Is non-clinical, personal use of binaural beats audio an effective stress-management strategy? A systematic review of randomised control trials

ABSTRACT

Objective

To explore the efficacy and safety of binaural beats (BB) for stress management in non-clinical contexts.

Method

Embase, Medline, PsycINFO, CINAHL, Web of Science and Cochrane databases were searched for RCTs that administered BB and reported ≥1 stress-related outcome in subjects with pre-existing stress, or in participants with laboratory-induced stress (PROSPERO registration: CRD42023428623). The risk of bias was assessed using the JBI critical appraisal tool and a purpose-designed tool to explore stress-specific methodological quality. Effects between groups were synthesised thematically, exploring physiological and psychological outcomes, duration of exposure and frequency (Hz).

Results

Twelve studies were included for review. Half of the studies used a single-session BB intervention and half used longer-term home-based protocols. Seven involved participants with pre-existing stress and five exposed healthy subjects to stressors. Eight studies reported ≥1 significant effect favouring intervention over control while four found no group effects. However, methodological heterogeneity, unclear randomisation and blinding procedures, and lack of oversight of stress mediators and moderators were found.

Discussion

BBs have promise as a self-care stress-management tool. However, evidence is not strongly in favour of widespread recommendation, there is little known about adverse effects, and there is not yet a clear optimal protocol (frequency and schedule) for stress-buffering effects.

KEYWORDS:

• Binaural beats
• stress
• anxiety
• brainwave entrainment
• non-pharmacological intervention

Introduction

Stress is a subjective term used to describe both situations and feelings, with definitions varying by research discipline (Epel et al., 2018). In psychology, it is traditionally defined as when the perceived demands of an environment exceed an individual’s ability to meet or control them (Lazarus et al., 1985). These demands or stressors, real or imagined, can induce adaptive physiological change in the body (Selye, 2013). Briefly, in the short-term, there is mobilisation of sympathetic, parasympathetic, and neuromuscular nervous systems. This is followed by activation of the neuroendocrine axis, commonly known as the ‘fight-or-flight’ response (Everly & Lating, 2019), with associated feelings of tension or worry often referred to as anxiety (APA, 2023). Stress is often used interchangeably with the term ‘anxiety’, which also has multiple meanings. ‘State anxiety’ best describes feelings triggered by a stressor, while ‘trait anxiety’ may represent one’s tendency to experience these feelings even in the absence of a stressor (Gidron, 2013). Furthermore, anxiety is also used as an umbrella term for a number of diagnoses such as generalised anxiety disorder and panic disorder (Mind, 2021).

Excessive stress and anxiety – in intensity and chronicity – can have long-acting impacts on the body with resulting physiological changes leaving some individuals at higher risk of physical and mental illness (Everly & Lating, 2019; Miller et al., 2011). Globally, some of the most prevalent morbidities such as cardiovascular disease, type 2 diabetes mellitus, and depression are associated with unmanaged stress and anxiety (Cohen et al., 2007; Ingrosso et al., 2023; Stojanovich & Marisavljevich, 2008). This is a growing problem with levels of mental health disorders – such as anxiety – rising 13% in the decade up to 2017 (WHO, 2023), with reports of significant further increases following the covid-19 pandemic (Salari et al., 2020). However, managing stress and maintaining mental wellbeing can buffer against these effects (Pressman & Cohen, 2005).

Currently, stress-management approaches involve pharmacological, psychological and lifestyle interventions. Approaches may vary depending on whether there is any clinical diagnosis, cause, and severity of stress. Drug treatments (i.e. benzodiazepines) have utility but come with the risk of adverse effects, rendering them less acceptable for long-term use (Bandelow, 2020). Psychological therapies (i.e. cognitive behavioural therapy) may be appropriate in some contexts, but the effectiveness depends on patient receptibility, while therapist access and costs can be barriers (Olthuis et al., 2016; Ruotsalainen et al., 2015). General lifestyle and self-care advice aim to equip the wider population with proactive self-stress-management tools such as exercise, social support and mindfulness (Cohen, 2004; Gordon et al., 2017; Gu et al., 2015). However, reduced capacity for self-care in the highly stressed, alongside skills and lifestyles can impede one’s ability to apply such strategies effectively (Luis et al., 2021). Expanding the stress self-care toolkit to include accessible, low-harm methods may allow more users to find an approach that suits their circumstances and needs.

Binaural beats (BB) are two tones presented at slightly different frequencies to both ears from which the brain perceives a third tone equal to the frequency difference of the two tones, inducing a phenomenon known as brainwave entrainment (Ross et al., 2014). Brainwave entrainment has been observed using electroencephalogram (EEG) (Schwarz & Taylor, 2005) with five frequency bands associated with different arousal states: delta (0.5–4 Hz), theta (4–7 Hz), alpha (7–13 Hz), beta (13–30 Hz) and gamma (30–50 Hz), with lower frequencies linked to lower arousal and greater relaxation, and the opposite for higher frequencies (Jirakittayakorn & Wongsawat, 2018). Since Oster (1973) first described BB’s potential clinical value, reviews of studies using BB as a therapeutic intervention for health and wellbeing have found evidence supporting its efficacy as an intervention for pain, cognition and anxiety (Basu & Banerjee, 2020; Chaieb et al., 2015; Garcia-Argibay et al., 2019; Maddison et al., 2022). Furthermore, several clinical trials have reported reductions in peri-operative anxiety associated with BB exposure (Isik et al., 2017; Loong et al., 2022; Parodi et al., 2021). However, most stressors exist outside of clinical settings such as at work or in the home (Harvey et al., 2017; Marsh et al., 2020), and the efficacy for the use of BB in these contexts is unknown despite there already being a population already self-prescribing them for relaxation (Barratt et al., 2022).

BB have potential as an accessible, low-harm, non-pharmacological approach to stress management in the general population. However, there are currently no systematic reviews that collate the findings on this subject. The following systematic review aims to address this gap by describing and critiquing evidence from randomised control trials (RCTs) with a focus on BB effects on both short-term stress reactivity and chronic stress in non-clinical contexts. In response to concerns surrounding the quality, appropriateness and consistency of stress research methodology (Crosswell & Lockwood, 2020; Epel et al., 2018), the review will also explore the use of stress-related methodologies with a discussion of implications.

Materials and methods

The protocol for this review was registered with PROSPERO (CRD42023428623). Reporting was conducted in adherence to the Preferred Reporting Items for Systematic Reviews (PRISMA) statement (Appendix 1) (Page et al., 2021) using EndNote (The EndNote Team, 2013) and Covidence (Covidence, 2023) platforms.

Search strategy

A search of Medline, Embase, PsycINFO, Cochrane, Web of Science and CINAHL for all peer-reviewed articles available in the English language and published from the inception of the database up until the search date was conducted on 6 June 2023. Additionally, reference lists of retrieved articles were manually searched for further studies.

The search strategy consisted of two elements: the intervention (BB) AND outcome (stress), utilising a range of specific and related terms to capture the anticipated disparate methodologies and language (Table 1).

Table 1. Example search strategy (Embase). Full strategy can be found in Appendix 2.

