Neurofeedback describes techniques of providing information about neuronal activity as measured by electroencephalogram (EEG) or functional magnetic resonance imaging (fMRI), to teach patients to self-regulate brain activity. The techniques attempt to normalize patterns of brain function in patients with a variety of disorders. These include but are not limited to autism spectrum disorder, learning disabilities, Tourette syndrome, traumatic brain injury, seizure disorders, menopausal hot flashes, panic and anxiety disorders, fibromyalgia, tinnitus, substance abuse disorders, eating disorders, depression, stress management, migraine headaches, stroke, Parkinson’s disease, premenstrual dysphoric disorder, posttraumatic stress disorder, and sleep disorders.
A number of EEG biofeedback systems (EEG hardware and computer software programs) have been cleared for marketing by the U.S. Food and Drug Administration (FDA) through the 510(k) process. For example, the BrainMaster™ 2E is indicated for relaxation training using alpha EEG Biofeedback. In the protocol for relaxation, the BrainMaster™ provides a visual and/or auditory signal that corresponds to the patient's increase in alpha activity as an indicator of achieving a state of relaxation. Although devices used during neurofeedback may be subject to FDA regulation, the process of neurofeedback itself is a procedure and, therefore, not subject to FDA approval.
Neurofeedback is considered EXPERIMENTAL/INVESTIGATIVE for all indications due to a lack of clinical evidence demonstrating an impact on improved health outcomes.
No additional statements.
Summary of Evidence
For individuals who have attention-deficit/hyperactivity disorder (ADHD) who receive neurofeedback, the evidence includes randomized controlled trials (RCTs) and meta-analyses. At least 5 moderately sized RCTs (N range, 90-102 patients) have compared neurofeedback with methylphenidate, attention skills training, or cognitive therapy. These trials found either small or no benefit of neurofeedback. Studies that used active controls have suggested that, at least part of the effect of neurofeedback may be due to attention skills training, relaxation training, and/or other nonspecific effects.
For individuals who have disorders other than ADHD (e.g., epilepsy, substance abuse, pediatric brain tumors, PTSD) who receive neurofeedback, the evidence includes case reports, case series, comparative cohorts, and small RCTs. For these other disorders, including psychiatric, neurologic, and pain syndromes, the evidence is poor and several questions concerning clinical efficacy remain unanswered. Larger RCTs that include either a sham or active control are needed to evaluate the effect of neurofeedback for these conditions.
Rationale
The largest body of evidence on neurofeedback addresses treatment of ADHD. Several meta-analyses and 5 additional moderately sized RCTs (N range, 144 to 202 patients) have compared neurofeedback with methylphenidate, biofeedback, cognitive behavioral therapy, cognitive training, or physical activity These studies found either small to moderate or no benefit of neurofeedback, and sustained long-term benefit (e.g., at 6 to 13 months) has not been consistently demonstrated. Studies using active controls have suggested that at least part of the effect of neurofeedback might be due to attention skills training, biofeedback, relaxation training, and/or other nonspecific effects. Two of the RCTs indicated that any beneficial effects were more likely to be reported by evaluators unblinded to treatment (parents), than by evaluators blinded (teachers) to treatment, which would suggest bias in the nonblinded evaluations. Moreover, a meta-analysis found no effect of neurofeedback on objective measures of attention and inhibition. Additional research with blinded evaluation of outcomes is needed to demonstrate the effect of neurofeedback on ADHD.
A 2014 systematic review of neurofeedback in children with ADHD included 5 studies with a total of 263 patients. The active treatment was theta/beta ratio training or slow cortical potential training. Control treatments included cognitive remediation, sham neurofeedback, or electromyography (EMG) biofeedback. Meta-analysis found a significant benefit from parent (probably not blinded) assessment on the overall ADHD score (standardized mean difference [SMD] = -0.49), the hyperactivity/impulsivity score (SMD = -0.34), and the inattention score (SMD = -0.46). This is considered to be a moderate effect size. For teacher assessment, which is more likely to be blinded, only the inattention score showed statistically significant improvement with neurofeedback (SMD = -0.30).
