What might an interactive music therapy service bring to the adult intensive care unit? A suggested service based on data from a scoping review

Abstract

Individual music listening in intensive care units (ICU) has more frequently reported benefits for reducing anxiety and pain. There may be benefits in reducing heart rate and respiratory rate. There is a lack of research in the ICU setting that examines the effects of active music-based interventions, for example, where patients play instruments as part of structured exercises for cognitive or physical rehabilitation or improvise music with a Health and Care Professions Council (HCPC) registered Music Therapist in ways that may benefit emotional well-being in addition to physical and cognitive function, including management of deconditioning.

The search string of the present scoping review was used on 25 January 2022 and then again for PubMed on 24 June 2023 and brought up any research using music or music therapy in the intensive care setting. Out of 139 titles, 45 full texts were included, with a total of 3,441 participants. Forty-three studies meeting the inclusion criteria used pre-recorded music, ranging from cassette to CDs, audio files, and MP3, most often delivered using headphones, some with audio pillows. All interventions were passive (listening) except for one, which delivered music played live. More research is needed into the workings of active and receptive music therapy in the adult ICU.

Introduction

Annually, more than 250,000 patients are admitted to the intensive care unit (ICU) in the United Kingdom alone (NHS Digital, 2023; NICE, 2022). While the age range of patients admitted to an adult ICU lies between 18 and 95 years (Zimmerman et al., 2013), reasons for admission can vary that is, elective surgery, emergency surgery and/or medical (Zimmerman et al., 2013). Patients suffering from critical illness or potentially recovering from life-threatening situations in the ICU are subject to different procedures, impacting length, intensity, and impact of the stay. Post Intensive Care Syndrome, general discomfort and other associated problems like delirium or depression are often related to an ICU stay, especially after discharge.

ICUs are described as hostile environments where patients are exposed to unfamiliar sounds and sensations while receiving life-saving care (NHS Digital, 2023). This can impact both the patient and next of kin during their stay. Reasons for this include the extreme medical environment, the uncertainty between life and death of the patient, and the visual and auditory stimuli impacting the patient. Senses connect us to the environment and are an important part of our interaction between the mind, body, and the world around us. The intensity of perception through the senses differs per-patient, however, and can have a significant impact on quality of stay, well-being, and recovery. The perception of sound and the associated auditory soundscapes created in the ICU through machines, procedures, and other patients have a significant impact on the way the patient is informed by the environment, judging whether it is safe, where they are safe, and how to react to stimuli (Truax, 1984). Unlike vision, auditory perception is permanent, since the patient cannot choose to close their ears as they can their eyes, therefore, being exposed to the complex soundscapes of the ICU (Luetz et al., 2019). Recent research indicates that changes in sound levels can have a more disruptive effect on patient well-being than continuous sound, in particular, differences between lower and higher noise pressure levels incorporating medical devices and alarms (Jaiswal et al., 2017).

Against this backdrop, the use of organized sound in the form of musical listening interventions has been associated with an increase in well-being and connectedness to the ‘real’ world outside of the ICU for both patients and next of kin (Almerud and Petersson, 2003; Chou and Vieira, 2020). A recent systematic review of RCTs and meta-analysis focused on the stress response of ICU patients and included 25 studies (Erbay Dalli et al., 2022). Music interventions were found to reduce anxiety levels concerning psychological stress. However, regarding physiological stress, it was found that music only had an effect on systolic blood pressure level (pressure in the arteries when the heart beats, i.e., the first number when measuring blood pressure) and no effect on diastolic blood pressure (pressure in the arteries when the heart rests between beats, i.e., the second number when measuring blood pressure), respiration rate, or heart rate. Multiple music sessions were reported to be more effective than a single session, and all but one delivered pre-recorded music for listening.

A Cochrane review that included 14 predominantly music-listening RCTs (805 participants) found the music intervention to have a positive effect on hemodynamics (how blood flows through the vessels), when compared to care as usual (Bradt and Dileo, 2014). Listening to music was found to consistently reduce respiratory rate and systolic blood pressure, suggesting a relaxation response. Furthermore, a reduced sedative and analgesic intake for patients was found in the music intervention group. The review concludes that music listening may have a beneficial effect on anxiety in mechanically ventilated patients.

Sufficient sleep as well as moments of calm are crucial for a good recovery (Luetz et al., 2019; Sayilan et al., 2020). As the ICU does not provide a calm and natural environment, sleep is often affected and associated with anxiety (Sayilan et al., 2020). Interventions aiming at alleviating sleep deprivation and anxiety often focus on pharmacological interventions, artificially sedating the patient in order to implement a sleep pattern (Hardin, 2009; Sayilan et al., 2020). While sleep can be induced, eliminating anxiety is a much tougher challenge (Hardin, 2009). Anxiety is seen as a key component of lack of sleep and the uncertainty of the patient’s medical outcome (Hardin, 2009) – music therapy can be seen as a non-pharmacological alternative.

Music-based therapies and interventions can profoundly improve patients’ anxiety and sleep quality. Receptive and interactive music therapy in the ICU, even though still lacking an evidence-based foundation in the ICU, has been promising in evoking empathy and resonance, strengthening positive emotions, relieving stress, psychological distorting, and tension, as well as evoking memories, physical and cognitive engagement (Browning et al., 2020; Erbay Dalli et al., 2022; Messika et al., 2019) across the life span.

Due to the nature of the ICU, an interactive music-based therapeutic intervention may prove difficult to implement, as patients are often immobile, non-responsive or sedated (Zimmerman et al., 2013), and so less able to access the instruments, respond with their playing to that of the therapist, and track musical interactions as coherent musical dialogue. Against this backdrop, receptive music therapy can be administered either live, by inviting Music Therapists to the ICU to play for the patients, or allowing Music Therapists to administer listening to music through headphones. Both methods must be patient-informed, as playing random music either through headphones or live can irritate or even aggravate the patient and associated pain and anxiety sensations (O’Callaghan, 1996). In cases where the patient is non-responsive, next of kin serve as a point of reference as to their preferred music. A recent systematic review and meta-analysis has shown that music interventions can have a reducing effect on anxiety and stress, as well as promote better sleep (Erbay Dalli et al., 2022). Their elegant study, however, even though defining music therapy, has not focused on music-based therapeutic interventions as provided by trained Music Therapists, and so does not help inform on what benefits there might be from their skill sets in the administration of music for ICU patients.

The aims of this scoping review are to identify any (interactive) music-based interventions that have been explored specifically in ICU settings, their intervention types, aims and outcomes, in order to inform clinical music therapy and research services at any hospital with an ICU. We aimed to identify how music-based interventions have been used in this setting, whether delivered by Music Therapists, and their potential benefits.

Method

Search strategy

The review was undertaken by two researchers (AS and ACJ). Searches were conducted, duplicates were removed, and the remaining papers were imported into Rayyan (an artificial intelligence [AI]-powered tool for systematic reviews). Initial screening of titles and abstracts was conducted, and any conflicts were resolved between the two reviewers – as was the case for the full-text reviews (See Figure 1 in appendices for flow diagram). Only articles published or available in English were included.

Figure 1.

Flow diagram for the review.

Using the search terms: ‘Music AND ICU AND Adult, ’ PubMed revealed 116 records, Web of Science – 27 records, CINAHL – 27 records, published between 1980 and 2022. This search string was used on 25 January 2022 and then again for PubMed on 24 June 2022 and brought up any research using music or music therapy in the intensive care setting. The Population, Intervention, Context, and Outcome (PICO) is shown in Table 1.

Table 1.

Population, Intervention, Context, Outcome (PICO) criteria for the review.

PICO	Inclusion Criteria	Exclusion Criteria
Population	Participants in the identified literature must be: (1) On an ICU (2) > 18 years (no upper age limit)	Literature will be excluded where participants are: (1) Not in ICU (2) < 18 years of age
Intervention	(1) Music listening, including passive or receptive auditory interventions, live or recorded music (2) Active music playing, including improvising or structured exercises (3) Group or individual	(1) metronome therapy (just pulses, no music) (2) vibroacoustic therapy
Context	Any ICU in a hospital setting	(1) Lab-based (2) Not in an ICU (e.g. home-based and day centre)
Outcome	Any outcome, including psychological, physiological, brain data, vital signs	Papers that did not report outcomes, for example, where feasibility data only are reported
Other – publication type and characteristics	Any accessible, published document reporting on the stated intervention, whether peer-reviewed or not Published/accessible in the English language	Conference abstracts Not published/accessible in the English language

Inclusion and exclusion criteria

Title and abstract screening excluded protocols, paediatric populations, literature reviews, thesis, foreign language and papers unavailable through our specified databases, inter-library loans, and ResearchGate.

