Blue Cross Blue Shield - Iowa biofeedback Coverage Update

inv-01: Investigational indications

Biofeedback is considered investigational for the following indications because evidence is insufficient to demonstrate improved net health outcome.

Investigational indications: Anxiety disorders; Asthma; Bell palsy; Depression; Hypertension; Insomnia; Movement disorders (eg, motor dysfunction after stroke, injury, or lower-limb surgery); Multiple sclerosis; Orthostatic hypotension in individuals with spinal cord injury; Pain management during labor; Posttraumatic stress disorder; Prevention of preterm birth; Raynaud disease; Sleep bruxism; Tinnitus; Home biofeedback devices and specified devices (FreeSpira, Heart Rate Variability, InTandem, StressEraser) are investigational.

List drawn verbatim from policy.

inv-02: Anxiety Disorders — Evidence Summary

Evidence for biofeedback in anxiety disorders summarized from systematic reviews and RCTs; conclusions depend on study quality and sample size.

Anxiety disorders evidence summary: Systematic review of HRVB (24 studies, N=484) reported large within- and between-group effects on anxiety but included many small studies with a median sample size of 14 and frequent high or unclear risk of bias; subsequent RCTs (eg, Chen et al) showed physiologic and symptom improvements in small samples, while at least one RCT (Ma et al) did not find significant effects possibly due to limited power.

Goessl et al meta-analysis and later RCTs; trial quality and small sample sizes limit confidence.

inv-03: Asthma — Evidence Summary

Evidence for biofeedback in asthma includes a systematic review and several RCTs with heterogeneous techniques and mixed clinical outcomes.

Asthma evidence summary: A systematic review identified 3 RCTs using different modalities (exhaled CO2 capnography, HRV physiograph, respiratory sinus arrhythmia/EMG); physiologic parameters changed in trials but pooled data showed no change in FEV1 and medication use decreased in only one trial. Individual RCTs (eg, Taghizadeh et al, Lehrer et al) reported physiologic improvements and some protective effects under stress, but trials were small and heterogeneous; HRVB should not be considered an alternative to asthma controller medications.

Yorke et al systematic review; Taghizadeh et al; Lehrer et al; results do not support substituting HRVB for controller therapy.

inv-04: Evidence-based indications

Summary evidence-based coverage considerations by indication

Asthma - evidence considerations: Randomized trials reported physiologic improvements for targeted parameters but did not consistently improve clinical outcomes (eg, medication use, FEV1); heterogeneity of techniques and small samples limit conclusions.

Systematic review of 3 RCTs plus later RCTs; HRVB not a substitute for controller medications.

Bell palsy - evidence considerations: Systematic review of 4 RCTs evaluated mirror and/or EMG biofeedback added to facial exercises; small sample sizes and heterogeneity limit the ability to draw conclusions about clinical benefit.

Cardoso et al review; heterogeneity and small trials noted.

Depression - evidence considerations: Two small RCTs published after systematic review reported short-term symptom improvement when respiratory/HRV biofeedback was added to usual care, but open-label designs, small samples, and short follow-up limit confidence in durable benefit.

Maynart et al and Park & Jung RCTs; preliminary findings only.

inv-05: Evidence summaries for considered indications

Evidence summaries and limitations relevant to coverage decisions

Depression evidence summary: Two small, primarily open-label RCTs reported improved depression scores at short-term follow-up (4–6 weeks) with respiratory-related biofeedback plus usual care versus usual care alone, but limitations include small sample sizes, lack of sham control, and short follow-up; long-term benefit is unknown.

Maynart et al (2021) and Park & Jung (2020)

Hypertension evidence summary: Systematic reviews and RCTs used heterogeneous biofeedback modalities and generally poor-quality trials; a recent 2024 meta-analysis of 20 RCTs reported modest mean reductions in systolic and diastolic blood pressure (~-4.52 / -5.19 mm Hg), but overall long-term efficacy and clinical significance remain uncertain.

Jenkins et al (2024) and earlier reviews; inconsistency across studies noted.