Element	Search terms
BB	‘binaural beat*’ OR ‘binaural sound*’ OR ‘binaural tone*’ OR binaural noise’ OR ‘binaural entrainment’ OR ‘binaural audi*’ OR ‘binaural stimulation’ OR ‘auditory entrainment’ OR ‘oscillatory entrainment’ OR ‘sensory entrainment’ OR ‘delta entrainment’ OR ‘theta entrainment’ OR ‘alpha entrainment’ or ‘beta entrainment’ OR gamma entrainment’ OR ‘brain* entrainment’
Stress	‘stress’ OR ‘anxiety’ OR ‘arousal’ OR ‘distress’ OR ‘mental suffer’ OR ‘nervousness’ OR ‘vigilance’ OR ‘autonomic nervous system’ OR ‘parasympathetic’ OR ‘sympathetic’ OR ‘cortisol’ OR ‘catecholamine’ OR ‘epinephrine’ OR ‘norepinephrine’ OR ‘heart rate’ OR ‘respiratory rate’ OR ‘hypertension’ OR ‘blood pressure’ OR ‘galvanic skin response’ OR ‘skin conduct*’ OR ‘body temperature’ OR ‘relaxation’ OR ‘eye*tracking’ OR ‘electroencephalogra*’ OR ‘electromyogra*

Retrieved articles were imported into EndNote and duplicates were removed digitally. The remaining articles were then uploaded to Covidence online software for title and abstract and full text screening, data extraction and quality assessment.

Inclusion and exclusion criteria

Two reviewers independently screened all titles and abstracts, and full texts, with any conflicts resolved through discussion. Inclusion and exclusion criteria were altered from those outlined in the study protocol (Platt & Hammond, 2023). During the screening phase, it was apparent that the included studies were too heterogenous to meaningfully review. Therefore, a decision was taken to further limit the review to RCTs and to remove EEG measurement as an outcome to focus on physically and psychologically felt stress and anxiety (see Table 2 for the eligibility criteria).

Table 2. Article inclusion and exclusion criteria. Italics denote criteria that were changed from the protocol.

Inclusion criteria	Exclusion criteria
Population:
Adolescents (10–19) and adults. Healthy (without stress) or with any physical or mental health condition not specified in the exclusion criteria Participants with a pre-existing stressor or indicator of stress (i.e. anxiety, chronic pain, insomnia) when there is no exposure to a stressor during the study	People with neurocognitive deficit (i.e. dementia, brain injury etc.) hearing impairment Children <10 years old Participants with no pre-existing stressor or indicator of stress (i.e. anxiety, chronic pain, insomnia) when there is no exposure to a stressor during the study
Intervention:
Studies where the BB intervention is self-administered by the participant Laboratory-based BB interventions with or without exposure to a non-clinical stressor (e.g. Trier social stress test) Participants awake during intervention (including pre-sleep relaxation) Pure-tone BBs or masked with another sound such as music or pink noise	BB therapy (e.g. audio-visual) where there is no standalone BB group Interventions applied in a medical context (e.g. peri-operative BB exposure) BB interventions applied during sleep
Control:
Non-BB control group or condition
Outcomes:
Studies with ≥1 stress-related primary or secondary outcome Physiological: Heart rate variability, blood pressure, pulse rate, respiratory rate, sweating, stress hormones Psychological: Questionnaires such as ‘State-Trait Anxiety Inventory’ (STAI) and ‘Perceived Stress Scale’ (PSS)	Studies that did not report any stress-related outcome measures EEG
Study:
Randomised control trials (RCTs) Randomised crossover trials (RXTs)	Quasi-experimental studies Observational studies Case series/reports Grey literature

Data extraction

One author completed data extraction with a second author independently checking the results. The following information and data were extracted from included studies into a pre-set template: author and year, journal name, study location, trial design (randomised control or randomised crossover trial), participant characteristics (sample size, age, sex, ethnicity, comorbidities and group characteristic), intervention details (setting, stressor type/description, BB source (i.e. commercially available or purpose-created), frequency, duration, schedule and delivery method), control conditions, outcome type, measurement, results, any reported side effects and intervention completion rate.

Quality appraisal for risk of bias

One author conducted a quality appraisal of all included articles while a second author independently checked the results. The JBI (2020) ‘Checklist for randomized control trials’ was used for this process. This tool comprises 13 methodology-based questions regarding topics such as randomisation, blinding and outcome measurement with results used to inform evidence synthesis and interpretation. No studies were excluded on a quality basis.

A second quality appraisal approach was employed by one author to explore further methodological questions for stress-related studies. No published quality appraisal tool for this was found. Therefore, a new form was devised to question whether authors have clearly described replicable methods and if these methods account for the intricacies of stress definition, induction and measurement raised by Epel et al. (2018) and Crosswell and Lockwood (2020). Results were summarised descriptively, and quality implications were discussed. However, it should be noted that this was exploratory and not an exhaustive or validated measure of quality.

Data synthesis

Results were assessed for meta-analysis suitability and due to methodological heterogeneity deemed unsuitable. Therefore, extracted data were synthesised narratively overall and using the following thematic groups: physiological and psychological outcomes (tabulated), duration of exposure, and BB frequency (Hz) in line with questions of optimal timing, duration and frequency raised by Maddison et al. (2022) and Garcia-Argibay et al. (2019). Effects were presented as significant or non-significant mean differences between groups.

Results

Study characteristics

From a total of 1131 retrieved articles, 12 met the inclusion criteria, of which seven were RCTs and five were randomised crossover trials (RXT) (see PRISMA flow chart in Figure 1 and study characteristics in Table 3). The evidence had a wide geographic scope, however, with one exception (Thanyawinichkul et al., 2022), and all studies were conducted in high-income countries such as the USA, UAE, and South Korea. Seven studies focused on people with a pre-existing stressor such as chronic pain, and five induced a stressor such as the ‘Trier Social Stress Test’ on healthy subjects. Healthy participants were often recruited from universities (Al-Shargie et al., 2021; Hautus et al., 2021; Katmah et al., 2023; Kelton et al., 2021), while studies on pre-existing stress had more varied groups such as clinical outpatients (Gkolias et al., 2020; Thanyawinichkul et al., 2022), military (Gantt et al., 2017) and again, students (Shalforoushan et al., 2021).

Figure 1. PRISMA flow chart detailing the screening process.

Table 3. Study characteristics.