Inclusion and data extraction

Out of 139 titles, 45 full texts were included (Table 1), with a total of 3,441 participants. One feasibility study was included, as it reported treatment-related effects on delirium (Johnson et al., 2018). The following data were extracted from the included articles: study design, aims and sample size, setting and country, whether delivery of music intervention by a Music Therapist was reported, intervention type, equipment, frequency and duration, outcome measures, and findings.

Results

Blood pressure

Blood pressure was reported as significantly reduced in seven studies (Almerud and Petersson, 2003; Barnason et al., 1995 Froutan et al., 2020; Johnson et al., 2018; Messika et al., 2019; Miller et al., 2021; Updike, 1990). A further study comparing music listening with an equivalent 30-min rest period observed significant differences at the end of the intervention (Wong et al., 2001). Six studies found no differences in blood pressure (Beaulieu-Boire et al., 2013; Dijkstra, Gamel, Van Der Bijl, and Bots Kesecioglu, 2010; Mateu-Capell et al., 2019; Phipps et al., 2009; White, 1999; Aktaş and Karabulut, 2016); one study used live harp music rather than pre-recorded playlists and found no significant effect on blood pressure (Chiasson et al., 2013).

Anxiety

Of 18 studies measuring changes in anxiety, 4 found no significant change or difference between groups (Ames et al., 2017; Barnason et al., 1995; Cooke et al., 2010; Messika et al., 2019) – with 1 focusing on discomfort during bed turning (Cooke et al., 2010). Thirteen studies found significantly lower anxiety levels (Chlan et al., 2013a, 2013b; Golino et al., 2019; Gullick and Kwan, 2015; Lee et al., 2017a, 2017b; Bolwerk, 1990; Miller et al., 2021; Phipps et al., 2009; Singh et al., 2009; Updike, 1990; White, 1999; Wong et al., 2001). Two studies did not report treatment effects on anxiety scores (Chlan et al., 2007; Pagnucci et al., 2019). One further study found music listening to improve relaxation during bedrest following open-heart surgery (Nilsson, 2008).

Heart rate

Eleven studies reported lowered heart rate (Chan, 2007; Chlan, 1998; Elay and Ozkaya, 2020; Froutan et al., 2020; Golino et al., 2019; Guzzetta, 1989; Johnson et al., 2018; Lee et al., 2017a; Phipps et al., 2009; Updike, 1990; Aktaş and Karabulut, 2016). Four studies found no effect (Barnason et al., 1995; Dijkstra et al., 2010; Mateu-Capell et al., 2019), including a live harp music study (Chiasson et al., 2013).

Respiratory rate

No significant reduction in respiratory rate was reported in eight studies (Beaulieu-Boire et al., 2013; Chan et al., 2009 Chiasson et al., 2013; Dijkstra et al., 2010; Mateu-Capell et al., 2019; Messika et al., 2019; Su et al., 2013; Wong et al., 2001). Seven studies found significant changes or more than in the control group (Chan, 2007; Chlan, 1998; Ecer and Saritaş, 2019; Elay and Ozkaya, 2020; Froutan et al., 2020; Golino et al., 2019; White, 1999).

Sleep

Sleep benefits were found in four studies (Hansen et al., 2018; Hu et al., 2015; Kim et al., 2020; Su et al., 2013).

Delirium

Prevention or reduction in delirium was reported in two studies (Browning et al., 2020; Johnson et al., 2018).

Pain

Of 13 studies measuring the effect on pain, a significantly lower pain score or reduced pain was found in 10 studies: (Ames et al., 2017; Chan, 2007; Chiasson et al., 2013; Elay and Ozkaya, 2020; Golino et al., 2019; Miller et al., 2021; Ajorpaz et al., 2014; Ryu et al., 2012 Shertzer and Keck, 2001; Aktaş and Karabulut, 2016); 2 studies found no effect (Mateu-Capell et al., 2019; Phipps et al., 2009) and 1 found a reduction in analgesic use (Beaulieu-Boire et al., 2013).

Deconditioning and muscle strength

One study comparing exercise performance with and without music found significant improvements in the music group in upper and lower limb muscle groups, including grip strength (Liang et al., 2022). A Music Therapist was consulted on the music selection but did not deliver the intervention.

Delivery of music

All data synthesis including listening duration, equipment and Music Therapist involvement can be seen in Table 2 (Appendices). Some of the included literature reported input from a Music Therapist in delivering live music for guided imagery or accompanied song, describing how the therapist was able to match tempo, volume and intensity to the patients’ heart and respiratory rate (Golino et al., 2019). There was input from Music Therapists reported on design aspects of interventions, such as music selection (Gullick and Kwan, 2015; Liang et al., 2022; Mateu-Capell et al., 2019), composition (Messika et al., 2019), or study design, but little on intervention delivery (Golino et al., 2019; Hansen et al., 2018).

Table 2.

All included full texts.