Steiner et al randomized 104 children with ADHD age 7-11 years to receive neurofeedback, cognitive training, or a no-intervention control condition in their elementary school. Both the neurofeedback and cognitive therapies were administered with commercially available computer programs (45-minute sessions 3 times per week), monitored by a trained research assistant. The neurofeedback EEG sensor was embedded in a standard bicycle helmet with the grounding and reference sensors located on the chin straps on the mastoids. No data were presented on the technical performance of this system. There were some differences in baseline measures between the groups, although these differences were not large. The slope of the change in scores over time was compared between groups. Children in the neurofeedback group showed a small improvement on the Conners 3-Parent Assessment Report (ES=0.34 for inattention, ES=0.25 for executive functioning, ES=0.23 for hyperactivity/impulsivity), and subscales of the Behavior Rating Inventory of Executive Function‒Parent Form (global executive composite, ES=0.23) when compared with baseline. Interpretation of these findings are limited by the use of a no-intervention control group and lack of parental blinding. Evaluator-blinded classroom observation (Behavioral Observation of Students in Schools) found no sustained change with a linear growth model but a significant improvement with a quadratic model. No between-group difference in change in medication was observed at the 6-month follow-up.
In 2012, Duric et al reported a comparative study of neurofeedback versus methylphenidate in 91 children with ADHD. The children were randomized into 3 groups, consisting of 30 sessions of neurofeedback, methylphenidate, or a combination of neurofeedback and methylphenidate. Parental evaluations found improvements in ADHD core symptoms for all 3 groups, with no significant differences between groups. Alternative reasons for improvement with neurofeedback include the amount of time spent with the therapist and cognitive-behavioral training introduced under neurofeedback. In a 2014 publication of self-reports from this study, all treatments resulted in significant improvements in attention and hyperactivity (P<0.001). (10) Only those in the neurofeedback group showed a significant improvement in self-reported school performance (P=0.042). The authors noted that influence of change in parental style, parents' expectations and satisfaction with treatment, might have affected the behavior and school performance and therefore confounded outcome variables. In addition, authors reported that a sham NF placebo was found unfeasible and that no consensus standard for number and frequency of sessions has not been established.
An RCT published in 2015 randomized 24 MS patients with primary fatigue and depression. Participants were randomized into two groups: neurofeedback training group (16 sessions of neurofeedback) or treatment as usual. Participants were evaluated at 3 time points (baseline, end of the treatment, and 2-month follow-up) using the Fatigue Severity Scale and Depression subscale of the Hospital Anxiety and Depression Scale as outcome measures. A repeated measures analysis of variance was used to examine differences between the groups. Neurofeedback significantly reduced symptoms of depression and fatigue in patients with MS patients, compared to treatment as usual (p <0.05), and these effects were maintained the 2-month follow-up (p < 0.05).
A controlled trial reported in 2014 utilized two experimental groups and a control group. Independent variables were neurofeedback and a transcutaneous electrical nerve stimulation (TENS; Cefaly® device). The Blanchard headache diary was used for assessment. Forty-five healthcare providers with primary headache were selected and randomly allocated to the neurofeedback, TENS, and control groups by block random assignment method. All three groups completed the headache diary during one week before and after the treatment period as pretest and posttests, respectively. Participants in the neurofeedback group were treated in the period between pretest and posttest with fifteen 30-minute treatment sessions three times a week and the TENS group was treated with fifteen 20-minute daily sessions. The control group received none of these treatments. Both neurofeedback and TENS resulted in a significant decrease in the frequency, severity, and duration of headache in experimental groups compared to the control group. Neurofeedback was more effective in reducing headache frequency and severity. Weaknesses of this trial include the small sample size, lack of random sampling and no controls for other treatments or lifestyle factors that may confound the results.
Preliminary studies on treatment of insomnia using neurofeedback have been published. In 2010, Cortoos et al compared tele-neurofeedback (n = 9) or an electromyography tele-biofeedback (n = 8) with 12 healthy controls, which were used to compare baseline sleep measures. A polysomnography was performed pre and post treatment. Total Sleep Time (TST), was the primary outcome variable. Sleep latency decreased pre to post treatment in both groups, but a significant improvement in TST was found only after the neurofeedback protocol. Furthermore, sleep logs at home showed an overall improvement only in the neurofeedback group, whereas the sleep logs in the lab remained the same pre to post training. Authors noted that the mixed results concerning perception of sleep might be related to methodological issues, such as the different locations of the training and sleep measurements.
