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 243
Guidelines for Clinical Practice: From Development to Use A Examples of Clinical Practice Guidelines and Related Materials1 This appendix, which is a collection of clinical practice guidelines and related materials,2 has three main purposes. First, for readers not familiar with guidelines, it presents samples that may make the text of this report more concrete. Second, it illustrates how guidelines can differ. Third, the appendix discusses the topic of formatting, which is in some ways a step between development and implementation. Apart from the sponsoring agency or organization, guidelines can vary (as noted in Chapter 1) in at least five key ways: Clinical orientation—whether the chief focus is a clinical condition, a technology (broadly defined), or a process. Clinical purpose—whether they advise about screening and prevention, evaluation or diagnosis, aspects of treatment, or other dimensions, or more discrete aspects of health care. Complexity—whether the guidelines are relatively straightforward in presentation and discussion or are marked by considerable detail, complicated logic, or lengthy narrative and documentation. For purposes of the descriptions in this appendix, complexity is indicated simply as high, medium, or low. 1 This appendix was compiled chiefly by Holly Dawkins, the IOM research assistant for this study. 2 For purposes of simplification, the term guideline is used quite broadly, and it encompasses materials that do not fit neatly into IOM's definition of practice guidelines.
OCR for page 244
Guidelines for Clinical Practice: From Development to Use Format—whether the guidelines are formatted as free text, tables, if-then statements, critical pathways, decision paths, or algorithms. Intended users—whether they are intended for practitioners, patients, or others. The next section of this appendix discusses the ways guidelines may be formatted. The rest of the appendix presents, in whole or in part, 16 guidelines and related items (see Table A-1). Each example is preceded by an annotation indicating the principal information for the five variables just noted. In addition, a brief introduction highlights especially salient points about the item in terms of purpose, content, or presentation. These notes should in no way be considered a complete analysis or evaluation of the item in the example. At the end of each write-up is the complete reference or citation to the guideline. Inclusion in this appendix does not imply endorsement of the content of these guidelines or of the process by which they were developed. Some of these materials are not, for example, the products of a systematic develop TABLE A-1 List of Examples, by Main Purpose Screening and Prevention 1. Screening for diminished visual acuity in children 2. Vaccination for pregnant women who are planning international travel Diagnosis and Pre-Diagnosis Management of Patients 3. Triage of the injured patient 4. Evaluating chest pain in the emergency room 5. Using erythrocyte sedimentation rate tests in diagnosis Indications for Use of Surgical Procedures 6. Indications for carotid endarterectomy 7. Indications for percutaneous transluminal coronary angiography 8. Managing labor and delivery after previous cesarean section Appropriate Use of Specific Technologies and Tests as Part of Clinical Care 9. Using autologous or donor blood for transfusions 10. Detecting or tracking deteriorating metabolic acidosis Guidelines for Care of Clinical Conditions 11. Using oral contraceptives to prevent pregnancy and manage fertility 12. Deciding on treatment for low back pain 13. Managing patients following coronary artery bypass graft 14. Guidelines for the management of patients with psoriasis 15. Acute dysuria in the adult female 16. Management of acute pain
OCR for page 245
Guidelines for Clinical Practice: From Development to Use ment process. Others may result from such a procedure but do not include references to the scientific literature used in development. FORMATTING GUIDELINES Formatting is a step beyond the development of a guideline. It can be executed in many ways and at many stages in the process of moving guidelines from development to application. Congress, for instance, recognized the importance of formatting by requiring the Agency for Health Care Policy and Research (AHCPR) to present its guidelines in formats that are appropriate for use by practitioners, medical educators, and medical care reviewers. Many formatting activities of most relevance to the persuasive and effective presentation of guidelines may occur after the initial formatting and dissemination of a set of guidelines by its sponsor or developer. For example, a professional society may initially present its guidelines in one format in a journal but then rework them into another format for use in continuing medical education. The initial guidelines may also be converted by target users--—hospitals, clinics, utilization review firms, health maintenance organizations, patient groups, and others—into formats ranging from a sophisticated computer-based algorithm to a simple chart that the patient can put on the refrigerator door or bathroom mirror. Voluntary associations such as the American Cancer Society and American Heart Association, as well as commercial firms, may reformat guidelines in various ways for dissemination to different groups. Guidelines can vary quite dramatically, both logically and graphically, in their modes of presentation. The major approaches are free-text and formalized presentations, including if-then statements, algorithms, flowcharts, and decision trees. More recently, some clinical researchers and medical informatics experts are moving to more complex computer-based approaches. These approaches are discussed briefly below. Free Text The most common format for guidelines is free text; this report, for example, is presented in free text. Generally, free text is the starting point for most other formats and itself has many variants. Shortened free-text versions of guidelines documents can be tailored for specific uses by specific types of practitioners. For instance, guidelines for the diagnosis and management of acute myocardial infarction might well be rendered into several different versions and formats depending on whether the target audience was to be emergency room physicians needing quick reference or cardiac specialists managing a patient over several weeks.