	Title/Authors	Aims/design/sample	Setting/Music Therapist (MT) Not Stated (NS)	Intervention/equipment	Frequency and duration	Outcome measure/s	Results
1	A musical intervention for respiratory comfort during non-invasive ventilation in the ICU. Messika et al., 2019	Determine the effect of a musical intervention on respiratory discomfort during NIV in patients with acute respiratory failure (ARF)/three-armed RCT/n = 113	Adult ICU. France/MT	(1) Music listening/BOSE AE2 headphones, Samsung Galaxy tablet. (2) Noise-cancelling headphones-no music. (3) Usual care	30 min listening to X 1	Primary endpoint = change in respiratory discomfort between the time before initiation and after 30 min of the first NIV session after randomization. Secondary endpoints included: (1) evolution of respiratory discomfort; (2) changes in respiratory parameters; (3) changes in cardiovascular parameters (heart rate and arterial pressure); (4) percentage of patients requiring endotracheal intubation at the end of a NIV session; (5) percentage of patients requiring physical restraint or sedative/anxiolytic treatments during NIV sessions and ICU stay; (6) anxiety/depression and health-related quality of life (HR QoL) measured by HADS and SF-36 scores at baseline and at day 90; (7) peri-traumatic stress measured with the PDI immediately on ICU discharge and at day 90; (8) patients’ overall assessment of NIV (discomfort, satisfaction, and trauma) at ICU discharge and day 90	No significant reduction in respiratory discomfort. A significant decrease in systolic and mean arterial blood pressure (BP) at the end of the first NIV session was associated with the musical intervention, along with a significant reduction in the PDI on ICU discharge. No significant difference between groups in anxiety (HADS)
2	A randomized controlled trial of the effects of listening to non-commercial music on quality of nocturnal sleep and relaxation indices in patients in medical intensive care unit. Su et al, 2013.	Examine the effects of non-commercial music on quality of sleep and relaxation indices, including heart rate, mean arterial pressure, and respiratory rate in patients in intensive care unit/two-armed RCT. N = 28	Large teaching hospital ICU. Taiwan/NS	Interventions consisted of four pieces of sedating piano music composed by two of the authors. The musical pieces all had similar characteristics, including a smooth rhythm to achieve a relaxing effect, a tempo of 60–80 beats/min, minor tonalities, smooth melody lines (Figure 2) and no dramatic changes in volume and rhythm. a Sony (CFD-S07CP) CD player	45 min. One night.	Continuous measurements of heart rate, mean arterial blood pressure, and respiratory rate. Sleep Quality data using PSG (Using EEG) for first 2 h of sleep, and questionnaire, Verran and Synder-Halpern Sleep Scale (VSH sleep scale)	Subjective sleep data showed music listening was beneficial for stages N2 and N3 sleep, and heart rate. Measurement of PSG may not have been long enough. NO SIGNIFICANT CHANGE IN RR.
3	Comparing effects between music intervention and aromatherapy on anxiety of patients undergoing mechanical ventilation in the intensive care unit: a randomized controlled trial. Lee et al., 2017a (1).	To test the effects of music intervention and aromatherapy on reducing anxiety for intensive care unit (ICU) patients undergoing mechanical ventilation. Three-armed RCT. N = 132	Academic medical centre. Taiwan/NS	Music listening. Western classical music (e.g. Erik Satie’s Trios Gymnopedies, Mozart’s Piano Concerto no. 26), Chinese classical music (e.g. bamboo flute, rain, and tears), music of natural sounds (e.g. Sylvan Spa, Relax Your Mood, and religious music, including Buddhist and Christian. Music characteristics: slow beat (60–80 bpm) corresponding to normal heart rate. MP3 player at bedside and headphones. Participants lay on bed, lights low, room temperature 26°C. Research nurse assisted with volume adjustment.	A 30 min session for each participant to listen to music was provided in the Music group between 4:00 and 4:30 pm	Stage-Trait Anxiety Inventory (C-STAI) and the Visual Analogue Scale for Anxiety (VAS-A). Heart rate, breathing rate, and blood pressure were measured every 10 min from baseline to the 30 min follow-up	The Music group had significantly better post-test VAS-A and C-STAI scores, and had lower heart rate and blood pressure than the control group
4	Does Music Influence Stress in Mechanically Ventilated Patients? Chlan et al., 2013 (1).	To determine the effects of patient-initiated music listening on anxiety during mechanical ventilatory support/three-armed RCT/N = 65	12 ICUs in five hospitals in the US/NS	(1) Experimental patient-directed music intervention (PDM), (2) active control condition of noise-cancelling headphones only, or (3) control group of usual ICU care	Patients could self-deliver music at will. Duration and frequency were not reported.	Cortisol-daily 24 h urine collections	No significant difference in urinary free cortisol (UFC) over time by study group. Subjects did not experience significantly less stress than those subjects in the control groups. The PDM group demonstrated less extreme values over time compared to the headphones group. Anxiety not reported in results.
5	Effects of earplugs and eye masks combined with relaxing music on sleep, melatonin and cortisol levels in ICU patients: a randomized controlled trial. Hu et al., 2015.	Determine the effects of using earplugs and eye masks with relaxing background music on sleep, melatonin and cortisol levels in ICU patients 2-armed RCT/N = 50	Cardiac surgical ICU. China/NS	Ear plugs and eye masks with relaxing music	Thirty minutes nighttime listening each night on the ward	Cortisol, melatonin. Richards-Campbell sleep questionnaire (RCSQ)	Sleep quality in the intervention group was significantly higher. Patients’ perceptions of nighttime noise were significantly lower in the experimental group than in the control group (p < .05). Repeated measures analysis of variance showed the 6-SMT levels of the first and second postoperative nights were significantly lower than those of the night before surgery (F = 7.53, p = .01). The cortisol levels of the first and second postoperative nights were significantly higher than those of the night before surgery (F = 88.63, p = .00).
6	Effects of music during daytime rest in the intensive care unit. Hansen et al., 2018	Investigate the effects of music on self-reported quality of sleep during daytime rest among patients in the intensive care unit/two-armed RCT/N = 37	Two-level three ICUs. Denmark/NS	The music was played by a loudspeaker, placed in the ceiling above the participants’ beds and is a newly developed method where the sound is centred around the patient (Soundfocus.dk). The music, MusiCure, was especially composed and developed for critically ill patients in ICU by Niels Eje (Musi-Cure.com). The intervention consisted of four pieces of soothing music, in terms of soft wind, bird twitter, ocean sound and music instruments.	30 min music listening during daytime-mean length of stay was 3 days.	Danish version of RCSQ was used to assess self-reported sleep quality	Participants who listened to music for 30 min during daytime rest had an improved self-reported quality of sleep
7	Effects of Music Intervention on State Anxiety and Physiological Indices in Patients Undergoing Mechanical Ventilation in the Intensive Care Unit: A randomized Controlled Trial. Lee et al., 2017b (2).	To explore the anxiety-reducing effects of music intervention in a sample of ICU patients on mechanical ventilation. All the participants and the research nurses were blinded to the randomized group assignment/two-armed RCT/N = 85	ICU in Academic medical centre, Taiwan/NS	Music group participants selected from: Western classical music (e.g. Erik Satie’s Trios Gymnopedies, Mozart’s Piano Concerto no. 26), Chinese classical music (e.g. bamboo flute, rain, and tears), music of natural sounds, Relax Your Mood, or religious music (Buddhist or Christian). All 60–80 beats per min. Control group participants had a 30 min rest between 4:00 and 4:30 pm while wearing headphones, while patients in the intervention group were listening to music	A single, 30 min music-listening session for each participant in the music group	Visual Analogue Scale for Anxiety (VAS-A) and the Chinese version of the C-STAI. e. Blood pressure and heart rate were automatically monitored and recorded (HP/Philips/Agilent M1205A)	All of our outcomes showed that a single 30 min session of music reduced anxiety for patients undergoing mechanical ventilation in an ICU
8	Effects of music on patients undergoing a C-clamp procedure after percutaneous coronary interventions: A randomized controlled trial. Chan, 2007.	To determine the effect of music on physiologic parameters and a psychologic parameter, namely, the level of pain in patients undergoing the application of a C-clamp after PCI/two-armed RCT/N = 66	Three ICUs in Hong Kong/NS	Three types of music were used, including Chinese classical music (e.g. bamboo flute), religious music (e.g. Buddha Bar IV–Tibet), and Western classical music (e.g. Mozart Piano Concerto no. 26). MP3 and headphones	45 min of music therapy or 45 min of an uninterrupted rest period During C-Clamp procedure	Physiologic parameters were blood pressure, heart rate, respiratory rate, and oxygen saturation. Psychologic parameters were measured using the University of California at Los Angeles universal pain score	In the experimental group there were statistically significant reductions in heart rate (p = .001), respiratory rate (p = .001), and oxygen saturation (p = .001), and a lower pain score (p = .001) than in the control group
9	Effects of music therapy on anxiety in ventilator-dependent patients. Wong et al., 2001	Assess the effectiveness of music therapy in decreasing anxiety/two-armed RCT/N = 20	ICU, Hong Kong/NS	Music listening. Audiocassette players and headphones. 1. Seven cassettes of relaxing music, including both Chinese music (Chinese folk song, music played by Chinese instrument, Chinese music played by Western instrument, Buddhist music) and Western music (Western classic, Western movie music, and piano music).	Either 30 min of uninterrupted rest and then 30 min of music therapy or the music first and then rest period, 30 min	Short version (6 items) of the Spielberger State-Trait Anxiety Inventory (STAI)	Findings indicated that music therapy was more effective in decreasing state anxiety than was an uninterrupted rest period (p < .01). Blood pressure and respiratory rate showed no significant differences in the two conditions over time. However, significant differences were observed at the end of the intervention (after 30 min) between the two conditions, with music therapy being superior to the rest period. The interval between the two conditions ranged from 6 to 9.5 h
10	Effects of patient-directed music interventions on anxiety and sedative exposure in critically ill patients receiving mechanical ventilatory support: a randomized clinical trial. Chlan et al., 2013 (2)	To evaluate if a patient-directed music (PDM) intervention could reduce anxiety and sedative exposure in ICU patients receiving mechanical ventilation/three-armed RCT/N = 373	12 ICUs at five hospitals in Minneapolis, US/NS	(1) PDM intervention, (2) active control with noise-cancelling headphones (NCH), or (3) usual care	Patients remained on protocol for a mean of 5.7 days.	Visual analogue scale for anxiety-daily, while ventilated. Sedative exposure was determined for all patients who received any of eight commonly administered sedative and analgesic medications in the ICU (midazolam, lorazepam, propofol, dexmedetomidine, morphine, fentanyl, hydromorphone, haloperidol) 24 h prior to enrolment and each day during the study. A data-logger system on the headphones captured each PDM session and total daily music-listening time	The PDM intervention decreased anxiety and sedative exposure over time more effectively than usual care or NCH. By the fifth study day, anxiety was reduced by 36.5%. PDM patients listened for a mean (SD) of 79.8 (126) min/day (median, 12 (range, 0–796) min/day). NCH patients wore the headphones for a mean (SD) of 34.0 (89.6) min/day (median, 0(range, 0–916) min/day).