Motor dysfunction after stroke - PICO and evidence gap: PICO framing provided for motor dysfunction after stroke; pooled short-term improvements in motor measures reported elsewhere, but this segment does not summarize trial-level results.

inv-06: Motor dysfunction after stroke — Evidence summary

Evidence-based findings relevant to coverage decisions for biofeedback in motor dysfunction after stroke:

Short-term motor improvement: Systematic reviews and pooled analyses (eg, Stanton et al) report a significant short-term improvement in motor function measures with biofeedback (pooled SMD 0.50; 95% CI, 0.3–0.7; pooled n=417 across 17 trials).short-term

Effects reported on laboratory/functional motor measures rather than patient-centered long-term outcomes.

Long-term clinical outcomes: Long-term outcomes were infrequently reported and did not consistently demonstrate benefit for clinical outcomes (eg, ADLs, fall rates, return to work); larger RCTs published since systematic reviews found no difference versus conventional rehabilitation in gait speed, balance, ADLs, fall rate, or return to work.long-term

Insufficient evidence to conclude durable clinical benefit.

Heterogeneity and limitations:

inv-07: Motor dysfunction after lower‑limb injury/surgery — Evidence summary

Findings relevant to coverage for biofeedback used after lower‑limb injury or surgery:

Population and intervention: Studies evaluated EMG biofeedback adjunctive to standard rehabilitation in patients after lower-limb injury or surgery (eg, knee rehabilitation); sample sizes were small and interventions heterogeneous.population_intervention

Silkman & McKeon; Xie et al meta-analysis included postsurgical knee rehab trials.

Outcomes assessed: Some meta-analyses reported improved range of motion (eg, SMD -0.48 favoring biofeedback in pooled ROM data), but individual trials often showed no difference for pain or physical function; long-term patient-centered outcome data are limited.outcomes

Meta-analysis may be driven by small trials; generalizability limited.

inv-08: Indication-specific evidence summaries

Findings summarized by indication

Motor dysfunction after lower-limb injury or surgery: Systematic reviews and RCTs yielded mixed results: some trials showed intermediate outcome benefits (eg, quadriceps contraction, ROM) but no consistent improvements in pain or physical function; larger/longer trials are needed to establish clinical benefit.small sample sizes, heterogeneous interventions

Xie et al and Silkman & McKeon reviews summarized evidence.

Multiple sclerosis: Evidence comprises small RCTs and crossover studies; some trials reported reductions in trunk sway or improvements in select measures, but overall sample sizes were small and between-group differences were generally not statistically significant.follow-up often short; sample sizes small

Miri et al, MacKay et al, van der Logt et al trials.

Orthostatic hypotension due to spinal cord injury:

inv-09: Evidence summaries by indication

Summaries of evidence by indication (presented as findings rather than explicit policy coverage rules):

Orthostatic hypotension after spinal cord injury: A systematic review identified 13 studies of nonpharmacologic interventions including 2 biofeedback studies (3 patients) reporting an average 39% increase in systolic blood pressure, but reviewers concluded clinical usefulness has not been proven.

Gillis et al review; evidence extremely limited.

Pain management during labor: A Cochrane review of 4 RCTs (N=186) judged trials highly variable and at high risk of bias; overall little difference between biofeedback and control interventions and evidence insufficient to support effectiveness.

Barragan Loayza et al Cochrane review.

Posttraumatic stress disorder (PTSD): Systematic reviews and updates identified small RCTs and uncontrolled studies; evidence is graded C (unclear/conflicting). A 2024 meta-analysis of uncontrolled HRVB studies in veterans (5 studies, N=95) reported moderate effect sizes but was limited by uncontrolled designs.

Wahbeh et al, Polak et al, Kenemore et al analyses.

inv-10: Coverage logic by indication

Coverage considerations based on available evidence for each indication:

Posttraumatic Stress Disorder (PTSD): Evidence includes small and inconsistent randomized and nonrandomized trials; uncontrolled studies report symptom improvements but randomized data are mixed; overall evidence grade is C (conflicting/unclear).

Systematic reviews and 2024 veteran-focused meta-analysis indicate mixed results and methodological limitations.

Prevention of Preterm Birth: One RCT (Siepmann et al, N=48) of HRVB reported reduced perceived chronic stress but no statistically significant differences in preterm birth rates, gestational duration, or birth weight.

Clinical primary outcomes not improved.