OCR for page 246
Guidelines for Clinical Practice: From Development to Use Narrative information may be collapsed into tables or graphs or summarized in highly technical terms with liberal use of acronyms, abbreviations, or symbols.3 To cite one example of a series of formatting steps, the U.S. Preventive Services Task Force took its guidelines to a commercial publisher, which created and copyrighted the graphic design and produced the guidelines for general publication. In addition to the basic text, that publication includes eight small, plasticized pull-out charts summarizing screening, counseling, and immunization schedules for different patient age groups. These tables include abbreviated references to nonroutine situations for which physicians may consider deviations from the schedules. Guidelines can also be rendered into much simpler, less technical documents for use by patients, consumers, and their families. Although the free-text approach is likely to be preserved, some use may be made of lay terms (including colloquialisms), simple drawings, and such heuristic devices as introductory questions. One would expect such versions of guidelines to differ, depending on the target audience. Formalized Presentations Apart from translations into shorter, simpler, nontechnical versions, which may still be in a free-text format, guidelines documents may be formatted into stylized graphic representations such as flowcharts or decision trees. These, in turn, may become programs for computer-based clinical decision making tools. One reason for translating free text into other formats is that some guidelines identify dozens, if not hundreds, of specific criteria for care; even creative free-text presentations may not allow practical, quick access to this volume of information. Instead, the free text may be reconfigured (or, less often, the initial guidelines may be drafted) using various related or overlapping formal approaches including flowcharts, decision tables, and if-then rules. 3 Guidelines may be translated from English into other languages to reach users, particularly patients and families, who do not use standard English comfortably. For instance, AHCPR plans a Spanish language translation, for at least the consumer versions, of certain of its guidelines. Guidelines may also diffuse internationally. Translation of guidelines into other languages may pose both technical and cultural difficulties because some terms and concepts may not have counterparts in other languages. For example, discussions of patient preferences and informed consent reflect policy and ethical concepts that are not equally salient across nations (see Miller, 1987). Not surprisingly, given the confusion over terminology in this country, it is also not immediately obvious how best to translate the term guidelines into other languages.
OCR for page 247
Guidelines for Clinical Practice: From Development to Use Algorithms Typically, a preliminary step in the development of many such formats is the development of a clinical algorithm. Algorithms were invented in the ninth century by a Persian mathematician, Al-khaforizmi, to solve arithmetic problems. They were first applied to practical problem solving by the U.S. Army in job-training manuals, and they have been fundamental to the programming of electronic computers. More specifically in the health field, algorithms have been used since at least the 1960s to aid clinical problem solving (Gottlieb, 1990). Strictly speaking, an algorithm is not a graphic representation but rather a presentation of information for decision making using step-by-step conditional logic rather than ordinary prose or lists of factors to be considered. This distinction is frequently ignored. The clinical algorithm, even when not used as the format of choice for disseminating guidelines, may be used to compare guidelines and to identify missing or conflicting decision "branches." In developing methods for analyzing and comparing guidelines, Margolis et al. (1991) have identified categories of logical error in guideline construction and have described the complexity of different guidelines in quantitative terms. Applying such a process could lead to the significant reworking of a set of guidelines, not just a repackaging or interpretation of the same content. Constructing algorithms and translating them into flowsheets and other tools can be a powerful learning process for practitioners. The process can (1) highlight differences in practice patterns and values that may need to be explored, (2) clarify key characteristics and weaknesses of processes of care, (3) identify gaps in clinical knowledge, and (4) contribute to redesign of systems of care. Free-text versions of guidelines or review criteria usually precede the development of algorithms and similar formats, but one exception is worth noting. The Health Standards and Quality Bureau of the Health Care Financing Administration (HCFA) is developing quality-of-care and appropriateness algorithms for collecting and analyzing clinical data in its Uniform Clinical Data Set (UCDS; Krakauer, 1990; Krakauer and Bailey, 1991). The agency believes that the application of these computer-based algorithms will be a major tool for quality review for the Medicare peer review organizations (PROs), improving on the manual review of hospital charts by PRO nurse reviewers. The UCDS development process began in the late 1980s with more than 3,000 complex software algorithms for direct collection of data and identification of hospital admissions that did not meet certain admission or quality-of-care criteria. However, little documentation of the programming or clinical logic was performed. Consequently, the procedures and rules against
OCR for page 248