11	Effects of relaxation and music therapy on patients in a coronary care unit with presumptive acute myocardial infarction. Guzzetta, 1989.	To determine whether relaxation and music therapy were effective in reducing stress in patients in a coronary care unit/three-armed RCT/N = 80	CCUs in three hospitals. US/NS	Music therapy (music listening) for 20 min Twice a day for a total of three sessions over a 2-day period. Three different music selection tapes of patient preferred music were screened and evaluated for their soothing and stress-reducing effects. A single taped music selection, 20 min in length, was selected to represent each of the three types of music: soothing classical music, soothing popular music and non-traditional music (defined as compositions having no vocalization or metre, periods of silence, and an asymmetric rhythm)	Music therapy for 20 min Twice a day for a total of three sessions over a 2-day period.	Psychophysiological stress was measured by using apical heart rates (auscultated for 1 full minute); peripheral temperatures (measured by taping calibrated mercury bulb finger thermistors) (Stress Temp Conscious Living, Inc’, Drain, Ore’) to the patient’s index finger; the incidence of cardiovascular complications occurring during the study (measured by a count of cardiac arrests, ventricular tachycardia requiring electrical cardioversion, congestive heart failure, pericarditis, pulmonary oedema, cardiogenic shock, temporary cardiac pacing, persistent cardiac pain, or unstable blood pressure requiring intravenous drug therapy); and via questions (e.g. asking how helpful participants thought the sessions were)	Relaxation and music therapy had a significant impact on apical heart rate, which was also lower after the third session than after the first two session, showing cumulative effect. Raising the finger temperature was more effective in the music therapy group than in the relaxation group. The incidence of cardiac complications was found to be lower in the intervention groups, and most intervention subjects believed that such therapy was helpful. Relaxation and music therapy are effective modalities to reduce stress in these Patients.
12	Effects of relaxing music on state anxiety in myocardial infarction patients. Bolwerk, 1990	To determine whether relaxing music would reduce anxiety/two-armed RCT/18 (control) 17 (music) completed (n = 35)	Five community hospitals. US/NS	Three compositions in each session: Largo (Bach), Largo (Beethoven), L’apres Midi D’une Faun (Debussy).	Three X 22 min sessions of listening on 3 consecutive days	State-Trait Anxiety Inventory (STAI) per-/post-	Significant decrease in anxiety for music group. HR and RR significantly lower 1 h post-intervention than controls.
13	Effects of relaxing music on cardiac autonomic balance and anxiety after acute myocardial infarction. White, 1999	Compare effects of relaxing music, quiet uninterrupted rest and treatment as usual on anxiety levels, physiological indicators of cardiac autonomic function/three-armed RCT/n = 45 (15 per-group) – AMI within 72 h	2-US hospitals. US/NS	Researcher selected classical music	Single session: 20 min mini-disc music listening or same of uninterrupted rest or usual activities	STAI, HR, HRV, respiratory rate, Systolic blood pressure, rate pressure product. Measures taken immediately after, then 1 and 2 h after	Music group had greater reductions in HR, respiratory rate, myocardial oxygen demand, state anxiety scores. No differences between groups in blood pressure.
14	Effectiveness of a music therapy intervention on relaxation and anxiety for patients receiving ventilatory assistance. Chlan, 1998	Test the effects of music therapy (music listening) on relaxation and anxiety reduction for patients receiving ventilatory assistance. two-armed RCT – receiving ventilation for average of 7.4 days at recruitment/n = 50 (completed)	4 ICUs. US/NS	Available tape collection consisted of a variety of non-lyric-containing music, with 60–80 beats per min, deemed to be of a relaxing nature. 22,23 Types of music represented in the collection included classical, New Age, country western, religious, and easy listening. Subjects were instructed to concentrate on the flow of the music with their eyes closed	Single, 30 min listening	State anxiety (pre-test and post-test), heart rate, and respiratory rate obtained every 5 min for 30 min	Music-listening group had significantly less anxiety post-test (10.1) than those subjects in the control group (16.2), and heart rate and respiratory rate decreased over time for them, too, compared with the control group subjects.
15	Effect of sleep-inducing music on sleep in persons with percutaneous transluminal coronary angiography in the cardiac care unit. Ryu et al., 2012	To compare the effect of earplug-delivered sleep-inducing music on sleep in persons with percutaneous transluminal coronary angiography/two-armed RCT/n = 58 (completed)	1 ICU. South Korea/NS	Sleep-inducing music listening included Nature Sounds (2 min and 8 s), Delta Wave Control Music (5 min 21 s), Goldberg Variations BWV. 988 (27 min and 3 s), Nature Sounds (4 min and 57 s), Goldberg Variations BWV. 988 (11 min and 1 s) and Nature Sounds (2 min and 25 s)	The MP3 music was supplied through earphones and played to the participants from 10:00 to 10:53 pm. Control wore earplugs	Quality of sleep and quantity of sleep questionnaires	The experimental group reported that the sleeping quantity and quality were significantly higher than the control group (t = 3.181, p = .002, t = 5.269, p < .001, respectively)
16	Impact of an Active music Therapy Intervention on intensive care Patients. Golino et al., 2019)	Examine the effect of an active music therapy intervention on physiological parameters and self-reported pain and anxiety levels/pre–postwithin-subject, single group/n = 52	ICU. US/MT	A relaxation/guided imagery experience with live music or a ‘song choice’ experience with live music (participant discusses song lyrics with the Music Therapist). The tempo of the music or beats per minute initially matched with the patient’s heart rate or respiratory rate; then, the tempo (as well as the volume and intensity) is gradually decreased in an attempt to synchronize the heart rate (or respiratory rate) with the music.	Single, 30-min session	HR, RR, and oxygen saturation, pain and anxiety (self-report)	Significant decreases (all p < .001) were found in respiratory rate (mean difference, 3.7 (95% CI, 2.6–4.7) breaths per minute), heart rate (5.9 (4.0–7.8) beats per minute), and self-reported pain (1.2 (0.8–1.6) points) and anxiety levels (2.7 (2.2–3.3) points). No significant change in oxygen saturation level was observed
17	Impact of therapeutic music listening on intensive care unit patients: A pilot study. Browning et al., 2020	To explore the association between therapeutic music listening as a nursing intervention for MV patients in the ICU and the proportion of time the pilot study patients were considered to have delirium/two-armed RCT/n = 6	34-bed MICU. US/MT	Two of the three MUSIC group patients selected bluegrass, country, and gospel genres, which are popular in the region where this pilot study was performed. Families assisted in music selection.	Music was delivered using free streaming applications on the computers. One hour increments, twice daily from 10:00 am to 11:00 am and from 9:00 pm until 10:00 pm	Confusion Assessment Method for the ICU (CAM-ICU) – per cent of time the patient was CAM +	MUSIC group (n = 3) experienced less proportion of time with documented ICU delirium (33%) than the control group (n = 3) did (67%). Therapeutic music listening can be applied as a nursing intervention, and it does not require intensive time, financial, or resource allocation, nor does it necessitate members of the health care team learning new skills
18	Influence of music on the stress response in patients receiving mechanical ventilatory support: a pilot study. Chlan et al., 2007	To explore the influence of music on serum biomarkers of the stress response in patients receiving ventilatory support/two-armed RCT/n = 10	One ICU. US/NS	Researcher selected, offering a range of genres on cassette (classical, new age, easy listening, country) – all selected classical	60 min of music, listening with headphones, or of resting quietly	Levels of corticotropin, cortisol, epinephrine, and norepinephrine were measured four times during the 60 min	The levels of the four biomarkers of the stress response did not differ significantly between patients who listened to music and patients who rested quietly, though the levels of corticotropin and cortisol showed interesting trends. In future studies, confounding factors such as endotracheal suctioning and administration of medications that influence the stress response should be controlled for
19	Investigating the physiological responses of patients listening to music in the intensive care unit. Chan et al, 2009	To determine whether definable subtypes exist within a cohort of patients listening to music with regard to their physiological patterns and to compare whether associated factors vary between subjects in groups with different profiles/single arm, repeated measures/n = 101	Three acute care hospitals in Hong Kong/NS	Researcher selected (based on previous local studies) – relaxing music defined primarily as being low-pitched, having a simple and direct musical rhythm and having a tempo of approximately 60–80 beats per minute. Plus: Chinese classical music, religious music, Western classical music and jazz	30 min, lights off, curtains closed, with headphones and instructed to concentrate on the sound	Heart rate, respiratory rate, SBP, DBP	No significant differences in RR and HR. A cluster analysis yielded two clusters. Patients in cluster 1 typically experienced relatively low therapeutic effects from listening to music. There were more males, of a younger age but more educated and employed than patients in cluster 2, and they represented 41.6% of the total respondents. Cluster 2 comprised almost 58.4% of this study sample and they reported high therapeutic effects of music. There were more females in this group as well as more older people and they mainly used a ventilator in the intensive care unit
20	Music as a therapeutic intervention on an inpatient neuroscience unit. Phipps et al., 2009	To determine the effect of a music intervention (MI) on physiological parameters, pain and mood states in the neuroscience patient/two-armed RCT/n = 53	One acute neuroscience care facility. US/NS	Researcher selected: ranging from light classical to relaxing music	30 min listening	Heart rate, blood pressure, respiration, peripheral skin temperature, pain perception, mood	Significant reductions in heart rate, respirations, perceived anxiety, depression and total mood score in subjects who received UC plus MI compared to UC. No differences in blood pressure. No significant differences in pain, BP, or skin temperature
21	Music and biological stress dampening in mechanically ventilated patients at the intensive care unit ward – a prospective interventional randomized crossover trial. Beaulieu-Boire et al., 2013	To evaluate the impact of slow-tempo music-listening periods in mechanically ventilated intensive care unit patients/RCT crossover (MP3 music or sham – headphones and no music)/n = 49	ICU in tertiary care hospital, Canada/NS	Researcher selected music based on previous research = 10 random and consecutive pieces: Bach (Air from Suite for Orchestra No. 3), Bach (Air for G string), Beethoven (Moonlight Sonata), Beethoven (Pathetic Sonata), Brahms (Lullaby), Chopin (Nocturne in G), Debussy (Clair de Lune), Pachelbel (Canon in D), St-Saens (The Swan), Tchaïkovsky (Panorama from Sleeping Beauty)	1 h X 2 per-day over 3 days	Pre-/post-arterial blood pressure, heart rate, respiratory rhythm. Blood tests for reference inflammation markers (i.e. IL-6, C-reactive protein (CRP)) and major hormonal stress sensors of the HPA axis (prolactin, cortisol, ACTH), were measured every morning before and after the listening period (but not during the evening listening period), as well as for leptin and MET-enkephalin in a post-analysis subset of biological responders to music	Music listening did not significantly alter overall VS over placebo control. A trend towards a reduction in narcotics was observed with music listening (fentanyl). Blood cortisol decreased after music listening (not the placebo). Blood prolactin decreased after music listening. Average leptin as well as MET-enkephalin levels did not change with music listening. Music listening tends to decrease pain-killer consumption. Music listening dampens stress hormone release.
22	Music and the PACU environment. Shertzer and Keck, 2001	Investigate the effect of soothing music and lowering noise levels on the pain experience of patients during their PACU stay/two-group quasi-experimental comparing music and controlled noise with usual environment (no randomization)/n = 97 (completed, 56 in experimental)	Large, veteran administration hospital, mid-west US. Same day surgery (except open-heart)/NS	Music listening. Researcher selected cassette: (1) Ocean music – continuous soothing ocean sounds, without structure. (2) Mozart, continuously soothing, melody is similar throughout	Cassette was 1 h long. Played as soon as the patient was admitted.	11-point Numerical Rating Scale. Pain intensity data were collected at 3-time intervals: on admission to the PACU, 30 min after admission, and at discharge. The 30 min time period was chosen because patients typically remained in the PACU for 1 h, and 30 min would be the midpoint of their stay	Pain ratings were lower for the experimental group than for the control group for both 30-min and discharge assessments but not statistically significant. The majority of patients in both groups reported no pain on admission, 30 min after arrival, and on discharge
