THE ROLE OF GREY LITERATURE IN META-ANALYSIS

Laura M McAuley, David Moher, and Peter Tugwell
Clinical Epidemiology Unit, Ottawa Civic Hospital, C4, 1053 Carling Ave, Ottawa, Ontario K1Y 4E9, Canada

Objectives: (l) To characterize the sources of grey literature, defined as difficult to find literature, and the prevalence of its citation in a random sample of meta-analyses in the medical literature. (2) To assess and compare the quality of (a) meta-analyses that include grey literature to those which do not, and (b) traditional and grey documents included in meta-analyses. (3) To determine if the inclusion of grey literature in meta-analysis influences the point estimate, the precision, or the heterogeneity of the reported results.

Design: A random sample of 135 meta-analyses was drawn from an existing database of meta-analyses published between 1983 and 1995. For each meta-analysis, data was collected a variety of parameters, including quality. For the subset of meta-analyses with grey literature, the included studies were retrieved, and the analyses repeated and replicated.

Results: To date, 132 meta-analyses have been retrieved. Of those, 28% include some form of grey literature. Twenty-two percent include grey literature from more than 1 source. Abstracts comprise the most frequent source of grey literature (56%).

Conclusions: There is much variability in the literature classified as meta-analyses. It ranges from publications with no formal literature search, to those with an explicit and finely detailed search of the literature. There does not appear to be any trend toward the inclusion or exclusion of grey literature in meta-analyses. This research will provide evidence to those involved in the peer-review process as to the relative importance of grey literature in meta-analysis.

Return to Session Information