Average criteria represent a community standard; they are often thought of as a range within which care Is considered acceptable. The lower bound [of average criteria] is likely tO be the minimal point.... The upper bound is likely to be a fairly high, 'ideal' point toward which the profession may simply aspire." Heather Palmer and Miriam Adams (forthcoming): Criteria are "prede- termined elements of care against which aspects of the quality of a medical service may be compared." Prescriptive criteria are "criteria written by asking providers tO decide what ought to be done in a given circumstance Physicians writing such criteria have been found sometimes to specie care at an unrealistically high level. Also called normative critena." (emphasis in the original) Hannl1 Vu on (1989:156~: "Criteria of good care are structural elements of care or procedures to be performed and information to be gathered when treating a patient with a given problem. . .or outcomes of care." Moreover, Vuori states that "criteria. . .identi~ the most important elements of good care...."

OCR for page 33
44 Cr. rlvIcAL PRACTICE GUIDELINES THE COMMITTEE'S DEFINITION: MEDICAL REVIEW CRITERIA Medical review criteria are systematically developed statements that can be used to assess the appropriateness of specific health care decisions, services, and outcomes. For medical review criteria, the committee definition stresses the eval- uanon of health care processes and outcomes rather than assistance to practitioners and patients in making decisions. The definition offered here is reasonably consistent with legislative language and most professional and common usage. It is also reasonably succinct and not tautological. As noted earlier in this chapter, the committee's definition of appropriate care does not require that judgments be made about the cost-effeciiveness of particular clinical practices.6 Neither does the definition preclude it. Medical review criteria have many different uses and users. They may be used (1) prospectively, for example, to review a proposed hospital admission or surgical procedure; (2) concurrently, for instance, to assess the need for continued hospitalization; or (3) retrospectively, for example, to make decisions about insurance claims. Critena-based reviews may focus on patterns of practice or on individual cases of care. For example, reviewers concerned primarily with assessing the quality of care may concentrate on retrospective analyses of patterns of care; they may also rely heavily on case-finding screens applied retrospectively to identify individual potential problems for further evaluation using more detailed criteria (IOM, 1990~. Users more concerned with cost management have increasingly emphasized review criteria that can be applied prospectively on a case-by~ase basis to certain relatively expensive procedures (IOM, 1989~. Traditional criteria for judging the process of care most often specify only the things that should be done and ignore things that should not be done. In contrast, criteria used for insurance clanns review tend to screen for the inappropriate service rather than list everything that is appropriate. (Claims review may also include a variety of nonclinical matters, such as whether the general category of service, setting, aIld provider was covered under a patient's benefit plan and whether the service was correctly coded.) As noted earlier, some practice guidelines may translate into or be used as medical review criteria in a straightforward fashion just as some foreign words and phrases for example, caveat emptor are easily used by English speakers). For practical or technical reasons, other guidelines may be more difficult or less suitable for use in this way. Of the initial conditions being 6A recent IOM report (1989) noted that most private payers and review organizations claim that their retrospective reviews of care for specific patients focus on clinical factom, not costs, and that their emphasis is on detecting clearly inappropriate care.

OCR for page 33
DEFINITIONS OF KEY TERMS 45 considered by the Forum for guidelines development (see the list presented in Chapter 1), the agency may find, for example, that medical review criteria are easier to develop from the cataracts guidelines than from the guide- lines for treatment of depression in the community setting. Similarly, the guidelines for managing incontinence may give rise to criteria for reviewing quality but none for making cavment decisions. ~ ~ , STANDARDS OF QUALITY The term standards is particularly difficult to clanfy. In common parlance it is often used synonymously with criteria; in the quality assurance lexicon, however, it has rather different connotations. Even in everyday use, standard has many different senses. COMMON USAGE: THE DICTIONARY The Random House Dictionary offers more than 25 meanings including the following: "1. something considered by an authority or by general consent as a basis of comparison; an approved model. . . 3. a rule or principle that is used as a basis for judgment. . . 4. an average or normal requirement, quality, quantity, level, grade, etc.... 24. of recognized excellence or established authonty; and 25. usual, common, customary." This dictionary suggests that a standard differs from a criterion in that the former implies a model against which the quality or excellence of other things may be determined ("she could serge as the standard of good breeding") whereas the latter does not ('~wealth is not a criterion of a person's worth"~. The OED sets forth the following different meanings and definition for standard (some of which are centuries old): "1. the exemplar of a measure or weight; 2. a normal uniform size or amount, a prescribed minimum size or amount; 3. an authoritative or recognized exemplar of correctness, perfection, or some definite degree of any quality (which can be construed as a rule, principle, or means of judgment or estimation, or a criterion or measure); and 4. a definite level of excellence, attainment, wealth, or the like, or a definite degree of any quality, viewed as a prescribed object of endeavour or as the measure of what is adequate for some purpose." The most troublesome aspect of common usage is the use of standard in quite different ways to describe either a minimum acceptable state or a state of high achievement and excellence. The same difficulty also shows up in the health services literature. 1

OCR for page 33
46 PROFESSIONAL AND TECHNICAL USAGE CONICAL PRACTICE GUIDELINES A~4 (J. Kelly, director of the Office of Quality Assurance, letter dated April 26, 1990~: A standard Is an "accepted principle for patient management." IOM (1974:283: "A'standard' is the degree of adherence to the defined criteria." Joint Commission on Accreditation of Healthcare Organizaiions (1989a): "expectations or rules relating to the structures or processes necessary for delivery of patient care." PSRO Manual (cited in Donabedian, 1981~: "Standards are profes- sionally developed expressions of the range of acceptable variation from a norm or criterion." Mark Chasszn (19~:43~39~: Chassin refers to "standards of care or practice guidelines" and goes on to say that "whey can also define several levels of quality of care, from the best attainable to a minimal level below which care is unacceptably poor. They can also be applied to groups of patients as well as to individuals." David Eddy (forthcoming): In distinguishing standards, guidelines, and options for "practice policies," Eddy makes the following point: "Apractice policy is considered a standard if the health and economic consequences of an intervention are sufficiently well known to permit meaningful decisions and there is virtual unanimity among patients about the desirability (or undesirability) of the intervention, and about the proper use (or nonuse) of the intervention." (That is, a standard is the most stringent form of a practice policy.) Heather Palmer and Miriam Adams (forthcoming): Standards are "pro- fessionally developed expressions of the range of acceptable variation from norms or criteria." Pahner goes on to say "commonly, all-or-nothing stan- dards (for example, 100 percent or 0 percent compliance required) are preferred. This is coupled with peer review of all cases that vary from the criteria." Virgil SZee (1974~: "The desired achievable (rather than the observed) performance or value with regard to a given parameter." (emphasis in orig- inal) Parameter is then defined as "an objective, definable, and measurable characteristic of the patient himself or of the process or outcome of his care. Each parameter has a scale of possible values...." Hanr~u Muon (1989:156~: "A standard is the value of a criterion mat indicates the border between acceptable and poor quality. Standards can be normative or empirical.7 A normative standard equates the acceptable performance with either the best imaginable performance or the best at- tainable performance under ideal circumstances. Empirical standards are 7Connotations derived from "norm" and "normative" cause endless problems. Some experts,

OCR for page 33
DEFINITIONS OF ~ TESS 47 often based on experiences gained in 'typical' situations." In addition, Vuori states that "standards. . .indicate which level [of a criterion] should be reached...." THE COMMITTEE'S DEFINITION: STANDARDS OF QUALITY Standards of quality are authoritative statements of (1) minimum levels of acceptable performance or results, (2) excellent levels of performance or results, or (3) the range of acceptable performance or results. Given the contradictory uses of the term, the committee decided it was pointless to try to overrule them. Instead, it takes the position that clarity about the nature of a standard is the key issue: does the standard cleary state that it sets a minimum level of performance or a level of excellence or a range? Preferably, specific labels such as "minimal standards" or "standards of excellence" should be employed. Defining standards of quality will be a major enterprise. Another IOM report, Medicare: A Strategy for Squalid Assurance (1990), provides an extensive overview of quality assurance activities, methods, and problems and recommends a Midyear strategy for improving the countIy's quality assessment and assurance efforts and results. In that report, quality of care is defined as "the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge" (p. 4~. One point made about this definition is that it "encompasses a wide range of elements of care." This suggests that standards of quality should be viewed primarily as mea rats for assessing patterns of care, including patterns that may extend across a range of clinical conditions, settings, and practitioners. Advice about whether the Forum should emphasize minimum stan- dards or standards of excellence or ranges is beyond the scope of this report. Among other things, such advice would entail an analysis of al- ternative views about how to assure quality of care. For example, the concept of minimum standards seems more consistent with the structure- processmutcome model of quality assurance than the continuous qualipr improvement model. The 1990 IOM report cited above discusses these models at some length, but that discussion provides no simple path for this committee to follow in specifically advising the Forum. Such advice including Vuori and Palmer, see norms as ideal or preferred states. Others treat norm and "av- erage" as essentially equivalent, and they equate the normative and the empirical because em- pirical data are needed to determine "not As." A 1974 TOM report, for instance, defined "norm" as `'merely a statistical average" (p. 28~.

OCR for page 33
48 CLINICAL PRACTICE GUIDELINES might also be premature, given that the Department of Health and Human Services and Congress are considering a varied of policy and strategy issues related to broad questions of quality assurance and assessment. Decisions on these issues presumably should influence the Forum's wore PERFORMANCE MEASURES This concept is the most unfamiliar of the four terms in OBRA 89. It is seldom used in the professional literature and has a variety of dictionary definitions. COMMON USAGE: THE DICTIONARY Random House offers for performance "the manner in which or the efficiency with which something reacts or fulfills its intended purpose." The TED provides two relevant definitions for performance: "1. the carrying out of a demand, duty, purpose, promise, etc.; 2. the accomplishment, execution, . . .of anything ordered or undertaken." The OLD gives many different definitions for measure. As a verb, these include "1. to form an estimate of (now especially to weigh or gauge the character or ability of something [e.g., a person]), with a view to what to expect [from that person]; 2. to ascertain or determine the spatial magnitude or quantity of "something] by application or comparison to some known or fixed unit; 3. to estimate the amount, duration, value, etc., of an immaterial thing [perhaps, performance] by comparison with some standard; 4. to judge or estimate the greatness or value of (a person, a qualify by a certain standard or rule; and to appraise by comparison with something else." As a noun, the definitions of measure include "1. an instrument for measuring [e.g., a vessel, graduated rod, line, tape, etc....~; and 2. a method of measuring, especially a system of standard denomination of units of length [etc.~." PROFESSIONAL AND TECHNICAL USAGE The term performance measure has been used by the Joint Commission on Accreditation of Healthcare Organizations, but the commission intends, in the future, to substitute the term indicator (OR. Marder, project manager, Indicator Development, personal communication, April 24, 1990~. Com- mittee staff found virtually no discussion of performance measures in the literature they reviewed. Joint Commzssion (1989b:~5~: "Performance measures provide data and information that serge as the basis for determining whether expectations are met." A clinical indicator is a "generic term. .intended to emphasize

OCR for page 33
DEFINITIONS OF ~ TEAS 49 the primary importance of (outcome) measures as monitors of performance. iThey] describe relatively precise measures of important aspects of care. . [They] are not clinical standards or practice guidelines." David Eddy (forthcoming): In discussing the distinction between prac- tice policies (see the earlier section) and performance policies, he notes that performance policies address the issue of ensuring that "whatever in- terventions are used, are used correctly.... They are intended to guide or review the performance of interventions, without concern for whether the interventions should have been performed in the first place" THE COMMITTEE'S PROVISIONAL DEFINITION: PERFORMANCE MEASURES Performance measures are methods or instruments to estimate or monitor the extent to which the actions of a health care practitioner or prouder conform to practice guidelines, medical review criteria, or standards of quality. Given the multiple common meanings of the terns and the paucity of professional and technical usage, the committee decided to oDer only a provisional definition of performance measures. As practice guidelines and related evaluation instruments are more widely developed and used, a different or more specific meaning for this term may emerge. The committee could have considered performance measures to be a combination of standards and criteria-measures of whether a given "performance" as judged by some set of criteria meets (or exceeds or falls short of) a standard. Considered in this way, the term verges on being synonymous with the other terms defined in this chapter. Therefore, the committee chose provisionally to view performance measures as instruments (for example, questionnaires, abstracting forms, measurement scales, or computer programs) for recording or measuring data on performance. In this sense, a distinction is made between measuring and judging. Thus, a performance measure is more like a bathroom scale than it iS Me a table of recommended weights by gender and body build. In sum, it is probably valuable to maintain a distinction between the instrument or means of measuring something and the judging of the results. Furthermore, performance measures should be sensitive to what is not done as well as what is done. CONCLUSIONS AND SUMMARY In this chapter, the committee has proposed definitions of the four key terms used in the legislation establishing the Office of the Forum

