If your mind races at night or sleep keeps slipping away, you are asking a question that has real research behind it. Short answer: two randomized trials found that non-invasive vagus nerve stimulation did better than a sham (a fake, inactive session) on anxiety measures, and a 2026 pooled review found lower anxiety-scale scores across trials in people with sleep disorders. Those results come from specific devices in specific groups, and none of the FDA approvals listed below is for anxiety.
If you are comparing neck stimulators, our guide to how Truvaga and gammaCore differ explains what each one is sold or cleared for. Neither is cleared to treat anxiety.
Affiliate disclosure: You will find affiliate links on this page. This site may receive a commission on purchases made through them.
Quick facts, dated 2026-11-11
| Question | What the source shows | Source |
|---|---|---|
| Does stimulation beat sham for anxiety? | Students with raised anxiety scores: Beck Anxiety Inventory averaged 9.3 before and 5.1 at two weeks in the active group, with no change in the sham group. Community adults: active stimulation beat sham on anxiety symptoms in the first 14-day phase. | Ferreira 2024 full text; Jackowska 2025 abstract |
| Pooled anxiety-scale result | Standardized mean difference -0.53 (95% CI -0.74 to -0.32) across trials in people with sleep disorders, not anxiety disorders | Lin 2026 review abstract |
| FDA approval for anxiety | None of the FDA approvals listed on this page covers anxiety | FDA VNS approval record |
| Consumer vagus devices | Sold as general wellness products for stress and sleep, according to their makers | Maker pages, see the device table below |
Where the vagus nerve is and why it matters
The vagus nerve is the tenth cranial nerve and the longest of the cranial nerves. It starts in the brainstem and runs down toward the abdomen. A September 2026 Utah State University Extension article says it branches out to organs such as the heart, lungs and stomach.
A 2022 review of vagus nerve stimulation describes the nerve as a two-way communicator between brain and body. A neurodevelopmental disorders review says that parasympathetic activation is linked to “rest-and-digest” responses, and that the vagus nerve carries information that helps the body move into calm states.
The 2008 pilot study abstract on implanted stimulation for anxiety disorders reasoned that the nerve sends information to brain regions important in anxiety regulation: the locus coeruleus, orbitofrontal cortex, insula, hippocampus and amygdala.
What the studies on anxiety found
Each card below names the people studied, the device type, the comparison, the length and what changed. Each result belongs to the device and group named in its card.
Students with raised anxiety scores (Ferreira 2024)
| Item | What the study reports |
|---|---|
| Study | Ferreira et al., Frontiers in Integrative Neuroscience, 2024; double-blind, randomized, sham-controlled |
| People | Health-science students aged 18 to 25 with a Beck Anxiety Inventory (BAI) score of 7 or higher. Students with raised anxiety scores, not people with an anxiety disorder. The abstract says 42 participants; the Discussion says 22 of them were randomized to active or sham; the BAI statistics use 20. |
| Device | Transcutaneous auricular vagus nerve stimulation (taVNS), non-invasive |
| Comparison | Sham stimulation |
| Length | One intervention week, with sessions on three alternate days; measured before, right after and 2 weeks later |
| Anxiety result | BAI average 9.3 before, 4.1 right after and 5.1 at 2 weeks in the active group (p < 0.05); the sham group did not change |
| Other outcomes | Main outcomes were jaw-muscle activity and pressure pain threshold; no adverse events reported |
| Authors’ limits | Self-reported scale; the authors call for larger samples and objective anxiety measures |
Community adults with subthreshold symptoms (Jackowska 2025)
| Item | What the study reports |
|---|---|
| Study | Jackowska et al., Biological Psychology, 2025; single-blinded randomized trial. Jackowska 2025 abstract |
| People | 70 community-dwelling adults (28 men), mean age 49.33, range 18 to 75, with subthreshold affective symptoms and perceived stress |
| Device | Transcutaneous vagus nerve stimulation (tVNS) |
| Comparison | Sham stimulation |
| Length | Two 14-day phases: one set of groups in days 0 to 13, the other in days 14 to 28 |
| Anxiety result | Active beat sham for anxiety symptoms and perceived stress in the early-phase groups, not the late-phase groups; no advantage for depressive symptoms |
| Authors’ wording | Scaling to population level “tentatively indicates” a possible help for stress vulnerability and anxiety |
Pooled review in people with sleep disorders (Lin 2026)
