Earplugs, eyemasks and combination of both

Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Hu, 2015

Cochrane review

PS., study characteristics and results are extracted from the SR (unless stated otherwise)

SR and meta-analysis of  (quasi-)RCTs

Literature search up to June 2014]

Setting and Country:

Medical University, Fujian, China

Source of funding and conflicts of interest:

No details

Inclusion criteria SR:

(quasi-) RCT

• evaluating the effects of non-pharmacological interventions for sleep promotion
• in critical care units (CCU) or intensive care units (ICUs) for critically ill adult participants
• published or unpublished
• any language

We excluded studies

enrolling

Exclusion criteria SR:

• participants who were diagnosed with obstructive sleep apnoea or dementia
• those who were terminally ill or required

palliative care.

8 studies included in the current summary

Facultative:

Authors conclusion:

The quality of existing evidence relating to the use of non-pharmacological interventions for promoting sleep in adults in the ICU was low or very low. We found some evidence that the use of earplugs or eye masks or both may have beneficial effects on sleep and the incidence of delirium in this population, although the quality of the evidence was low. Further high-quality research is needed to strengthen the evidence base.

Earplugs

Lin, 2021

Type of study:

RCT

Setting and country:

Single hospital, ICU, Taiwan

Funding and conflicts of interest:

‘This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors; The authors declare that there is no conflict of interest’

Inclusion criteria:

• Aged ≥20 years,
• conscious,
• able to read and communicate
• have a stay in the ICU that was expected to be more than 48 h.

Exclusion criteria:

• treated with sedatives,
• had cognitive or hearing impairment
• used mechanical ventilation or physical restraints

N total at baseline:107

Intervention: 55

Control:52

Important prognostic factors²:

ages ranged from 28 to 87 years (mean = 59, SD = 12.89)

Sex:

Most participants were male (n = 81, 75.7%)

Groups comparable at baseline? No, more males in the control group and more females in the intervention group

slept with earplugs between 10 pm and 7 am on the second night of their intensive care unit stay + standard of care

The research intervention was the use of earplugs. The intervention was delivered on Day 2 between 10 pm and 7 am. The earplugs used in the study were disposable polyurethane foam earplugs with a pushin, bell shape feature that matches the contours of the inner ear to protect hearing, with a maximum sound level reduction of 33 dB (MAX-30, Howard Leight). The earplugs were selected and administered by patients' primary nurses at nocturnal sleep duration.

No earplugs, standard of care

Participants in the control group received the usual care.

Standard of care:

Half the rooms had a large sliding window, and each bed had a clock. A digital sound level metre on the wall of the ICU continuously monitored sound levels. The lighting was controlled to dim at nocturnal sleep duration.

Length of follow-up:

3 days

Loss-to-follow-up &

incomplete outcome data:

Intervention: 0%

Control: 0%

Sleep quality (critical)

Chinese version of Richards–Campbell Sleep Questionnaire,  consists of five self-reported items (perception of depth of sleep, sleep onset latency, number of awakenings, time spent awake and overall sleep quality), with each VAS running from 0 to 100 mm. The mean score of the items provided a global sleep score, 0-100mm, higher scores indicates better sleep quality, mean ± SD

RCSQ total score (Day 0)

I: 53.35 ± 27.30

C: 57.81 ± 28.53

RCSQ total score (Day 1)

I: 60.64 ± 25.56

C: 56.40 ± 25.35

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

Litton, 2017

Type of study:

RCT

Setting and country: 10-bed ICU of a large, private hospital in Perth; Australia

Funding and conflicts of interest: funding not reported

“Competing interests: None declared.”

Inclusion criteria:

• listed for surgery that included a planned ICU admission and were

expected to be undergoing mechanical ventilation (MV) on admission to the ICU

Exclusion criteria:

•  age < 18 years,
• pre-existing hearing difficulties requiring the use of a hearing aid,
• suspected or confirmed tympanic membrane rupture,
• unwillingness to have earplugs placed for sleep promotion in the ICU
• treating clinician deeming that enrolment was not in the best interests of the patient

N total at baseline: 40

Intervention: 20

Control: 20

Important prognostic factors²:

For example

Age, median (IQR):

I: 70 (63–74)

C: 66 (57–74) , 

Male sex, n (%)

I: 14/20 (70%)

C: 17/20  (85%)

Groups comparable at baseline?

earplugs

 Participants assigned to receive earplugs had earplugs placed on admission to the ICU, according to a standard operating procedure (see Appendix). Once extubated, participants assigned to receive earplugs were offered earplug placement between the hours of 10 pm and 6 am for their first night in the ICU (see Appendix).

No earplugs

For participants assigned to standard care alone, earplugs were not provided and the use of earplugs was considered a protocol violation

Length of follow-up:

Intervention and assessment after the first night in the ICU once they were extubated

Loss-to-follow-up &  incomplete outcome data:

Intervention: 0%

Control:  0%

Sleep quality (critical)

Median RCSQ total score* (IQR) after their first full night during which they were not undergoing MV.

I: 43 (20–58)

C: 45 (29–64)

Median diff 2 (–21 to 25)

P=0.580

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Surrogate for hyperactive delirium: Median days receiving antipsychotic medication while in ICU

I:  0

C: 0

Duration of hospital/ICU stay (important)

Median ICU length of stay, days (IQR)

I: 4 (3–5)

C: 3 (3–5)

Median diff 1 (0–1)

P=0.251

Median hospital length of stay, days (IQR)

I: 9 (8–14)

C: 9 (7–11)

Median diff 0 (–4 to 3)

P=0.376

From:

Hu, 2015

Cochrane review

Scotto 2009;

2-arm, parallel group design RCT

2 critical care units of a Midwestern US teaching hospital, USA

N, mean age

Number randomized: 100

Number analysed: 88 (39 in the control group, 49 in the intervention group)

Mean age: 63.1 years

Sex: 53 men, 35 women?

used earplugs during regular night-time sleeping hours for 1 night

no earplugs

End-point of follow-up:

1 night

For how many participants were no complete outcome data available?

12/100

Sleep quality (critical)

VSH scores (Verran/Snyder-Halpern Sleep Scale)

“Scotto 2009 also assessed sleep quality using the VSH

sleep score. The author reported that use of earplugs improved the sleep quality (P value < 0.05), but no mean scores were reported.”

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported in SR

Duration of hospital/ICU stay (important)

Not reported in SR

• 12/100 patients not in analysis
• Only t-scores provided;

From:

Hu, 2015

Cochrane review

Van Rompaey 2012;

3-arm, parallel group RCT

ICU, Antwerp University Hospital, Belgium

Total number randomized: 136 (69 in the experimental group, 67 in the control group)

Total number analysed: 136 in night 1, 71 in night 2, 27 in night 3, and 12 in night 4

Mean age: 59 years (range = 18 to 84)

Sex: 66% were men

participants sleeping with earplugs during the night

participants sleeping without earplugs during the night

End-point of follow-up:

4 days

For how many participants were no complete outcome data available?