Author, (year),study type & location	Setting & Participants	Stressor	Binaural beat stimuli (content & source)	Intervention schedule	Control description	Outcome(s)	Results
Al-Shargie et al. (2021) RXT UAE	Laboratory; n = 15; 2 female; students/researchers	Induced: SCWT	Beta (16 Hz); pure tones; created using MATLAB	1 session; 10 × 30-sec exposures during SCWT	No intervention	SAA (cortisol)	Significant reduction (p < 0.05) in favour of BB condition
Bang et al. (2019) RCT South Korea	Home; n = 43; 32 female; mean age 34.3 years; community	Pre-existing: sub-clinical insomnia	Theta (6 Hz); embedded in music; Dlogixs audio apparatus	2-weeks; 1 × 30 min/day	Music w/o BB	STAI-S	No significant difference between groups (p = 0.56)
Gantt et al. (2017) RCT USA	Home; n = 74; 21 female; mean age 38.11 years; military	Pre-existing: post-deployment stress; induced: hand-eye coordination challenge (operation)	Theta (≈4–7 Hz); embedded in music; Hemi-sync dreamland audio (Monroe institute)	4-weeks ≥3 consecutive nights/week; ≥30-mins	Music w/o BB	HRV; Daily mood logs	Significant differences in HF (p < 0.01), LF (p < 0.001), & total power (p < 0.05) HRV favouring BB intervention, and consistently lower scores for the BB group in mood logs
Gkolias et al. (2020) RXT Greece	Laboratory/Home; n = 21; 52.38% female; mean age 58.76 years; chronic pain outpatients	Pre-existing: chronic pain syndrome	Theta (5 Hz); embedded in music; created using MATLAB	Lab: 1 × 30-min session; Home: 1-week on-demand	Music w/o BB	STAI; BP; PR	STAI significantly reduced in BB group only (p < 0.001) after 1-week intervention. No differences between groups in physiological outcomes
Hautus et al. (2021) RXT New Zealand	Laboratory n = 92; 68 female & 1 undisclosed; mean age 23.10 years; students	Induced: PASAT	4 conditions: Delta (2 Hz), Theta (5 Hz), Alpha (10 Hz) & Beta (24 Hz); pure tones; created using Lab-VIEW & Audition 1.5	1 session; 4 × 4-mins during task recovery	Silence	PR; RR; RD; SC	SC significantly reduced in delta and beta BB conditions compared to silence (p < 0.05). No other significant effects
Katmah et al. (2023) RXT UAE	Laboratory n = 30; 5 females; university community	Induced: SCWT Enhancement condition: slow presentation & no time limit Mitigation condition: time limit, feedback, and performance comparison indicator	Beta (16 Hz); no further details reported	1 session; 2 × 10-mins during SCWT	No intervention	SAA	Levels significantly reduced in both BB conditions: enhancement (p < 0.001) & mitigation (p < 0.01) compared to their controls
Kelton et al. (2021) RCT USA	Laboratory n = 64; 32 female; mean age 19 years; students	Induced: TSST	Theta (7 Hz); embedded in pink sound & carrier tones; Monroe Institute file (exact file not specified)	1 session; 20-mins pre-TSST	Pink noise	VAS; High-frequency (HF) HRV	HF HRV was significantly higher in the BB group than control (p = 0.03). No differences were found between groups in VAS scores
Lee et al. (2022) RCT South Korea	Home n = 76; 61 female; mean age 46.12 years; community	Pre-existing: non-acute stress (PSS, BDI-II, ISI & STAI-S)	Night: Theta (5 Hz), Day: Alpha (8 Hz); embedded in music; Dlogixs application	3-weeks; 1 × 15-mins (day); alpha & 1 × 30-mins (night)	ASMR	PSS; STAI-S	No significant differences between groups for PSS (p = 0.49) or STAI-S (p = 0.72)
Leistiko et al. (2023) RXT USA, UK, Jordan & Hong Kong	Home (online) n = 58; 30 female; mean age 26 years	Induced: Attention network test	Gamma (40 Hz); embedded in pink sound; created using Ableton Live 10	1 session; 5 mins before and during task	340 Hz stereo tone	Mental states questionnaire	No significant difference between groups (p = 0.399)
Mallik and Russo (2022) RCT Canada	Home (online) n = 163; 120 female; mean age 28.29 years; anxiolytics users	Pre-existing: Moderate- and high-trait anxiety (STICSA)	2xTheta (Hz not reported), 1. embedded in music, 2. pure-tone; LUCID research application	1 session; 24-mins	Pink noise, music only	STICSA S: Somatic C: Cognitive	BB alone least effective of all conditions. Music + BB most effective for moderate- (S: p = 0.009; C: p = 0.03) & music alone for high-trait anxiety (S: p = 0.04; C: p = 0.04) (p-values vs BB alone).
Shalforoushan et al. (2021) RCT Iran	Home n = 24; 12 female; mean age 24.14 years; students	Pre-existing: Poor sleep quality	Alpha (10 Hz); not reported if pure-tone/embedded; created using Audacity software	30-days; 1 × 30-mins	No intervention (waiting list) with the same bedtime conditions	PSS; RRS	Significant reductions (p < 0.001) in PSS and RRS in BB group but not control.
Thanyawinichkul et al. (2022) RCT Thailand	Home n = 22; 16 female; mean age 52.5 years; pain clinic patients	Pre-existing: Moderate to severe chronic back pain	Theta (6 Hz) (reducing from 20 Hz over 2 mins, final minute increasing back to 20 Hz); embedded in music; created using Gnaural & Audacity	14-days 1 × 20 mins	Acoustic piano music	Thai-HADS-A	No significant difference between groups (p = 0.226)

ASMR: Autonomous Sensory Meridian Response; BDI-II: Beck Depression Inventory-II; BP: Blood Pressure; HADS-A: Hospital Anxiety and Depression Scale – Anxiety; HRV: Heart Rate Variability; Hz: Hertz; ISI: Insomnia Severity Index; PASAT: Paced Auditory Serial Addition Task; PR: Pulse Rate; PSS: Perceived Stress Scale; RD: Respiration Depth; RR: Respiration Rate; RRS: Rumination Response Scale; SAA: Salivary Alpha-Amylase; SC: Skin Conductance; SCWT: Stroop Colour Word Task; STAI-S: State-Trait Anxiety Inventory – State; STICSA: State-Trait Inventory for Cognitive and Somatic Anxiety; TSST: Trier Social Stress Test; VAS: Visual Analogue Scale.

Six studies involved a longer-term home-based BB protocol, six used a single BB session (four laboratory-based and two online) and one study included both approaches (Table 3). Generally, single sessions induced a stressor, while longer protocols did not. Two exceptions were the study by Mallik and Russo (2022), which exposed a single BB session to participants with pre-existing anxiety, and the study by Gantt et al. (2017), which tested the effectiveness of a home-based protocol with a laboratory-induced stressor.

Most studies investigated a single BB frequency – most commonly theta – with a music-only condition as the most used control (Table 3). Eight studies used sound-embedded BB (i.e. with music), two studied pure-tone BB, while two did not clarify the content of the BB stimuli. Similarly, sources of BB varied with five assessing a commercially available BB (i.e. Monroe Institute file), and six using purpose-created BB for the study. One study failed to describe the BB stimuli source. In the studies testing commercially available BB stimuli, all provided source references indicating some reproducibility. However, it was unclear whether they were verified as true BB. In the studies that used purpose-created BB, sound characteristics and generation procedures were described with varying detail. For example, Katmah et al. (2023) described the BB frequency only, while Al-Shargie et al. (2021) detailed frequency (for each ear and the BB), sampling rate and sound pressure level. Furthermore, there was no clear reporting of any pilot or third-party testing of purpose-created BB stimuli.

