Abstract and Introduction
Little is known about the effectiveness of placebo interventions in patients with nonspecific low back pain (LBP). This systematic review assessed the magnitude of the effects of placebo interventions as compared to no intervention in randomized controlled trials (RCTs) including patients with LBP. Embase, MEDLINE (Ovid), and Cochrane CENTRAL databases were searched from inception to December 5, 2019. Randomized controlled trials comparing placebo intervention vs no intervention in adult patients with nonspecific LBP were included. Pain intensity, physical functioning, and health-related quality of life measured at short-term, medium-term, and long-term follow-up were the outcomes of this review. Twenty-one randomized controlled trials were included; one concerning acute LBP and one subacute LBP, whereas 19 studies reported on chronic LBP. In chronic LBP, placebo interventions were more effective than no intervention at short-term follow-up for pain intensity (standardized mean difference = −0.37, 95% confidence interval [CI] = −0.55 to −0.18, moderate-quality evidence), physical functioning (standardized mean difference −0.19, 95% CI = −0.39–0.01, moderate-quality evidence), and physical quality of life (mean difference = −2.71, 95% CI = −4.71–0.71, high-quality evidence), respectively. These effects were not significant at medium-term follow-up, and no data were available at long-term follow-up. These results show placebo interventions are more effective than no intervention at short-term follow-up in patients with chronic LBP. However, the magnitude of the effects is probably not clinically relevant (approximately 8 points on a 0–100 pain scale). Future research should identify effect modifiers and causal mechanisms explaining the short-term effects of placebo interventions in patients with chronic LBP.
Placebo effects capture the imagination of patients, clinicians, and researchers, but their definitions and clinical relevance have been a topic of debate for decades. In 1986, Grünbaum theorized that the observed effects of any intervention may be attributable to characteristic and contextual factors; he defined placebo as one of these factors.[21,29] From 2001 onward, Hróbjartsson and Gøtzsche published a series of systematic reviews concerning placebo interventions in clinical trials for all clinical conditions.[30–33] They included trial designs that included both a placebo intervention group and a no intervention group (eg, waiting lists) and were thus able to estimate the magnitude of placebo effects compared with no intervention in trials. In their most recent update, Hróbjartsson and Gøtzsche concluded that, generally speaking, placebo interventions do not have important clinical effects, with the sole exception of patient-reported outcomes, such as pain (standardized mean difference [SMD] = −0.28). Nevertheless, another research team systematically reviewing only randomized controlled trials (RCTs) conducted in patients with pain complaints (and including articles from an earlier version of the Hróbjartsson and Gøtzsche literature search) found that the pooled magnitude of placebo effects is very small (3.2 points reduction on a scale of 100). The magnitude of the placebo effect can depend on the method of delivery. Studies using placebo as an intervention have shown greater effects than studies using placebo as a control. Both situations are included in our study.
In the field of low back pain (LBP, the most disabling condition worldwide[13,24]), conservative interventions are often administered to patients, despite evidence suggesting that such interventions have modest effects(0–9 points on a scale of 0–100) at best when compared with placebo.[11,55,57] It has been suggested that sham oral medication could produce clinically meaningful change in pain scores in patients with LBP without a specific nociceptive source for their complaints (eg, LBP), thus contradicting earlier evidence regarding the magnitude of placebo interventions for pain. However, studies with a no intervention group were not included in this systematic review, which makes it impossible to actually estimate the magnitude of the placebo effect vs no intervention in LBP. Comparison of placebo intervention vs no intervention is important to correct for phenomena such as regression to the mean and natural history of the disease, which do not represent true placebo effects. Thus, the current knowledge about actual magnitude of placebo effects in LBP is still limited. As the literature search of Hróbjartsson and Gøtzsche is now over 10 years old and because uncertainty remains about the effectiveness of placebo interventions for LBP, it is time to gain further insight into the magnitude of placebo effects in the LBP field. The main objective of this systematic review is to assess the magnitude of effects of placebo interventions in comparison with no intervention for patients with LBP.
Pain. 2021;162(12):2792-2804. © 2021 Lippincott Williams & Wilkins