A systematic review of the processes used to link clinical trial registrations to their published results

Abstract

Background: Studies measuring the completeness and consistency of trial registration and reporting rely on linking registries with bibliographic databases. In this systematic review, we quantified the processes used to identify these links.

Methods: PubMed and Embase databases were searched from inception to May 2016 for studies linking trial registries with bibliographic databases. The processes used to establish these links were categorised as automatic when the registration identifier was available in the bibliographic database or publication, or manual when linkage required inference or contacting of trial investigators. The number of links identified by each process was extracted where available. Linear regression was used to determine whether the proportions of links available via automatic processes had increased over time.

Results: In 43 studies that examined cohorts of registry entries, 24 used automatic and manual processes to find articles; 3 only automatic; and 11 only manual (5 did not specify). Twelve studies reported results for both manual and automatic processes and showed that a median of 23% (range from 13 to 42%) included automatic links to articles, while 17% (range from 5 to 42%) of registry entries required manual processes to find articles. There was no evidence that the proportion of registry entries with automatic links had increased (R ² = 0.02, p = 0.36). In 39 studies that examined cohorts of articles, 21 used automatic and manual processes; 9 only automatic; and 2 only manual (7 did not specify). Sixteen studies reported numbers for automatic and manual processes and indicated that a median of 49% (range from 8 to 97%) of articles had automatic links to registry entries, and 10% (range from 0 to 28%) required manual processes to find registry entries. There was no evidence that the proportion of articles with automatic links to registry entries had increased (R ² = 0.01, p = 0.73).

Conclusions: The linkage of trial registries to their corresponding publications continues to require extensive manual processes. We did not find that the use of automatic linkage has increased over time. Further investigation is needed to inform approaches that will ensure publications are properly linked to trial registrations, thus enabling efficient monitoring of trial reporting.

Keywords: Clinical trials as topic; Publication bias; Reporting bias; Systematic reviews as topic; Trial registration.

Fig. 1

PRISMA flow diagram of study selection for a search and screening process that resulted in the inclusion of 81 studies

Fig. 2

The processes used to identify links in 81 included studies, including studies that examined automatic links only (red), both automatic and manual processes (purple), manual processes only (blue), and studies that did not report the processes used (grey)

Fig. 3

The proportions of published articles identified in cohorts of registry entries (top, 43 studies, ranging from 34 to 8907 registry entries) and the proportions of registry entries found in cohorts of published articles (bottom, 39 studies, ranging from 54 to 698 articles), with studies that only considered automatic links (red) and all other studies (blue). The circle areas are proportional to the study size

Fig. 4

The proportions of published articles found in cohorts of registry entries (12 studies, top) and the proportions of registry entries found in cohorts of published articles (16 studies, bottom), by automatic links (grey) and manual processes (blue)