Quality of studies

Overall, studies were of mixed quality (see Table 4) with numbers of criteria met ranging from 2 to 12/13. Note for RXTs and one-session experiments, not all criteria were applicable. The main limitations of the included studies were unclear group allocation and blinding procedures. For example, five studies clearly used true randomisation methods, and only one study evidently blinded their outcome assessors. Oppositely, most studies maintained equal treatment between groups or conditions, and with three exceptions, were judged to have used appropriate methods.

Table 4. Quality appraisal results (reproduced from JBI, 2020).

		Al-Shargie et al.	Bang et al.	Gantt et al.	Gkolias et al.	Hautus et al.	Katmah et al.	Kelton et al.	Lee et al.	Leistiko et al.	Mallik & Russo	Shalforoushan et al.	Thanyawinichkul et al.
Group allocation	1. True randomisation
	2. Allocation concealment
	3. Similar groups at baseline
Blinding	4. Participants
	5. Those delivering treatment
	6. Outcome assessors
Equal treatment of groups	7. Intervention
	8. Follow-up
	9. Group analysis
	10. Outcome measurement
Appropriateness of methods	11. Outcome measurement
	12. Statistical analysis
	13. Trial design

Key: = yes; = unclear; = no; = not applicable.

In the stress-specific quality appraisal, studies were also judged to be of mixed quality (see Table 5). In the six stress induction studies, the numbers of criteria met ranged from 3 to 9/11. Only two studies met all ‘stressor’ criteria including justification of stressor selection. In the seven pre-existing stress studies, the numbers of criteria met ranged from 2 to 8/9. Of these, four were judged to adequately describe the stressor (i.e. for comparability and replicability). For all studies, the main area of concern was the lack of consideration of potential stress mediators and moderators in their designs.

Table 5. Results of a purpose-designed tool to explore methodological quality in the context of stress research. Adapted from comments made by Crosswell and Lockwood (2020) and Epel et al. (2018).

	Al-Shargie et al.	Bang et al.	Gantt et al.	Gkolias et al.	Hautus et al.	Katmah et al.	Kelton et al.	Lee et al.	Leistiko et al.	Mallik & Russo	Shalforoushan et al.	Thanyawinichkul et al.
"Stress’ or ‘anxiety’ a primary (P) or secondary (S) outcome?	P	S	P	S	P	P	P	P	S	P	S	S
Stressor
Induced stressor? (yes (Y) or no (N))	Y	N	Y	N	Y	Y	Y	N	Y	N	N	N
Was the procedure clearly described?
If applicable, was the procedure/equipment used appropriately referenced?
Did the authors justify stressor selection in the context of the population/procedure?
Existing stressor? (yes (Y) or no (N))	N	Y	Y	Y	N	N	N	Y	N	Y	Y	Y
Was the stressor adequately described?
Potential stress moderators/meditators
Were additional sources of stress accounted for? (i.e. recent life events)
Did the authors state participants abstained from common substances that may interfere with ANS activity/feelings of stress around the procedure or maintained their usual lifestyle patterns? (i.e. alcohol, caffeine)
Did researchers account for sleep and circadian rhythm in study design? (i.e. recent sleep pattern/quality & time of procedure/measurements)
Outcome measurement
Were measurement procedures comprehensively described?
If applicable, were measurement resources/equipment used appropriately referenced?
Was the timing of outcome measurement appropriate for the question? (i.e. anticipatory stress, stress during a task, stress recovery, changes over time)
Were strengths/limitations/implications of measurement procedures considered? (i.e. comparability with other measures, conscious perception in self-report methods)
Did the authors justify outcome measurement selection in the context of the stressor, population, and procedure?

Key: = yes; = unclear/partially; = no; = not applicable.

Summaries of effects

Outcomes of BB intervention on 15 unique stress-related outcomes are reported in Table 6 (physiological) and Table 7 (psychological). Overall, eight studies reported at least one significant effect in favour of BB intervention, and four reported no significant differences between groups. One study assessed both pure-tone and music-combined theta BB finding significant effects in favour of combined but reported pure-tone BB as the least effective condition for short-term anxiety reduction (Mallik & Russo, 2022). Note for this synthesis, the combined condition in this study is counted as having an effect in favour of BB.

Table 6. Effects of physiological outcomes.

Author	Outcome 1 effect	Outcome 2 effect	Outcome 3 effect	Outcome 4 effect	Adverse effects
Al-Shargie et al.	Salivary cortisol/alpha-amylase Significant increases (p < 0.05) from baseline to control, and control to stress. Significant decrease (p < 0.05) between stress and mitigation. No significant difference between control and mitigation.				Authors did not report
Gantt et al.	HRV (ms^2/Hz) LF HRV statistically significant difference (p < 0.01) – BB −2.5; Control +7.99 HF HRV statistically sig. difference (p < 0.01) – BB +2.5; Control −7.64 Total power HRV – control group showed astatistically significant decrease (−1,113.64) compared to the BB group (−26.68) (p < 0.05)				Authors did not report
Gkolias et al.	Systolic BP (mmHg) Brief application: BB – pre: 126.9 ± 16.1; post: 124.9 ± 14.9 (p = 0.69) Control: pre: 128.4 ± 14.3; post: 126.2 ± 15.9 (p = 0.63)	Diastolic BP (mmHg) Brief application: BB – pre: 77.5 ± 10.4; post: 76.8 ± 8.9 (p = 0.94) Control – pre: 77.2 ± 7.8; post 76.3 ± 9.3 (p = 0.66)	Pulse rate (BPM) Brief application: BB – pre: 69 (63–73); post: 65 (61–70) (p < 0.05) Control – pre: 73 (63–76); post: 68 (60–76) (p < 0.05)		Authors did not report
Hautus et al.	Skin conductance (µS) change from baseline Silence: −1.73 (1.93) 2 Hz: −2.28 (2.10) 5 Hz: −2.20 (1.97) 10 Hz: −2.35 (2.15) 24 Hz: −2.65 (2.15) F = 3.45 (p < 0.05). Significant difference between silence and 2 Hz, silence and 24 Hz.	Pulse rate (BPM) (1–20s) mean change from baseline Silence: 1.69 (3.43) 2 Hz: 1.84 (4.15) 5 Hz: 1.34 (3.72) 10 Hz: 2.10 (4.11) 24 Hz: 3.12 (3.43) F = 0.525	Respiration depth (mm) mean change from baseline Silence: 0.43 (3.97) 2 Hz: −0.50 (3.81) 5 Hz: −0.09 (5.26) 10 Hz: 0.10 (4.58) 24 Hz: 1.05 (4.69) F = 0.672	Respiration rate (bpm) mean change from baseline Silence: −2.49 (3.26) 2 Hz: −1.60 (2.73) 5 Hz: −1.79 (3.11) 10 Hz: −1.02 (3.47) 24 Hz: −1.86 (2.95) F = 0.090	Authors did not report
Katmah et al.	Salivary alpha-amylase Baseline to vigilance ± 227.83% (p < 0.001) Vigilance to stress ± 32.09% (p < 0.001) Vigilance to enhancement – −40.29% (p < 0.001) Stress to mitigation – −30.29% (p < 0.01)				Authors did not report
Kelton et al.	HRV Intervention group experienced significantly higher HF HRV during TSST F (1,55) = 4.84 p = 0.03 n^2= 0.08				Authors did not report

BPM: Beats per minute; bpm: breaths per minute; Hz: hertz; mmHg: mm: millimetres; millimetres of mercury; ms: milliseconds; S: Siemens.