23	Music intervention to prevent delirium among older patients admitted to a trauma intensive care unit and a trauma orthopaedic unit. Johnson et al., 2018	Evaluate music listening for delirium prevention among patients admitted to a Trauma Intensive Care and Trauma Orthopaedic Unit/Feasibility two-arm RCT/n = 40 comparing music listening with usual care	Trauma ICU. US/NS	Music listening: included: (a) simple repetitive rhythm, (b) self-selection and (c) slow tempo (60–80 bpm), designed to alter physiological responses including SBP, HR, and RR	60 min, two-times per-day (2 pm and 8 pm), over 3 days. Headphones and iPod shuffle	Delirium measured by the CAM-ICU. Secondary outcomes included physiologic signs measured by SBP, HR, and RR. Physiologic signs were collected in the clinical setting on admission to the TICU and TOU during the initial assessment of the patient, and every 4 h over a 3-day period	There were statistically significant differences in Systolic blood pressure with the ML group compared to the UC group. A multiple post hoc analysis revealed statistically significant differences between D2 T1 heart rate (HR) post-ML and D1T1HR pre-ML (p = .010); D2T1HR post-ML from D1T1HR post-ML (p = .005); and D2T2HR post-ML and D1T1HR post-ML (p = .039). All screened negative for delirium.
24	Music therapy – a complementary treatment for mechanically ventilated intensive care patients. Almerud and Petersson, 2003.	Determine if music therapy relaxes patients temporarily on a respirator and after completion of respirator treatment. Investigate those patients’ experiences of the music therapy/twenty patients were included using consecutive selection. Physically stable needing mechanical ventilation. Music with headphones – headphones without music	ICU, Sweden/NS	Music listening. Researcher selected (based on previous studies) Beethoven (Suite 1 from Sonata in C-sharp minor-Moonlight sonata) Pachebel (Canon) Debussy (Claire de lune), Bach (Air from suite for orchestra no. 3) Vivaldi Spring (Largo) Seymer Solöga (Suneye), Marcello Concerto for oboe in D minor: Adagio	Cassette of classical music played for 30 min in conjunction with night sleep on two occasions	Each patient received sedatives during respirator treatment. Each patient received sedatives during respirator treatment recorded at 5 min intervals during the period of Intervention. Interview questions concerning recollections and experiences of respirator treatment and music therapy	Patients remembered very little of their time in ICU. Quantitative data showed a significant fall in systolic and diastolic blood pressure during the music therapy session and a corresponding rise after cessation of treatment. All changes were found to be statistically significant.
25	Music therapy results for ICU patients. Updike, 1990	Examine the effects of specifically chosen music on physiological and emotional state in critically ill patients/pre-/post-convenience sample/n = 20	ICU, US/NS	Replicates the Bonny method (1978). Patients chose one of eight classical or contemporary listening programmes: 21 classical titles including Pachebel (Canon), Debussy (Claire de Lune), Popular, jazz, country titles (10), including Cat Stevens (Morning has broken) and Willies Nelson (Moonlight in Vermont)	Cassette, 30 min listening	Heart rate, mean arterial pressure, blood pressure, double product index, electrocardiogram, analgesic dosage. Non-directive, open-ended questionnaire (emotional condition rating scale)	Significant reduction in systolic blood pressure, anxiety, depression. EKG = significant rhythm pattern change. Significant change in mood to more positive and other physiological measures
26	Music to reduce stress in hospitalized patients. Miller et al., 2021	To determine the impact of music on the physiologic and psychological stress/single-armed pre-/post-test/n = 50 (completed)	ICU, Northern California, US/NS, but important	Easy listening, classic rock, inspiring instruments, classic country, and jazz. It also allows users to select tempo categories such as energy, relax, sleep, and wake (the ‘relax’ tempo setting was utilized for all participants)	30 min	Wong-Baker FACES pain rating scale, Face anxiety scale, Richmond agitation sedation scale (RASS), BP, HR, and respiratory rate. HR and respiratory rate collected at 5 min intervals for 30 min from a remote central monitor. BP was not taken during the session, only at the 30 min mark	Patients reported significantly lower pain and anxiety. Statistically significant reduction in pain and anxiety and diastolic blood pressure and RASS and HR reduction (pre/post) after the music intervention.
27	Patients’ perception of music versus ordinary sound in a post-anaesthesia care unit: a randomized crossover trial. Fredrikson et al., 2009	Test the hypothesis that patients will experience a higher degree of well-being if they listen to music compared to ordinary PACU sounds during their early postoperative care, (ii) to determine if there is a difference over time, and (iii) to evaluate the importance of the acoustic environment and whether patients prefer listening to music during their stay/single-blind crossover two-armed RCT/n = 44 (completed)	PACU, Sweden/NS	Group A (n = 23) experienced music – ordinary sound—music; Group B (n = 21) experienced ordinary sound – music – ordinary sound. Each period lasted 30 min. after each period the patients assessed their experience of the sound. Music used was MusiCure (MusiCure, 2009)	Thirty minute listing using an Ergonomic audio pillow, 65 cm × 30 cm (Wellness Music pillow, 2009).	After each period the patients assessed their experience of the sound. The questionnaire was study-specific and developed by the authors based on knowledge of the research field and professional experience of PACU car	Significant difference (p < .001) between groups in the proportions of patients reporting that the acoustic environment was of great importance for their well-being. most participants (n = 36 versus n = 8) noticed that they were exposed to different sounds during the PACU period. No differences between the groups regarding the distribution of patients’ sense of well-being (Mann–Whitney, p > .05) after each intervention period. ordinary sound – music – ordinary sound. Significant change in well-being in the OMO group (p > .01) after the second intervention (music)
28	Patient-directed music therapy reduces anxiety and sedation exposure in mechanically ventilated patients: a research critique. Gullick and Kwan, 2015	To determine whether self-directed, self-initiated music therapy, compared to either noise-cancelling headphones or usual care reduced self-reported anxiety or sedation use/three-armed RCT/n = 373	12 ICUs in 5 hospitals in the US/NS	Music listening. Participants reviewed a starter set of generic, relaxing tracks with the research nurse. Within 24 h of randomization, a Music Therapist then compiled a list of music preferences with the participant using the Music Assessment Tool (3-page, yes/no music preference instrument designed for use with mechanically ventilated patients).	Music = 2 X per-day; Noise-cancelling headphones = whenever they wished to block out noise or have quiet time. Usual care recipients received no study-directed music or NCH and were managed according to local unit protocols. Mean study duration of 5.7 days	Anxiety was measured using the VAS-A	The (Patient-directed music) PDM group listened to music for a mean of 79.8 min (median 12 min, range 0–796 min) per day. NCH were worn for a mean time of 34 min (median 0 min, range 0–916 min) per day. More PDM participants were extubated at the end of the study (p = .02). PDM participants recorded significantly lower VAS-A (anxiety) scores compared to usual care, and significantly larger decreases in sedation frequency compared to both NCH (p = .04) and usual care (p = .01)
29	Promoting nighttime sleep in the intensive care unit: Alternative strategies in nursing. Pagnucci et al., 2019	To identify if complementary interventions impacted on conscious intensive care patients’ perception of stress factors and quality of sleep/two-armed RCT/n = 74	ICU, central Italy/NS	(1) Each patient chose his/her favourite music from a digital music library or listening to relaxing sounds of nature. A digital MP3 player was used to reproduce the sounds, allowing each patient to listen through headphones. (2) Receptive musical sounds plus a feet and leg massage	30 min (from second night admitted). Listening for 7 h, from 11 pm to 7 am	Vital signs; utilized pain, anxiety, and agitation rating scales; and verbal feedback. ‘Stress Factors in Intensive Care Unit Questionnaire’ (SEDAICU), ‘Modified Richards-Campbell Sleep Questionnaire] (Modified RCSQ)	Receptive musical sounds and massage using aromatherapy improved the quality of patients’ sleep (t = 2.01, p = .047)
30	Sound isolation and music on the comfort of mechanically ventilated critical patients. Mateu-Capell et al., 2019	Assess the effect of sound isolation versus music on the comfort of mechanically ventilated patients/two-armed RCT crossover/Group A: sound isolation (1 h) followed by music (1 h); Group B: music (1 h) followed by sound isolation (1 h)/n = 82	ICU, Spain/MT approved	Music listening. Researcher selected: one piece of New Age (Zen) music: Reiki – The Light Touch from the Album Merlin’s Magic by Andreas Mock (60 min)	Single, 4 h session with four phases	The hypnotic level, the level of sedation and the alteration of behaviour because of pain was assessed using the bispectral index (BIS), the Ramsay scale and the Behavioural Pain Scale (BPS). Secondary outcome measures included systolic and diastolic blood pressures, heart, and respiratory rates and dose of sedatives	Decrease of 4–5 points out of 100 during sound isolation and music therapy was recorded in BIS in both groups; the differences were not statistically significant compared with baseline scores. No changes in secondary measures.
31	The effects of musical stimulation on the level of consciousness among patients with head trauma hospitalized in intensive care units: A randomized control trial. Yekefallah et al, 2021	To evaluate the effects of musical stimulation on level of consciousness (LOC) among patients with HT/two-armed RCT/54	ICUs in two hospitals in Iran/NS	‘Beach walk’ music from the ‘Top-Hits zum Entspannen’ album (Volume 1) – 60–80 bpm. MP3 and headphones	15 min Music listening X 1 daily for 7 consecutive days (MP3 and headphones) Control = headphones only	LOC, Richmond agitation scale, GCS	Significant between-group differences respecting the post-test mean score of LOC in the third, fourth, fifth, sixth, and seventh days of the study intervention (p < .05), the post-test mean score of LOC in the intervention group significantly increased in the intervention group (p < .0001), while it did not significantly change in the control group (p > .05). Musical stimulation is effective in significantly improving LOC among hospitalized patients with HT.
32	The effects of music interventions on anxiety in the patient after coronary artery bypass grafting. Barnason et al., 1995	Examine the influence during the early postoperative period of selected nursing interventions on mood and anxiety of patients undergoing heart surgery/three-armed RCT/n = 96	ICU, US/NS	Music listening. Cassette and headphones: subjects were able to select from Country Western Instrumental, Fresh Aire by Mannheim Steamroller, Winter Into Spring by George Winston, and Prelude and Comfort Zone, both by Steven Halpern. 30 min videocassette was used, which played soft instrumental music combined with visual imaging on the television screen of different scenes to accompany the music. The subjects had a choice of one of two videocassettes by Steven Halpern: Summer Wind or Crystal Suite (1) music therapy, (2) music-video therapy, or (3) scheduled rest group	Subjects in the groups received their assigned 30 min intervention at t w o episodes on postoperative days 2 and 3	Mood (numeric rating scale), State-Trait Anxiety Instrument (STAI), Physiologic measures of anxiety and mood include blood pressure and heart rate	Significant improvement in mood among subjects in the music intervention group after the second intervention when controlling for the preintervention rating of mood, F(2, 87) = 4.33, p = .016. No significant differences were reported for anxiety significant main effects over time for heart rate and systolic and diastolic blood pressure
