How completely are randomized controlled trials of non-pharmacological interventions following concussion reported? A systematic review.

Abstract

Purpose: The study aimed to examine the reporting completeness of randomized controlled trials (RCTs) of non-pharmacological interventions following concussion.

Methods: We searched MEDLINE, Embase, PsycInfo, CINAHL, and Web of Science up to May 2022. Two reviewers independently screened studies and assessed reporting completeness using the Template for Intervention Description and Replication (TIDieR), Consensus on Exercise Reporting Template (CERT), and international Consensus on Therapeutic Exercise aNd Training (i-CONTENT) checklists. Additional information was sought my study authors where reporting was incomplete. Risk of bias (ROB) was assessed with the Cochrane ROB-2 Tool. RCTs examining non-pharmacological interventions following concussion.

Results: We included 89 RCTs (n = 53 high ROB) examining 11 different interventions for concussion: sub-symptom threshold aerobic exercise, cervicovestibular therapy, physical/cognitive rest, vision therapy, education, psychotherapy, hyperbaric oxygen therapy, transcranial magnetic stimulation, blue light therapy, osteopathic manipulation, and head/neck cooling. Median scores were: TIDieR 9/12 (75%; interquartile range (IQR) = 5; range: 5-12), CERT 17/19 (89%; IQR = 2; range: 10-19), and i-CONTENT 6/7 (86%; IQR = 1; range: 5-7). Percentage of studies completely reporting all items was TIDieR 35% (31/89), CERT 24% (5/21), and i-CONTENT 10% (2/21). Studies were more completely reported after publication of TIDieR (t₈₇ = 2.08; p = 0.04) and CERT (t₁₉ = 2.72; p = 0.01). Reporting completeness was not strongly associated with journal impact factor (TIDieR: r_s = 0.27; p = 0.01; CERT: r_s = -0.44; p = 0.06; i-CONTENT: r_s = -0.17; p = 0.48) or ROB (TIDieR: r_s = 0.11; p = 0.31; CERT: r_s = 0.04; p = 0.86; i-CONTENT: r_s = 0.12; p = 0.60).

Conclusion: RCTs of non-pharmacological interventions following concussion demonstrate moderate to good reporting completeness, but are often missing key components, particularly modifications, motivational strategies, and qualified supervisor. Reporting completeness improved after TIDieR and CERT publication, but publication in highly cited journals and low ROB do not guarantee reporting completeness.