Table 7. Effects of psychological outcomes.

Author	Outcome 1 effect	Outcome 2 effect	Outcome 3 effect	Outcome 4 effect	Adverse effects
Bang et al.	STAI-S Active – Pre-38.95 ± 9.54; post-36.15 ± 6.76 (p = 0.09) Sham – Pre-39.19 ±  9.01; post-34.95 ± 9.42 (p = 0.018) Independent t test – p = 0.56 (95% CI −3.26,5.92)				No adverse events occurred
Gantt et al.	Mood logs (tense at work/home) BB group consistently scored lower across all time points				Authors did not report
Gkolias et al.	STAI Brief application: BB – pre: 46.8 ± 13.3; post: 37.6 ± 9.4 (p < 0.001) Control – pre: 43.8 ± 13.3; post: 37.9 ± 10.1 (p < 0.05). One week on-demand: BB – pre: 46.8 ± 13.3; post: 39.8 ± 11.5 (p < 0.001) Control – pre: 43.8 ± 13.3; post: 45.2 ± 14.7 (p = 0.45)				Authors did not report
Kelton et al.	VAS No significant difference between intervention and control				Authors did not report
Lee et al.	PSS BB group pre: 22.89 ± 2.70 post: 16.74 ± 3.46 (p =< 0.001) ASMR group pre: 22.63 ± 3.54 post: 16.16 ± 3.28 (p =< 0.001) Significant effect of time (p < 0.01) No significant effect of group (BB not sig. different from ASMR) (p = 0.49)	STAI-S BB group pre: 53.42 ± 7.42 post: 42.68 ± 8.93 (p =< 0.001) ASMR group pre: 54.55 ± 10.14 post: 2.82 ± 9.22 (p < 0.001) Significant effect of time (p < 0.01) No significant effect of group (BB not sig. different from ASMR) (p = 0.72)			1 (BB) & 2 (ASMR) participants complained of mild headache which improved after follow-up without treatment.
Leistiko et al.	Mental states questionnaire – anxiety No significant difference in change between groups: Z = 593, p = 0.399				Authors did not report
Mallik & Russo	Somatic state anxiety (moderate-trait participants) Mean reductions Combined: 4.36 ± 3.49 Music: 3.43 ± 3.23 BB: 1.86 ± 2.20 Pink: 2.10 ± 1.64 Effects (Cohen's d) Combined vs BB: 0.86 (p = 0.009) Combined vs pink: 0.83 (p = 0.04)	Cognitive state anxiety (moderate-trait participants) Mean reductions Combined: 6.38 ± 4.14 Music: 3.35 ± 3.20 BB: 3.13 ± 4.18 Pink: 4.42 ± 3.83 Effects (Cohen's d) Combined vs BB: 0.78 (p = 0.03) Combined vs music: 0.82 (p = 0.01) Combined vs pink: 0.49 (p = 0.05)	Somatic state anxiety (high-trait participants) Mean reductions Combined: 7.00 ± 6.04 Music: 8.80 ± 6.25 BB: 4.72 ± 4.96 Pink: 6.60 ± 8.86 Effects (Cohen's d) Music vs BB: 0.72 (p = 0.04)	Cognitive state anxiety (high-trait participants) Mean reductions Combined: 8.71 ± 6.70 Music: 10.13 ± 6.42 BB: 5.72 ± 5.52 Pink: 8.65 ± 5.92 Effects (Cohen's d) Music vs BB: 0.74 (p = 0.04)	Authors did not report
Shalforoushan et al.	PSS Experimental group: pre-31.16 ± 11.02; post-21.01 ± 9.82 p < 0.001 Control group: pre-33.50 ± 13.31; post-35.83 ± 12.44 p = 0.200 alpha binaural beat intervention could have a protective effect on stress (OR = 0.69, P < 0.0001)	RRS Experimental group: pre-46.83 ± 15.67; post – 33.91 ± 12.58 p < 0.001 Control group: pre-41.08 ± 17.99; post-42.08 ± 17.80 p = 0.150 Alpha binaural beat intervention could have a protective effect on rumination (OR = 0.85, P < 0.0001),			Authors did not report
Thanyawinichkul et al.	HADS-A Pre–post (BB) – effect 0.37 (p = 0.064) pre-6.90 ± 4.2 post-5.60 ± 2.7 Pre–post (Control) – effect 0.29 (p = 0.260) pre-8.75 ± 3.7 post-7.58 ± 24.4 Pre–post intergroup mean differences pre-1.850 (−1.627–5.327) (p = 0.280) post-1.983 (−1.330–5.297) (p = 0.226)				Authors did not report

Physiological and psychological outcomes

Two RCTs examined the effect of theta BB on heart rate variability (HRV) both finding significantly higher high-frequency HRV during stress tasks in intervention conditions than controls (pink noise and music) indicating higher parasympathetic activity (Gantt et al., 2017; Kelton et al., 2021). Two RXTs reported salivary hormone levels (cortisol and alpha-amylase) to significantly reduce with beta BB exposure during a stressful task (Al-Shargie et al., 2021; Katmah et al., 2023). However, there is limited confidence in both studies due to the high risk of carryover effects and lack of blinding. One RXT measured skin conductance and found a reduction in subjects exposed to 4 minutes of delta and beta BB during stressor recovery, when compared with silent control (Hautus et al., 2021). No significant effects between groups were reported in any of the remaining autonomic indicators: pulse rate, respiration rate and depth, and blood pressure (Gkolias et al., 2020; Hautus et al., 2021).

Psychological outcomes were all measured using self-report methods. Five studies assessed anxiety using measures including State-Trait Anxiety Inventory (STAI) (Bang et al., 2019; Gkolias et al., 2020; Lee et al., 2022), State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA) (Mallik & Russo, 2022) and Hospital Anxiety and Depression Scale-Anxiety (HADS-A) (Thanyawinichkul et al., 2022) questionnaires reporting mixed results. For example, Gkolias et al. (2020) reported a significant reduction in STAI scores in chronic pain outpatients following a 1-week on-demand theta BB protocol but not in the sham control group. However, Thanyawinichkul et al. (2022) used the Thai-HADS-A also in subjects with chronic pain finding no significant effects. Alternatively, one study measured pre-sleep rumination (Rumination Response Scale) as a proxy for anxiety in students with poor sleep quality (Shalforoushan et al., 2021) and reported significant reductions and group effects with pre-bedtime alpha BB exposure.

Perceived stress was also evaluated in several ways. Two studies used the Perceived Stress Scale (PSS) and both reported scores to reduce pre–post-intervention (Lee et al., 2022; Shalforoushan et al., 2021). This effect was significant when compared to a waiting-list control by Shalforoushan et al. (2021) but similarly effective when compared to an ASMR control by Lee et al. (2022). Studies using the Visual Analogue Scale (VAS) and a mental states questionnaire reported no significant effects (Kelton et al., 2021; Leistiko et al., 2023). Finally, Gantt et al. (2017) used an unvalidated questionnaire that asked participants to reflect on the previous day’s stress observing consistently lower stress reporting in the intervention group.

