Searches for evidence mapping: effective, shorter, cheaper

Abstract

An evidence map is an overview of a broad research field that describes the volume, nature, and characteristics of research in that field [1]. Evidence maps can indicate research links, gaps, and strengths in a broad clinical context, while systematic reviews focus on a single clinical question [2, 3]. Maps complement systematic reviews by engaging stakeholders to identify and prioritize questions that may be informed by research evidence, highlighting the applicability of research evidence to different populations and contexts, and identifying important gaps that can inform further primary research or systematic reviews [3]. Evidence mapping can be a labor-intensive process, as it must provide an overview of the nature and characteristics of all research in any given field (Figure 1, online only). There are many more potentially useful references to review when mapping a topic area than when answering a specific clinical question, as in a systematic review. The searching challenges for evidence mapping are similar to those of a rapid review, which also needs to produce high-quality evidence with reduced resources. In the case of the rapid review, the time needed is limited [4]. The Global Evidence Mapping (GEM) Initiative was funded in 2007 by the Victorian Neurotrauma Initiative (Australia) to map the research addressing important questions about treatments, diagnostic tests, prognosis, and cost effectiveness in the broad clinical areas of traumatic brain injury (TBI) and spinal cord injury (SCI). The project involved identifying the scope of clinical issues in pre-hospital, acute, and rehabilitation (long-term care) for TBI and SCI; identifying key questions in the area; and developing evidence maps to support strategic allocation of research funds. Although a number of sources, including, but not confined to, electronic databases [5, 6], were routinely checked as part of this project, this paper describes an evaluation of just the electronic database-searching procedures and methodologies that were developed to populate the evidence maps. Pragmatic changes to otherwise rigorous and accepted procedures were done only after consideration of time and benefit. At the end of the project, a newly developed evidence mapping search method was compared with highly sensitive systematic review searches using PubMed MEDLINE as the test searching platform. The evaluation of the effectiveness of these two methods was based on the following parameters: yield, sensitivity, time requirements, and resource use.