Total number analysed: 136 in night 1, 71 in night 2, 27 in night 3, and 12 in night 4

Sleep quality (critical)

“136 participants compared sleep perception using five dichotomous questions on the self-reported sleep quality of the participant, which they categorized as: bad sleep (sum < 2), moderate sleep (sum 2 < 4), and good sleep (sum ≥ 4).

More participants perceived good sleep in the intervention group than those in the control group after the first night (P value = 0.042, no Chi2 test value reported).”

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported in SR

Quality of life (critical)

Not reported in SR

Delirium (important)

Incidence of delirium and confusion; validated Neelon and Champagne Confusion Scale (NEECHAM)  based on the nurses' 24-hour assessment of the level of processing information, the level of behaviour, and the physiological condition,

rating the participant on a 30 to 0 scale

I: 24/69

C: 40/67

Duration of hospital/ICU stay (important)

Not reported in SR

Eyemasks

Mahran, 2020

Type of study:

RCT

Setting and country:

an 11-bed postoperative

unit at Assiut University Cardiac Center, Assiut, Egypt

Funding and conflicts of interest:

“Financial Disclosures

None reported.”

Inclusion criteria:

• aged 18 years or older,
• having undergone cardiac surgery requiring immediate postoperative care

in the ICU, ability to communicate verbally and complete the outcome measures,

• length of ICU stay of at least 72 hours

Exclusion criteria:

•  history of neurological or psychiatric disorders,
• Richmond Agitation-Sedation Scale score of 3 or lower for more than 50% of their ICU stay,
• eye disease,
• unwillingness to use an eye mask.

N total at baseline: 70

Intervention: 35 (31 in analysis)

Control: 35

Important prognostic factors²:

Age, mean (SD), y

C: 46.91 (7.27)

I: 48.03 (7.43)

Sex Male

I: 23/31 (74%)

C: 11/35 (31%)

Groups comparable at baseline? Yes, more mail in intervention group, more females in control group.

nocturnal eye masks

The intervention

group received single-use eye masks (Flight Eye Masks; Dreaming) during nocturnal sleep on all 3 nights of the trial period. Participants were advised to wear the eye masks every night during their stay in the ICU and were instructed on how to use the masks  properly. During the ICU stay, ICU nurses assisted patients with wearing eye

masks from 9 pm to 7 am the next morning

No nocturnal eye masks

Standard of care:

The control group received routine care during the

nighttime period; this consisted of reductions in light,

noise, and awakening for medications

Length of follow-up:

First 3 nights

Loss-to-follow-up & incomplete outcome data:
I: 4/35 (refused to wear

eye masks because of discomfort or feeling anxious about not seeing anything)

C: 0/35

Sleep quality (critical)

Arabic version of the Richards-Campbell Sleep Questionnaire

(RCSQ; each item was scored using a 100-mm visual

analog scale (VAS), which ranged from 0 mm (worst

sleep) to 100 mm (optimal sleep). The total score was

calculated by dividing the sum of all scores by 5. The

RCSQ was administered by the researcher daily at 7 am.

Total RCSQ score for each night, mean (SD)

I:

1: 46.82 (8.49)

2: 60.44 (6.87)

3: 70.12 (10.08)

C:

1: 36.79 (7.31)

2: 50.23 (6.04)

3: 59.19 (7.62)

Mean (SD) RCSQ score for 3 nights I: 59.13 (6.48)

C: 48.74 (6.04)

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Days in intensive care unit

I: 4.32 (0.54)

C: 4.97 (1.01)

Earplugs & eyemasks

Akpinar, 2022

Type of study:

RCT

Setting and country: coronary intensive

care unit (CICU) of a university hospital,

Turkey

Funding and conflicts of interest:

“This study did not receive any specific grant from

funding agencies in the public, commercial, or not-for-profit sectors.

CONFLICT OF INTEREST

The author declare no potential conflicts of interest with respect to

the authorship and/or publication of this article.”

Inclusion criteria:

•  18 years of age and older
•  not having received any sedative medications or opioids for the past 24 hours;
• not having received any general anaesthesia for the previous 24 hours;
• not being mechanically ventilated;
• not suffering from any auditory/visual problems or any conditions making verbal communication difficult;
• absence of pain;
• being healthy enough to independently insert and remove the earplugs besides wearing an eye mask and taking it off when necessary;
• absence of a previously diagnosed sleep disorder;
• having sufficient cognitive acumen to provide informed consent (GCS > =13 and ICDSC ≤4) at the beginning of the study

Exclusion criteria:

• being under 18 years of age,
• patients whose condition suddenly deteriorated,
• those who could not effectively use earplugs and eye masks during the night
• those who voluntarily withdrawn from the study,
• those who were transferred from the intensive care unit to another department

N total at baseline: 84

Intervention: 42

Control:42

Important prognostic factors²:

For example

age ± SD:

I: 64.04 ± 8.64

C:63.73 ± 6.92

Sex, n %:

I: 24; 57.1% M

C: 22; 52.4% M

Groups comparable at baseline? yes

earplugs and an eye mask

The patients in the experimental group were asked to use earplugs and an eye mask overnight (from approximately 22:30 to 6:30 the next morning) on the second (Time 1) and third nights (Time 2). They were told that the earplugs and eye mask should be

removed only for a short time if an intervention or communication was necessary.

Standard of care

The patients in the control group received only routine care.

Length of follow-up:

3 days

Loss-to-follow-up & Incomplete outcome data:

Intervention: 0

Control: 0

Sleep quality (critical)

Richards-Campbell sleep questionnaire (RCSQ), mean SD (mean rank)

Baseline

I: 40.11 ± 16.55 (36.52)

C: 44.07 ± 7.30 (48.48)

Time 1 (second night)

I: 64.09 ± 14.07 (58.79)

C: 46.97 ± 9.22 (26.21)

Time 2 (third night)

I: 72.07 ± 11.75 (61.83)

C: 47.04 ± 11.53 (23.17)

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

ICDSC

Baseline

I: 0.47 ± 0.50 (43.50)

C: 0.42 ± 0.50 (41.50)

Time 1 (second night)

I: 0.33 ± 065 (39.52)

C: 0.50 ± 0.70 (45.48)

Time 2 (third night)

I: 0.19 ± 039 (36.12)

C: 0.57 ± 0.66 (48.88)

Duration of hospital/ICU stay (important)

Not reported

Note: Unclear what was meant with: “To ensure that all the patients were equally affected by the  environmental conditions

that were likely to affect sleep quality in the ICU, such as noise, light, and heat, the same number of patients was included in the

experimental and control groups each night.”

Demoule, 2017

Type of study:

RCT

Setting and country:

16-bed adult general ICU

within a 1600-bed hospital in Paris, France

Funding and conflicts of interest:

“Funding

This study was supported by a grant from the French Ministry of Health,

Assistance Publique – Hôpitaux de Paris (PHRC Regional 2009) and by the

Investissement d’Avenir ANR-10-AIHU 06 program of the French government.”

Competing interests were described in detail.