Raynaud Disease: Systematic review and RCTs (including a 313-patient Raynaud's Treatment Study) found no consistent benefit of thermal biofeedback; pooled analyses favored sham control for reduction in attack frequency.

Largest trial showed nifedipine beneficial but thermal biofeedback did not differ from EMG control; pooled analysis favored sham.

inv-11: Evidence-based coverage considerations by indication

Coverage considerations based on available evidence summaries and RCTs

Raynaud disease: Systematic review identified 5 RCTs; pooled analysis (4 trials) showed that reduction in frequency of attacks favored sham control rather than active biofeedback.

Malenfant et al systematic review; pooled results favored sham.

Sleep bruxism: Systematic reviews and an RCT published since those reviews provide mixed evidence: pooled analyses of bruxism episodes/hour did not show significant benefit; one RCT reported short-term reductions in pain or event duration but not sustained reductions in episodes/hour.

Wang et al; Jokubauskas et al; Bergmann et al.

Tinnitus: A single RCT (Weise et al, N=130) of a biofeedback-based behavioral intervention (12 sessions over ~3 months) reported reductions in tinnitus annoyance, loudness diary ratings, improved controllability, and improvements in depressive symptoms; additional confirmatory trials are needed.single RCT evidence

inv-12: Indication-specific evidence summaries

Evidence-based summaries for specific indications

Tinnitus - evidence summary: One randomized trial (Weise et al, N=130) of a biofeedback-based cognitive-behavioral intervention (12 sessions over ~3 months) reported improvements in tinnitus annoyance, loudness ratings, controllability, coping cognitions, and depressive symptoms; evidence is limited to this single RCT and requires confirmation.single RCT evidence

Documented trial-level improvements in patient-reported tinnitus outcomes.

Home biofeedback: No relevant clinical effectiveness studies of biofeedback delivered solely via home equipment were identified; therefore home biofeedback is not supported by the evidence base.no studies identified

Claims relying on home equipment should be scrutinized.

Biofeedback delivered via home equipment (including but not limited to FreeSpira, Heart Rate Variability devices, InTandem, and StressEraser) is designated investigational. The policy states that evidence is insufficient to determine that home-delivered biofeedback improves net health outcomes, and no relevant clinical effectiveness studies of biofeedback using home equipment were identified in the review.

Heart rate variability biofeedback (HRVB) is not an alternative to asthma controller medications. The policy cites randomized trial data indicating HRVB did not provide consistent improvements in clinical outcomes and therefore should not replace established controller therapy.

Trials of biofeedback in asthma produced physiologic changes for targeted measures (eg, exhaled CO2, HRV) but these did not reliably translate into improved clinical outcomes such as FEV1 or reduced controller medication use. Consequently, use of biofeedback as a substitute for established asthma controller therapy is not supported by the trial data and should not be considered a replacement for standard pharmacologic management.

Many trials reviewed lacked sham biofeedback comparators and had limited duration of follow-up. Where evidence consisted of short-term, open-label, or noncontrolled studies, claims of sustained benefit are unreliable and should not be the basis for coverage decisions without higher-quality, longer-term comparative data.

Heterogeneity across biofeedback interventions (modalities, dose, feedback type) and outcome measures, together with methodological limitations in many studies, restricts the ability to generalize positive findings from one technology or protocol to others. These evidence limitations reduce confidence in broad coverage of disparate biofeedback technologies.

For the miscellaneous indications reviewed, there is no consistent, high-quality evidence demonstrating clear clinical usefulness. Absent indication-specific, robust comparative data showing improved patient-centered outcomes, biofeedback for these conditions should not be assumed to meet coverage criteria.

Although the policy does not enumerate procedural exclusions by name for every condition, it repeatedly emphasizes that the clinical usefulness of biofeedback for the listed indications is unproven. This language should be interpreted conservatively when making coverage determinations: in the absence of supportive, high-quality evidence, requests may be denied as investigational.

Pooled randomized trial data for Raynaud disease did not demonstrate benefit from thermal biofeedback and, in pooled analyses, outcomes favored sham control. While no explicit procedural exclusion is listed, these findings imply insufficient evidence to support coverage for Raynaud disease.

The evidence base for biofeedback in sleep bruxism is small and at moderate-to-high risk of bias. Pooled analyses of episodes/hour did not show significant benefit, and trials are heterogeneous and generally underpowered; these limitations warrant caution when considering coverage for sleep bruxism.