33	The effects of music on physiological responses and sedation scores in sedated, mechanically ventilated patients. Dijkstra et al., 2010	Determine the effects of music on physiological responses and sedation scores in sedated, mechanically ventilated patients/two-armed RCT/n = 20	ICU, Netherlands/NS	Researcher selected: classical (Anton Bruckner: Quintet F-Dur: Adagio and Gustav Mahler: Symphony Nr.4 G-Dur: Ruhevoll) and easy listening (film music: Vangelis: 1492, songs without vocals were chosen)	Music three times for 30 min over 2 days, or three times 30 min rest periods (between 8:00 pm and 10:00 pm)	Systolic, diastolic and mean arterial blood pressure and heart and respiratory rate were assessed	Physiological parameters did not show a significant difference between the groups. Patients in the experimental group showed significantly higher Ramsay (sedation) scores than patients in the control group after the first session. The higher scores indicate that patients were less responsive to external stimuli
34	The Effects of music on the Life Signs of Patients in the Reanimation Unit/Recovery Room After Laparoscopic Cholecystectomy. Ecer and Saritaş, 2019	Determine the effects of music on the life signs of patients in the post-anaesthesia care unit after laparoscopic surgery/2-armed RCT/n = 148	Post-op. care unit, Turkey/NS	Not stated	20-min music listening	Life/vital signs: (1) baseline (on admission), 20 mins music after registration, (2) second data collection, (3) third measure post-operation	After music treatment (second measurement), there was a significant difference only in the respiratory rates (p < .05)
35	The effects of music therapy in endotracheal suctioning of mechanically ventilated patients. Aktas and Karabulut, 2016	Determine the effect of music therapy on pain intensity, sedation level and physiological parameters during endotracheal suctioning of mechanically ventilated patients in cardiovascular surgery intensive care unit (post-heart surgery)/single-blind 2-arm RCT experimental survey/n = 66	ICU, Turkey/NS	Music listening. Exp group = Instrumental reed flute music with a low pace (60–80 rhythms/min) and without strong beat and fluctuating rhythms (all instrumental), sufi music (instrumental and vocal) Control = no music	20 min before and after suctioning – MP3 and cre-atone music pillow	Patient Information Form, Critical-Care Pain Observation Tool, Ramsay Sedation Scale and Form of Physiological Parameters	Ramsay Sedation Scale during endotracheal aspiration difference between groups was statistically significant (p = .003). Critical-Care Pain during endotracheal suctioning in the music group was statistically lower than control (p < .001). No significant differences before, during and 20 min after suctioning between the two groups in systolic and, diastolic blood pressure, heart rate, and oxygen saturation.
36	The effect of live spontaneous harp music on patients in the intensive care unit. Chiasson et al., 2013	Determine whether live harp music would serve as a relaxation or healing modality (pre-or postoperative)/2-armed RCT/n = 100	ICU, US/NS	Improvised harp music	10 min live music or equivalent rest period	Pain, heart rate, respiratory rate, blood pressure, oxygen saturation, and heart rate variability	Harp music significantly decreased patient perception of pain by 27% but did not significantly affect other measures
37	The Effect of music and Massage on the Pain Scales and Vital Signs of ICU Patients with Haemodialysis Catheter. Elay and Ozkava, 2020	To evaluate the effect of music and massage on the pain level during haemodialysis catheterization/4-arm RCT/n = 220	ICU, Turkey/NS	Type of music was chosen by the patient if s/he was conscious; otherwise, the family members chose the type of music	58 = music listening (headphones), 58 = hand massage, 56 music listening and hand massaging, 50 = no intervention 30 min music listening. Hand massage was performed for 5 min on the right palm, and 5 min on the back of the hand	Sedation levels were evaluated using Richmond Agitation Sedation Scale (RASS), Ventilator-supported patients were evaluated using the Behavioural Pain Scale (BPS), WONG, and BPS were used for conscious and unconscious patients, respectively	The respiratory rate and heart rate of Music and massage (Mm) group was lower than the starting (p = .027, p = .043). There was a decrease in Behavioural Pain Scale of intubated Mm and music group (p = .001, p = .000). Considering non-intubated patients, Mm, only music and only massage groups showed a significant decline in terms of WONG (Wong-Baker faces pain rating scale) (p = .000, p = .001, p = .000). The Mm group showed that patients were more sedatized in terms of RASS (Richmond Agitation-Sedatation Scale) (p = .046)
38	The effect of music on discomfort experienced by intensive care unit patients during turning: a randomized crossover study. Cooke et al., 2010.	Identify the effect of music on discomfort experienced by ICU patients during turning/Single-blind 2-arm RCT crossover/n = 17	2xhospital ICUs, Australia/NS	Participants were invited to bring a CD, or they could choose pre-operatively from a selection of classical, jazz, country, and western, new age, easy listening or ‘other’ (mostly by contemporary artists). Control wore headphones with no music	15 min before and during turning	Discomfort and anxiety were measured 15 min before and immediately after two turning procedures	Listening to music 15 min before and during turning did not significantly reduce discomfort or anxiety
39	The effect of music therapy on physiological parameters of patients with traumatic brain injury: A triple-blind randomized controlled clinical trial. Froutan et al., 2020	Investigate the effect of music therapy integrated with family recollection on physiological parameters. 2-armed RCT/n = 60	ICUs Iran/NS	The music listening audio files should include two parts: at first the selected family member should introduce himself/herself and clarify the time and place, and also briefly describe what has happened to the patient. In the second part, they were asked to recall and tell good and positive memories such as pleasant family gatherings or trips for about 4 min. And finally, they were asked to state positive and encouraging comments about the patient’s recovery, the patient’s return to home and also to discuss some future plans for the patient. The audio files were mixed along with a mild instrumental music (a mild and relaxing soundtrack by Arnd Stein about the sounds of birds and ocean waves) for 5 min. Control group wore headphones with no music.	For 6 consecutive days, the patients received a combination of music and auditory stimulation twice a day for 15 min	Physiological parameters were measured before the intervention, and then 10 min and finally 30 min after the intervention	The two-level multiple linear models which were performed for 6 consecutive days indicated a significant decrease in systolic blood pressure, diastolic blood pressure, respiratory rate, and heart rate for the patients in the intervention group
40	Self-Managed Music-Guided Exercise Intervention Improved Upper and Lower Extremity Muscle Strength for ICU Survivors – A Pilot Randomized Controlled Study. Liang et al., 2022	To evaluate effects of a self-managed music-guided exercise intervention on muscle strength among intensive care unit (ICU) survivors/two-armed RCT/n = 26	ICU, US/MT consulted	Playlist that consists of rhythmic musical cues with verbal prompts that facilitate five targeted upper and lower extremities movements Music group was taught to self-manage upper and lower extremity exercise movements by listening to an individualized music-guided playlist. Control group was provided an exercise brochure and advised to perform the same exercises at the same interval. MP3 player and speaker for each participant	Twice daily for 5 days. 10–15 min sessions	Upper extremity muscle strength assessments (left and right): handgrip strength, elbow flexion, and shoulder abduction; (2) lower extremity muscle strength assessments (left and right): plantar flexion of the foot and knee extension. Jamar Smart Hand Dynamometer	The music group showed significant improvements in handgrip, plantar flexion, leg extension, elbow flexion, and shoulder adduction strengths on left and right sides. Additionally, left and right leg extensor and left plantar flexor strengths showed significant post-differences and small to moderately large effect sizes between the music group and control group
41	Comparison of the effectiveness of music and progressive muscle relaxation for anxiety in COPD – A randomized controlled pilot study. Singh et al., 2009	Evaluate the acute effects of music and progressive muscle relaxation (PMR) in hospitalized COPD subjects after a recent episode of exacerbation/two-armed RCT/n = 72	ICU, India/NS	Music group listened to a self-selected music of 60–80 beats per minute. PMR group practised relaxation through a pre-recorded audio of instructions on 16 muscle groups. Subjects in music group selected one record from six Indian classical instrumental music, including flute, sitar, mandolin, and mixed instrumentals. Headphones and CD player	30 min in two sessions (morning and afternoon) on the same day	Spielberger’s state anxiety inventory (SSAI), Spielberger’s trait anxiety inventory (STAI), dyspnoea, systolic blood pressure (SBP), diastolic blood pressure(DBP), pulse rate (PR) and respiratory rate (RR)	Statistically significant main effect across the sessions for state anxiety, trait anxiety, dyspnoea, SBP, PR, and RR. There was a statistically significant interaction effect between the two groups for state anxiety, trait anxiety, dyspnoea, SBP, PR, and RR
42	Soothing music can increase oxytocin levels during bed rest after open-heart surgery: a randomized control trial. Nilsson, 2008	To evaluate the effect of bed rest with music on relaxation for patients who have undergone heart surgery on postoperative day 1/two-armed RCT/n = 40	ICU, US/NS	The music, MusiCure, (Musi Cure), was soft, relaxing, and included different melodies of 60 to 80 beats per minute (bpm). All patients also received oxygen support. Music pillow and MP3. Control = bed rest	30 min	Plasma oxytocin, heart rate, mean arterial blood pressure, PaO2, SaO2, and subjective relaxation levels	Music group, levels of oxytocin increased significantly in contrast to the control group for which the trend overtime was negative that is, decreasing values. Subjective relaxation levels increased significantly more, and there were also significantly higher levels of PaO2 in the music group compared to the control group. There was no difference in mean arterial blood pressure, heart rate, and SaO2 between the groups
43	Music Listening Among Postoperative Patients in the Intensive Care Unit: A Randomized Controlled Trial with Mixed-Methods Analysis. Ames et al., 2017	Evaluate the effects of music listening on opioid use, self-reported pain anxiety, distress/two-armed RCT/n = 41	ICU, US/NS	MusiCure Dreams album, noise-cancelling headphones	50 min listening or rest. Frequency: initially 3 X per-day, then on request – 1–8 sessions in total depending on length of stay (48 h on ICU). Settling routine for both groups, including dimming lights.	Pain = visual analogue scale and NRS; STAI; GAD-7; emotional thermometers (ET)	No significant difference in pain, opioid intake, anxiety, distress, or anxiety during first 4 time points. Mixed-modelling analysis of pre/post-scores at the first time-point found significant interaction with NRS for pain, with reduction in music group
44	Effects of Patient-Directed Interactive Music Therapy on Sleep Quality in Postoperative Elderly Patients: A Randomized Controlled Trial. Kim et al., 2020	Investigate the effects of self-directed music listening on saliva melatonin levels and sleep quality in elderly ICU patients/3-armed RCT/133	ICU, Korea/MT	1. Interactive music listening. 2. Passive music listening. 3. Usual care. PASSIVE: classical music including Air for G string, Allemande, Canon, Nocturne and Swan, from which patients could choose	30 min between 8 and 10 pm. MP3 + earphones. Music Therapist assessed if each patient was willing to listen each time ACTIVE: individual MT with certified therapist. 15–20 min between 10 am and 12 pm (breathing exercise, interactive music making, breathing exercise), with patient-preferred music. PASSIVE: researcher selected music listening	Saliva melatonin and cortisol three times pre-op. ay and post-op at 11 pm. Richards-Campbell Sleep Questionnaire and Quality of Recovery-40 questionnaire on post-op day, plus PODS 1 and 2	RCSQ significantly improved in the IMT compared to control on POD2. No difference on QoR40. Saliva melatonin for IMT in POD1 elevated compared to control. Cortisol: no significant differences between groups
45	Effect of music on postoperative pain in patients under open-heart surgery. Ajorpaz et al. (2014)	Investigate the effects of instrumental music listening on pain in heart surgery patients/two-armed RCT/n = 60	ICU, Iran/NS	Researcher selected, instrumental, sedative music. Control = routine care	30 min, MP3 and headphones	Pain intensity – visual analogue scale	Significant post-intervention reduction in postoperative pain for music group compared to control

