No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 25 October 2005
Objectives: As part of a review of guidelines for sleep laboratories, this study summarizes the quality and relevance of evidence cited in support of recommendations on investigations for individuals with sleep disorders.
Methods: Quality of evidence in support of each recommendation was rated as A (from well-conducted, prospective controlled studies), B (from controlled studies and case series, with minor shortcomings), or C (from case series or case series plus controlled studies with substantial limitations). Relevance of the cited evidence was also rated from A to C, in decreasing order of merit.
Results: Of eighty-one recommendations from thirty-seven guidelines, forty-six were supported by evidence from primary studies. Six recommendations were supported by category A evidence, fifteen by category B, and twenty-five recommendations by category C. The cited evidence was highly relevant to eighteen recommendations, of some relevance for twenty-two and of little or no relevance to six. Four recommendations were informed by an absence of available evidence. For the other thirty-one, no evidence was provided in support.
Conclusions: The approach used provided a concise overview of recommendations and supporting evidence for decision-makers. Guidelines on the use of sleep laboratory investigations contain much detailed information, but evidence supporting several recommendations is of limited quality and relevance. Furthermore, good-quality studies of many sleep laboratory applications are needed to assist both health policy formulation and clinical practice.
Sleep disorders are common complaints, particularly among middle-aged and elderly people. For those with severe symptoms, the approach to diagnosis and management of the sleep disorder may involve observation in a sleep laboratory, typically using polysomnography (PSG) and/or the multiple sleep latency test.
In Canada, there has been an increasing demand for sleep laboratory services. Sleep laboratories are expensive, specialized facilities and health policy-makers require advice on which sleep disorder investigations are appropriate, as input to decisions on future resource allocation. Guidelines on sleep laboratory investigations prepared by professional organizations are a source of information on indications for and against testing individuals who have sleep disorders. At the request of health ministries, the Canadian Coordinating Office for Health Technology Assessment carried out a review of recommendations made by these guidelines and the evidence on which they are based (5).
The material cited in support of the guideline recommendations is diverse and covers many sleep disorders and clinical situations. One of the challenges for the CCOHTA review was to provide a concise presentation of this evidence for decision-makers, as a readily accessible overview of the more detailed appraisal of sleep laboratory investigations that forms the bulk of the report. In this article, we describe the approach used to present summaries of the quality and relevance of the evidence cited in support of guideline recommendations and illustrate this presentation with reference to some of the findings from the review.
Published guidelines and associated reviews on the use of sleep laboratory investigations for various sleep disorders were identified through a search of electronic data bases and appropriate professional Web sites that is described elsewhere (5). We selected publications for review when they included advice and recommendations on the use of examinations in sleep laboratories for persons with sleep disorders. Those reports that considered only the operation of equipment were excluded.
For each recommendation that related to a clinical application of sleep laboratories, we independently identified those studies and other publications that were cited by the guideline as providing support. Where a primary study was cited in support, the paper was retrieved and reviewed independently with the aid of a structured form. Consideration was given to the type and design of the study, study population, quality of the evidence, and relevance of the evidence and conclusions to the recommendation.
For each recommendation, an indication of the overall quality of the evidence cited in support was obtained by assigning ratings on a three point scale on the basis outlined in Table 1. We also noted instances where no evidence was offered in support or where the recommendation reflected the absence of evidence. In addition, judgments were made on the relevance of studies cited in support of recommendations, again using a three point scale (Table 1). In considering relevance of evidence, we took into account the relationship of each study to the routine operation of sleep laboratories and to the management of the sleep disorder. Ratings of quality and relevance of the cited evidence were made independently, any differences being resolved by consensus.
Reviews and general descriptive articles cited in support of recommendations were used as background information but not reviewed in detail. No attempt was made to retrieve and assess earlier information that was cited in these secondary sources.
Thirty-seven guideline publications were selected for review and eighty-one recommendations were identified, related to eighteen sleep laboratory applications. Forty-six of the recommendations were supported by evidence from primary studies. Of these, only six had support that corresponded to category A quality, with most or all of the cited evidence from well-conducted, prospective controlled studies. For fifteen recommendations, cited evidence was category B, and twenty-five were supported by category C evidence.
Four of the remaining recommendations were supported by an absence of available evidence that had been identified by the guideline. For the other thirty-one, no supporting evidence was provided; in some cases, the guideline indicated that there was support by consensus.
Judgments on the relevance of the cited evidence from primary studies suggested that it was highly relevant to the recommendation in eighteen cases, of some relevance in twenty-two, and of little or no relevance in six. Factors that decreased relevance of studies included use of methods, such as multi-variate analysis, that would not be used in routine sleep laboratory operation. In some cases, cited studies provided evidence of associations between sleep characteristics and a sleep disorder but did not address the practical clinical use of such information.
Examples of the presentation of summary findings from the review are shown in Table 2. For each application, recommendations related both to indications for and against use of sleep laboratory investigations are given, with indications of the quality of supporting evidence. For some applications, individual guidelines had different recommendations or emphases on indications for and against use of sleep studies. These differing conclusions are also identified in the table.
The approach taken to summarizing and presenting sleep lab recommendations and related evidence has the merit of giving a concise overview of the field, with an immediate indication of areas where there may be uncertainty or disagreement. This seemed to us to be a useful approach in addressing an area where there were several clinical applications and a large number of studies of variable quality and scope. It provides a starting point for persons who want advice on what the clinical consensus is on sleep laboratory investigations and the level of evidence supporting that consensus. Those wishing to follow-up such information can then consider the wealth of additional detail that is given in the full review (5).
The publications on the use of sleep laboratory investigations prepared by professional organizations are clinical guidelines. They contain much information for health professionals concerned with the management of sleep disorders. However, the level of evidence for many applications is of limited quality, and some cited studies are not directly relevant to the recommendations that are made. Many recommendations reflect consensus positions, and no evidence is cited.
Studies cited in the guidelines in support of recommendations are mostly small and often noncomparative. There is a need for further, good-quality studies of many sleep laboratory applications.
The material in the guidelines is helpful to policy areas in identifying positions on appropriate clinical practice. However, the value of this information is qualified by the limited quality and relevance of much of the supporting evidence and lack of agreement between guidelines on some issues.
There is also the practicality that the guidelines, as documents intended primarily for health professionals, cannot address various issues that will be of interest to some policy areas. In most cases, guidelines are not specific on what proportions of patients with particular conditions will require sleep laboratory studies. Some selection rules have been developed but may require further validation or not be readily applicable in a routine situation. In some areas, approaches and technologies are still developing. Longer-term outcomes associated with sleep laboratory interventions are rarely considered in any detail and economic issues are not addressed. The guidelines can provide useful input into policy considerations, but additional sources of information will also be needed to address other clinical and resource issues.
David Hailey, PhD (dhailey@ozemail.com.au), Adjunct Professor, Department of Public Health Sciences, University of Alberta, 13-130 Clinical Sciences Building, Edmonton, Alberta T6G 2G3, Canada
Khai Tran, MSc, PhD (khait@ccohta.ca), Research Officer, HTA Directorate, Canadian Coordinating Office for Health Technology Assessment (CCOHTA), 600-865 Carling Avenue, Ottawa, Ontario K1S 5S8, Canada
This project was funded by the Canadian Coordinating Office for Health Technology Assessment (CCOHTA). We are grateful to Shaila Mensinkai and Lynda McGahan, CCOHTA, and to Robert Dales, University of Ottawa, for their support in other areas of the project.
Classification of Study Quality and Study Relevance
Examples of Guideline Recommendations and Supporting Evidence for Sleep Laboratory Applicationsa
You have
Access