This policy found no clinical effectiveness studies that evaluated biofeedback delivered solely via home equipment. Requests relying on home-device delivery should be considered unsupported by the evidence presented and treated as investigational.

The extract provided does not contain explicit, itemized clinical coverage criteria beyond the investigational designations; instead the document summarizes evidence by indication and notes policy history with revisions from 2001 through 2025. Coverage decisions should therefore follow the policy stance that listed indications are investigational unless higher-quality evidence specific to an indication is provided.

The policy explicitly designates the listed miscellaneous conditions (including anxiety disorders, asthma, Bell palsy, depression, hypertension, insomnia, movement disorders, multiple sclerosis, orthostatic hypotension, labor pain, PTSD, prevention of preterm birth, Raynaud disease, sleep bruxism, and tinnitus) as investigational because current evidence is insufficient to demonstrate improved net health outcomes.

Although some trials in asthma demonstrated changes in physiologic measures targeted by biofeedback, these physiologic improvements did not translate into consistent, meaningful clinical benefits such as improved FEV1 or reduced need for controller medications, undermining claims of clinical effectiveness for asthma.

Evidence for biofeedback in Bell palsy is limited to a small number of randomized trials with modest total sample sizes (N≈132) and heterogeneity in techniques and treatment duration. Reviewers concluded the current evidence does not support routine use for this indication.

Systematic reviews of biofeedback for hypertension report inconsistent results across many small, heterogeneous trials. While one meta-analysis found modest average reductions in blood pressure, overall trial quality and inconsistency limit confidence, and biofeedback is not supported for hypertension without stronger, long-term data.

Although pooled short-term motor function outcomes after stroke sometimes favored biofeedback, larger and more recent randomized trials did not demonstrate consistent long-term improvements in activities of daily living, fall rates, or return-to-work. Long-term functional benefits are therefore not consistently demonstrated.

Across the miscellaneous indications reviewed, long-term functional outcomes and quality-of-life improvements were infrequently reported and when measured were not consistently improved by biofeedback interventions. This lack of durable, patient-centered benefit is a recurrent limitation in the evidence base.

Evidence for orthostatic hypotension (reviewed studies included only a few patients for biofeedback) and for pain management during labor (4 small RCTs at high risk of bias) is insufficient to support clinical effectiveness; trial limitations preclude reliable coverage conclusions for these indications.

Prevention of preterm birth, PTSD, and Raynaud disease have not demonstrated consistent benefit on primary clinical outcomes in randomized trials and systematic reviews: a small RCT in preterm labor did not significantly change preterm birth rates, PTSD trials are small and mixed, and pooled Raynaud analyses favored sham. These findings do not support coverage for these prevention or treatment claims.

Pooled evidence for sleep bruxism did not demonstrate a significant reduction in EMG-measured episodes per hour; studies are small, heterogeneous, and at moderate-to-high risk of bias. While one newer RCT reported reductions in pain with a biofeedback splint at 3 months, no sustained reduction in bruxism episodes was shown.

The policy reiterates the evidence gap for home equipment: no relevant clinical effectiveness studies were identified for biofeedback delivered solely via home devices. This absence of evidence is cited to justify considering home-device claims investigational.

Prior authorization, documentation, and billing guidance

Prior Authorization

Prior approval — benefit verification and authorization expectations

Providers must verify member benefits and contract exclusions at the time of service. There is no universal prior authorization requirement for biofeedback across all contracts; however, prior authorization is likely for investigational uses and when clinical benefit is unproven. Confirm benefit eligibility, any plan-specific prior authorization rules, and applicable coding before initiating therapy.

Verify member benefits and contract exclusions at time of service.
Prior authorization may be required for investigational indications or per contract.
Use listed CPT/HCPCS/Revenue/ICD coding to support benefit verification.

Documentation Required

Evidence standard for efficacy

To establish efficacy for biofeedback, the strongest evidence comes from randomized controlled trials (RCTs) and comparative prospective trials. When RCTs are unavailable, prospective comparative observational studies are preferred. Single‑arm studies may inform long‑term safety or rare outcomes but are low on the hierarchy for efficacy assessment.