COPD: chronic obstructive pulmonary disease; PACU: ponst-anaesthetic care unit; NRS: numerical ratings scale; NIV: noon-invasive ventilation; MV: mechanically ventilated; ML: music listening; IMT: interactive music therapy; HPA: hypothalamic pituitary adrenal; FACES: refers to a scale with facial expressions representing degrees of main or of mood; PSG: polysomnogram; TOU: trauma orthopaedic unit; TICU: trauma intensive care unit; UC: usual care; VS: vital signs.

Duration of delivery was not always clearly reported, but when stated ranged from 10 min of live harp music (Chiasson et al., 2013) to 60 min music listening broken into 20 min before, during and after suctioning (Mateu-Capell et al., 2019) and twice daily with an average listening time of 79.8 min (Gullick and Kwan, 2015). Many studies delivered single sessions, and several spanned a number of days, including 7 consecutive days for head trauma (Yekefallah et al., 2021) and 6 consecutive days with traumatic brain injury (Froutan et al., 2020), 5.7 days for stress reduction (Gullick and Kwan, 2015) and twice daily for 3 days (Beaulieu-Boire et al., 2013).

Listening equipment ranged from cassettes played through headphones (Barnason et al., 1995; Updike, 1990) to MP3 players and audio pillows (Mateu-Capell et al., 2019). Not all papers stated the device or whether headphones or a speaker were used.