Inclusion criteria:

• no sedation for > 24 h,
• sedation level < 3 on the Ramsay Sedation Scale,
• expected remaining ICU stay > 48 h,
• morphine < 0.01 mg/kg/minute and norepinephrine < 0.3 μg/kg/minute

Exclusion criteria:

• history of sleep disorders such as sleep-related breathing disorders, insomnias, or sleep movement disorders;
• psychiatric illness requiring chronic medication;
• a known diagnosis of central neurological impairment;
• liver disease with encephalopathy; uncontrolled sepsis; severe hearing impairment;
•  blindness.
• age < 18 years

N total at baseline: 64

Randomized

I: 32

C: 32

Included in ITT analysis:

I: 23

C: 28

Included in per protocol analysis

I: 15

C: 28

Important prognostic factors²:

age median (IQR):

I: 64 (54–74)

C: 65 (58–74)

Sex:

I: 20/30 (67%) M

C: 18/31 (58%) M

Groups comparable at baseline? Yes

earplugs and eye mask

No earplugs and eye mask

Length of follow-up:

90 days

Loss-to-follow-up& incomplete outcome data:

Randomized

I: 32

C: 32

Included in ITT analysis:

I: 23

C: 28

Included in per protocol analysis

I: 15

C: 28

Reasons:

I: n=7, data problems, n=9 did not adhere to intervention

C: n=3, data problems

I: N=23, C: N=22

Sleep quality (critical)

self-reported comfort and sleep quality were assessed daily using a simplified visual analogue scale (VAS; 10 cm horizontally) from zero for worst possible comfort or sleep quality to 10 for best possible comfort or sleep quality; median (IQR), a at ICU discharge

I: 70 (50–70)

C: 60 (25–80)

P= 0.63

Pittsburgh Sleep Quality Index

At 90 days follow up

I: 8 (5–11)

C: 5 (5–8)

P=0.25

Sleep duration

Duration of sleep

Total sleep duration per 18 h,minutes

I: 290 (146–410)

C: 301 (229–398)

P=0.91

Total sleep duration during nighttime, minutes

I: 286 (120–392)

C: 274 (177–329)

P= 0.77

EEG sleep parameters

Polysomnography could not be scored accurately, owing to poor signal quality in seven patients in the intervention group and three patients in the control group. These patients were included in the analysis but did not contribute to the primary outcome. Data for the primary outcome variable were subsequently collected for 23 patients in the intervention group and 28 patients in the control group.  Sleep measurements are detailed in Table 2. The N3 proportion was not different between the two groups (21 [7–28]% in the intervention group vs. 11 [3–23]% in the control group, p = 0.09). Prolonged awakenings were less frequent in the intervention group (21 [19–26] %) than in the control group (31 [21–47] %, p = 0.02). Sleep quality following the first night after inclusion was similar in the two groups. We compared the 31 patients in the control group with the 21 patients in the intervention group who actually wore earplugs all night long in a per-protocol analysis. Among them, polysomnogramphy could be scored in 15 patients (Additional file 3: Table S3). N3 sleep duration was higher and prolonged awakenings were less frequent in intervention group patients who wore earplugs all night long than in the control group patients.

Quality of life (critical)

Not reported

Delirium (important)

Delirium, n (%) at ICU discharge, CAM-ICU

I: 2 (7)

C: 2 (6)

Duration of hospital/ICU stay (important)

ICU length of stay, days, median (IQR)

I: 7 (4–11)

C: 7 (5–26)

P=0.18

Hospital length of stay, days, median (IQR)

I: 24 (12–47)

C:  26 (14–86)

P=0.76

Leong, 2021

Type of study:

RCT

Setting and country:

Singapore General Hospital, a 1700-

bed tertiary hospital

Funding and conflicts of interest:”The study was funded with a grant from the

Anaesthesiology and Peri-operative Sciences Academic

Clinical Programme Pilot Research Grant 2018. Registered

on clinicaltrials.gov before patient enrolment

(NCT03702296).”

“No

competing interests declared.”

Inclusion criteria:

•  aged > 21 y
• undergoing elective major colorectal surgery
• with a Glasgow Coma Scale (GCS) of at least 10

Exclusion criteria:

• known hearing impairment,
• dementia,
• confusion,
• delirium,
• pre-existing tracheostomy
• who returned postoperatively to the ward after 22.00

N total at baseline:

Intervention: 48

Control: 45

Important prognostic factors²:

Age, median, IQR, range:

I:60 (56–69) [30–85)

C:  67 (58–74) [39–88]

Sex:

I 25/48 (52.1%)

C: 26/45 (57.8%)

Groups comparable at baseline?

More individuals with higher ASA score (worse health) in control group

ASA physical status

1: I: 1; C: 1

2: I: 36; C: 25

3: I: 11; C: 18

NB: Difference in anaesthetic technique seems to be a typo (table 1); text states no difference

Given earplugs and eye masks

No earplugs and eye masks

Length of follow-up:

3 days

Loss-to-follow-up & Incomplete outcome data:

Intervention:0

Control: 0

Sleep quality (critical)

RCSQ; assesses patient’s

perceived sleep quality over six domains based on a score

from 0 to 100. These domains cover the aspects of sleep:

depth; latency; awakenings; proportion of time awake;

quality of sleep; and noise perception [15]. A zero value

reflects a negative perception of sleep and 100 a positive

perception

Mean, IQR, range

POD1

I: 60 (44–82 [18–100])

C: 64 (38–74 [0–100])

P=0.310

POD2

I: 56 (42–76 [0–100])

C: 60 (42–78 [0–100])

P=0.797

POD3

I: 62 (48–78 [12–100])

C: 66 (50–80 [0–100])

P=0.270

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Neelon and Champagne Confusion Scale

“There was no difference in the incidence of delirium between the two groups overall over the three

postoperative days (Fig. 2).”

Duration of hospital/ICU stay (important)

Not reported

Obanor, 2021

Type of study:

RCT

Setting and country:

Surgical ICU within the University of Texas Southwestern

Medical Center, USA

Funding and conflicts of interest:

Not reported

Inclusion criteria:

• age ≥ 18 yr;
• female patients
• admitted to the ICU following plastic surgical

breast free flap procedures

• requiring hourly postoperative assessments.

Exclusion criteria:

• pregnancy,
• current incarceration,
• diagnosis of sleep apnea, insomnia, or other sleep disturbance

N total at baseline:

Intervention: 44

Control: 43

Important prognostic factors²:

For example

age ± SD:

I: 51.4 (9.3)

C: 50.7 (8.7)

Sex:

I: 100% F, 0% M

C: 100% F, 0% M

Groups comparable at baseline? Yes

standard postoperative care plus earplugs and eye mask

standard postoperative care

Length of follow-up:

All patients completed the Richards-Campbell Sleep

Questionnaire (RCSQ) following every night in the

ICU and a modified version of the Family Satisfaction

in the ICU (FS-ICU) survey upon ICU discharge.