Hammer et al published a 2011 pilot study to compare an SMR protocol and a sequential, quantitative EEG (sQEEG)-guided, individually designed (IND) protocol, to alleviate insomnia and associated daytime dysfunctions of participants with insomnia. Both protocols used instantaneous Z scores to determine reward condition administered when awake. Twelve adults with insomnia, free of other mental and uncontrolled physical illnesses, were randomly assigned to the SMR or IND group. Eight completed this randomized, parallel group, single-blind study. Both groups received fifteen 20-min sessions of Z-Score neurofeedback. Pre- and post-assessments included sQEEG, mental health, quality of life, and insomnia status. ANOVA yielded significant post-treatment improvement for the combined group on all primary insomnia scores: Insomnia Severity Index (ISI p<.005), Pittsburgh Sleep Quality Inventory (PSQI p<.0001), PSQI Sleep Efficiency (p<.007), and Quality of Life Inventory (p<.02). Baseline EEGs showed excessive sleepiness and hyperarousal, which improved post-treatment. Both Z-Score neurofeedback groups improved in sleep and daytime functioning. Post-treatment, all participants were normal sleepers. There were no significant differences in the findings between the two groups.
A pilot study sought to evaluate the neurofeedback training outcomes in childhood obesity management. The study involved 34 overweight and obese children, age 6-18 years (12 patients in the intervention group, 22 in the control group). Complete assessment of children was done before the intervention and 3 and 6 months after the intervention; eating behavior and quality-of-life questionnaires were assessed at study start and 6 months after. All children received lifestyle recommendations for weight management, while the intervention group also had 20 neurofeedback sessions (infra-low-frequency training). The neurofeedback intervention was associated with less weight loss compared with classic weight management. The mean change in body-mass index standard deviation score at 3 months was -0.29 for the intervention group and -0.36 for the control group (p=0.337); after 6 months, the changes were -0.30 and -0.56, respectively (p=0.035). Quality of life improved similarly for both groups. Subjective outcomes reported by patients in the intervention were less snacking, improved satiety, enhanced attention capacity, ameliorated hyperactivity, and better sleep patterns. The authors noted that larger studies, with training methods involving both the left and right cortices, should further clarify the role of neurofeedback training in obesity management.
In a double-blind, parallel design RCT, 20 inpatients with OCD underwent 25 sessions of neurofeedback or sham feedback (SFB). Neurofeedback was aimed at reducing EEG activity in an independent component previously reported abnormal in this diagnosis. Resting-state EEG recorded before and after the treatment was analyzed to assess its posttreatment changes, relationships with clinical symptoms and treatment response. Overall, clinical improvement in OCD patients was not accompanied by EEG change as assessed by standardized low-resolution electromagnetic tomography and normative independent component analysis. Pre- to posttreatment comparison of the trained component and frequency did not yield significant results; however, in the neurofeedback group, the nominal values at a downtrained frequency were lower after treatment. The neurofeedback group showed significantly higher percentage reduction of compulsions compared to the SFB group (p = 0.015). Pretreatment higher amount of delta (1-6 Hz) and low alpha oscillations as well as a lower amount of high beta activity predicted a worse treatment outcome. Source localization of these delta and high beta oscillations corresponded with previous EEG resting-state findings in OCD patients compared to healthy controls. Authors concluded that Independent component neurofeedback in OCD proved useful in percentage improvement of compulsions. Based on correlation analyses, they hypothesize that they had targeted a network related to treatment resistance.
A systematic review of neurofeedback as a treatment for substance abuse disorders described difficulties in assessing the efficacy of this and other substance abuse treatments, including the lack of clearly established outcome measures, differing effects of the various drugs, presence of comorbid conditions, absence of a criterion standard treatment, and use as an add-on to other behavioral treatment Regimens The authors concluded that alpha-theta training, when combined with an inpatient rehabilitation program for alcohol dependency or stimulant abuse, would be classified as level 3 or “probably efficacious.” This level is based on beneficial effects shown in multiple observational studies, clinical studies, wait-list control studies, or within-subject or between-subject replication studies. The authors also noted that few large-scale studies of neurofeedback in addictive disorders have been reported, and a shortcoming of the evidence for alpha-theta training is that it has not been shown to be superior to sham treatment.