Effects by duration of exposure

In the studies examining the effects of single-session BB interventions, total durations ranged from 5 minutes (10 × 30 seconds) to 24 minutes (continuous) and were mostly around 20 minutes. Of these studies, all except Leistiko et al. (2023), who presented gamma BB during an attention task, found significant reductions in stress/anxiety. Comparing two studies with similar designs, Al-Shargie et al. (2021) presented a total of 5-minute beta BB, and Katmah et al. (2023), 20 minutes. While both achieved significant reductions in salivary stress hormones, the 20-minute protocol yielded more significant results (p < 0.05 vs p < 0.0001 & p < 0.01).

Longer protocols ranged from 1 week (on-demand) to 30 days (1 × 30 minutes every day). No dose–response effect was apparent with significant effects found in both the shortest (1 week) and longest (30 days) interventions. Demographic differences limit comparability between single-session and longer protocols. However, Gkolias et al. (2020) studied both a theta BB brief 30-minute and 1-week application in an RXT design. Both groups had reduced anxiety (STAI) following the brief application, but after a week, only the intervention significantly reduced anxiety levels.

Effects by BB frequency (Hz)

The distribution of studies for each BB frequency is shown in Figure 2. There were insufficient numbers of studies across frequency bands with only single papers for delta, alpha, gamma and combined (alpha-theta) protocols. Thus, it was not possible to make any conclusions on effectiveness by frequency. Furthermore, of seven studies examining a theta protocol, four reported results in favour of BB intervention while similarly, three reported no significant effects, which is again inconclusive. One study compared multiple BB frequencies (delta, theta, alpha and beta) and silence for short-term stressor recovery (Hautus et al., 2021) finding only delta (lowest Hz) and beta (highest Hz) BBs to significantly reduce skin conductance compared to silence. However, no patterns in frequency effectiveness were apparent.

Figure 2. Distribution of studies per BB frequency band (Hz) with numbers reporting and not reporting one or more significant effects favouring BB intervention.

Adverse effects and acceptability

Ten studies did not explicitly report any adverse effects associated with the intervention (Tables 6 and 7). One study stated that there were no adverse events (Bang et al., 2019) and one reported mild headache, which improved without treatment (Lee et al., 2022). However, this also occurred in the control condition, so it was not attributed to BB specifically. Additionally, all studies reported that above 90% completion rate inferring BB intervention was acceptable for involved populations.

Discussion

Excessive stress is a recognised threat to health and wellbeing (Everly & Lating, 2019; Selye, 2013). In contrast, subjective wellbeing has been associated with longevity and health (Diener & Chan, 2011), and studies have shown potential for physiological improvement from stress-management interventions such as meditation (Carlson et al., 2004; Jacobs et al., 2011). BB is now a widely available technology for which existing evidence also suggests potential for wellbeing (Basu & Banerjee, 2020). Yet, this review is the first to explore the efficacy of BB as a self-care stress-management tool.

Non-clinical use of BB appears to be safe and acceptable with no harms attributed to exposure and high (>90%) intervention completion rates. However, reporting of adverse effects was largely overlooked, as it was in other reviews on BB interventions (Basu & Banerjee, 2022; Garcia-Argibay et al., 2019; Maddison et al., 2022) perhaps indicating an assumption of safety. Given that concerns have been raised elsewhere for potential adverse effects on hearing (Musiek et al., 2012) and mental health (Wahbeh et al., 2007a; Wahbeh et al., 2007b), readers should be cognisant of this lack of reporting.

Overall, the evidence for efficacy was equivocal with a third of studies not finding BB to be associated with any significant reductions in stress or anxiety and inconsistent results within some studies. For example, Hautus et al. (2021) reported reductions in skin conductance associated with 2 and 24 Hz BB but no other physiological measures or BB frequencies. Methodological heterogeneity contributed to difficulties in making conclusions with 15 unique outcomes used, inconsistent application of BB stimuli and intervention schedules, and varied methodological quality. Moreover, it was generally unclear whether the BB stimuli were verified as true BB with no clear reporting of any pilot or third-party testing procedures. Nonetheless, there were some interesting findings as discussed below.

BB and the stress-response

All five studies that measured physiological responses to stressors found a significant effect between BB and non-BB groups, indicating that BB can have acute effects on the stress response. The strongest evidence comes from two studies that found significantly higher HRV with theta BB exposure (Gantt et al., 2017; Kelton et al., 2021). Higher HRV represents a balanced autonomic nervous system able to respond to stressors, while lower HRV indicates that the body is working to adapt to or recover from a stressor, and it has clinical relevance as a predictor of illness such as cardiac mortality (Kim et al., 2018; Sztajzel, 2004). Apart from some questions surrounding blinding procedures by Kelton et al. (2021), these studies were judged as having a lower risk of bias and their findings are consistent with evidence from relaxation studies exposing healthy subjects also to theta BB (Bakaeva et al., 2022; Chockboondee et al., 2023).

Significant effects on salivary stress-indicators (alpha-amylase and cortisol) were found by two similarly designed studies that mitigated against induced stress with beta BB (Al-Shargie et al., 2021; Katmah et al., 2023). Alpha-amylase is a recognised surrogate marker of sympathetic nervous system activation (Petrakova et al., 2015), while cortisol is released, circulates in response to stressors, and is a well-used measure of stress reactivity (Everly & Lating, 2019). These effects were interesting as beta BB is usually associated with enhanced arousal and cognition rather than relaxation (Garcia-Argibay et al., 2019). One possible explanation is that the perception of good or bad performance may influence stress-reactivity in a similar way to active and passive coping (Everly & Lating, 2019). However, the above should be interpreted with caution due to the high risk of bias in both studies.

Apart from some findings for skin conductance (Hautus et al., 2021), BB had no effects on the remaining physiological outcomes in the included studies. However, effects have been seen in clinical contexts with several studies reporting reductions in blood pressure and pulse rate in patients exposed to BB peri-operatively (i.e. cataract surgery) (Loong et al., 2022; Opartpunyasarn et al., 2022; Wiwatwongwana et al., 2016). This contrast in findings may reflect contextual differences. For example, experiments conducted in clinical settings where that type of data collection is routine and environments are naturalistic have higher ecological validity (Andrade, 2018). For example, an unfamiliar laboratory environment could interfere with stress responses and participants’ abilities to relax as they might at home. In contrast, laboratory studies may have more relevance for less routine situations such as public speaking and job interviews.

BB for chronic stress management

While the above results indicate the potential effectiveness of BB on stress, to address the question of its efficacy as a personal stress-management tool, it is useful to look to longer duration, home-based interventions for more ecologically valid studies that explore if effects persist beyond initial exposure.

One week of BB exposure was sufficient to observe significant differences in STAI scores in chronic pain outpatients (Gkolias et al., 2020) while a daily intervention of 30 days also significantly reduced perceived stress and rumination in students with poor sleep quality (Shalforoushan et al., 2021). However, interventions of 2 weeks reported no significant group effects (Bang et al., 2019; Thanyawinichkul et al., 2022). Similar inconsistencies exist throughout the literature with significant reductions in trait anxiety recorded during a pilot trial of a 60-day delta BB intervention (Wahbeh et al., 2007) and only slight, non-significant effects found in a 4-week study (Le Scouarnec et al., 2001). Thus, it is unclear what duration of exposure one should aim towards for meaningful outcome assessment.