Guidelines for Clinical Practice: From Development to Use which practitioners and hospitals may be judged to have provided poor or unnecessary care cannot be easily explained to clinicians or institutions. HCFA has sponsored some work to prepare free-text versions of the fraction of these algorithms that can generate "flags" about potential quality-of-care deficiencies. Flowcharts and Similar Formats Graphic representations of algorithms are of various kinds. They include flowcharts, decision tables (sometimes used to indicate appropriate health screening), protocol charts (e.g., for handling medical problems by telephone), and so-called influence diagrams (a decision clarification tool imported from the business world and now being adapted to medicine). As used to assist clinical problem solving, flowcharts (which are commonly called algorithms) begin with a clinical condition or patient symptom and lead the reader through a series of branching, dichotomous choices based on the patient's risk status, medical history, or clinical findings. They also include action steps such as testing, treating, or scheduling further examinations. This appendix presents two flowcharts, one for patients (Example 12 on the treatment of low back pain) and one for practitioners (Example 15 on the treatment of dysuria). In clinical practice, flowcharts may help practitioners choose from among alternative actions the most efficient sequence (as in a diagnostic workup); they may also aid in reducing the likelihood of overlooking uncommon but important elements of care for specific patients. The basic elements of flowcharts include boxes and arrows; the latter connect the boxes or direct the user to other parts of the algorithm. The boxes may be numbered and have internal text, and they may come in several shapes, depending on whether they describe a clinical state, ask a question (diagnostic assessment), or describe an action to be taken. Prose may be used to annotate the boxes. Experts recommend that flowcharts read from left to right and top to bottom and that only two arrows exit from a given box, corresponding to a yes or no response to the clinical question posed. (Further rules for creating flowcharts and algorithms are proposed below.) Practitioners differ widely in their attitude toward flowcharts, decision trees, and other shorthand, visual formats for guidelines. Some see them as helpful reminders and a useful tool for assimilating or quickly locating information. Others are emphatic in their dislike of these formats, maintaining that the sequences do not represent the experienced clinician's mental processes or that the necessary complexity makes them impractical to use during care of a patient. Alternatively, they may argue that simplified formats do not adequately account for all the factors present in patient care.
OCR for page 249
Guidelines for Clinical Practice: From Development to Use Some practitioners are concerned that algorithms make patient care appear cut and dried and that they will become a series of rules to be applied, with mindless rigidity, by those without clinical expertise. As can happen with any format, some physicians may view any guidelines for familiar clinical conditions as insults that imply that without such a guideline they would not know what care to provide. Proposed Standards As noted, the clinical algorithm map, a type of practice guideline, has received increased attention in recent years and appears more frequently in various medical journals. However, format, style, graphics, and uses vary widely, posing an obstacle to widespread use and dissemination. To overcome this difficulty, Margolis, Gottlieb, and their associates (Margolis et al., 1991) advanced some suggestions for standardization of algorithms, which are briefly presented here. The proposals involve use of boxes, including clinical state boxes, decision boxes, action boxes, and link boxes; arrows; a numbering scheme; pagination; abbreviations; and various aspects of annotations. TITLE The title should define the clinical topic and intended users. Under it, the authors, their degrees, and institutional affiliations should be listed. The date of publication and revision (if applicable) should be specified. A footnote to the title should state the process by which the algorithm logic was decided; this might be, for instance, group consensus after literature review, individual recommendation based on clinical experience, or some other technique. BOXES Clinical state box—rounded rectangle or elliptical box. This box defines the clinical state or problem. It has only one exit path and may or may not have an entry path. This box always appears at the beginning of an algorithm. The initial clinical state box should describe the clinical problem to be addressed. Clinical state boxes in the body of the algorithm are used to clarify the status of the patient or diagnosis along the path of the algorithm (i.e., to describe a subset of patients with a particular clinical condition). Decision box—hexagon. This box requires a branching decision, whose response will lead to one of two alternative paths. It always has an entry path and two exit paths. Statements in decision boxes should be phrased as questions punctuated with question marks. If two assessments are to be determined, then the developers should specify whether both (''and") or one
OCR for page 250
Guidelines for Clinical Practice: From Development to Use ("or") must be positive for a "yes" response. Multiple questions can be asked in one box, and criteria are specified for a "yes" response to the entire box—for instance, whether two of three criteria must be present, whether all must be present, or whether any must be present. Action box—rectangle. This box indicates an action, commonly either therapeutic or diagnostic. Several rules are suggested for action boxes, as follows: A single phrase within a box should not be punctuated with a period. Multiple actions that do not need to be sequenced in time may be listed in one box. When multiple actions are presented in one box, each action is to be listed on a separate line (preceded with an optional number, dash, or bullet). When two statements are to be joined by "and" or "or," the conjunction should be placed on a separate line for emphasis. Link box—small oval. This box is used in place of an arrow, to link boxes