Loss-to-follow-up & incomplete outcome data:

Intervention: 2/45

Control: 1/45

Reasons: 2 withdrew, 1 did not require frequent flap checks in the ICU

Sleep quality (critical)

RCSQ, following the first night in the ICU

Sleep depth:

My sleep last night was:

0 Light → 50 Normal → 100 Deep

I: 57.0 (27.1) CI, 49.0–65.0

C: 41.3 (28.9) CI, 32.7–49.9

P=0.0157

Sleep latency:

I: 73.0 (27.0) CI, 65.0–81.0

C: 60.5 (30.6) CI, 51.4–69.6

P=0.0595

Number of awakenings:

→ 100 Awake very little

I: 59.0 (26.4) CI, 51.2–66.8

C: 45.5 (22.3) CI, 38.8–52.2

P=0.0174

Sleep efficiency:

I: 74.5 (27.5) CI: 66.4–82.6

C: 53.8 (30.2) CI, 44.8–62.8

P=0.0022

Sleep quality:

I: 59.1 (28.1) CI, 50.8–67.4

C: 35.3 (27.8) CI, 26.9–43.6

P=0.0003

Overall perception of sleep:

Average RCSQ score

I: 64.5 (21.1) CI, 58.3–70.7

0.0007

C: 47.3 (21.9) CI, 40.8–53.8

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Overall Confusion Assessment Method for the ICU score, n (%) – CAM ICU

No delirium

C: 43 (49.4%)

I: 44 (50.6%)

Duration of hospital/ICU stay (important)

ICU length of stay, d, mean (sd)

I: 1.66 (0.50)

C: 1.61 (0.45)

0.6538

The RCSQ is a five-item questionnaire, previously validated against polysomnography, used to evaluate perceived sleep depth, sleep latency, number of awakenings, efficiency, and sleep quality (24). Each RCSQ question is measured on a 0 (minimum) to 100 (maximum) Visual Analogue Scale, with higher scores representing better sleep. The average score of the five items is the total score and represents the overall perception of sleep

Meaning of sub scale scores:

• Sleep depth: My sleep last night was: 0 Light → 50 Normal → 100 Deep
• Sleep latency: Last night, the first time I got to sleep, I: 0 Couldn’t fall asleep → 50 Fell asleep normally → 100 Fell asleep immediately
• Number of awakenings:

Last night, I was: 0 Awake all night long → 50 Awake sometimes → 100 Awake very little

• Sleep efficiency: Last night, when I woke up or was awakened, I: 0 Couldn’t get back to sleep → 50 Fell back to sleep normally → 100 Got back to sleep immediately
• Sleep quality: I would describe my sleep last night as: 0 Bad night’s sleep → 50 Normal night’s sleep → 100 Good night’s sleep

From:

Hu, 2015

Cochrane review

Xie 2011

Quasi-RCT

medical intensive care unit, China

N, mean age

Numbers randomized: 75 (42 in the experimental group, 33 in the control group)

Mean ages: 56.4 ± 10.2 years

Sex:

Sex: 43 men, 32 women

Groups comparable at baseline: unclear

used earplugs and eye masks

did not use earplugs and eye masks

End-point of follow-up:

3 days

For how many participants were no complete outcome data available?

Incomplete outcome data were not described

Sleep quality (critical)

PSQI score (0 = best sleep, 21 = worst sleep)

mean diff intervention versus control:

-7.25 (95% CI -8.46 to -6.04).

Sleep duration

Duration of sleep

Total sleep duration, hours, mean (SD)

I: N=42; 7.8 (0.8)

C: N=33; 4.86 (1.04)

MD 2.94 [95% CI 2.51,3.37]

EEG sleep parameters

One quasi-RCT of 75 ICU patients, Xie 2011, compared the use of earplugs and eye masks versus usual care on objective sleep variables, as measured by EEG. There was a greater improvement in the mean number of hours of SWS in the intervention group compared with the control group (SWS: post-test mean = 2.18, SD = 0.34 versus post-test mean = 1.43, SD = 0.28) (P value < 0.01) (REM: post-test mean = 2.09, SD = 0.28 versus post-test mean = 0.71, SD = 0.36) (P value < 0.01). A greater reduction in the mean number of hours of waking time was also reported in the intervention group compared with the control group (post-test mean = 1.79, SD = 0.75 versus post-test mean = 3.8, SD= 0.79) (Pvalue < 0.01); no significant difference in NREM time was observed between groups (P value > 0.05).

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

High risk of bias (selection bias, performance bias; unclear whether there was also detection bias of reporting bias)

Music

Hansen, 2018

Type of study:

RCT

Setting and country:

two level three multidisciplinary ICUs (The Intensive

Care Society, 2009) with 18 beds at Aarhus University Hospital between February and

April 2016, Denmark

Funding and conflicts of interest:

Not reported

ICU; acute care of

medical and surgical patients’ one of the ICUs also has patients recovering from elective surgery

Inclusion criteria:

• aged ≥18 years;
• non-sedated;
• ventilated or non-ventilated;
• they had a Glasgow

Coma Scale ≥14;

• able to communicate,

understand and complete the sleep questionnaires;

able to give informed consent to participate

Exclusion criteria:

• history of sleep disorder before admission to ICU;
• diagnosed with cognitive dysfunction;
• history of postoperative complications;
• sedated;
• in a coma
• in delirium

N total at baseline:

Intervention:  18

Control: 19

Important prognostic factors²:

age (SD):

I: 60 (18)

C:  65 (16)

Sex:

I: 10/19 (52.6%)

C: 10/18 (55.6%)

Groups comparable at baseline? yes

listen to music for 30 min during daytime rest

Participants in the intervention group listened to music for 30 min during daytime rest in the ICU. The duration of interventions with music in other studies varied between 20 and 45 min per session (de Niet et al., 2009), therefore, we chose the duration of 30 min in this study. The music, MusiCure, was especially composed and developed for critically ill patients in ICU by Niels Eje (MusiCure.com). The intervention consisted of four pieces of soothing music, in terms of soft wind, bird twitter, ocean sound and music instruments. There is a clear distinction between interventions with music; as music therapy implies, trained music therapists administer the interventions (active music). The intervention in this study is categorized as music medicine (passive music), which refers to listening to pre-recorded music administered by the health care professionals or the patient (Bradt and Dileo, 2014). 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 centerd around the patient (Soundfocus.dk). The volume of the music was set at a comfortable level for each participant.

rested without music

The control group received standard care during their daytime rest (no music).

Length of follow-up:

Not specified; Data were collected shortly after the participants’ daytime rest

Loss-to-follow-up & incomplete outcome data:

Not reported; 5 patients excluded after inclusion and before randomization, reasons not specified

Sleep quality (critical)

Richards-Campbell Sleep Questionnaire; Danish version of RCSQ was used to assess self-reported sleep quality The original version of RCSQ has five items: (1) sleep depth, (2) latency, (3) awakenings, (4) time awake and (5) perceived quality of sleep using a 0–100 mm visual analogue scale (VAS) for each question. The total score for the RCSQ is calculated by adding the score for each item divided by five. High scores indicate good quality of sleep. The Danish version of the RCSQ has six items where item six is about the perceived nocturnal noise level; this item is an option in the RCSQ. High scores indicate low perceived noise level (Vestergaard et al., 2016). The Danish version of RCSQ has not been validated but it is the only  questionnaire developed to measure sleep in the ICU that has been translated into WHO guidelines.

