Design: Systematic review and meta-analysis of six sham-controlled randomised trials
Population: Adults with depressive disorders using home-based tDCS
Primary outcome: Depressive symptom change versus sham
Effect: g 0.36 (95% CI 0.06–0.66); non-significant after removing the largest positive trial

This systematic review included 12 studies of transcranial direct current stimulation (tDCS) given at home with varying remote supervision; six were sham-controlled randomised trials and formed the meta-analysis.

Active stimulation beat sham with a small pooled effect (Hedges g 0.36, 95% CI 0.06–0.66; I² = 34%). Removing the single largest positive trial left the effect non-significant (g 0.39, −0.12 to 0.91). The two largest trials — one unsupervised with 210 participants, one self-administered with 141 — were negative on their primary outcomes, while the largest trial with real-time supervision (174 participants) was positive. The pooled effect sat at or near the minimal clinically important difference. Adverse events were mostly mild; one pilot trial stopped early for skin lesions and one non-fatal suicide attempt occurred in an unsupervised trial.

The authors conclude the evidence does not support routine use, and that is the right reading. Patients asking about devices sold for home use should hear that the benefit, if any, is small and uncertain.

Tell patients asking about home tDCS devices that the evidence is inconclusive
Do not substitute home tDCS for an adequate antidepressant or psychological treatment
If a patient is already using a device, ask about skin burns and review suicide risk
Refer interested patients to a trial rather than an unsupervised purchase

Why it matters

Home stimulation devices are marketed directly to patients, and the controlled evidence does not yet support them.

The statistics, in plain English

Hedges g of 0.36 is a small standardised effect, roughly a few points on a depression rating scale. When removing one study moves the interval to cross zero, the result is fragile: the pooled number is being carried by that study rather than by consistent findings across trials.