for graphic clarity. This might be useful at page breaks or between separated nodes to maintain path continuity. The box itself should read "Go to Page ... Box ...." ARROWS Several rules for arrows have also been advanced. The flow should be from top to bottom. In general, the flow should be from left to right, except when a side branch rejoins the main stem. Arrows should never intersect. Link boxes (see above) can be used to avoid crossing paths. Arrows originating from decision boxes should be labelled "yes" or "no." No other text should be used over an arrow. Wherever possible (i.e., where clinical content will not be obscured), "yes" arrows should point to the right, and "no" arrows should point down. NUMBERING SCHEME Clinical state boxes, decision boxes, and action boxes should be numbered sequentially from left to right and from top to bottom. Link boxes are not numbered. PAGING Whenever possible, it is advisable to consolidate the algorithm so that it can be presented on one page. Page breaks are inserted where clinical logic indicates, and a single box should not be isolated on a page. For complex algorithms, the first page could best serve as a directory to clinical subsets of patients. In this case, each subset is identified as a clinical state box. ANNOTATIONS Citation of the annotation. Annotations are an intrinsic part of the algorithm. They are used to clarify the rationale of the decisions, cite the supporting literature, and expand on less essential details of the clinical
OCR for page 251
Guidelines for Clinical Practice: From Development to Use information contained in the box. Annotations would be cited following a single phrase using a capital letter at the end of the phrase. When multiple statements are contained in a single box, annotation(s) should appear at the end of the phrase(s) to which it is applicable. If an annotation is applicable to the entire box with multiple statements, then it should be cited using a capital letter centered on a separate line at the bottom of the box. Annotation format. Annotations should be written in text format and should appear on pages that are separate from the algorithm. They should be referenced according to standard medical reference format, with references numbered using superscripts within the text. ABBREVIATIONS Except for units of measurement, abbreviations are discouraged. Computer-based Formats Some clinical researchers argue that clinical flowcharts are inefficient representations of algorithms because they are limited by yes/no branch points to arbitrary sequences and thus cannot accommodate the richer choices common to medicine. They have been looking for ways to avoid flowcharts and to move toward a "meta-language" by using a standard syntax to convert algorithms to different kinds of computer-based decision aids. The Arden Syntax, developed in medical centers with private-sector support, is an effort to create such a system (McDonald et al., 1991; Megargle, 1991). Using this syntax, a well-defined algorithm can be transformed into various kinds of computer programming statements such as if-then statements or for-loop statements. Use of the Arden Syntax allows easy transfer of understandable contraindication alerts, management suggestions, data interpretations, treatment protocols, and similar aids from one computer system to another. Example 10, on detecting deteriorating metabolic acidosis, and Example 11, on the use of oral contraceptives, illustrate specific computer-based formats. The rest of the appendix presents 16 examples of guidelines and related materials. A range of topics, formats, sponsors and users, clinical orientations and purposes, and levels of complexity are presented and discussed; cross-comparisons within the appendix are noted. As stated above, these discussions should not be taken as complete analyses of the items in question, nor should inclusion in this appendix be taken as endorsement of the content or development process of these guidelines.
OCR for page 252
Guidelines for Clinical Practice: From Development to Use Example 1 SCREENING FOR DIMINISHED VISUAL ACUITY Clinical orientation: Clinical condition Clinical purpose: Screening and prevention Complexity: Medium Format: Free text and a stand-alone reference chart Intended users: Practitioners, perhaps patients In 1989, the U.S. Preventive Services Task Force (USPSTF) published a 419-page document intended mainly for primary care providers. The task force's objective was to develop comprehensive recommendations addressing preventive services for all age groups for 60 target conditions, using a systematic process and explicit criteria to review evidence and develop recommendations. The work of the task force was discussed extensively in Chapter 2 and elsewhere in the report. This particular guideline is presented, in the USPSTF book, as part of a larger course of preventive care. It is one of 169 guidelines for specific preventive interventions, each of which may include recommendations for preventive care by age group (e.g., in favor of vision screening for children of younger ages and possibly for the elderly but not for adolescents and adults). Reproduced here are (1) the specific recommendations for vision screening and (2) the plasticized reference card for preventive care of children ages 2-6, which recommends vision screening. In its concern with reliability of a particular test, this guideline is similar to the one on erythrocyte sedimentation rate tests (Example 5). As is true of several items in the appendix, this guideline cites the literature on which it is based. Educated patients might also make use of this guideline, as they could for the guidelines on, for instance, deciding what to do about low back pain or managing labor and delivery after a previous cesarean delivery (see Examples 12 and 8, respectively). SOURCE: Reprinted (public domain) from: U.S. Preventive Services Task Force. Guide to Clinical Preventive Services: An Assessment of the Effectiveness of 169 Interventions. Baltimore, Md.: Williams & Wilkins, 1989.