| Item | What the study reports |
|---|---|
| Study | Lin et al., Psychology, Health & Medicine, published July 29, 2026; systematic review and meta-analysis. Lin 2026 review abstract |
| People | People with sleep disorders, not anxiety disorders; 14 randomized trials included, 11 in the pooled analysis |
| Device | Transcutaneous auricular vagus nerve stimulation (taVNS) |
| Length | Seven databases searched to January 10, 2025 |
| Anxiety result | Pooled anxiety-related scale scores: standardized mean difference -0.53 (95% CI -0.74 to -0.32; p < 0.00001). A standardized mean difference compares effects across different scales, so it is not a number of points. |
| Depression result | Pooled depression-related scale scores: standardized mean difference -0.67 (95% CI -1.09 to -0.25; p = 0.002) |
| Safety | Only mild side effects, mostly resolved on their own |
| Authors’ wording | Heterogeneity across studies; findings “should be considered preliminary” |
Adults with treatment-resistant anxiety disorders (George 2008)
| Item | What the study reports |
|---|---|
| Study | George et al., Brain Stimulation, 2008; open-label pilot. University of Minnesota research record |
| People | 11 adult outpatients with treatment-resistant OCD, panic disorder or PTSD; all had failed several medications and CBT |
| Device | Surgically implanted VNS (VNS Therapy System), added to stable medication |
| Comparison | None; open-label, no sham |
| Length | 12-week acute phase, then long-term follow-up; four patients were followed four years after implant |
| Anxiety result | Among 7 OCD patients, 3 were acute responders on the Y-BOCS; statistically significant Y-BOCS improvement at 7 of 18 quarters in follow-up; four years after implant, four patients (two OCD, one panic disorder, one PTSD) still on VNS with sustained improvement in anxiety scores compared with baseline |
| Funding | Data collected by Quintiles; statistical analyses by staff of Cyberonics, maker of the VNS Therapy System |
| Authors’ wording | VNS “generally tolerated”; “open data” suggest further double-blind studies “may be warranted” |
Stress response in healthy young adults (Sanchez-Perez 2023)
| Item | What the study reports |
|---|---|
| Study | Sanchez-Perez et al., Frontiers in Neuroscience, 2023; single-blind, sham-controlled crossover. 2023 crossover study full text |
| People | 19 young healthy volunteers |
| Device | Transcutaneous auricular VNS (taVNS), compared with sham and with stimulation of another nerve (median nerve) |
| Length | Four lab visits, with three lab stressors |
| Result | Two markers of sympathetic (“fight or flight”) output changed with taVNS versus sham (p = 0.015 and p = 0.029) |
| Anxiety scale | None; the study measured body signals, not a rating scale |
A study plan, not a result (Lai 2025 protocol)
| Item | What the study reports |
|---|---|
| Study | Lai et al., Heliyon, 2025; study protocol. Generalized anxiety disorder protocol |
| Plan | 60 adults with generalized anxiety disorder, randomized to active or sham transcutaneous VNS with electrical stimulation |
| Planned length | 8 weeks |
| Planned primary outcome | Treatment response on the Hamilton Anxiety Rating Scale; fMRI to examine brain mechanisms |
| Status | A protocol is a plan; this paper reports no results |
Trauma and depression research
A 2025 systematic review of trauma-related disorders screened 322 abstracts, included 7 randomized studies and reported no pooled numbers. Its authors describe the state of VNS as treatment for PTSD as “limited and does not meet clinical expectations,” with overall certainty of evidence “very low.” PTSD sits in the trauma-related group of disorders, separate from generalized anxiety and panic. In one included trial of a cervical device against sham, there was no significant difference on the clinician-rated PTSD scale (CAPS-5), and somatic anxiety on the Hamilton Anxiety scale was lower than sham.
A UT Dallas news report describes a nine-patient Phase 1 trial in treatment-resistant PTSD. Patients were symptom-free up to six months after completing traditional therapy paired with VNS. The news report says a double-blind, placebo-controlled Phase 2 pilot is ongoing.
For depression, a 2023 depression meta-analysis pooled 12 studies and 838 participants. Its authors say taVNS reduced Hamilton Depression scale scores, with low to very low certainty (GRADE) evidence. A Wu 2018 meta-analysis is a meta-analysis of trials in people with depression, with anxiety scales as secondary results. It pooled 4 studies and 222 people. Against sham, the Self-Rating Anxiety Scale mean difference was -6.57 (95% CI -9.30 to -3.84), while the clinician-rated Hamilton Anxiety scale showed no significant difference (-1.12, 95% CI -2.56 to 0.32, P = .13).
