*2.1. Search Strategy*

We used the PubMed database to identify studies investigating DBS for MDD and screened all studies for inclusion. We used the following search terms to identify relevant studies: ("deep brain stimulation"[MeSH Terms] OR ("deep"[All Fields] AND "brain"[All Fields] AND "stimulation"[All Fields]) OR "deep brain stimulation"[All Fields] OR "DBS"[All Fields]) AND ("depressive disorder"[MeSH Terms] OR ("depressive"[All Fields] AND "disorder"[All Fields]) OR "depressive disorder"[All Fields] OR "depression"[All Fields] OR "depression"[MeSH Terms]). All studies were considered, including studies written in other languages. The search was conducted on 10/16/2019, and the analysis followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines.

### *2.2. Study Inclusion and Exclusion Criteria*

Only studies that investigated the e fficacy of DBS for MDD were included. We excluded studies that utilized other therapies to treat MDD (e.g., ECT, epidural stimulation, vagal nerve stimulation, transcranial magnetic stimulation, or tDCS). We also excluded studies that investigated comorbid depression in the context of other disorders, such as epilepsy, dystonia, Tourette syndrome, anorexia nervosa, obsessive–compulsive disorder, schizophrenia, headache, ET, and PD. We excluded all non-human studies. Of the studies relevant to DBS as a treatment for MDD, we excluded case reports, non-systematic reviews, perspectives, commentaries, editorials, and opinions. We included the remainder of the studies for our qualitative review. For the quantitative meta-analysis, we only included studies in which sham stimulation was compared to active stimulation in a blinded fashion (either single- or double-blind). The clinical trial designs were varied and included both crossover and parallel studies.

### *2.3. Data Extraction and Outcome Measures*

Our primary outcome was the e fficacy of DBS as a treatment for depression as assessed by changes in the Hamilton Depression Rating Scale (HDRS) or Montgomery–Åsberg Depression Rating Scale (MADRS) scores. We compared sham stimulation scores to active stimulation scores. The data were extracted from tables when provided. If tables with the raw data were not provided, the WebPlotDigitizer tool was used to extract data from published graphs. We also extracted the number of patients, stimulation target, side e ffects of treatment, adverse events, study design, and depression rating scale used.