OCR for page 253
Guidelines for Clinical Practice: From Development to Use Screening for Diminished Visual Acuity Recommendation: Vision screening is recommended for all children once before entering school, preferably at age 3 or 4 (see Clinical Intervention). Routine vision testing is not recommended as a component of the periodic health examination of asymptomatic schoolchildren. Clinicians should be alert for signs of ocular misalignment when examining all infants and children. Vision screening of adolescents and adults is not recommended, but it may be appropriate in the elderly. Screening for glaucoma is discussed in Chapter 32. Burden of Suffering About 2-5% of American children suffer from amblyopia ("lazy eye") and strabismus (ocular misalignment), and nearly 20% have simple refractive errors by age 16.1-4 Amblyopia and strabismus usually develop between infancy and ages 5-7.3 Since normal vision from birth is necessary for proper eye development, failure to treat amblyopia and strabismus before school age may later result in irreversible visual deficits, permanent amblyopia, loss of depth perception and binocularity, cosmetic defects, and educational and occupational restrictions.1,4,5 In contrast, refractive errors such as myopia become common during school age but rarely carry serious prognostic implications.1,3,6,7 Experts disagree on whether uncorrected refractive errors cause diminished academic performance among schoolchildren. 1,3,5,7,8, The majority of vision disorders occur in adults; over 8.5 million Americans suffer from visual impairment 9 Visual disorders such as presbyopia (decreased ability to focus on near objects) become more common with age10 and therefore the prevalence of visual impairment is highest in those over age 65. Preliminary statistics from recent surveys suggest that nearly 13% of Americans age 65 and older have some form of visual impairment, and almost 8% of this age group suffer from severe impairment: blindness in both eyes or inability to read newsprint even with glasses.11 Vision disorders in the elderly may be associated with injuries due to falls and motor vehicle accidents, diminished productivity, and loss of independence.12 Many older adults are unaware of changes in their visual acuity, and up to 25% of them may be using an incorrect lens prescription.12 Efficacy of Screening Tests Although screening for strabismus and amblyopia is most critical at an early age, screening tests to detect occult vision disorders in children under age 3 have
OCR for page 335
Guidelines for Clinical Practice: From Development to Use ACUTE DYSURIA IN THE ADULT FEMALE A primary goal of this algorithm is to separate women with acute uncomplicated UTI that can be treated with single dose antibiotic therapy from women with complicated UTI that will require further evaluation or longer duration of therapy. Therefore, women who have symptoms longer than 2 or 3 days, women who have fever or flank pain, pregnant women and women with frequent recurrences or other underlying medical problems need to be eliminated from this algorithm. Initial steps in their management are suggested at branch points of this algorithm, but other algorithms will be necessary to more fully address the management of these groups of patients. Stamm, W., Causes of the Acute Urethral Syndrome in Women, NEJM 1980; 303; 409-415. Choices for multiple dose Rx include 7-10 day course of: Trimethoprim sulfa DS BID (contraindicated in pregnancy, known G6PD deficiency or allergic Hx). Amoxicillin 250 mg po tid (1st choice in pregnancy). Nitrofurantoin 50 mg QID (alternative for patient with multiple allergies or pregnant patient with Hx Pen allergy). Prophylaxis is usually continued for 6 months. Options for prophylaxis include: Trimethoprim sulfa 12 regular strength tab, QHS. Nitrofurantoin 50 mg QHS (in pregnant patient or patient with Hx T/X allergy or known G6PD deficiency). Ronald, A. and Harding, G., Urinary Infection Prophylaxis in Women, Annals Int. Med. 1981; 94(2) 268-269. Options for single dose Rx include: Trimethoprim sulfa DS 2 tabs x 1. Amoxicillin 3 gm po x 1. Kamaroff, A., Acute Dysuria in Women, NEJM 1984; 310; 368-375. Patients who have failed single dose Rx should be considered to have upper tract infection and treated per pyelo protocol.