What FDA approval covers, and what it does not
| Device | FDA record | What the record covers |
|---|---|---|
| Implanted VNS Therapy System (surgically implanted) | P970003, epilepsy decision July 16, 1997; depression supplement letter dated July 15, 2005 | Epilepsy in the 1997 decision; in the 2005 letter, adjunctive long-term treatment of chronic or recurrent depression in adults 18 and older who have not had an adequate response to four or more antidepressant treatments |
| Vivistim System (implanted) | PMA P210007, decision August 27, 2021 | Stimulation during rehabilitation therapy to reduce upper-limb motor deficits in chronic ischemic stroke patients with moderate to severe arm impairment |
| gammaCore Sapphire (non-invasive, prescription) | 510(k) K182369, decision November 27, 2018, “Substantially Equivalent” | Adjunctive preventive treatment of cluster headache in adults; acute pain of episodic cluster headache in adults; acute pain of migraine headache in adults |
FDA uses different words for different routes. Its device approvals and clearances page calls the PMA “the most stringent type of device marketing application.” It describes a 510(k) as showing a device is “as safe and effective, that is, substantially equivalent” to a legally marketed device. Medicare’s national coverage rule for VNS covers FDA-approved VNS devices for treatment-resistant depression only inside a CMS-approved double-blind, randomized, placebo-controlled trial with at least one year of follow-up.
Consumer vagus devices: what the makers say
These devices are sold as general wellness products for stress and sleep. Each maker’s own wording is below.
| Device | What the maker says | Regulatory wording on the maker’s pages | Link |
|---|---|---|---|
| Truvaga Plus and Truvaga 350 | Electrical stimulation through the neck; the company says it has “not performed clinical testing on any other part of the body”; results vary by individual, and some people need 2 to 3 weeks of consistent use | “Truvaga devices are considered a wellness product and are not covered by health insurance” | Product Page |
| Pulsetto | “a smart neckband that is worn by users in order to assist them in relaxing”; the company says it “has been clinically tested” | “considered a general wellness device”; “The Pulsetto device is FCC-certified” | Product Page |
| Nuropod (Parasym) | A wearable the maker says is “scientifically proven to modulate the vagus nerve via precise electrical impulses” | “Third party certified device” | Product Page |
Does Pulsetto actually work?
Pulsetto’s own pages say the neckband is a general wellness device, and that the company does not claim to diagnose, treat or prevent any medical condition. A Pulsetto-funded paper printed on its website tested 15 healthy people. Each had one session with stimulation and one without, on different days. The only outcome was heart rate variability. The paper reports that the LF/HF ratio fell by an average of 64.5% after stimulation and 6.8% after the session with no stimulation. The paper says “Pulsetto Proof of Concept gave funding for the project.” Its abstract says five patients, while its body says 15. No anxiety questionnaire was used, so the paper measures heart-rate signals, not anxiety.
Apollo Neuro, Nurosym and the FCC label
Apollo Neuro’s vibration help article says the wearable “generates the vibration patterns and sends them to the body.” The company calls the wearable “a consumer wellness device” that “does not require FDA approval for sale” in its FDA help article. Nurosym’s help center article says its device is “a certified device (CE) in Europe.”
Pulsetto says its FCC certification means it is approved for well-being use. The FCC equipment authorization page covers radio frequency devices that can cause interference, and says such a device must be authorized before it can be marketed, imported or used in the United States.
Side effects, and who the makers say should skip these devices
| Source | Side effects named |
|---|---|
| Implanted VNS, per the Cleveland Clinic article | Coughing, hoarseness, voice change, neck or throat pain, shortness of breath, trouble speaking or swallowing, headache; serious ones include vocal cord paralysis, infection and device complications |
| Truvaga, per the maker | In clinical studies, less than 5% of participants had mild site discomfort or redness, muscle twitching, localized face, head or neck pain (including toothache), mild headache or mild dizziness |
| taVNS in sleep-disorder trials, per the Lin 2026 review | Only mild side effects, mostly resolved on their own |
| taVNS in students, per Ferreira 2024 | No adverse events reported |
| Cardiac caution, per Utah State University Extension, citing the Cleveland Clinic | Fainting or bradycardia (slow heart rate); people with heart conditions or low blood pressure should be cautious |
| Device | The maker says it should not be used by |
|---|---|
| Truvaga | Anyone with an active implantable medical device, such as a pacemaker, hearing aid implant, or any implanted electronic device |
| Pulsetto | Anyone under 18; people with a heart condition, irregular heartbeat, high or low blood pressure, or a past serious cardiac event; people with a history of seizures or fainting or a diagnosed neurological condition; people with neck surgery, serious head injury or unusual neck anatomy; anyone with an active electronic implant or metal implant, surgical hardware or stent in or near the neck; pregnancy, cancer treatment, or broken or damaged skin on the neck |
| Nuropod | People who have had surgery to cut the vagus nerve; people with a severely low resting heart rate; people with a permanent implanted metal or electronic device; pregnant women; children; people with severe coronary disease or a heart attack within 5 years; anyone with an active implanted device, unless a health professional advises otherwise |
| gammaCore Sapphire (prescription) | Patients with carotid atherosclerosis, prior cervical vagotomy, pregnancy, clinically significant high or low blood pressure, or abnormal heart rate, for whom safety and efficacy have not been evaluated; people with an active implanted device; people with a metal device such as a stent, bone plate or bone screw at or near the neck; people using another device at the same time |
If one of these applies to you, ask your primary care clinician before you try any of these devices.