Preferred study designs: RCTs and comparative prospective trials.
Acceptable when higher‑quality trials lacking: prospective comparative observational studies.
Single‑arm studies useful for long‑term safety or adverse event detection only.

Denial Risk

Risk of bias and evidence limitations that may affect coverage

Risk of bias and study limitations in the biofeedback literature frequently affect coverage determinations. Common issues include small sample sizes, inadequate or unclear randomization and allocation concealment, lack of blinding (including lack of sham controls), high attrition, short follow‑up, and inconsistent or surrogate outcomes. These limitations may reduce confidence in efficacy and can lead to denial of coverage for investigational uses.

High/unclear risk of bias (inadequate randomization, allocation concealment) reduces evidentiary weight.
Lack of blinding or sham‑control weakens causal inference; sham controls are important where feasible.
Small sample sizes, heterogeneity in interventions/outcomes, high attrition, and short follow‑up may prompt denial.

Documentation Required

Evidence expectations and comparator design concerns

When requesting authorization for biofeedback, submit trial‑level evidence and documentation addressing study design limitations. Lack of sham control or inadequate comparator design in submitted studies should be noted and justified; studies without appropriate comparators may be insufficient to demonstrate clinical benefit.

Provide RCTs or prospective comparative trials when available; explain relevance if only nonrandomized data exist.
Highlight whether studies used sham biofeedback, active comparators, or usual care.
If comparators are inadequate, provide rationale and any additional supporting data.

Denial Risk

Trials for biofeedback in labor were judged high risk of bias

Many indications (for example, pain management during labor) have trials judged to be at high risk of bias or otherwise insufficient. Labor pain RCTs were small, heterogeneous in methods and outcomes, and were assessed as having high risk of bias; these limitations increase the likelihood that authorization for labor‑pain biofeedback will be denied.

Cochrane review (labor pain) identified 4 RCTs (N=186) with high risk of bias.
Heterogeneity in modalities and outcomes prevented pooled analysis.
Due to high risk of bias and small sample sizes, requests for labor‑pain biofeedback should be scrutinized.

Documentation Required

Required documentation and supporting evidence for authorization

Prior authorization requests should include comprehensive documentation for evidence review: population, intervention (modality and session plan), comparator, outcomes measured (symptoms, function, QOL), duration of follow‑up, and study limitations. Provide trial‑level citations or concise summaries of higher‑quality trials (RCTs or systematic reviews) that specifically support the requested indication.

Required evidence elements: population, intervention details (modality, number/duration of sessions), comparator, measured outcomes, and duration of follow‑up.
Submit citations or concise summaries of RCTs and systematic reviews that support the requested use.
If available, include study limitations (risk of bias, sample size, follow‑up) and applicability to the member.

Documentation Required

Key definitions and modalities

Biofeedback device (FDA definition)

Definition (FDA)An instrument that provides a visual or auditory signal corresponding to the status of one or more of a patient’s physiologic parameters (e.g., brain alpha wave activity, muscle activity, skin temperature) so the patient can voluntarily control those parameters.

Regulatory statusDevices cleared by FDA via 510(k), designated class II with special controls.

Primary feedback modalitiesProvides visual or auditory signals tied to physiologic measures for patient self-regulation.

Example physiologic parametersBrain alpha wave activity, muscle activity (EMG), skin temperature, heart rate variability, respiratory measures.

Heart rate variability biofeedback (HRVB)

Core definitionA biofeedback technique that trains patients to modify cardiac vagal tone by providing feedback on heart rate variability (HRV), often via paced breathing to influence autonomic regulation.

Clinical associationLow HRV is associated with anxiety and stress-related states; HRVB aims to increase HRV and improve symptoms.

Evidence contextEvaluated in systematic reviews and meta-analyses (e.g., Goessl et al 2017: 24 studies, N=484) and studied in PTSD trials (small/some uncontrolled studies).

Typical interventionPaced breathing training to augment respiratory sinus arrhythmia and cardiac vagal modulation.

Diaphragmatic breathing relaxation

DefinitionA relaxation technique using diaphragm muscle contractions to draw air downward, increasing diaphragm length and breathing efficiency.

Use caseEmployed as a relaxation/biofeedback intervention in anxiety trials; outcomes measured include anxiety scales and physiologic measures (skin conductivity, peripheral blood flow, heart rate, breathing rate).