The limited available evidence for exposure schedule suggests daily listening may not be necessary, but exposures should be within 1-week of each other to maintain stress-buffering effects. A washout period analysis by Gkolias et al. (2020) indicated effects from a 1-week intervention did not persist for more than 1-week afterwards. Gantt et al. (2017) observed significant differences in HRV up to 1-week post-intervention where the protocol did not involve daily exposure (>3 times/week). Longer post-intervention follow-ups could explore this further.

Stress-methodology implications

The issues highlighted in the stress-specific methodology were consistent with those raised in the wider stress literature and limit the trustworthiness of some of the reviewed studies (Crosswell & Lockwood, 2020; Epel et al., 2018). Changes in physiological stress-markers have been observed alongside variables including sleep quality, caffeine consumption and life events such as bereavement (Al-Amin et al., 2016; Fagundes et al., 2018; van Dalfsen & Markus, 2018). Yet, not accounting for these was the largest concern drawn from the stress-method exploration. Controlling for these, either by exclusion, or recording and reporting as characteristics would strengthen findings in future BB stress research.

Further to the above, outcome selection (particularly of self-report methods) in stress measurement requires careful consideration of population characteristics and circumstances, and the mechanisms of stress in those contexts (Crosswell & Lockwood, 2020). For example, overwhelm may be relevant to student stress during exam periods.

Similarly, cultural norms in definitions and expressions of mental health in China lean towards the use of somatic measures whereby headache may be a better stress indicator (Ryder & Chentsova-Dutton, 2012). This is a consideration for global studies such as that by Leistiko et al. (2023). However, the comparability of variable outcomes is limited as relationships between measures are weak (Epel et al., 2018). Thus, context-specific measures limit comparability between populations.

Strengths and limitations

This review was rigorously conducted, harnessing the strengths of a double-reviewer process and the thoroughness of a sensitive search strategy. However, there were some limitations. Primarily, it was difficult to define stress due to there being many related constructs (i.e. relaxation) and closely linked topics such as sleep quality (van Dalfsen & Markus, 2018). Papers on closely linked topics were included as stress management was considered pertinent in these contexts. However, this contributed to methodological heterogeneity, and it was difficult to disentangle the effects of stress from these other factors. Secondly, in the absence of an existing tool, the review used an unvalidated measure of stress-specific methodological quality. While this provided awareness of challenges in stress research, its findings should be interpreted within context. Finally, the review was altered from protocol to limit to RCTs. Excluding quasi-experimental studies potentially misses findings, which may have provided desired evidence on efficacy, frequency and dose–response effects of BB. However, it would have been challenging to meaningfully review that scope of evidence with the resources available.

Several limitations of the included studies may impact the interpretation of this review’s conclusions. Firstly, reporting of randomisation and blinding procedures was generally unclear. Thus, the risk of exaggerated effect reporting was unclear (Forbes, 2013). Secondly, all studies except one had modest sample sizes (<100). High variability in stress responses due to factors including personality and childhood experiences (Bibbey et al., 2013; Brindle et al., 2022) may be problematic for clinical studies with small samples. Finally, there was a methodological and demographic divide in studies. Healthy subjects were only exposed to single sessions while subjects with a pre-existing stressor tended to partake in home-based protocols with no physiological outcome measurement. This meant it was not meaningful to compare effectiveness between healthy and ‘stress’ subjects. Moreover, healthy subjects tended to be younger. Some evidence suggests BB elicits a greater response in younger people (Grose & Mamo, 2012). Therefore, results cannot be generalised to older populations.

Recommendations for future research

The findings of this review could be quantified with meta-analyses with a narrower methodological focus (for example, laboratory assessments of BB on physiological stress outcomes). Future primary research should address the limited knowledge of adverse effects with a systematic approach to identification, analysis and reporting of harms associated with BB exposure, such as that outlined in ‘CONSORT Harms 2022’ (Junqueira et al., 2023). Research should also aim to define and verify BB stimuli (i.e. embedded vs pure-tone, frequency (Hz), and intervention schedule) to work towards standardised methodologies with empirically validated stimuli. Recruitment of larger and more diverse samples could offer better generalisability and account for large variability in stress responses. This would also allow for more subgroup analyses of different characteristics (i.e. trait anxiety) and enable testing of multiple BB stimuli and intervention schedules, to disentangle methodological and demographic variables from effects. These samples could be achieved with the use of online applications. Two studies provided proof of this concept successfully by adapting to remote/online evaluations following COVID restrictions (Leistiko et al., 2023; Mallik & Russo, 2022). Additionally, the use of wearable technologies could enhance remote studies with the inclusion of objective physiological outcomes such as HRV (Hinde et al., 2021). Finally, as part of this review, we developed a stress-specific quality assessment tool. However, the development of a validated tool could address many of the concerns raised here, and in the wider stress literature. This could provide guidance around procedures, outcomes and stressor selection to minimise the impact of external factors such as acute stressful events.

Implications for practice

Globally, there is a growing burden of stress and stress-related disease, despite awareness and existing stress-management strategies. BB may have some stress-buffering effects, but there is currently no evidence for adverse effects, and they are widely accessible on mainstream media platforms. However, this review did not find strong evidence for their efficacy in non-clinical contexts, and it is unclear what protocol may be optimal to see benefits. On this evidence, they are not yet recommended for widespread use for stress management but may have use complementary to more evidence-based approaches.

Conclusion

Unmanaged stress is a growing problem with significant public health implications; however, some evidence indicates that BB has a stress-buffering effect. This review explored the effects of BB interventions on non-clinical stress to ascertain their efficacy as a self-care stress-management strategy. Twelve included studies provided equivocal results with promising findings weakened by unclear randomisation and blinding procedures and oversight of the intricacies of stress measurement. Furthermore, methodological heterogeneity led to difficulties in making conclusions. Online and wearable technologies may allow future research to access larger, more diverse samples to address questions surrounding optimal frequencies and schedules and generate more robust results.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Appendices