OCR for page 336
Guidelines for Clinical Practice: From Development to Use Example 16 MANAGEMENT OF ACUTE PAIN Clinical orientation: Clinical condition Clinical purpose: Management of acute pain due to operative procedures, medical procedures, or trauma Complexity: Medium Format: Free text, flow charts, tables, graphic pain scales Intended users: Practitioners and patients The following pages are excerpts from the 1992 guideline, Acute Pain Management: Operative or Medical Procedures and Trauma, commissioned by the Agency for Health Care Policy and Research (AHCPR). The guideline addresses the "widespread inadequacy of pain management" (AHCPR, February 1992, p. 1) by providing clinicians and patients with approaches and tools for assessing and managing pain. An interdisciplinary panel developed the guideline. They focused on the need for pain management, clinical practice patterns, and clinical and technological options for pain management. The development process included an extensive literature search, evaluation of the quality of clinical data, peer review of drafts of the guideline, tests of the guidelines in clinical situations, an open meeting for testimony, and use of external consultants. The guideline is available in four forms. In addition to the full guideline cited above, there are two "quick reference guides for clinicians" Acute Pain Management in Infants, Children, and Adolescents: Operative and Medical Procedures and Acute Pain Management in Adults: Operative Procedures-and a patient's guide, Pain Control After Surgery, available in both English and Spanish. The guideline will also be incorporated into data bases at the National Library of Medicine and the National Technical Information Service and into the computer-based information systems. The complete guide discusses: the need for aggressive postoperative control of pain; pain assessment and reassessment; options for preventing and controlling postoperative pain; control of site-specific pain; management of pain in infants, children, adolescents, and other patients with special needs (e.g., the elderly, known or suspected substance abusers); and institutional responsibility for effective pain relief. It also contains a significant list of references and appendices.
OCR for page 337
Guidelines for Clinical Practice: From Development to Use SOURCES: Reprinted (public domain) from: Acute Pain Management Guideline Panel. Acute Pain Management: Operative or Medical Procedures and Trauma. Clinical Practice Guideline. AHCPR Pub. No. 92-0032. Rockville, Md.: Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services. February 1992. Acute Pain Management Guideline Panel. Acute Pain Management in Adults: Operative Procedures. Quick Reference Guide for Clinicians. AHCPR Pub. No. 92-0019. Rockville, Md.: Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services. 1992.
OCR for page 338
Guidelines for Clinical Practice: From Development to Use
OCR for page 339
Guidelines for Clinical Practice: From Development to Use Options to Prevent and Control Postoperative Pain Patient education and reduction of any preexisting pain should occur before the operation. Because the goal of the treatment plan is to prevent significant postoperative pain from the outset, treatment alternatives, potential risks, dosage adjustments, and adjunctive therapies should be described to the patient and family. Teaching emphasizes what the patient is likely to experience postoperatively, including the specific method(s) of pain assessment, intervention(s) the staff will employ, and the level of patient participation required. Staff also should inform patients that it is easier to prevent pain than to "chase" or treat it once it has become established, and that communication of unrelieved pain is essential to its relief. Pain control options include: Cognitive-behavioral interventions such as relaxation, distraction, and imagery; these can be taught preoperatively and can reduce pain, anxiety, and the amount of drugs needed for pain control; Systemic administration of nonsteroidal anti-inflammatory drugs (NSAIDs) or opioids using the traditional "as needed" schedule or around-the-clock administration (American Pain Society, 1989); Patient controlled analgesia (PCA), usually meaning self-medication with intravenous doses of an opioid; this can include other classes of drugs administered orally or by other routes; Spinal analgesia, usually by means of an epidural opioid and/or local anesthetic injected intermittently or infused continuously; Intermittent or continuous local neural blockade (examples of the former include intercostal nerve blockade with local anesthetic or cryoprobe; the latter includes infusion of local anesthetic through an interpleural catheter); Physical agents such as massage or application of heat or cold; and Electroanalgesia such as transcutaneous electrical nerve stimulation (TENS).