Quality of sleep, mean (SD)

I: 72 (30)

C: 50 (27)

Mean diff 22 (2⋅97 to 41⋅02)

Sleep duration

Duration of sleep

Not reported

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

Kim, 2020

Type of study:

RCT

Setting and country:

Severance hospital,

tertiary university hospital with 2500 beds.

Korea

Funding and conflicts of interest:

This work was supported by a grant from the National

Research Foundation of Korea (2016R1A 2B1012708). / The authors report no conflicts of interest in this work.

postoperative elderly patients admitted to

ICU

Inclusion criteria:

• scheduled for ICU admission after elective major surgery,
• confirmed their intention to participate in this RCT
• obtained written consent

Exclusion criteria:

• neuropsychiatric history,
• alcoholism,
• dementia, impaired cognitive dysfunction,
•  cerebral infarction,
• transient cerebral ischemic attack,
• severe liver disease,
• kidney disease,
• diabetes,
• sleep disorders before admission
• emergency surgery

N total at baseline:

Randomized

Intervention: 64

Control: 79

Analyzed:

I: 44

C: 45

Important prognostic factors²:

For example

age ± SD:

I: 72.3 ± 4.7

C: 74.1 ± 6.7

Sex (M:F) 

I: 32:12

C: 35:10

Groups comparable at baseline? Baseline APACHE score higher in control group; melatonin level higher in control group.

Music listening

Passive Music-Listening Intervention

The PL group listened to music at night without music intervention during the day. As per the IMT group, the duration of music-listening was 30 mins. The MP3 player was set to automatically turn off after 30 mins. The selected music had been judged as relaxing in prior studies. Classical pieces of music or contemporary instrumental music, including Air for G string, Allemande, Canon, Nocturne, and Swan were selected. Prior to music listening, a list of music pre-selected by the therapist was shown to each patient. If a patient opted not to listen to the music provided, another list of pre-selected music was

presented. If the patient requested to stop listening to music, they were withdrawn from the study.

No music listening

Length of follow-up:

2 postoperative days

Loss-to-follow-up & incomplete outcome data:

Intervention: 20/64 (31.3%)

Control: 34/79 (43.0%)

Analyzed:

I: 44

C: 45

Sleep quality (critical)

RCSQ; 5-items; 100-mm visual-analogue scale per item; higher scores representing better sleep

POD1

46.95 (4.12)

54.22 (4.08)

MD −7.27 (5.80)

P=0.4242

POD2

71.50 (3.72)

56.89 (3.68)

MD 14.61 (5.23)

P=0.012**

Sleep durationDuration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

No data provided;

“Patients’ outcomes including ICU/hospital stay, mortality, and delirium were not significantly different among the groups.”

Duration of hospital/ICU stay (important)

No data provided;

“Patients’ outcomes including ICU/hospital stay, mortality, and delirium were not significantly different among the groups.”

From:

Hu, 2015

Cochrane review

Sha 2013

2-arm, parallel group design RCT

ICU, cancer hospital of Tianjin Medical University, China

Lung cancer participants in ICU after thoracotomy;

N, mean age

Number randomized: 240 (120 in each group)

Number analysed: 112 in the control group, 107 in the intervention group

Mean age: 55.5 years

Sex: 146 men, 73 women

Groups comparable at baseline? yes

individualized music  intervention (12.30 p.m. to 1.30 p.m. and 8.30 p.m. to 9.30 p.m.)

usual care

End-point of follow-up:

more than 7 days

For how many participants were no complete outcome data available?

Less than 15% of participants were excluded (N = 21 - 13 in the intervention

group, 8 in the control group)

Sleep  quality (critical)

Pittsburgh Sleep Quality Index, day of the ICU discharge “In Sha 2013, the sleep quality, sleep duration, sleep efficiency, and total PSQI scores were significantly improved in the intervention group compared with the control group (P value < 0.05). Additionally, the incidence of sleep disorder in the music intervention group was significantly lower than that in the control group (P value = 0.036).”

Sleep duration

Duration of sleep

Not reported

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported in SR

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

From:

Hu, 2015

Cochrane review

Su 2013

2-arm, parallel group RCT

a 45-bed medical ICU, a 650-bed multispecialty teaching hospital located in Taipei, single-bed

rooms, Taiwan

N, mean age

Total number randomized: 28 (14 in both groups)

Total number analysed: 28

Mean age: 61.68 ± 9.82

Sex: 17 men, 11 women

Groups comparable at baseline? yes

45-minute music listening intervention

Music intervention consisted of 4 pieces of sedating piano music composed by 2 of the authors; the music was played on a Sony (CFD-S07CP) CD player

usual care without music

End-point of follow-up:

Not reported in SR

For how many participants were no complete outcome data available?

There were no losses

Sleep quality (critical)

Verran/Synder-Halpern Sleep Scale (VSH sleep scale) Chinese version; this VSH consists of 15 self-reported scales with each visual analogue scale running from 0 to 100 mm. The sum of the scores provided a global sleep quality score ranging from 0 to 1500

“Similarly, sleep quality was improved in participants receiving music intervention plus eye masks and earplugs versus usual care  (SMD 1.37, 95% CI 0.79 to 1.94; N = 58; Su 2013).”

Sleep duration

Duration of sleep

No differences found

EEG sleep parameters

Intervention group shorter stage 2 sleep duration and longer stage 3 sleep duration compared with control group.  “PSG sleep was recorded for the first 2 hours of the nocturnal sleep between the hours of 9.30 p.m. to 11.30 p.m. “One study examined the effects of listening to music (versus usual care) on PSG sleep variables in 28 ICU patients (Su 2013). The authors reported that participants in the music group had “shorter stage two sleep duration (P value = 0.014) and longer stage three sleep duration (P value = 0.008) in the first two hours of the nocturnal sleep as calculated by generalized estimating equation analysis. No statistically significant differences in the mean total sleep duration, sleep efficiency, and stage one sleep durations were reported between groups (P value > 0.05).”

Quality of life (critical)

Not reported in SR

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

Earplugs & eye masks & music

Hu, 2015

Type of study:

RCT

Setting and country:

21-bed Cardiac Surgical Intensive Care Unit (CSICU)

of Fujian Medical University Union Hospital, Fuzhou,

China

Funding and conflicts of interest:

“This study was partly funded by National Natural Science Foundation of China (81201500) and Ministry of National Education of China (11YJC190008).

