Economic Sources for Systematic Reviews of Health Policy

Abstract(s)

Purpose: To explore and assess the utility of consulting non-biomedical sources in searches for the systematic review in health care policy. Setting: An academic health policy research centre, specifically the research group focusing on pharmaceutical policy. Methods: Case studies of two literature searches for systematic reviews of health policy interventions. All citations retrieved for the reviews were tagged with their source(s) of origin (typically database name). Absolute number and percentage of relevant citations per database were tracked after the initial “weed,” after the full-text weed, and after the reviewers had assessed full-text articles for review inclusion criteria. Citation sources were analyzed for number of results used in the review, number of unique results used in the review (found in only one source), and “missed” results (turned up in another database search and “should” have been duplicated in that database as well). Results: Initial results indicate that Medline and EMBASE, while highest in number of citations, had very few unique citations. Non-biomedical sources, such as economic and business databases, on the other hand, garnered unique, relevant results not indexed in biomedical databases. Citation tracing “snowballing” garnered more unique results than any single biomedical database. Discussion: Investigators seeking to complete a systematic review of health policy should consider the inclusion of non-biomedical databases, particularly economic databases, in their search strategies. Citation tracing is reaffirmed as a critical element of the systematic review search process.