Family Practice, 2018, Vol. 35, No.

4, 365–375
doi:10.1093/fampra/cmx122
Advance Access publication 29 November 2017

Systematic Review

Sleep hygiene education as a treatment

of insomnia: a systematic review and
meta-analysis

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

Ka-Fai Chunga,*, Chit-Tat Leeb, Wing-Fai Yeungc, Man-Sum Chand,
Emily Wing-Yue Chunge and Wai-Ling Linf
a
Department of Psychiatry, University of Hong Kong, Hong Kong SAR, China, bDepartment of Psychiatry, Queen Mary
Hospital, Hong Kong SAR, China, cSchool of Nursing, Hong Kong Polytechnic University, Hong Kong SAR, China,
d
Maternal Mental Health, Waitemata District Health Board, Takapuna, Auckland, New Zealand, eChinese University
of Hong Kong, Hong Kong SAR, China, fHong Kong Institute of Integrative Medicine, Chinese University of Hong Kong,
Hong Kong SAR, China

*Correspondence to Ka-Fai Chung, Department of Psychiatry, University of Hong Kong, Pokfulam, Hong Kong SAR, China;
E-mail: kfchung@hku.hk

Abstract
Background. Sleep hygiene education (SHE) is commonly used as a treatment of insomnia in
general practice. Whether SHE or cognitive-behavioural therapy for insomnia (CBT-I), a treatment
with stronger evidence base, should be provided first remains unclear.
Objective. To review the efficacy of SHE for poor sleep or insomnia.
Methods. We systematically searched six key electronic databases up until May 2017. Two
researchers independently selected relevant publications, extracted data and evaluated
methodological quality according to the Cochrane criteria.
Results. Twelve of 15 studies compared SHE with CBT-I, three with mindfulness-based therapy, but
none with sham or no treatment. General knowledge about sleep, substance use, regular exercise
and bedroom arrangement were commonly covered; sleep-wake regularity and avoidance of
daytime naps in seven programs, but stress management in only five programs. Major findings
include (i) there were significant pre- to post-treatment improvements following SHE, with small
to medium effect size; (ii) SHE was significantly less efficacious than CBT-I, with difference in
effect size ranging from medium to large; (iii) pre- to post-treatment improvement and SHE-CBT-I
difference averaged at 5% and 8% in sleep-diary-derived sleep efficiency, respectively, and two
points in Pittsburgh Sleep Quality Index; (iv) only subjective measures were significant and (v) no
data on acceptability, adherence, understanding and cost-effectiveness.
Conclusions. Although SHE is less effective than CBT-I, unanswered methodological and
implementation issues prevent a firm conclusion to be made on whether SHE has a role in a
stepped-care model for insomnia in primary care.

Key words: Sleep hygiene education, cognitive-behavioural therapy, psychological intervention, systematic review,
meta-analysis, insomnia.

© The Author(s) 2017. Published by Oxford University Press. All rights reserved.
365
For permissions, please e-mail: journals.permissions@oup.com.
366 Family Practice, 2018, Vol. 35, No. 4

Introduction June 2015 were searched without language restriction using the
search terms: (sleep hygiene OR sleep education OR sleep health)
Insomnia is a highly prevalent condition that is associated with sub-
AND (random* OR controlled trial OR clinical trial OR RCT)
stantial distress, psychosocial impairment and medical and psychi-
AND (sleep OR insomnia OR dyssomnia) in titles or abstracts. An
atric morbidity (1). Patients with sleep problems consult their general
updated search was conducted in June 2017 for publications up to
practitioners more frequently than other health professionals (2) and
31 May 2017. Reference lists of the included studies and relevant
typically prefer non-pharmacological treatments (3), among which
reviews were examined for additional articles. As a forward search,
sleep hygiene education (SHE) is the most commonly used (4). The
we used the Ovid MEDLINE to identify all papers that have cited
term ‘sleep hygiene’ was first used by Peter Hauri in 1977 in the con-
the included studies.
text of providing recommendations for patients with insomnia (5,6).
The list of sleep hygiene recommendations was updated in 1991 (5),
and many versions are now available (7). In a recent review (8), Irish Study selection
et al. reported that caffeine, tobacco and alcohol use, exercise, stress, Studies included in this review are randomized controlled trials
noise, sleep timing and daytime napping are the areas commonly cov- that examined participants with a complaint of poor sleep or in-
ered during SHE. Whether SHE should be given priority for treating somnia who received SHE in comparison with no treatment, rou-

