Investigation of The Extent of Adverse Events Reports Among Systematic Review: A Case Study of Medical Cannabis

Article type
Authors
Watanabe AH1, Hong H1, Lai NM2, Chaiyakunapruk N1
1Department of Pharmacotherapy, University of Utah, Salt Lake City
2School of Medicine, Taylor's University, Selangor
Abstract
Background: Several systematic reviews (SRs) have summarized the potential effectiveness of cannabinoids but it is unclear to what extent safety-related outcomes were incorporated. The PRISMA harms checklist was published in 2016 to improve reporting of safety-related outcomes among SRs but its impact on safety evidence reporting remains unclear.

Objectives: To investigate the extent of adverse events reports among SRs.

Methods: A search for SRs investigating the use of cannabis for medical purposes was conducted in PubMed, EMBASE, CENTRAL, and AMED for publications up until June 2019. Studies were included if they assessed clinical outcomes for a cannabis-based intervention. The outcome of interest was the inclusion of safety information in different sections of the SR. Data were extracted by two independent reviewers. Study quality was assessed using the Cochrane risk of bias.

Results: A total 1,491 records were identified and 121 SRs published between 1999 and 2019 were included. A total of 28 (23.1%) SRs included safety information in the title, 103 (85.1%) SRs included safety information in the abstract section, 71 (58.7%) SRs included safety as part of their objective, 83 (68.6%) SRs described safety outcomes in the method section, 112 (92.6%) SRs reported safety-related outcomes in the results section, 109 (90.1%) SRs included safety in the discussion section, and 86 (71.1%) SRs included safety in the conclusion section. The proportion of SRs reporting safety related outcomes in the discussion sections has increased after 2016 from 84.4% (95% CI: 75.5% - 93.3%) to 96.5% (95% CI: 91.7% - 101.3%, p = 0.03).

Conclusions: This study provides a better understanding of how adverse event was reported among systematic reviews over time. Our findings may highlight areas for improvement for the conduct and reporting of systematic review works to improve patient’s safety.