Should Systematic Reviewers Search for Randomized, Controlled Trials Published as Letters?
- Alla E. Iansavichene, MLIS;
- Margaret Sampson, MLIS;
- Jessie McGowan, MLIS; and
- Isola S.Y. Ajiferuke, PhD
- From London Health Sciences Centre and University of Western Ontario, London, Ontario, Canada, and Children's Hospital of Eastern Ontario Research Institute and Institute of Population Health, University of Ottawa, Ottawa, Ontario, Canada.
Background: A report by Deeks and Altman (1) found randomized, controlled trials (RCTs) published as letters to be incompletely reported and noncompliant with the Consolidated Standards of Reporting Trials (CONSORT). The authors recommended that journals reexamine their policies of publishing RCTs as short reports.
We examined RCTs published as letters before and after these recommendations. Our interest stems from our involvement in information retrieval for systematic reviews. A rigorous search strategy is a key component in identifying evidence (2), because an inadequate search strategy could result in missed studies and misleading conclusions. Yet, these searches often exclude certain publication types, such as letters and editorials (3). This is thought to improve the search precision and reduce reviewer burden by eliminating irrelevant citations.
Objective: To determine whether RCTs would be missed by excluding letters from the …
This 100-word excerpt has been provided in the absence of an abstract.
Most Read
