U.S. Veterans with low back pain benefit from transcranial direct current stimulation: Results from a cognitive behavioral therapy randomized controlled trial
Published September 5, 2026. In 115 veterans receiving chronic-pain CBT, the group assigned real tDCS improved faster in average pain and pain-related functioning than the sham group. By the end of treatment, however, the groups did not differ, and opioid-misuse scores improved similarly in both groups. This suggests a possible effect on the pace of improvement, not proof of a larger final benefit or an effect on opioid misuse.
Douglass MA, Santa Ana EJ, Mappin GM, et al.
The Journal of Pain · Publication date: 2026-09-05
Added to site: 2026-11-22
DOI: 10.1016/j.jpain.2026.106434
Published online September 5, 2026. Added to Neurofeedback.com November 22, 2026 after a November 20 university report surfaced it for comparison with fresh and retained candidates. The source publication date is separate from the site addition date.
- Modalities
- Transcranial direct current stimulation, Noninvasive brain stimulation, Cognitive behavioral therapy
- Clinical focus
- Chronic low back pain, Prescription opioid use disorder, Treatment trajectory and functional impairment
- Population
- Treatment-seeking U.S. veterans with chronic low back pain and DSM-5 prescription opioid use disorder
- Sample
- 115 randomized, 64 to CBT-CP plus real tDCS and 51 to CBT-CP plus sham tDCS; the planned enrollment target was 150
- Intervention / method
- Participants completed eleven weekly 60-minute cognitive behavioral therapy for chronic pain sessions. During each session they received 30 minutes of real or sham tDCS targeting the left dorsolateral prefrontal cortex. Pain, functional impairment and opioid misuse were assessed repeatedly through six-month follow-up. The registered trial was NCT02483468.
- Comparator
- The same CBT-CP course plus sham tDCS, with participants, therapists and study personnel blinded to allocation
Clinician summary
Mixed-effects models favored real tDCS for the rate of change in average pain, time-by-group B=0.061, 95% CI 0.037 to 0.086, p<.001, and functional impairment, B=0.069, 95% CI 0.037 to 0.101, p<.001. End-of-treatment levels did not differ. Opioid misuse decreased over time in both groups without a treatment interaction, B=-0.002, 95% CI -0.011 to 0.007, p=.612. No serious adverse event related to tDCS or CBT-CP was reported.
How to interpret the findings
The trial enrolled 115 participants rather than the planned 150 and produced imbalanced groups of 64 and 51. The specialized veteran sample with chronic pain and prescription opioid use disorder may not generalize to civilians, other pain conditions or people without opioid misuse. Faster model-predicted improvement did not yield a between-group difference at treatment completion. Follow-up attrition occurred, and a complete allocation-specific accounting of non-serious adverse events was not identified in the checked source sections. The results do not establish standalone tDCS efficacy, a durable final advantage or reduced opioid misuse.
Equipment and disclosures
Coded tDCS devices delivered real or sham stimulation during CBT-CP. The source does not establish superiority for a commercial device or protocol, and no equipment is attributed to BrainMaster or another site sponsor.
The authors declared no competing interests. Morgan Douglass was supported by National Institute on Aging grant T32AG044296, and the study was supported by 1R01DA038971. The authors state that the funder had no role in study design, data collection, analysis, interpretation or manuscript preparation. No correction, retraction or expression-of-concern notice tied to the DOI appeared in the checked PubMed, Crossref, publisher and bounded DOI and title searches on November 22, 2026; this is not assurance of absence. AI-assisted source review for Neurofeedback.com, not independent human clinical review.
Related evidence
Compare findings from the same clinical area, including mixed results.
Summary prepared with AI assistance. Source verification does not imply an independent replication or a clinician's endorsement. Read our review method.