APPENDIX E
Guideline Development Process

The Guideline for the Management of Stroke Rehabilitation is the product of many months of diligent effort and consensus building among knowledgeable individuals from the Veterans Administration (VA), Department of Defense (DoD), academia, and guideline facilitators from the private sector. An experienced moderator facilitated the multidisciplinary Working Group that included internists, physiatrists, neurologists, geriatricians, nurse practitioners, occupational therapists, physical therapists, recreational therapists, speech and language pathologists, psychologists, social workers, kinesiotherapists, pharmacists, and rehabilitation/clinic coordinators, as well as consultants in the field of guideline and algorithm development.

Development Process

"Only well-focused questions and search terms will lead to a successful search for evidence" (AHCPR, 1996). The process of developing this guideline was evidence-based whenever possible. Evidence-based practice integrates clinical expertise with the best available clinical evidence derived from systematic research. Where evidence is ambiguous or conflicting, or where scientific data are lacking, the clinical experience of the multidisciplinary Working Group was used to guide the development of consensus-based recommendations. The developers incorporated the evidence and recommendations into a format that would maximally facilitate clinical decision-making (Woolf, 1992). The review of the literature, evaluation of evidence, and development of the guideline proceeded in sequential steps.

The following three guidelines were identified by the Working Group as appropriate seed guidelines. They served as the starting point for the development of questions and key terms.

• Agency for Health Care Policy and Research (AHCPR) Post-Stroke Rehabilitation (1995)
• Royal College of Physicians (RCP) National Clinical Guidelines for Stroke (2000)
• Scottish Intercollegiate Guidelines Network (SIGN) Management of Patients with Stroke, 20 (1997)

Fifty-one researchable questions and associated key terms were developed by the Working Group after orientation to the seed guidelines and to goals that had been identified by the Working Group. The questions specified:

• Population - characteristics of the target population
• Intervention - diagnostic, screening, therapy, and assessment
• Control - the type of control used for comparison
• Outcome - the outcome measure for this intervention (morbidity, mortality, patient satisfaction, and cost)

A systematic search of the literature was conducted. It focused on the best available evidence to address each key question, and ensured maximum coverage of studies at the top of the hierarchy of study types: evidence-based guidelines, meta analyses, and systematic reviews (Cochrane, EBM, and EPC reports). The seed guidelines evidence was carefully reviewed.

The search continued using well-known and widely available databases that were appropriate for the clinical subject. Limits on language (English) and type of research (randomized controlled trials [RCTs]) were applied. The search included MEDLINE and additional specialty databases (DARE), depending on the topic.

The search strategy did not cast a wide net. Once definitive clinical studies that provided valid relevant answers to the question were identified, the search stopped. It was extended to studies/reports of lower quality (observational studies) only if there were no high quality studies.

The results of the search were organized and reported using reference manager software. At this point, additional exclusion criteria were applied. Typical exclusions were studies with physiological endpoints, or studies of populations that were not comparable to the population of interest (e.g., studies of rehabilitation of patients with other diseases).

Evidence Appraisal Reports for each of the 51 questions were prepared by the Center for Evidence-based Practice at the State University of New York, Upstate Medical University, Department of Family Medicine (these reports are available by request). Each report covered:

• Summary of findings
• Methodology
• Search terms
• Resources searched
• Articles critically appraised
• Findings

The Working Group suggested some additional references. Copies of specific articles were provided to participants on an as-needed basis. This document includes references through January 2002.

The clinical experts and research team evaluated the evidence for each question according to criteria proposed by the U.S. Preventive Services Task Force (USPSTF) (2001). See "Rating the Evidence," below.

The Working Group participated in two face-to-face sessions to reach a consensus about the guideline recommendations and to prepare a draft document. The draft was revised by the experts through numerous conference calls and individual contributions to the document. The guideline presents evidence-based recommendations that have been thoroughly evaluated by practicing clinicians.

The final draft was reviewed by experts from the VA and DoD in physical medicine and neurology. Their feedback was integrated into the final draft. Nonetheless, this document is a work in progress. It will be updated every two years, or when significant new evidence is published.

Rating the Evidence

Evidence-based practice involves integrating clinical expertise with the best available clinical evidence derived from systematic research. The Working Group reviewed the evidence and graded it using the rating scheme developed by the USPSTF (2001). The experts themselves, after an orientation and tutorial on the evidence grading process, formulated Quality of Evidence ratings (see Table 1), a rating of Overall Quality (see Table 2), a rating of the Net Effect of the Intervention (see Table 3), and an overall Recommendation (see Table 4).

• Once assembled, admissible evidence is reviewed at three strata: (1) the individual study, (2) the body of evidence concerning a single linkage in the analytic framework, and (3) the body of evidence concerning the entire preventive service. For each stratum, the Task Force uses explicit criteria as general guidelines to assign one of three grades of evidence: good, fair, or poor.
• Good or fair quality evidence for the entire preventive service must include studies of sufficient design and quality to provide an unbroken chain of evidence-supported linkages that generalize to the general primary care population and connect the preventive service with health outcomes. Poor evidence contains a formidable break in the evidence chain such that the connection between the preventive service and health outcomes is uncertain.
• For services supported by overall good or fair evidence, the Task Force uses outcomes tables to help categorize the magnitude of benefits, harms, and net benefit from implementation of the preventive service into one of four categories: substantial, moderate, small, or zero/negative.
• The Task Force uses its assessment of the evidence and magnitude of net benefit to make a recommendation, coded as a letter: from A (strongly recommended) to D (recommend against). It gives an "I" recommendation in situations in which the evidence is insufficient to determine net benefit (Harris et al., 2001).

Algorithms

The overall view of the Stroke Rehabilitation guideline is presented in an algorithmic format. There are indications that this format improves data collection and clinical decision-making and helps to change patterns of resource use. It allows the clinician to follow a linear approach to critical information needed at the major decision points in the clinical process, and includes:

• An ordered sequence of steps of care
• Recommended observations
• Decisions to be considered
• Actions to be taken.

A clinical algorithm diagrams a guideline into a step-by-step decision tree. Standardized symbols are used to display each step in the algorithm (SMDMC, 1992). Arrows connect the numbered boxes indicating the order in which the steps should be followed.

	Rounded rectangles represent a clinical state or condition.
	Hexagons represent a decision point in the guideline, formulated as a question that can be answered Yes or No. A horizontal arrow points to the next step if the answer is YES. A vertical arrow continues to the next step for a negative answer.
	Rectangles represent an action in the process of care.
	Ovals represent a link to another section within the guideline.

A letter within a box of an algorithm refers the reader to the corresponding annotation. The annotations elaborate on the recommendations and statements that are found within each box of the algorithm. Included in the annotations are brief discussions that provide the underlying rationale and specific evidence tables. The reference list includes all the sources used-directly or indirectly-in the development of the annotation text. A complete bibliography is provided at the end of the document.

REFERENCES