OCR for page 340
Guidelines for Clinical Practice: From Development to Use Examples of Pain Intensity and Pain Distress Scales
OCR for page 341
Guidelines for Clinical Practice: From Development to Use Institutional Responsibility for Pain Management The institutional process of acute pain management begins with the affirmation that patients should have access to the best level of pain relief that may safely be provided. (See Table 3 for a summary of the scientific evidence for interventions to manage pain in adults.) Each institution should develop the resources necessary to provide the best and most modern pain relief appropriate to its patients and should designate who and/or which departments are responsible for the required activities. Optimal application of pain control methods depends on cooperation among different members of the health care team throughout the patient's course of treatment. To ensure that this process occurs effectively, formal means must be developed and used within each institution to assess pain management practices and to obtain patient feedback to gauge the adequacy of pain control. The institution's quality assurance procedures should be used periodically to assure that the following pain management practices are being carried out: • Patients are informed that effective pain relief is an important part of their treatment, that communication of unrelieved pain is essential, and that health professionals will respond quickly to their reports of pain. They are also told that a total absence of pain is often not a realistic or even a desirable goal. • Clear documentation of pain assessment and management is provided. • There are institution-defined levels for pain intensity and relief that elicit review of current pain therapy, documentation of the proposed modifications in treatment, and subsequent review of their efficacy. • Each clinical unit periodically assesses a randomly selected sample of patients who have had surgery within 72 hours to determine their current pain intensity, the worst pain intensity in the first 24 hours, the degree of relief obtained from pain management interventions, satisfaction with relief, and satisfaction with the staff's responsiveness.
OCR for page 342
Guidelines for Clinical Practice: From Development to Use Table 3 Scientific Evidence for Interventions to Manage Pain in Adults Pharmacologic Interventions Intervention1 Type of Evidence Comments NSAIDs Oral (alone) LB, IV Effective for mild to moderate pain. Begin preoperatively. Relatively contraindicated in patients with renal disease and risk of or actual coagulopathy. May mask fever. Oral (adjunct to opioid) Ia, IV Potentiating effect resulting in opioid sparing. Begin preop. Cautions as above. Parenteral (ketorolac) Ib, IV Effective for moderate to severe pain. Expensive. Useful where opioids contraindicated, especially to avoid respiratory depression and sedation. Advance to opioid. Opioids Oral IV As effective as parenteral in appropriate doses. Use as soon as oral medication tolerated. Route of choice. Intramuscular Ib, IV Has been the standard parenteral route, but injections painful and absorption unreliable. Hence, avoid this route when possible. Subcutaneous Ib, IV Preferable to intramuscular for low-volume continuous infusion. Injections painful and absorption unreliable. Avoid this route for long-term repetitive dosing. Intravenous Ib, IV Parenteral route of choice after major surgery. Suitable for titrated bolus or continuous administration (including PCA), but requires monitoring. Significant risk of respiratory depression with inappropriate dosing. PCA (systemic) Ia, IV Intravenous or subcutaneous routes recommended. Good, steady level of analgesia. Popular with patients but requires special infusion pumps and staff education. See cautions about opioids above. Epidural and intrathecal Ia, IV When suitable, provides good analgesia. Significant risk of respiratory depression, sometimes delayed in onset. Requires careful monitoring. Use of infusion pumps requires additional equipment and staff education. Local Anesthetics Epidural and intrathecal Ia, IV Limited indications. Expensive if infusion pumps employed. Effective regional analgesia. Opioid sparing. Addition of opioid to local anesthetic may improve analgesia. Risks of hypotension, weakness, numbness. Use of infusion pump requires additional equipment and staff. Peripheral nerve block Ia, IV Limited indications and duration of action. Effective regional analgesia. Opioid sparing
OCR for page 343
Guidelines for Clinical Practice: From Development to Use Nonpharmacologic Interventions Intervention1 Type of Evidence Comments Simple Relaxation (begin preoperatively) Jaw relaxation Progressive muscle relaxation Simple imagery Ia, IIa, IIb, IV Effective in reducing mild to moderate pain and as an adjunct to analgesic drugs for severe pain. Use when patients express an interest in relaxation. Requires 3-5 minutes of staff time for instructions. Music Ib, IIa, IV Both patient-preferred and ''easy listening" music are effective in reducing mild to moderate pain. Complex Relaxation (begin preoperatively) Biofeedback Ib, IIa, IV Effective in reducing mild to moderate pain and operative site muscle tension. Requires skilled personnel and special equipment. Imagery Ib, IIa, IIb, IV Effective for reduction of mild to moderate pain. Requires skilled personnel. Education/ Instruction (begin preoperatively) Ia, IIa, IIb, IV Effective for reduction of pain. Should include sensory and procedural information and instruction aimed at reducing activity related pain. Requires 5-15 minutes of staff time. TENS Ia, IIa, III, IV Effective in reducing pain and improving physical function. Requires skilled personnel and special equipment. May be useful as an adjunct to drug therapy. 1 Insufficient scientific evidence is available to provide specific recommendations regarding the use of hypnosis, acupuncture, and other physical modalities for relief of postoperative pain. Type of Evidence - Key Ia Evidence obtained from meta-analysis of randomized controlled trials. Ib Evidence obtained from at least one randomized controlled trial. IIa Evidence obtained from at least one well-designed controlled study without randomization. IIb Evidence obtained from at least one other type of well-designed quasi-experimental study. III Evidence obtained from well-designed nonexperimental descriptive studies, such as comparative studies, correlational studies, and case studies. IV Evidence obtained from expert committee reports or opinions and/or clinical experiences of respected authorities. Note: References are available in the Guideline Report. Acute Pain Management: Operative or Medical Procedures and Trauma. AHCPR Pub. No. 92-0001. Rockville, MD: Agency for Health Care Policy and Research, Public Health Service, U.S. Department of Health and Human Services. In press.