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

insomnia remains a controversy due to its low cost and easy avail- tine care, placebo or sham treatment or any forms of psychological,
ability. A review paper published in the American Family Physician pharmacological, complementary or alternative medicine treatment.
(9) placed SHE equivalent to cognitive-behavioural therapy for Supplementary Table 1 presents the population intervention com-
insomnia (CBT-I), but the recommendation was based on consensus parison outcome (PICO) protocol. SHE was defined as any advice
and usual practice (Strength-of-Recommendation Taxonomy grade provided to patients with an intention to help their sleep without
C), while the American Academy of Sleep Medicine Report in 2006 any elements of CBT-I (including stimulus control, sleep restriction,
(10) and a clinical guideline published in the Journal of Clinical Sleep relaxation training and cognitive therapy) or other complementary
Medicine in 2008 (11) did not support SHE as a single therapy due and alternative medicine components (e.g. Taichi, qigong, massage,
to insufficient evidence (No recommendation level). Recent literature acupressure). We did not set any specifications for delivery modality,
further supports the effectiveness of CBT-I, e.g. an evidence report by treatment content and duration, outcome measure or study quality.
the American College of Physicians considered CBT-I as an effective Two investigators selected relevant publications independently
intervention for insomnia disorder (moderate-strength evidence) (12) according to the eligibility criteria. Any disagreement was resolved
and the Australasian Sleep Association guideline placed CBT-I as a by thorough discussion and consultation with the senior author
first line treatment (Level I evidence from meta-analyses) (13). (KC). When a study had more than one patient group (e.g. one group
General practitioners seldom conduct CBT-I or refer patients of primary insomnia and another group of comorbid insomnia), we
with insomnia for psychological treatment (14,15). Although verbal considered it twice as two different comparisons. When the same
advice and a sleep hygiene sheet are often used (14), they are seen group of authors published more than one article using data from
to be insufficient to address the sleep problem by most general prac- the same group of subjects, we considered it as one set of comparison
titioners (15). A stepped-care model has been proposed by Espie as and used the largest dataset that was available.
a solution to the high demand of CBT-I services (16). The model is
often conceptualized as a pyramid, of which high patient volume Data extraction and quality assessment
is managed at the base of the pyramid using low intensity treat- One investigator extracted the data and another checked the extracted
ments, e.g. self-help CBT-I, with progressively smaller volumes and data. For each study, the following variables were extracted: study
greater expertise in assessment and treatment towards the top step. design, subjects’ characteristics including age, gender, duration and
Although self-help CBT-I has a strong evidence base for its effective- diagnosis of insomnia, components and procedure of SHE, com-
ness (17), it contains more information and may be harder to under- parison intervention and outcome parameters. Primary outcome
stand than sleep hygiene recommendations; hence is worthwhile to was sleep questionnaire score, but other outcomes, such as sleep
explore whether SHE can be a starting point for the treatment of diary, actigraphy and polysomnography-derived variables were also
insomnia. recorded if available. We analyzed the quality of studies using the
To our knowledge, there has been no systematic review on SHE. Cochrane’s risks of bias assessment (19), which has six domains:
The last review was published in 2003 and did not follow system- random sequence generation, allocation concealment, blinding of
atic protocol (7). Since SHE is commonly used in healthcare settings participants, personnel and outcome assessors, incomplete outcome
and many studies may have been published on SHE, the aim of this data, selective outcome reporting and other sources of bias. The rat-
systematic review and meta-analysis is to examine whether SHE is ings of each domain can be ‘yes’ (low risk of bias), ‘no’ (high risk of
an effective treatment and how SHE compares to CBT-I and other bias) or ‘unclear’ (uncertain risk).
forms of treatments for insomnia.
Data synthesis and analysis
We used the Comprehensive meta-analysis software version 3.0 for
Method
statistical analysis. The summary measures were the mean difference
Literature search and its 95% confidence interval (CI) and effect size, calculated as
The meta-analysis was conducted with reference to the preferred Hedges’s g. We analyzed the pre- to post-treatment improvements
reporting items for systematic reviews and meta-analyses (PRISMA) and between-group differences in outcomes. Due to differences in
(18). The protocol was registered at the International prospect- demographic characteristics and inclusion and exclusion criteria be-
ive register of systematic reviews (CRD42015024995). The Ovid tween studies, it was expected that there was heterogeneity a priori;
MEDLINE, EMBASE, CINAHL plus, PsycINFO and Dissertation & hence the random-effects model and inverse-variance method were
Thesis A&I and Cochrane Library from inception through 30 employed to calculate summary estimates (20). Heterogeneity was
Sleep hygiene education for insomnia 367