The authors declare that they have no competing interests”

ICU patients; primary and elective cardiac surgery

Inclusion criteria:

•  primary and elective cardiac surgery;
• age ≧40 years;
• with normal liver, kidney and lung preoperative function
• without history of diabetes;
• no history of neurological or

psychiatric disorders;

• ability of patients to communicate verbally and understand the sleep questionnaires administered before surgery and after being transferred from the
• ICU;
• length of ICU stay ≥48 hours;
• (GCS) >10 in the first and second postoperative days;
• stable hemodynamics postoperatively

Exclusion criteria:

• :severe sleep disorder requiring daily treatment before surgery;
• patients with severe postoperative complications;
• presence of postoperative renal failure;
• presence of thoracic aortic dissection;
• postoperative unconsciousness, coma or delirium;
• cardiac valve replacement or congenital heart disease requiring sedation

and analgesics after surgery

N total at baseline:

Intervention: 25

Control:25

Analysed:

I: 20

C: 25

Important prognostic factors²:

For example

age ± SD:

I:56.6 ± 11

C:  56.8 ± 11.2

Sex:

I: 11/20 M

C: 16/25 M

Groups comparable at baseline?

earplugs and eye masks combined with relaxing music

After randomization, earplugs (3 M Corporation,

Beijing, China) and eye masks were provided 2 to 3 days before surgery and patients in the intervention group were asked to wear them. Meanwhile, the researcher explained to them that they should wear the earplugs and eye masks during their postoperative stay in ICU to ensure rest and instructed patients to use them properly. The patients chose from three types of eye mask provided. Providing the earplugs and eye masks preoperatively allowed patients to adapt to wearing earplugs and eye masks, and it also helped to play a role in establishing a time cue. During the postoperative ICU stay, ICU nurses assisted patients with wearing earplugs and eye masks from 9:0 pm every night until the next morning. Pieces of music for relaxing and implying time of day were collected and recorded on an MP3 player. Sounds of nature and bird songs were selected to imply morning. Sounds of frogs and waves were selected to imply evening. Pieces of classical music, including Blue Danube, Morning Song, Lofty Mountains and Flowing Water, Clouds Chasing the Moon, Lotus Emerging out of Water, and Moonlight Sonata, et cetera, were selected as relaxing music. Patients used earphones to listen to the corresponding music at 8:00 to 9:00 pm and 7:30 to 8:30 am every day after surgery. The duration of listening to music was 30 minutes. The music volume was set at a comfortable level for each participant. The MP3 music was supplied through earphones to the participants. Sometimes listening to music had to be stopped due to need for immediate care; when this occurred, the period of listening to music was shifted, although the range remained within the period of 9:00 pm and 8:30 am. During the night when care-givers needed to interact with the patients, whether or not the earplugs and eye masks were retained was left up to the nurses’ judgment, patients’ request and specific circumstances. For patients who did not like the music that was provided, we reselected other pieces of music for them on the basis of the requirement to relax the patients and help them sleep. Those who were strongly disinclined to listen to music were withdrawn from the study.

No earplugs and eye masks combined with relaxing music

 In the control

group, no interventions mentioned above were offered to

the patients and routine preoperative and postoperative

medical care was provided.

Length of follow-up:

2 days following transfer out of the ICU

Loss-to-follow-up:

Intervention: 5/25  (20%)

Reasons (describe: disinclined to listen to music)

Control:  0 (0%)

Incomplete outcome data:

Not reported

Analysed:

I: 20

C: 25

Sleep quality (critical)

Sleep quality: Chinese version of the Richards-Campbell sleep questionnaire (RCSQ). The original RCSQ had six items and evaluated aspects of night-time sleep including: (1) depth; (2) latency (time to fall asleep); (3) number of awakenings; (4) efficiency (percent of time awake); (5) quality; and (6) perceived night-time noise measured on a 100-mm visual-analogue scale (VAS) [29].

Depth

I: 26.7 ± 21.5

C: 55.5 ± 27.4 0.00

Latency (time to fall asleep)

I: 23.7 ± 17.4

C: 60.4 ± 25.9 0.00

Number of awakenings

I: 25.3 ± 16.2

C: 51.2 ± 26.7 0.00

Efficiency (percent of time awake)

I: 21.7 ± 20.9

C: 63.4 ± 21.9 0.00

Perceived quality

I: 23.7 ± 20.6

C: 54.0 ± 25.5 0.00

Perceived night-time noise

I: 25.0 ± 24.0

C: 40.2 ± 28.8 0.047

Sleep duration

Duration of sleep

Not reported

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Length of ICU stay, hours, mean ± SD

I: 53.0 ± 16

C: 58.9 ± 20

P=0.29

Length of hospital stay, days, mean ± SD

I: 20.7 ± 6.1

C: 22.6 ± 10.8

P=0.5

Nurse intervention

Diaz-Alonso, 2018

Type of study:

RCT

Setting and country:

 Asturias Central

University Hospital (Spain)

Funding and conflicts of interest:

“The authors have disclosed that they have no significant relationship with,

or financial interest in, any commercial companies pertaining to this article”

ICU following valvular cardiac surgery

Inclusion criteria:

• admitted to the Asturias Central University Hospital with prior notice between February and April 2016 for valve cardiac surgery
• with planned admission to the ICU,
• had postsurgical sedation with propofol,
• extubated prior to 24 hours after surgery

Exclusion criteria:

• cognitive state or clinical condition that would prevent patient from responding adequately to measuring instruments

N total at baseline:

Intervention: 24

Control:26

Important prognostic factors²:

For example

Age, n (%)

<65y:

C: 7 (35)

I: 5 (25)

65-74y:

C:  4 (20)

I:  8 (40)

+/>75y:

C:  9 (45)

I: 7 (35)

Sex:

I: 12/20 (60%) M

C: 12/20 (60%) M

Groups comparable at baseline? Yes

Nursing intervention; a brief intervention on

the previous day to surgery and ICU admission

The intervention consisted of a visit from a trained nurse who, for 10 to 15 minutes, explained to the patients those aspects with a potential to generate anxiety, stress, or discomfort during ICU hospitalization.

The intervention:

• was designed to anticipate those aspects of the unit’s environment usually perceived as alien and hostile by patients, such as electronic noises, intermittent lights, alarms, monitors, beds, ventilator, and infusion equipment.
• focused in explained the professionals’ assistance, including the common procedures and  communication difficulties with them and talking about the presumable sense of fear or uncertainty regarding clinical evolution.
• was designed with a clinical psychologist’s counseling and was based on Roy Adaptation Model. This model considers individuals as adaptive systems who develop adaptive behaviors in the presence of stressful events.15 
• attempted to anticipate the stressful stimulus for the patients in order to develop, with help of a nurse, functional adaptive behaviors. A set of photographs and videos was used to illustrate the environment and assistance in the ICU.

In order to gain consistency and to ensure the reproducibility of the interventions, a single nurse made all the presurgery visits.

Standard of care

Habitual care in this ICU involves some protocolized interventions to preserve sleep quality, including the attenuation of lights and alarms and the avoidance of unnecessary care tasks during the night.

Length of follow-up:

unclear

Loss-to-follow-up & incomplete outcome data:

I: 4/24

C: 6/26

I: N=20

C: N=20

Sleep quality (critical)

RCSQ; This is an interviewer-assisted questionnaire formed by 5-point visual analog scale scoring from

0 to 100 (highest sleep quality). It measures specifically sleep depth, frequency of awakenings, sleep latency, returning to sleep, and overall quality of sleep. As has been done in similar studies, a sixth visual analog scale concerning perception of noise was included.16-18 A RCSQ

total score was calculated averaging the sum of the 6 partial scores

RCSQ total, mean (SD)

I: 59.0 (27.2)

C:  63.0 (34.9)

P=.688

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

From:

Hu, 2015

Cochrane review

Li, 2011

coronary care

units, China

N, mean age

Number randomized: 52 (26 in both groups)

Mean age: 64 years

Sex: 29 men, 23 women

Groups comparable at baseline?

nursing intervention programme using the Roy

Adaptation Model as a guide

2 weeks

conventional care

End-point of follow-up:

Study duration: 2 weeks

For how many participants were no complete outcome data available?