OCR for page 33
so CONICAL PRACTICE GUIDELINES for Quality and Effectiveness in Health Care. The committee sought definitions that were, insofar as possible, parsimonious, consistent with common and professional usage, practically and symbolically acceptable to important interests, not easily misused, and not tautological. The list below recapitulates the committee's definitions of practice guidelines, medical review critena, standards of quality, and performance measures. PRACTICE GUIDELINES are systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances. MEDICAL REVIEW CRITERIA are systematically developed state- ments that can be used to assess the appropriateness of specific health care decisions, services, and outcomes. STANDARDS OF QUALITY are authoritative statements of (1) minimum levels of acceptable performance or results, (2) excellent levels of performance or results, or (3) the range of acceptable performance or results. PERFORMANCE MEASURES (provisional) are methods or in- struments to estimate or monitor the extent to which the actions of a health care practitioner or provider conform lo practice guidelines, medical review cntena, or standards of quality. For other organizations and other purposes, different terms and defi- nitions may serve better as long as they are clearly stated. This committee neither expects nor unlashes to dictate terminology Its objective is modest: to provide clear definitions of key terms that the Forum can use in carrying out its legal responsibilities with a minimum of confusion or provocation. Definitions, however, are only the starting point. The Forum also needs to distinguish good guidelines from bad and to communicate its expectations to contractors, expert panels, and others. The next chapter is a first step in that process. REFERENCES Chassin, M.R Standards of Care in Medicine. Inquay 25(Wlnter):437050, 1988. Compact Edition of the Oxford English Dictiorwy. Oxford, England: Oxford University Press, 1971. Donabedian, A. Criteria, Noes and Standards of Quality: What Do They Mean? American Journal of Public Health 71:40g 412, 1981. Eddy, D. A Manual 1~ Assessing Health Practices and Designmg P~racizce Policies (draft dated May 31, 1989~. American College of Physicians, forthcoming. See also the articles cited in Chapter 1 of this report. Havighurst, C. Practice Guidelines for Medical Cam: The Policy Rationale. St Louis University kzw lownal, forthcoming. Institute of Medicine. Advancing the 01i~y of Health Care. Washington, D.C.: National Academy Press, 1974.

OCR for page 33
DEFINITIONS OF KEY TERMS 51 Institute of Medicine. Controlling Costs anti Changing Panent Care? The Role of Utilization Management, B. Gray and M. Field, eds. Washington, D.C.: National Academy Press, 1989. Institute of Medicine. Medicare: A Strategy for (dually Assurance, vole. 1 and 2, K Lohr, ea., Washington, D.C.: National Academy Press, 1990. Joint Commission on Accreditation of Healthcare Organizations. A National Invitational Forum on Clinical Indicator Development, Chicago, March 1989a. Joint Commission on Accreditation of Healthcare Organizations. Statement before the House Subcommittee on Health, Committee on Ways and Means, concerning the Medical Care Quality Research Act of 1989. Washington, D.C., May 24, 1989b. Leape, L Practice Guidelines and Standards: An Overview. Malay Review Bullean 16:42A9, 1990. Lohr, K.N., and Brook, R.H. Quality Assurance and Clinical Phalmapy: Lessons from Medicine. Drug.Intellit~nce and Clinical Pharmacy 15:75~765, 1981. Meyer, H. Medicine Debates Parameters (Or Are They Guidelines?) American Medical News, December 15, 1989, p. 36. Palmer, R.H., and Adams, M. Considerations in Defining Quality of Health Care. In Perspectn~es on (duality Assurance. Ann Arbor, Mich.: Health Administration Press, forthcoming. Park, R.E., Fink, A., Brook, R.H., et al. Physician Rating; of Appropriate Indications for SO Medical and Surgical Procedures. R-3280-CWF/HF/PMT/RWJ. Santa Monica, Calif.: The RAND Corporation, 1986. Physician Payment Review Commission. Improving the Quality of Care: Clinical Research and Practice Guidelines. Appendix 1, Background Paper for Conference, October 1988; draft dated September 28, 1988a. Physician Payment Review Commission. Increasing Appropriate Use of Services: Practice Guidelines and Feedback of Practice Patterns. Chapter 13 in Annual Report to Confess. Washington, D. C, 1988b. Random House Dictionary of the English L~n`yAage, 2nd ea., unabridged. New York: Random House, 1987. Slee, V. PSRO and the Hospital's Quality Control. Annals of Internal Medicine 81:97-106, 1974; cited in Donabedian (1981~. U.S. Preventive Services Ask Force. Guide lo Clinical Preventive Services: An Assessment of the Fif~ectiveness of 169 Interventions. Baltimore, Md.: Williams & Wilkins, 1989. Vuori, H. Research Needs in Quality Assurance. Malay Assurance in Health Care 1:147-159, 1989.