Categorical response rate differences between the intervention groups according to MADRS score did not reach statistical significance for partial response (
p = 0.08), response (
p = 0.36), or remission (
p = 0.34) criteria (Supplementary Table
2). A MADRS item-level analysis showed that compared to sham, active tsDCS induced a greater decrease in LS mean (±SE) MADRS “reported sadness” item (−1.8 ± 0.4 vs. −3.2 ± 0.4,
p = 0.012). A statistical trend in the same direction was observed for “pessimistic thoughts” item (−0.8 ± 0.5 vs. −1.9 ± 0.4, 0.094), as well as week-8 clinical global impression-improvement (CGI-I) scale scores (2.0 ± 0.3 vs. 1.3 ± 0.3,
p = 0.091). Although greater numerical decreases in the active group were observed on all MADRS items (except “reduced sleep”), no other statistically significant difference was observed between groups. There were no significant differences in change from baseline to week 8 between intervention groups on self-reported PHQ-9, MAIA, 4-DSQ, and BES scales (Table
2).