Open Access Archiving and Article Citations within Health Services and Policy Research

Abstract

Objective: The Canadian Institutes of Health Research (CIHR) is now among the many funders that require grant recipients to make research outputs Open Access (OA). This study describes OA-archiving practises within journals containing Canadian health services and policy research, and examines the association between OA status and citations to an article. Methods: We employed an article-level analysis comparing citation rates for articles drawn from the same, purposively selected journals. We used descriptive statistics to describe archiving practises, and a two-stage analytic approach designed to test whether OA is associated with likelihood that an article is cited at all and total number citations that an article receives, conditional on being cited at least once. Results: Adjusting for several potential confounders, OA archived articles were 60% more likely to be cited at least once, and, once cited, were cited 29% more than non-OA articles. The majority of articles that were made OA were archived in just one location, and archived copies found were primarily on websites rather than institutional or subject-based repositories.Discussion: It appears that there may be a citation “advantage” associated with articles made open access in this field. Whether this advantage is solely a result of OA status cannot be confirmed from this data alone. Regarding archiving practises, it is concerning that the vast majority of archived copies are on web sites, as findability and preservation of these archived copies may be inferior to those in centralized institutional or subject-based repositories.