Does vagus nerve stimulation slow your heart rate?
Vagus activity is closely tied to heart rate variability. A 2020 article on VNS and the heart says “the link between vagus nerve activity and the high-frequency component of heart rate variability (HRV) has been well established.” Its examples of heart slowing come from animals. In 1960, Sarnoff et al. found that VNS led to bradycardia (a slow heart rate) and decreased contractility of the atria in canines. The Cleveland Clinic caution above is the human safety point to keep in mind.
Autism and vagus nerve stimulation
The neurodevelopmental disorders review describes VNS as a potential adjunctive therapy for individuals with autism. None of the FDA approvals listed above is for autism.
Vagus nerve stimulation at home
Home options are consumer devices sold as wellness products, as the table above shows. The approved uses listed above are implanted systems, a prescription headache device, or a stroke-rehabilitation system used alongside rehabilitation therapy. The Cleveland Clinic says external devices “can’t match the frequencies and voltages used in implanted devices” and that “The long-term side effects of commercial VNS devices are yet to be fully known.” The 2018 510(k) record for gammaCore Sapphire marks prescription use.
What the research does not cover
- The anxiety-specific studies above enrolled between 11 and 70 people.
- None of the FDA approvals above is for anxiety.
- The implanted approvals involve surgery, and the Cleveland Clinic lists serious side effects for implanted systems.
- The Pulsetto-funded paper measures heart rate variability and uses no anxiety questionnaire.
Common mix-ups
- Instant results. The same article says VNS therapy for epilepsy and depression “often takes weeks or months to notice improvements.” Truvaga’s maker says some people notice effects “relatively quickly” while others may need 2 to 3 weeks of consistent use.
- Humming and gargling. The extension article says these practices “are not a standalone treatment for anxiety or depression.”
- Vibration is not stimulation. Apollo Neuro’s help article describes vibration patterns, not electrical current.
- Other conditions. A result for PTSD, stroke rehabilitation or headache is a result for those conditions, not for anxiety.
What health agencies list first for anxiety
The NHS says a GP will usually suggest talking therapies before medicine or specialist referral for generalised anxiety disorder. Talking therapy usually means cognitive behavioural therapy (CBT), and medicine is usually an SSRI, according to the NHS generalised anxiety treatment page. The same page lists self-help options: self-help books or online courses, talking to someone, calming breathing exercises, yoga or pilates, regular exercise, enough sleep, a regular diet, and avoiding caffeine, alcohol, cigarettes and drugs.
For panic disorder, the NHS panic disorder page lists talking therapies, especially CBT, and medicines such as SSRIs or tricyclic antidepressants. It says antidepressants can take 2 to 4 weeks to start working and up to 8 weeks to work fully. During an attack, it advises breathing slowly and deeply.
The National Institute of Mental Health lists two main approaches, often used alone or together: psychotherapy (CBT and acceptance and commitment therapy) and medication (SSRIs and SNRIs, plus benzodiazepines and buspirone).
Questions to take to your clinician
- Which first-line option fits my anxiety: talking therapy, medicine, or both?
- Do I qualify for referral to a community mental health team?
- Does any device I am considering suit my health history, including a heart condition, blood pressure problems, an implant, or pregnancy?
- Before changing any medicine, I will ask my prescriber.
For non-prescription options, our guides on natural anxiety relief supplements and mental health supplements cover the research on each.
By the Global MH Summit Research Team
This article is for general information purposes only and does not constitute medical advice. Consult your doctor or qualified healthcare provider before making changes to your health routine.