(intervention/control)

Sleep quality (critical)

PSQI (0 = better sleep, 21 =worse sleep); Chinese version

“significantly higher sleep quality in the intervention group than in the control group (post-test mean = 5.57, SD = 2.62 versus post-test mean = 10.03, SD = 2.62) (P value < 0.05)

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

Not reported

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

Note of selective reporting / risk of bias

“The trial did not provide information about general characteristics before randomization between the groups or the baseline sleep scores”

Mou, 2020

Type of study:

RCT

Setting and country:

ICU of the Neurology

Department of The First Affiliated Hospital of

Xi’an Jiaotong University , China

Funding and conflicts of interest:
“Acknowledgements

This work was supported by the Fund for

“Protection of Edaravone on Intracerebral

Hemorrhagic Injury in Rats and Molecular

Mechanism” (No. 2016SF-053).

Disclosure of conflict of interest

None”

Inclusion criteria:

• 35-75 years old;
• patients firstly admitted to ICU;
• core family members of patients aged between 25 and 55 years old;
• informed consent  signed by patient or their families.

Exclusion criteria:

• needed surgical treatment,
• mental illness;
• severe cognitive impairment;
• coma or could not complete the questionnaire independently when
• severe dysfunction of important organs;
• mental diseases
• various acute and chronic diseases in core family members.

N total at baseline:

Intervention: 53

Control: 53

Important prognostic factors²:

For example

age ± SD:

I: 57.8±5.4

C: 58.2±5.9

Sex:

I: 26/53 M

C: 29/53 M

Groups comparable at baseline? Yes

systematic nursing

intervention

The study group was given systematic nursing intervention.

Firstly, nursing care for the patients: (1) Psychological counseling: Responsible nurses frequently communicated with patients and gave targeted psychological counseling according to patients’ psychological changes. They patiently told patients about successful cases to enhance patients’ self-confidence, and to alleviate patients’ bad psychological emotions. Besides, nurses patiently listened to patients’ real ideas to eliminate patients’ pessimism. (2) Sleep intervention: Nurses provided a good sleep environment for patients by adjusting indoor light, mattress, etc., carried out psychological counseling to relieve the ideological shackles of patients, and played soft music before going to bed to promote patients’ sleep.

(3) Early rehabilitation nursing: Responsible nurses helped patients turn over on time to avoid pressure sores and lift legs or joints passively to prevent venous thrombosis of lower limbs. Patients with language disorders were encouraged to speak aloud, carry out pronunciation training and communicate with nurses and core family members.

(4) Diet care: Patients had food with high protein, low salt and low-fat; with more meals a day but less food at each meal. For patients with dysphagia such as weak chewing, choking and other dysphagia, they were given nasogastric feeding accordingly.

Secondly, nursing care for core family members: (1) Establishing trust and mutual assistance relationships: Nurses allowed family members to enter the ICU ward and actively introduced themselves, to establish a trusting relationship. Nurses actively asked if there was any place where they needed help and promised to provide help. (2) Establishing care and expressing wishes: For comatose patients, nurses told families about the day of patients, such as the condition, vital signs, medication, diet and excretion, so that the family members can better understand the patient’s condition. Also, nurses expressed hope that the patient will recover as soon as possible [8].

routine nursing

The control group received routine nursing,

including strict disinfection of the ward and

24-hour monitoring of the changes of vital signs

of patients. To prevent infection, only a core family member of each patient could be allowed to enter the ICU ward and they were required to change clothes before entering.

Length of follow-up:

Until discharge

(Length of stay in ICU

I: 18.9±2.8

C: 19.4±3.0)

Loss-to-follow-up & incomplete outcome data

I: 1/53

C: 3/53

Reason unknown

Sleep quality (critical)

Pittsburgh sleep quality index (PSQI) on the next day after hospitalization and the day of discharge, respectively, to evaluate the sleep quality of patients before and after the intervention, with a total of 18 points. The scale was collected on the spot after filling out, and the recovery rate was 100%.

N=53 and 53

I: 7.78±2.35

C: 10.95±2.08

Sleep duration

Duration of sleep

EEG sleep parameters

Not reported

Quality of life (critical)

health status questionnaire

(SF-36);  on the next day after hospitalization

and the day of discharge,

contains eight dimensions;

higher score, better quality of life

N: I: 52, C: 50

Physiological function

I: 67.70±6.69

C: 64.40±5.86

Social function

I: 72.10±6.30

C: 68.98±6.76

Energy

I: 65.50±6.05

C: 59.97±4.50

Physiological function

I: 70.07±5.40

C: 66.70±6.04

Emotional function

I: 81.10±4.30

C: 75.50±5.55

Mental health

I: 70.08±4.30

C: 64.40±5.40

Physical pain

I: 73.30±4.07

C: 66.60±5.44

General health

I: 67.78±5.66

C: 64.40±5.55

Delirium (important)

Not reported

Duration of hospital/ICU stay (important)

Not reported

Hu, 2015

Cochrane review

Yes

Yes

Yes

Yes

Not applicable

Yes

Yes

Yes

Yes

Study reference

(first author, publication year)

Was the allocation sequence adequately generated?

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure

LOW

Some concerns

HIGH

Lin, 2021

Definitely yes;

Reason: draw a lot at random from an

opaque jar

Definitely yes;

Reason: The first author was responsible to oversee the enrollment and treatment allocation process

Probably yes;

Reason: The first author was responsible to oversee the enrollment and treatment allocation process, and the researchers who

performed the analyses or performed the HPLC for the urinary MT6s were blinded to the randomization.

No information about the health care providers and data collectors

Definitely yes;

Reason: All randomized patients included in analysis

Probably yes;

Reason: All relevant and announced outcomes reported; registered: ISRCTN94971645

Probably yes

Reason: A per-protocol analysis was used for analysing the data; all patients treated as randomized

LOW

Litton, 2017

Definitely yes

Reason:  The randomisation sequence  was generated by an online randomisation resource (http://

www.randomization.com)”

Probably yes

Reason: “Allocation concealment was

maintained by using permuted block randomisation and

sealed, opaque, consecutively numbered envelopes.” No further details

Probably no

Reason: “By

necessity, the study was open label.”

No further blinding of outcome assessor or data analysts reported

Definitely yes

Reason: all patients included in analysis, no lost to FU

Probably yes

Reason: registered

ACTRN12615001125516

Probably yes

Reason: no other info; funding not reported

LOW

Mahran, 2020

Probably yes

Reason: “Treatment assignment codes were randomly generated using a computerized random number generator and concealed from the researcher. A study coordinator screened and enrolled participants using

sequentially numbered, sealed opaque envelopes.”

Probably yes

Reason: “Treatment assignment codes were randomly generated using a computerized random number generator and concealed from the researcher. A study

coordinator screened and enrolled participants using

sequentially numbered, sealed opaque envelopes.”

