Manual for ACC/AHA Guideline Writing Committees

Methodologies and Policies from the ACC/AHA Task Force on Practice Guidelines

Section II: Tools and Methods for Creating Guidelines

• Step Two: Define and Conduct Appropriate and Comprehensive Literature Searches

Finding and Managing the Evidence
Once the scope of the guideline has been determined, comprehensive searching of the published literature takes place. A key component of the ACC/AHA guideline methodology is the creation of recommendations based on the entirety of the evidence currently available. The Institute of Medicine describes literature searching as the key step in developing valid guidelines.

It has been estimated that over 2 million articles and more than 17,000 biomedical books are published annually. The challenge of finding relevant articles among the millions is compounded by the availability of multiple electronic databases, all of which offer different but partially overlapping pools of information.

Because a guideline based on an incomplete or biased evaluation of the literature can lead to inappropriate recommendations, the search for relevant research should be comprehensive, research should be selected using explicit criteria, and the validity of the results should be judged in a rigorous and reproducible fashion. - Cook, 1997

The current resources for ACC/AHA guideline development allow for searching in MEDLINE (via PubMed), EMBASE, the Cochrane Library, and Best Evidence. The Research Analyst assigned to the guideline will manage a computerized database of all citations relevant to the guideline topic. The Research Analyst and Librarian conduct searches and forward relevant citations to the writers.

Literature Search Methodology
Figure 2 outlines the ACC/AHA process for conducting comprehensive literature searches for the guidelines. Initial literature searching focuses on published meta-analyses and systematic reviews. If high quality, relevant, and up-to-date meta-analyses or systematic reviews are found, these articles allow writers to focus on critiquing and updating an existing review as opposed to creating one. For the majority of topics, literature searching also includes randomized clinical trials, and is expanded to non-randomized studies, case studies, and opinion documents until the evidence-base is sufficient for each clinical question identified in Step One. Each article should be critically evaluated as to quality and clinical limitations, as discussed in Step Three.

Documentation of Searching
All literature searching for guideline development must be documented by the searcher and stored at the ACC offices. This allows the chair and Research Analyst to construct the text of the guideline describing the literature search criteria, thereby allowing guideline users to assess the comprehensiveness of the searching.

In addition to searches conducted by staff, writing committee members are welcome to conduct their own literature searches, including search criteria beyond what the ACC/AHA resources are able to provide (see below: Standard Search Criteria for ACC/AHA Guidelines). The documentation for all literature searches must be forwarded to the Research Analyst using the Template: Literature Searches for ACC/AHA Guidelines included in this section.

Standard Search Criteria for ACC/AHA Guidelines

• Literature searching includes the following on-line databases:
  • MEDLINE/PubMed
  • EMBASE
  • Cochrane Library
  • Best Evidence
• Searches are limited to English language. (Searches will be expanded to languages other than English on a case-by-case, as requested basis.)
• Searches are limited to human subjects.
• In the case of a guideline update, searches are limited to the time period following the publication of the last version of the guideline.
• In the case of a new guideline or full revision, no time limits on searches are imposed, unless the writing committee determines that a different time frame appropriate (for example, a guideline on a diagnostic that did not exist before a certain date).
• Gender and age are not limited, except when a specific clinical objective applies only to a particular sex or age group.
• Publication type is initially limited to meta-analyses and systematic reviews. Publication type is expanded on an as-needed basis to include randomized controlled trials, non-randomized studies, case studies, and opinion documents.
• If an acceptable systematic review or meta-analysis is identified, searches to update it are typically limited to the time period following the search cut-off date reported in the review.