Appendix 1. Completed PRISMA checklist

Table

Section and Topic	Item #	Checklist item	Location where the item is reported
TITLE
Title	1	Identify the report as a systematic review.	Pg 4
ABSTRACT
Abstract	2	See the PRISMA 2020 for Abstracts checklist.	Pg 2
INTRODUCTION
Rationale	3	Describe the rationale for the review in the context of existing knowledge.	Pgs 3–4
Objectives	4	Provide an explicit statement of the objective(s) or question(s) the review addresses.	Pg 4
METHODS
Eligibility criteria	5	Specify the inclusion and exclusion criteria for the review and how studies were grouped for the syntheses.	Pg 5
Information sources	6	Specify all databases, registers, websites, organisations, reference lists and other sources searched or consulted to identify studies. Specify the date when each source was last searched or consulted.	Pg 5
Search strategy	7	Present the full search strategies for all databases, registers and websites, including any filters and limits used.	Pg 5 (Table 1) & Appendix 2
Selection process	8	Specify the methods used to decide whether a study met the inclusion criteria of the review, including how many reviewers screened each record and each report retrieved, whether they worked independently, and if applicable, details of automation tools used in the process.	Pg 5
Data collection process	9	Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and if applicable, details of automation tools used in the process.	Pg 5
Data items	10a	List and define all outcomes for which data were sought. Specify whether all results that were compatible with each outcome domain in each study were sought (e.g. for all measures, time points, analyses), and if not, the methods used to decide which results to collect.	Pg 5 (Table 2)
	10b	List and define all other variables for which data were sought (e.g. participant and intervention characteristics, funding sources). Describe any assumptions made about any missing or unclear information.	Pg 5 (Table 2)
Study risk of bias assessment	11	Specify the methods used to assess the risk of bias in the included studies, including details of the tool(s) used, how many reviewers assessed each study and whether they worked independently, and if applicable, details of automation tools used in the process.	Pg 6
Effect measures	12	Specify for each outcome the effect measure(s) (e.g. risk ratio, mean difference) used in the synthesis or presentation of results.	Pg 6
Synthesis methods	13a	Describe the processes used to decide which studies were eligible for each synthesis (e.g. tabulating the study intervention characteristics and comparing against the planned groups for each synthesis (item #5)).	Pg 6
	13b	Describe any methods required to prepare the data for presentation or synthesis, such as handling of missing summary statistics, or data conversions.	N/A
	13c	Describe any methods used to tabulate or visually display the results of individual studies and syntheses.	Pg 6
	13d	Describe any methods used to synthesize results and provide a rationale for the choice(s). If meta-analysis was performed, describe the model(s), method(s) to identify the presence and extent of statistical heterogeneity, and software package(s) used.	Pg 6
	13e	Describe any methods used to explore possible causes of heterogeneity among study results (e.g. subgroup analysis, meta-regression).	N/A
	13f	Describe any sensitivity analyses conducted to assess the robustness of the synthesized results.	N/A
Reporting bias assessment	14	Describe any methods used to assess the risk of bias due to missing results in a synthesis (arising from reporting biases).	N/A
Certainty assessment	15	Describe any methods used to assess certainty (or confidence) in the body of evidence for an outcome.	N/A
RESULTS
Study selection	16a	Describe the results of the search and selection process, from the number of records identified in the search to the number of studies included in the review, ideally using a flow diagram.	Pgs 6–7 (Figure 1)
	16b	Cite studies that might appear to meet the inclusion criteria, but which were excluded, and explain why they were excluded.	N/A
Study characteristics	17	Cite each included study and present its characteristics.	Pg 7 (Table 3)
Risk of bias in studies	18	Present assessments of risk of bias for each included study.	Pg 7 (Table 4)
Results of individual studies	19	For all outcomes, present, for each study: (a) summary statistics for each group (where appropriate) and (b) an effect estimate and its precision (e.g. confidence/credible interval), ideally using structured tables or plots.	Pg 8 (Tables 6 and 7)
Results of syntheses	20a	For each synthesis, briefly summarise the characteristics and risk of bias among contributing studies.	Pgs 8&9
	20b	Present results of all statistical syntheses conducted. If meta-analysis was done, present for each the summary estimate and its precision (e.g. confidence/credible interval) and measures of statistical heterogeneity. If comparing groups, describe the direction of the effect.	N/A
	20c	Present results of all investigations of possible causes of heterogeneity among study results.	N/A
	20d	Present results of all sensitivity analyses conducted to assess the robustness of the synthesized results.	N/A
Reporting biases	21	Present assessments of risk of bias due to missing results (arising from reporting biases) for each synthesis assessed.	N/A
Certainty of evidence	22	Present assessments of certainty (or confidence) in the body of evidence for each outcome assessed.	N/A
DISCUSSION
Discussion	23a	Provide a general interpretation of the results in the context of other evidence.	Pgs 10–12
	23b	Discuss any limitations of the evidence included in the review.	Pgs 13–14
	23c	Discuss any limitations of the review processes used.	Pg 13
	23d	Discuss the implications of the results for practice, policy, and future research.	Pg 14
OTHER INFORMATION
Registration and protocol	24a	Provide registration information for the review, including the register name and registration number, or state that the review was not registered.	Pg 4
	24b	Indicate where the review protocol can be accessed, or state that a protocol was not prepared.	Pg 4
	24c	Describe and explain any amendments to the information provided at registration or in the protocol.	Pg 5
Support	25	Describe sources of financial or non-financial support for the review and the role of the funders or sponsors in the review.	N/A
Competing interests	26	Declare any competing interests of review authors.	N/A
Availability of data, code and other materials	27	Report which of the following are publicly available and where they can be found: template data collection forms; data extracted from included studies; data used for all analyses; analytic code; any other materials used in the review.	Pgs 5 & 6

Appendix 2. Embase search strategy

Table

Search strategy
1	binaural beat*.mp.	248
2	binaural sound*.mp.	140
3	binaural tone*.mp.	80
4	binaural noise.mp.	56
5	binaural entrainment.mp.	2
6	binaural audi*.mp.	222
7	binaural stimulation.mp.	319
8	auditory entrainment.mp.	43
9	oscillatory entrainment.mp.	92
10	sensory entrainment.mp.	21
11	gamma entrainment.mp.	50
12	theta entrainment.mp.	13
13	alpha entrainment.mp.	34
14	beta entrainment.mp.	6
15	delta entrainment.mp.	10
16	brain* entrainment.mp.	58
17	stress.mp. or exp physiological stress/	1,681,836
18	exp anxiety/or anxiety.mp.	507,798
19	arousal.mp. or arousal/	75,601
20	distress.mp. or exp distress syndrome/	302,548
21	mental suffer*.mp.	490
22	nervousness.mp. or exp nervousness/	15,998
23	vigilance.mp. or exp alertness/	37,231
24	autonomic nervous system.mp. or autonomic nervous system/	94,742
25	parasympathetic.mp. or parasympathetic tone/	32,790
26	sympathetic.mp. or sympathetic tone/	141,456
27	cortisol.mp.	94,693
28	catecholamine/ or catecholamine.mp.	90,833
29	epinephrine.mp. or epinephrine/	75,033
30	norepinephrine.mp. or noradrenalin/	182,682
31	heart rate.mp. or exp heart rate/	384,926
32	respiratory rate.mp. or exp breathing rate/	69,832
33	hypertension.mp. or hypertension/	1,158,028
34	blood pressure.mp. or blood pressure/	784,313
35	galvanic skin response.mp. or exp electrodermal response/	8938
36	skin conductance/ or skin conduct*.mp.	10,481
37	body temperature.mp. or exp body temperature/	98,862
38	relaxation.mp.	178,790
39	exp eye movement/or exp eye tracking/or eye*tracking.mp.	68,710
40	electroencephalogram/ or electroencephalogra*.mp. or electroencephalography/	295,002
41	exp electromyogram/or exp electromyography/or electromyogra*.mp.	148,016
42	17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41	5,053,500
43	1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16	1289
44	42 and 43	345
45	limit 44 to english language	339