Typical programIndividualized training courses (example: an 8-week course with sessions), with physiologic and symptom assessments pre/post intervention.

Exhaled CO2 capnography biofeedback

DefinitionA biofeedback technique that measures exhaled carbon dioxide (capnography) to guide respiratory patterns.

Clinical trial useUsed in asthma RCTs (one pooled trial with exhaled CO2 capnography; physiologic differences observed but not consistently linked to clinical benefits such as FEV1).

Outcome measures in studiesCommon outcomes included peak expiratory flow, respiratory impedance, and medication use; physiologic changes did not consistently translate into improved FEV1.

HRVB — heart rate variability biofeedback

DefinitionHRV biofeedback provides real-time feedback on heart rate and respiratory sinus arrhythmia to train paced breathing and enhance autonomic regulation.

Training targetsPaced slow breathing to maximize respiratory sinus arrhythmia and increase cardiac vagal tone (improving HRV).

Clinical contexts studiedStudied for anxiety, PTSD, and asthma as an adjunctive therapy; trials vary in size and quality.

EMG biofeedback

DefinitionElectromyography (EMG) biofeedback provides patients with real-time feedback on muscle electrical activity to facilitate neuromuscular retraining.

ApplicationsUsed in neuromuscular retraining contexts such as Bell palsy (with mirror therapy) and postsurgical knee rehabilitation to target quadriceps activation.

Typical outcome measuresIntermediate measures include muscle contraction values and EMG amplitudes; many trials assessed range of motion or strength rather than long-term function.

Biofeedback modalities

Modalities listedThermal (skin temperature) biofeedback, galvanic skin response (GSR), pulse wave velocity monitoring, heart rate variability (HRV) training, respiratory sinus arrhythmia, device-guided slow breathing, EMG, vibrotactile feedback, visual and auditory feedback from force platforms and gait sensors.

Feedback channelsVisual, auditory, vibrotactile (tactile), or combined modalities used depending on the physiologic target.

Typical trial varietyTrials used single or multiple modalities and varied widely in session number (e.g., 4–48 sessions reported in hypertension reviews).

Depression outcome measures

Common scalesBeck Depression Inventory (BDI) and Hamilton Depression Rating Scale (HAM-D).

Use in trialsDepression RCTs reported baseline and follow-up BDI or HAM-D scores to measure symptom change (examples: Maynart et al 2021, Park & Jung 2020).

Typical follow-up timingShort-term follow-up at 4–6 weeks was common in small depression RCTs assessing biofeedback plus usual care.

Hypertension definition used in reviews

Hypertension threshold used in reviewsEssential hypertension defined in some trials as at least 140/90 mm Hg.

Meta-analysis findingsA 2024 meta-analysis reported mean reductions of approximately -4.52 mm Hg systolic and -5.19 mm Hg diastolic with biofeedback, though studies were heterogeneous.

Trial characteristicsTrials varied in sessions (4–48) and follow-up (2 weeks to 12 months); many trials were small and single-center.

Biofeedback techniques (motor rehabilitation)

Measurement modalities for motor rehabWeight-distribution force platforms or sensors, EMG for muscle activity, linear gait parameters, joint-angle measurement with goniometers, and trunk-sway sensors (e.g., SwayStar); feedback delivered visually, auditorily, or vibrotactilely.

Outcomes assessedOutcomes often include short-term laboratory-based measures (e.g., postural sway, gait speed, range of motion) rather than long-term ADLs or fall rates.

Evidence limitationsThough pooled short-term effects showed benefit (SMD ~0.50), long-term functional outcome data are limited and heterogeneous across trials.

biofeedback — general definition

Short definitionTherapeutic approach that uses physiologic monitoring and real-time feedback to teach voluntary control of bodily processes.

Typical program lengthPrograms commonly described as 10–20 sessions of about 30 minutes each (used alone or with other therapies).

Examples of monitored signalsHeart rate variability, skin temperature, muscle EMG, respiratory measures, and other physiologic parameters.

EMG and vibrotactile biofeedback

EMG biofeedbackReal-time electromyographic feedback to improve targeted muscle activation (used in knee rehab, facial retraining).

Vibrotactile biofeedbackProvides direction-specific tactile cues (e.g., for trunk sway reduction) to aid balance and posture control.

