Abstract
ObjectiveExamine how well non-pharmacological interventions are reported following concussion and whether reporting completeness has improved over time.DesignSystematic review.Data SourcesMEDLINE, Embase, PsycInfo, CINAHL, Web of Science up to May 2022.Eligibility CriteriaRCTs in English or French examining non-pharmacological interventions following concussion. We contacted authors to provide unreported information. Two reviewers independently rated reporting completeness using Template for Intervention Description and Replication (TIDieR), Consensus on Exercise Reporting Template (CERT), and international Consensus on Therapeutic Exercise aNd Training (i-CONTENT) checklists. Risk of bias was assessed with the Cochrane RoB- 2 Tool.Main ResultsWe screened 7456 studies and included 89 RCTs (n=46 high risk-of-bias), representing 9714 participants with concussion. Studies examined 11 different interventions, including sub- symptom threshold aerobic exercise, cervicovestibular therapy, physical and/or cognitive rest, vision therapy, education, psychotherapy, hyperbaric oxygen therapy, transcranial magnetic stimulation, blue light therapy, osteopathic manipulation, and head/neck cooling. The percentage of items completely reported was 80% (95%CI,79.5–80.5) (TIDieR), 83% (95%CI,81.7- 84.3) (CERT), and 81%(95%CI,80.7–81.3) (i-CONTENT). All studies reported TIDieR items 1, Brief name, 4, What procedures, 8, When and how much, and i-CONTENT item 1, Patient selection. The item reported most commonly on the CERT was 14a, Generic or tailored (89%). TIDier items 10, Modifications (33%;95%CI,32.9–33.1) and 11, How well (planned adherence) (42%;95%CI,41.9–42.1) were the least reported. Only CERT reporting completeness increased over 26 years.ConclusionsNon-pharmacological interventions following concussion are moderately-well reported and of high risk-of-bias. Incomplete published description of interventions potentially limits replication of findings and translation of evidence into clinical practice.