OCR for page 344
Guidelines for Clinical Practice: From Development to Use Summary Summary recommendations 1-5 and 7, below, should be implemented in every hospital where operations are performed on inpatients. The Acute Pain Management Guideline Panel recommends that any hospital in which abdominal or thoracic operations are routinely performed offer patients postoperative regional anesthetic, epidural or intrathecal opioids, PCA infusions, and other interventions requiring a similar level of expertise, under the supervision of an acute pain service as described in summary recommendation 6, below. For pain management to be effective, each hospital must designate who or which department will be responsible for all of the required activities. There are a number of alternative approaches to preventing or relieving postoperative pain, many of which can give good results if attentively applied. The following elements, however, apply to most cases and might serve as a focus for assessing the results of these guidelines: Promise patients attentive analgesic care. Patients should be informed before surgery, verbally and in printed format, that effective pain relief is an important part of their treatment, that talking about unrelieved pain is essential, and that health professionals will respond quickly to their reports of pain. It should be made clear to patients and families, however, that the total absence of any postoperative discomfort is normally not a realistic or even a desirable goal. Chart and display assessment of pain and relief. A simple assessment of pain intensity and pain relief should be recorded on the bedside vital sign chart or a similar record that encourages easy, regular review by members of the health care team and is incorporated in the patient's permanent record. The intensity of pain should be assessed and documented at regular intervals (depending on the severity of pain) and with each new report of pain. The degree of pain relief should be determined after each pain management intervention, once a sufficient time has elapsed for the treatment to reach peak effect. A simple, valid measure of intensity and relief should be selected by each clinical unit. For children, ageappropriate measures should be used. Define pain and relief levels to trigger a review. Each institution should identify pain intensity and pain relief levels that will elicit a review of the current pain therapy, documentation of the proposed modifications in treatment, and subsequent review of its efficacy. This process of treatment review and followup should include participation by physicians and nurses involved in the patient's care. Survey patient satisfaction. At regular intervals defined by the clinical unit and quality assurance committee, each clinical unit should assess a randomly selected sample of patients who have had surgery within 72 hours. Patients should
OCR for page 345
Guidelines for Clinical Practice: From Development to Use be asked their current pain intensity, the worst pain intensity in the past 24 hours, the degree of relief obtained from pain management interventions, satisfaction with relief, and their satisfaction with the staff's responsiveness. Analgesic drug treatment should comply with several basic principles: Non-opioid "peripherally acting" analgesics. Unless contraindicated, every patient should receive an around-the-clock postoperative regimen of an NSAID. For patients unable to take medications by mouth, it may be necessary to use the parenteral or rectal route. Opioid analgesics. Analgesic orders should allow for the great variation in individual opioid requirements, including a regularly scheduled dose and "rescue" doses for instances in which the usual regimen is insufficient. Specialized analgesic technologies, including systemic or intraspinal, continuous or intermittent opioid administration or patient controlled dosing, local anesthetic infusion, and inhalational analgesia (e.g., nitrous oxide) should be governed by policies and standard procedures that define the acceptable level of patient monitoring and appropriate roles and limits of practice for all groups of health care providers involved. The policy should include definitions of physician and nurse accountability, physician and nurse responsibility to the patient, and the role of pharmacy. Nonpharmacological interventions: Cognitive and behaviorally based interventions include a number of methods to help patients understand more about their pain and to take an active part in its assessment and control. These interventions are intended to supplement, not replace, pharmacological interventions. Staff should give patients information about these interventions and support patients in using them. Monitor the efficacy of pain treatment: Periodically review pain treatment procedures as defined in summary recommendations 1-4 above, using the institution's quality assurance procedures.
Representative terms from entire chapter: