Clinical improvements due to specific effects and placebo effects in conservative interventions and changes observed with no treatment in randomized controlled trials of patients with chronic nonspecific low back pain: a systematic review and meta-analysis

Abstract Little is known about the contribution of placebo effects and changes observed with no treatment in interventions for nonspecific low back pain (NSLBP). This systematic review assessed the proportions of the overall treatment effect that may be attributable to specific treatment effects, placebo effects, and changes observed with no treatment in randomized controlled trials (RCTs) in patients with NSLBP. Trials published before 2019 were identified from a published systematic review, and the search was updated in MEDLINE, Embase, and Cochrane Central for trials published between January 2019 and March 2023. Three-arm RCTs comparing the effects of experimental interventions vs placebo control vs no intervention reporting pain intensity, physical function, and/or health-related quality of life (HRQoL) were included. Sixteen RCTs with 1436 adults with chronic NSLBP testing conservative and mainly passive interventions were included. For pain intensity (16 studies), 33%, 18%, and 49% of the overall short-term treatment effect was attributable to specific treatment effects, placebo effects, and changes observed with no treatment, respectively. For physical function (11 studies) and HRQoL (6 studies), these proportions were 34%, 13%, and 53%, and 11%, 41%, and 48%, respectively. These results show that approximately half of the overall treatment effect of conservative and mainly passive interventions for patients with chronic NSLBP is attributable to changes observed with no treatment, rather than specific or placebo effects of treatments. However, the certainty of evidence was very low to low, suggesting that the true effects might be markedly different from the effect sizes underlying these estimates.