Article type
Year
Abstract
Introduction: Since 1993, the French Federation of Comprehensive Cancer Centers has been producing clinical practice guidelines aimed at improving the quality of health care and outcomes for cancer patients (the SOR project). These guidelines, which cover all aspects of patient management, are based on the critical appraisal of the literature by a multidisciplinary group of experts, with feedback from specialists in cancer care delivery.
Objectives: To optimize literature searching by defining and using standardized search strategies, to improve the identification of rigorous studies designs and the reproducibility of the literature search and to facilitate the updating process.
Methods: We identified previous studies in this field by searching Medline, the Cochrane Library, the Health Information Research Unit and other Cochrane web sites. These data were synthesised and adapted to the SOR methods and the field of oncology. These strategies are being tested on several topics, by various experts to assess and validate them.
Results: The strategies will be presented for different aspect of healthcare (e.g. diagnosis, therapeutics) and by type of publication (e.g. guidelines, review of the literature) or study designs (randomized controlled trials, cohort studies). The successive steps of a search can be defined starting with a focus on existing guidelines, consensues or reviews. Different base strategies can be combined (e.g. therapeutics + RCT, prognosis + cohort studies) if the use of a methodologic filter is relevant to the aspect studied, or they can be run separately. The results of the tests on different topics will be presented and discussed.
Discussion: The search of the literature is an essential step in the elaboration of recommendations and so this process should be traceable and reproducible. The development of standardized strategies is a means of achieving this goal, and strategies are being developed and published for all types of studies, including economic analyses and quality of life assessments. One remaining problem is the standardization of searching terms across different databases and software used.
Objectives: To optimize literature searching by defining and using standardized search strategies, to improve the identification of rigorous studies designs and the reproducibility of the literature search and to facilitate the updating process.
Methods: We identified previous studies in this field by searching Medline, the Cochrane Library, the Health Information Research Unit and other Cochrane web sites. These data were synthesised and adapted to the SOR methods and the field of oncology. These strategies are being tested on several topics, by various experts to assess and validate them.
Results: The strategies will be presented for different aspect of healthcare (e.g. diagnosis, therapeutics) and by type of publication (e.g. guidelines, review of the literature) or study designs (randomized controlled trials, cohort studies). The successive steps of a search can be defined starting with a focus on existing guidelines, consensues or reviews. Different base strategies can be combined (e.g. therapeutics + RCT, prognosis + cohort studies) if the use of a methodologic filter is relevant to the aspect studied, or they can be run separately. The results of the tests on different topics will be presented and discussed.
Discussion: The search of the literature is an essential step in the elaboration of recommendations and so this process should be traceable and reproducible. The development of standardized strategies is a means of achieving this goal, and strategies are being developed and published for all types of studies, including economic analyses and quality of life assessments. One remaining problem is the standardization of searching terms across different databases and software used.