ApplicationsUsed to improve muscle recruitment, reduce trunk sway, and provide sensory cues during rehabilitation tasks.

HRVB — HRV biofeedback evaluated in PTSD

HRVB in PTSDHRV biofeedback trains paced breathing to influence autonomic regulation; evaluated in PTSD studies and meta-analyses, but randomized data are small and mixed.

Evidence gradeReviews gave biofeedback evidence for PTSD a grade C (unclear/conflicting); some uncontrolled studies report moderate effects but RCTs are inconsistent.

Population notedSeveral studies involve military veterans; trials are often small or uncontrolled limiting conclusions.

Biofeedback — therapy definition

Therapeutic premiseUses physiologic monitoring (e.g., HRV, thermal, EMG) to help patients gain voluntary control over bodily processes and reduce symptom burden.

Delivery formatsMay be delivered as standalone biofeedback sessions or as part of a biofeedback-based behavioral/cognitive program.

Typical targetsTargets vary by indication and include autonomic regulation, muscle activation, respiratory patterns, and temperature regulation.

Populations of interest

Populations reviewedIndividuals susceptible to preterm birth, patients with PTSD, Raynaud disease, and adults with sleep bruxism were among populations of interest.

Other populationsAlso includes motor dysfunction after stroke, lower-limb injury/surgery, multiple sclerosis, orthostatic hypotension in spinal cord injury, tinnitus, and depression.

Evidence characteristicsStudies across populations were often small, heterogeneous, and varied in modality and outcome measures.

Sleep bruxism measurement

Primary outcome (sleep bruxism)Number of sleep bruxism episodes per hour detected by EMG recording.

Measurement methodsEMG recording during sleep is the standard method used in trials; some studies used bite-force or wireless devices for related measures.

Evidence summarySystematic reviews found small, heterogeneous trials and pooled analyses did not show significant reductions in episodes/hour.

Tinnitus outcome measures

Tinnitus outcome measuresGlobal tinnitus annoyance (Tinnitus Questionnaire total score) and daily diary ratings (visual analog scales) are primary patient-reported outcomes.

Scoring range exampleThe Tinnitus Questionnaire total score ranges from 0 to 84 (Weise et al preassessment mean 54.7, postassessment mean 32.5).

Study formatOutcomes assessed after a course of sessions (example: 12 sessions over ~3 months) in RCTs of biofeedback-based behavioral interventions.

Biofeedback-based behavioral intervention

Program definitionA therapeutic program combining physiologic monitoring (biofeedback) with behavioral or cognitive strategies delivered over multiple sessions (example: 12 sessions over 3 months for tinnitus).

Documentation needsClinical documentation should include modality, number/duration of sessions, baseline and post-treatment outcome measures to demonstrate benefit.

Use casesDelivered for tinnitus, depression, and other indications as a structured multi-session intervention rather than single-device home use.

90875	Individual psychophysiological therapy incorporating biofeedback training by any modality (face-to-face with the patient), with psychotherapy; 30 minutes.
90876	Individual psychophysiological therapy incorporating biofeedback training by any modality (face-to-face with the patient), with psychotherapy; 45 minutes.
90901	Biofeedback training by any modality.
90912	Biofeedback training, perineal muscles, anorectal or urethral sphincter, including EMG and/or manometry, when performed; initial 15 minutes of one-on-one physician or other qualified health care professional contact with the patient.
90913	Biofeedback training, perineal muscles, anorectal or urethral sphincter, including EMG and/or manometry, when performed; each additional 15 minutes of one-on-one physician or other qualified health care professional contact with the patient (List separately in addition to code for primary procedure).

E0746	Electromyography (EMG), biofeedback device.
E1399	Durable medical equipment, miscellaneous (may be used for home biofeedback devices which may include the following: FreeSpira, Heart Rate Variability, or StressEraser).
E3200	Gait modulation system, rhythmic auditory stimulation, including restricted therapy software, all components and accessories, prescription only (InTandem).

OpenPayer

Biofeedback for Miscellaneous Indications

Coverage criteria and evidence summaries

Prior authorization, documentation, and billing guidance

Billing codes and coding guidance

Background and scope

Key definitions and modalities

Policy history and references