An RCT by Gabrielsen et al (2022) randomized adults with substance abuse disorders enrolled in outpatient abuse programs to either 20 sessions (30 minutes each) of infralow (ILF) neurofeedback plus standard of care, or standard of care alone, over a mean of 5 months.(21) At the end of the intervention period, both groups demonstrated a significant improvement in quality of life scores from baseline, but there was no difference between groups. Restlessness was reportedly significantly lower in the ILF-neurofeedback group compared to standard of care post-treatment, but this was a secondary endpoint, meaning the study was not powered to find differences only in this endpoint. Individuals were not stratified based on drugs of abuse and there was a lack of sham neurofeedback, limiting results.
A systematic review by Melo et al (2019) included 7 RCTs of biofeedback techniques, including neurofeedback, in the treatment of chronic insomnia. The authors identified conflicting results in comparisons of neurofeedback with other cognitive behavioral therapy techniques, placebo, and no treatment. A majority of outcomes demonstrated no significant differences between comparison groups.
De Ruiter et al (2016) reported on a multicenter, triple-blind RCT of neurofeedback in 80 pediatric brain tumor survivors who had cognitive impairments. The specific neurofeedback module was based on individual EEG, and participants, parents, trainers, and researchers handling the data were blinded to assignment to the active or sham neurofeedback module. At the end of training and 6-month follow-up, there were no significant differences between the neurofeedback and sham feedback groups on the primary outcome measures for cognitive performance, which included attention, processing speed, memory, executive functioning, visuomotor integration, and intelligence
Hong and Park (2022) conducted a meta-analysis of 7 RCTs of adults with PTSD treated with neurofeedback. Three studies used functional magnetic resonance imaging (fMRI) based neurofeedback and 4 studies used EEG-based neurofeedback. The overall effect of all studies pooled together demonstrated a significant improvement in PTSD symptoms with neurofeedback compared to sham neurofeedback, no treatment, of other treatment. When analyzed by type of neurofeedback, the significant improvement in PTSD symptoms remained with EEG-based neurofeedback, but not with fMRI. Five studies overall assessed anxiety and depression with various validated scales. Overall, there was no significant impact on anxiety and depression with neurofeedback compared to control group. Two studies demonstrated a high risk of performance or detection bias, while all other studies demonstrated overall low risk of bias.
In 2019, the American Academy of Pediatrics (AAP) published a guideline update to the 2011 guideline for the treatment of attention-deficit/hyperactivity disorder (ADHD) in children and adolescents. The guideline states that electroencephalogram (EEG) biofeedback is one of several nonmedication treatments that have either too little evidence to support their recommendation for use or have little or no benefit.
The Society for Development and Behavioral Pediatrics (SDBP) published a guideline in 2020 on the assessment and treatment of children and adolescents with complex ADHD. Regarding neurofeedback, the guidelines state: "Additional nonpharmacological ADHD interventions have been developed such as cognitive training (e.g., working memory training) and neurofeedback. Although these approaches have shown some improvement in laboratory-based, task-specific outcomes, none have demonstrated sufficient evidence of effectiveness in real-world domains of functioning (e.g., behavior at home and school, academic performance, peer relationships) to recommend them for use in practice with children and adolescents with ADHD."
Reference List
Blue Cross and Blue Shield of Minnesota medical policies apply generally to all Blue Cross and Blue Plus plans and products. Benefit plans vary in coverage and some plans may not provide coverage for certain services addressed in the medical policies. When determining coverage, reference the member’s specific benefit plan, including exclusions and limitations.
Medicaid products may provide different coverage for certain services, which may be addressed in different policies. For Minnesota Health Care Program (MHCP) policies, please consult the MHCP Provider Manual website.
Medicare products may provide different coverage for certain services, which may be addressed in different policies. For Medicare National Coverage Determinations (NCD), Local Coverage Determinations (LCD), and/or Local Coverage Articles, please consult CMS, National Government Services, or CGS websites.
Note that services with specific coverage criteria may be reviewed retrospectively to determine if criteria are being met. Retrospective denial of claims may result if criteria are not met.
Blue Cross and Blue Shield of Minnesota reserves the right to revise, update and/or add to its medical policies at any time without notice. Codes listed on this policy are included for informational purposes only and are subject to change without notice. Inclusion or exclusion of a code does not constitute or imply member coverage or provider reimbursement.
These guidelines are the proprietary information of Blue Cross and Blue Shield of Minnesota. Any sale, copying or dissemination of the medical policies is prohibited; however, limited copying of medical policies is permitted for individual use.
Acknowledgements:
CPT® codes copyright American Medical Association® 2023. All rights reserved.
CDT codes copyright American Dental Association® 2023. All rights reserved.