Discussion

Music interventions increasingly find their way into the adult intensive care unit. While providing passive listening interventions seems to be the predominant music use, there is a lack of research in the ICU setting that examines the effects of active music-based interventions, that is, patients playing instruments as part of structured exercises for cognitive or physical rehabilitation, or improvised music with a Music Therapist in ways that may benefit emotional well-being in addition to physical and cognitive function, including management of deconditioning. In addition, there is a lack of live-performed music therapy studies, which can be seen as an active form of patient – therapist interaction and participation.

Summary discussion of physiological outcomes

The included literature indicates that one major benefit of individual music listening in ICU is for pain reduction. There may be additional benefits in vital signs, including heart rate, respiratory rate, and blood pressure, the latter being reported as significantly reduced in seven music-listening studies, but six found no effect, as did a further live one, making an equal number that either reported an effect or no effect. Eight studies reported no significant reduction in respiratory rate, while seven found significant changes or more than in the control. With more standardized music-listening procedure protocols for the different patient populations and ICU settings, further research would help determine any recommendations for those patient groups, potentially resulting in quicker and/or more favourable recovery, patient experience, and a reduction in pharmacological interventions that bring unwanted side-effects.

Summary discussion of psychological outcomes

Along with pain management, the other major area with reported benefits was anxiety. Anxiety has been the most researched psychological field, showing a decrease in anxiety levels as well as overall discomfort in the Unit. Furthermore, sleep seemed to be positively impacted through music therapeutic interventions, which in turn may lead to a better recovery time and shorter stays in the ICU.

Summary discussion of neurological outcomes

While there is a large amount of literature investigating the benefits of active music therapy on multiple neurological outcomes (Magee et al., 2017; Sihvonen et al., 2017), the literature is rather thin when applying this to the adult ICU. The reasons may be that Music Therapists are not yet represented as caregivers within this healthcare setting and, therefore, have not had the opportunity to deliver active music therapy in a research setting. Against this backdrop, delirium prevention and treatment have been identified as having significant importance (Field and Wall, 2013; Inouye et al., 2014). While few music-based studies for delirium have been conducted, there could be potential for using music listening in complementing and/or reducing pharmacological use and addressing some of its mechanisms, including physiological stress, inflammation and neurotransmitter imbalance (Johnson et al., 2018).

What might Music Therapists bring to ICU?

More research could be conducted that draws on Music Therapists’ unique training and skills, helping to identify which music-based interventions need these skills and which could be delivered by other healthcare professionals, as well as identifying the types of training materials needed to support them. One example of the latter might be in rhythmic entrainment – used in a study outside of this review for pain management on a burns unit (Prensner et al., 2001), which might be achievable using emerging music-listening technology that can map music in ways that enable it to meet existing arousal levels and either maintain them or adjust them in the desired direction, for example, to increase relaxation and reduce anxiety (P et al., 2020). Some of the included literature did not clearly report whether a Music Therapist was involved, but made frequent references to music therapy (Chlan, 1998), suggesting that some principles of the profession may be employed, but these clinicians are not actively involved. One study provided some background to differentiate between the skill sets required for interventions delivered by a Music Therapist and other professionals, such as nurses, who could implement therapeutic music listening, rather than what the authors defined as music therapy (Browning et al., 2020).

Other charities and research funders (National Institute for Health Research, Medical Research Council, for example) could fund more innovative music therapy interventions, such as active music-based exercises to help reduce and manage deconditioning, delirium, and enhance rehabilitation. Single case designs could be suitable, depending on cohorts – whether long or short-term inpatients – and study aims. In this design, patients are their own control, receiving baseline and washout periods (no intervention) between experimental interventions, which could be single or multiple delivery over days, weeks or months (Krasny-Pacini and Evans, 2018). Active music interventions, where patient and therapist create music together, or the therapist devises structured exercises, for example, to improve muscle conditioning, stamina, and cognition (attention, executive function, and comprehension), generally require a higher level of musicianship and clinical skills using music, more often found in HCPC registered Music Therapists. This is not to say that clinicians from other disciplines such as occupational therapy could not access such interventions, provided they had adequate support and/or training from a Music Therapist – this is another area of research to explore. It has been reported that music-based interventions delivered to brain injury survivors in neurorehabilitation by a Music Therapist may bring added benefits in some areas (Magee et al., 2017), and that their skills bring added value to multidisciplinary teams with that patient population (Magee, 2019). One study included in this review stated the importance of having a Music Therapist in ICU but did not describe their input in their study (Miller et al., 2021).

Potential for active music therapy for rehabilitation and/or reducing deconditioning

Rehabilitation on ICU using active music-based interventions could be explored for helping to prevent and treat deconditioning, including muscle tone and strength and respiratory weakness (Gosselink et al., 2012). Some stroke rehabilitation research has reported equal benefits in sensory-motor recovery compared to usual care, in addition to less fatigue and enhanced mood and cognitive function that are not found in usual care (Grau-Sánchez et al., 2018). Such interventions have been delivered as part of usual care in the acute and subacute setting, where feasibility and acceptability were well rated, and there was possibly an improvement in patient mood (Street et al., 2020; Thompson et al., 2022). Based on this, a reasonable hypothesis might be to try some music-based exercises with ICU patients to help manage deconditioning and combine the benefits of passive (music listening) and active approaches to also help manage stress and anxiety.

Music listening as part of usual care: feasibility and implementation research

As can be seen from the scoping review and other literature reviews, music interventions have not been explored on ICUs in the United Kingdom, so it is not known how they would fit into their health systems, or what effect they might have on health service delivery and patient outcomes, including length of stay. However, the literature and related research provide enough detail to inform on the setting up of a service delivering patient-preferred music, targeting a number of factors experienced by patients on ICU, including anxiety and pain.

In the form of feasibility and implementation research, or service evaluation and audit, a Music Therapist could deliver music-listening interventions but also train research nurses or other nursing staff to compile and deliver patient-preferred music playlists. They could collect data on patient experience, undertake co-design work with patients, their relatives and staff to develop delivery methods in a range of ICU settings and measure effects on pain and anxiety, delirium and sleep, using standardized measures, which could be done as part of usual care depending on availability of research nurses.

Bio-data-driven music listening for arousal adjustment

Systems have been developed to deliver pre-recorded music in medical settings that are driven by bio-data such as heart rate (Sice et al., 2020). Music is selected by these systems that have been processed using algorithms and mapped onto an arousal valance. Arousal can then be adjusted or maintained based on a live heart rate stream from the user, feeding the system, starting with music that is matched to the user’s existing heart rate. X-system delivers music in this way and could be costed for research purposes.

Conclusion

Some music-listening research has been reported in ICU settings, but little on guided and active music-based therapeutic interventions administered by a Music Therapist. This makes it somewhat problematic to fully establish, based on this review, what a service implemented by Music Therapists using active interventions would look like. Some of the single-session studies included in this review indicate that there are benefits from this potentially cost-effective passive listening approach, but more needs to be revealed about the benefits from this and active music therapy interventions for patients in short- and longer-term ICU stays. While we believe – and some evidence in this review suggests – that receptive music listening is better than no music intervention at all, it may be of importance for a Music Therapist to be part of the ICU and hospital environment. Their expertise and training may compliment medical professionals in providing the best care for the patient and in turn improving well-being, recovery, and patient experience.