Reproducibility of resistance exercise therapy interventions in trials for chronic low back pain is challenging: A systematic review

Determine reproducibility of resistance exercise regimens in trials for CLBP and determine if recently available checklists are effective.Four databases (Medline, PubMed, Cochrane and CINAHL) were searched for keywords related to back pain and resistance exercise. Reproducibility was assessed using two checklists, the 12-item Template for Intervention Description and Replication (TIDieR) and the 19-item Consensus on Exercise Reporting Template (CERT). The proportion reporting was analysed, with additional comparison of trials pre- and post-availability of each checklist. A generalised linear regression was conducted with checklist items as the dependent variable and year of publication as the independent (PROSPERO ID = #CRD42020186036).Overall, details that facilitate reproducibility were under-reported. No trials reported all checklist items, while only 18 trials (35.5%) and 5 trials (9.8%) reported 75%+ of checklist items for the TIDieR and CERT, respectively. A median of 8 (IQR 2) of 12 TIDieR criteria were reported and a median of 9 (IQR 7) of 19 criteria were reported for the CERT. There was no difference pre/post checklist publication (TIDieR median before = 8 (IQR 2), after = 8 (IQR 2.25); CERT mean before = 9 (IQR 5.25), after = 9 (IQR 7)). Regression failed to support improved reporting over time. The majority of studies (86.3%) were scored as having an elevated risk of bias.Reproducibility of resistance exercise in CLBP trials appears questionable due to low levels of reporting. The publication reporting checklists have not resulted in improvement. Real-world reproducibility is questionable. There is a need to improve reporting to maximise reproducibility.The present results reveal a demand in improved reporting to ensure both enhanced clinical translation in the real-world and replicability to enhance knowledge of best-practice for resistance exercise in the CLBP population.