Probably yes

Reason: “Although it was not possible to blind participants and

clinical staff members to the assigned interventions,

the assessments were undertaken by a researcher who

was blinded to group allocation”

Probably no

Reason: 4/35 patients in the intervention group did not want to wear eyemask (11% of group)

Probably yes

Reason: registered NCT03213860

Probably no

Reason: funding not reported; no intention-to-treat analysis performed  (4/35 patients in intervention group did not want to wear eyemask, not included in analysis)

Some concerns

Reason: 11% of intervention group did not receive intervention and was excluded from analysis

Akpinar, 2021

c

Reason: “were included in the study and

then assigned to the experimental and control groups via the block randomization method. A research team member not involved in either

subject recruitment or the intervention processes prepared the random

assignment schedule, which was generated by the computer software

Research Randomizer.”

Definitely yes

Reason: “Group assignments were placed in sealed envelopes and revealed sequentially at the time of randomization.”

Probably no

Reason: not described; patient blinding not possible

Definitely yes

Reason: all allocated patients included in analysis

Probably yes

Reason: announced outcomes reported; trial not registered

No information

Reason: funding and conflicts described; unclear whether intention-to-treat protocol was used and whether data was incomplete

Some concerns

Reason: no registration, blinding not reported; protocol for missing data and analysis unclear

Obanor, 2021

No information

Reason:

“patients were randomly

assigned in a 1:1 ratio”

No information

Reason:

“patients were randomly

assigned in a 1:1 ratio”

Probably no

Reason: impossible to blind patients; other blinding not described

Definitely yes

Reason: all allocated patients included in analysis

Probably yes

Reason: registered

NCT03957317

No information

Reason: no other info; funding not reported

Some concerns

Reason: randomization, allocation concealment and blinding not reported

Leong, 2021

Definitely yes

Reason: “A randomisation sequence by permuted blocks of 4 and 6 with an allocation

of 1:1 was generated by a computer programme and concealment was maintained by an independent investigator with no clinical involvement in the trial.”

Probably yes

Reason: “A randomisation sequence by permuted blocks of 4 and 6 with an allocation

of 1:1 was generated by a computer programme and concealment was maintained by an independent investigator with no clinical involvement in the trial.”

Probably yes

Reason: patients and care providers not blinded; outcome assessor blinded (except for satisfaction but that outcome is not included in this summary); data analyst blinded

Definitely yes

Reason: all allocated patients included in analysis

Probably yes

Reason: registered

NCT03702296

Probably no

Reason: funding and conflicts described; intention-to-treat protocol was used for primary analysis but unclear which results come from primary analysis; part of results possibly retrospectively assessed

Some concerns

Reason: missing data possibly retrospectively assessed

Demoule, 2017

Definitely yes

Reason: Randomization was performed using a computer-generated sequence provided through a website

No information

Probably no

Reason: impossible to blind patients; other blinding not described except for the objective sleep measures

Definitely no

Reason: high drop out due to technical problems (ITT analysis; drop out higher in intervention group) and additionally to not adhering to protocol (per protocol analysis)

Probably yes

Reason: NCT02292134. Registered on 21 Nov 2013.

Probably yes

Reason: no other info; funding not reported

Some concerns

Reason: allocation concealment, missing data, high rate of excluded patients

Hu, 2015

No information

Reason: “Patients were randomly assigned to two different groups

using the closed-envelope method.”

Probably yes

Reason: “Patients were randomly assigned to two different groups

using the closed-envelope method.”

Probably no

Reason: impossible to blind patients; other blinding not described

Definitely no

Reason: 20% (5/25) patients excluded from intervention as they did not want to listen to music

Probably yes

Reason: registered Chinese Clinical Trial Registry: ChiCTR-IOR-14005511.

Definitely no

Reason: no intention to treat protocol for analysis although high drop out was present in intervention group due to subjective reason (not wanting to listen to music)

HIGH RISK of bias

Reason: method of randomization unclear, no blinding, high drop out rate in intervention group and no analysis according to ITT

Hansen, 2018

No information

Reason: “Participants were randomly assigned

to an intervention group or a control group

using block randomization´

No information

Probably no

Reason: “The researchers collecting data and performing statistical analysis were not blinded

to which subjects were assigned to the

intervention or control group, respectively.

In addition, it was not possible to blind

the participants due to the nature of the

intervention.”

Probably no

Reason: 5/42 (11.9%) excluded after inclusion; reasons not explained

Probably yes

Reason: announced outcomes reported; trial not registered

No information

Reason: no other info; funding and conflicts of interest not reported

HIGH RISK of bias

Reason: method of randomization and allocation concealment unclear, no blinding, unclear why 11% of patients were excluded after inclusion and before randomization; funding and conflicts of interest not reported

Kim, 2020

Definitely yes

Reason: Patients were randomly allocated to three groups (patient-directed interactive music therapy group, passive listening group, control group) by an investigator using a reproducible set of computer-generated random numbers via a web-based system (REDCap).

Definitely yes

Reason: … reproducible set of computer-generated random numbers via a web-based system (REDCap). The allocation was thus concealed until the last minute of the process of obtaining consent.

Probably no

Reason: blinding not described

Definitely no

Reason: 33.5% of randomized patients not included in follow-up analysis

Probably no

Reason: registered

NCT03156205; results not reported separately for both intervention groups; passive listening seems to be pooled with the control group or the total number of patients in the analysis do not add up.

Definitely no

Reason: no intention to treat protocol for analysis although high drop out was present in all groups;  patients that did not want to proceed the intervention, were withdrawn from the study.

HIGH RISK of bias

Reason: no blinding reported, 33.5% lost to FU, no analysis according to ITT

Mou, 2020

No information

Reason: “… were recruited and randomly

divided into a study group and a control group”.

No information

Probably no

Reason: blinding not described

Probably  yes

Reason: almost all allocated patients included in analysis

Probably yes

Reason: announced outcomes reported; trial not registered

Probably yes

Reason: not all outcome data available for all patients; reason unclear

Some concerns

Reason: no information about randomization, allocation concealment, blinding and handling missing data; no protocol available

Diaz-Alonso, 2018

No information

Reason: “ Finally,

40 patients were included and were randomly assigned to

the 2 studied groups (20 for CG and 20 for EG).”

No information

Probably no

Reason: patient and care provider not blinded, outcome assessor blinded (“ an independent nurse

performed the postintervention measures in the ICU, to reduce bias in data collection.”); data analyst unknown.

Probably yes

Reason: 4/24 and 6/26 lost to FU; lost or intubated

Probably yes

Reason: announced outcomes reported; trial not registered

Probably yes

Reason: number of patients lost to FU and reasons comparable between treatment groups

Some concerns

Reason: no information about randomization, allocation concealment; blinding

Borji, 2017

No information

Reason: The patients were randomly assigned into

experimental and control groups with 30 individuals each.

No information

No information

No information

Probably yes

Reason: announced outcomes reported; trial not registered

Probably yes

Reason: number of patients lost to FU and reasons comparable between treatment groups

High

No information was provided about randomisation, concealment of allocation and blinding. The study was quasi-experimental by design.