In the past decade, clinical practice guidelines have proliferated rapidly as professional organizations, academic institutions, insurers, hospitals and governments have promoted them in the hope that they will facilitate more consistent, effective and efficient medical practice. Unfortunately, the rapid proliferation of guidelines has not been matched by the consistent use of sound guideline development and reporting methods. Clinicians, patients and policy makers need new information tools to help them seek, sort and appraise high quality guidelines. The investigators have developed a standardized format for abstracts of practice guidelines to help readers quickly screen the applicability, importance, and validity of these documents. This is approved for online abstracts in National Library of Medicine (NLM) databases and has not been implemented or evaluated in any existing guideline dissemination initiative. The primary objective of this project is to evaluate the reliability and validity of guideline abstraction methods using a full set of preventive care guidelines and representative examples of guidelines in other clinical domains. Secondary objectives are to evaluate the character and properties of abstracted guidelines, assess the information needs of potential database users, develop and evaluate user interfaces, integrate the database contents with NLM databases, and to consider the feasibility of initiating and maintaining large scale guidelines databases. A computerized database for registering, storing and updating standardized guideline abstracts will be populated with recommendations from preventive care guidelines together with information about the objective and targets of the guideline, the main practice options and outcomes considered, the nature of the evidence and values that were analyzed, the main benefits and harms expected from implementation, the results of any attempt to validate the guideline by clinical testing or external review, and the groups who participated in guideline development or endorsed the product. The main questions addressed by this project are: 1) What information would clinicians and policy makers seek from a database of structured guideline abstracts and how would they want to access that information? 2) What criteria must the database satisfy to permit its integration with existing bibliographic and full-text databases? 3) How accurate are more informative guideline abstracts prepared by trained abstractors? 4) How reliable are standardized guideline abstracts prepared by trained abstractors? 5) Do guideline abstracts prepared by trained, independent abstractors differ systematically from abstracts prepared by guideline developers? 6) How do clinicians and policy makers use a prototype database and how do they rate the usefulness of more informative compared to conventional guideline summaries? 7) What resources would be required to initiate and sustain a comprehensive database of more informative abstracts of preventive, diagnostic, therapeutic and palliative guidelines? Placed in the public domain, the perfected guideline abstraction instrument, computer database and user-interfaces could serve groups interested in developing other guideline-based information tools. The database structure and contents could also serve clinicians and policy makers interested in guideline development, evaluation, dissemination and implementation. The database will reflect the state-of-the-art of preventive care guidelines.
