Treatment- and sham-related improvement and associated factors in non-invasive brain stimulation studies in treatment-resistant schizophrenia: a systematic review and meta-regression analysis.

BACKGROUND

The magnitude and determinants of treatment effects in non-invasive brain stimulation (NIBS) trials for treatment-resistant schizophrenia (TRS) remain unclear, and sham improvements are suggested to dilute them. We aimed to quantify sham-related improvement, NIBS-related improvement, NIBS-sham differences, and factors associated with each.

METHODS

We conducted a systematic review and meta-analysis of rater-blinded randomized controlled trials comparing NIBS with sham in TRS. Random-effects models estimated pooled standardized mean changes (SMC) from baseline for sham and NIBS arms, and standardized mean differences for NIBS-sham differences. Meta-regressions examined associations and sources of heterogeneity.

RESULTS

65 trials with 2949 participants were included. Sham-related improvement was moderate (SMC = 0.53, 95% CI 0.42-0.65), with no clear differences among sham modalities. Sham-related improvement was correlated with NIBS-related improvement (β = 0.63, 95% CI 0.41-0.85), but not with their NIBS-sham differences (β = -0.11, 95% CI -0.37-0.15). Chinese mainland trials were associated with greater NIBS-related improvement (β = 0.54, 95% CI 0.30-0.79) and greater treatment effects (β = 0.46, 95% CI 0.20-0.72). Similarly, shorter illness duration was associated with greater NIBS-related improvement (β = -0.40 per 10-year increase, 95% CI -0.70--0.10) and larger treatment effects (β = -0.32 per 10-year increase, 95% CI -0.64--0.00). Some concerns in randomization were associated with larger treatment effects (β = 0.38, 95% CI 0.10-0.66).

CONCLUSIONS

Sham improvement in NIBS trials for TRS was moderate and consistent, indicating substantial non-specific symptom improvement. Future trials are needed to distinguish, minimize these effects in trials, or potentially leverage them in clinical contexts.