Current treatments for chronic pain have limited effectiveness and tolerability. With growing interest in the potential of cannabinoids, there is a need to inform risk-benefit considerations. Thus, this focused systematic review assesses the quality of safety assessment and reporting in chronic noncancer pain cannabinoid trials.
The protocol for this review has been published, and, registered in PROSPERO. We searched MEDLINE, Embase, The Cochrane Library, Scopus, and PsychINFO for double-blind, placebo-controlled, randomized controlled trials of cannabinoids for chronic pain, with a primary outcome related to pain. The primary review outcome is adherence to the 2004 Consolidated Standards of Reporting Trials (CONSORT) Harms extension. Secondary outcomes included type, reporting method, frequency and severity of adverse events (AEs), trial participant withdrawals, and reasons for withdrawals.
In total, 43 studies (4436 participants) were included. Type of cannabinoid (number of studies) included nabiximols (12), dronabinol (8), nabilone (7), oral cannabis extract preparations (5), smoked tetrahydrocannabinol (5), vaporized tetrahydrocannabinol (3), novel synthetic cannabinoids (2), sublingual cannabis extract preparations (1). The median CONSORT score was 7. On average, 3 to 4 recommendations of the CONSORT guidelines were not being met in trials. Seventeen trials did not provide their method of AE assessment, 14 trials did not report on serious AEs and, 7 trials provided no quantitative data about AEs.
Better harms assessment and reporting are needed in chronic pain cannabinoid trials. Improvements may be achieved through: expanded education/knowledge translation increased research regulation by ethics boards, funding agencies and journals, and greater emphasis on safety assessment and reporting throughout research training.