evaluated using the Cochrane’s Q statistic, with P value < 0.10 indi- Results
cating significant heterogeneity. The I2 statistic was computed as a
Identification of studies
compliment to the Q statistic. As suggested by Higgins et al. (21),
I2 of 0%, 25%, 50% and 75% indicate no, low, moderate and high Figure 1 presents the flowchart of the systematic review. A total of
heterogeneity, respectively. If there were 10 or more studies in a com- 2361 entries were included for title and abstract screening and 133
parison, publication bias would be examined by visual inspection papers were selected for full-text screening. Fifteen studies met the
of the funnel plot, which is a scatterplot of treatment effect against eligibility criteria and were included in this review (23–37).
sample size. Sensitivity analysis was performed using the leave-one-
out method in order to investigate the influence of outlying stud- Overview of the included studies
ies on the synthesized effect size in the random-effects model (22). Table 1 summarizes the characteristics of the 15 included studies.
Subgroup analyses were performed to determine the impact of in- Sample size was typically small, ranging from 20 to 159, with a
somnia nature (primary versus comorbid), delivery modality (in-per- total of 1194 subjects. About 52.8% were female and the mean age
son versus printed material) and the number of SHE sessions (1–2 was 65.6 years. CBT-I was the most common comparator (n = 12),
versus ≥ 3). The chosen factors were considered having potential im- followed by mindfulness-based treatment (n = 3), while no studies

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

pact on treatment outcome. compared SHE with placebo or sham treatment, treatment as usual,

Figure 1. Selection of trials for inclusion in the review (+ indicates updated review).
Table 1. Characteristics of included studies of a systematic review and meta-analysis of sleep hygiene education as a treatment of insomnia (up to May 2017)
368

First author, Country/type of case Source of Mean age, Diagnostic Study Baseline Treatment Follow-up Sample size Major Results Risks of biasa
year participants year/% female criteria design severity, mean duration outcomes reported
score

Comorbid insomnia
 Edinger USA/PI and CMI Postal survey PI: RDC and Two-parallel PI: PSQI: 8 weeks Immed and 81 PSQI, DBAS, PI and CMI: U, U, L, L, L,
et al. (23) and physician 54.2/12.5% DSM-IV arms (CBT-I; CBT- 6 months Sd, actig CBT-I > SHE L, L
referral CMI: SHE) I: 11.0/ SHE: in Sd-SOL,
54.2/14.6% 11.6 CMI: WASO,
PSQI: CBT-I: TST and SE
13.7/ (immed) and
SHE: 12.0 Sd-WASO
and TST
(6 months)
 Epstein USA/breast cancer Mass media 58.2/100% Sd-SOL/ Two-parallel Sd-SE: CBT-I: 6 weeks 2 weeks 72 Subj rating, CBT-I > SHE L, U, H, L,
et al. (24) survivors with in- and support WASO ≥ arms (CBT-I; 69.0/ SHE: Sd, actig in subj rat- L, L, L
somnia > 3 months group 30 min for SHE) 72.2 ings on SOL,
≥ 3 nights/ WASO, TST
weeks and quality
of sleep.
 Martínez Spain/25–60 years Specialized 47.6/100% DSM-IV Two-parallel PSQI: CBT-I: 6 weeks Immed, 3 and 59 PSQI CBT-I > SHE L, U, H, L,
et al. (25) with fibromyalgia clinic arms (CBT-I; 15.3/ 6 months in PSQI L, L, L
SHE) SHE: 14.9
 Nakamura USA/cancer Specialized MBB: MOS-SS ≥ 35 Three- MOS-SS: 3 weeks Immed and 57 MOS-SS MBB and L, U, H, U,
et al. (26) survivors clinic, mass 55.4/68.4% parallel arms MBB: 58.0/ 2 months MM > SHE L, L, L
media and MM: (MBB; MM; MM: 63.3/ in MOS-SS
support 50.8/80.0% SHE) SHE: 54.9
group SHE:
51.6/77.8%
Insomnia disorder
 Alessi et al. USA/ >60 years Postal 72.2/3.1% ICSD Two-parallel PSQI: CBT-I: 6 weeks 1 week, 159 PSQI, ISI, Sd, CBT-I > SHE L, L, L, L, L,
(27) with insomnia > arms (CBT-I; 9.4/ 6 months and actig in Sd-SOL, L, L, L
3 months SHE) SHE: 8.3 12 months TWT and SE,
PSQI and ISI.
 Bjorvatn Norway/insomnia > Mass media 50.0/58% BIS Two-parallel PSQI: CBT-I: 3 months 3 months 155 BIS, PSQI CBT-I > SHE U, U, H, U,
et al. (28) 6 months and website arms (CBT-I; 12.9/ in BIS, PSQI L, L, L
SHE) SHE: 12.8 and DBAS
 Black et al. USA/ >55 years with Mass 66.3/67% PSQI >5 Two-parallel PSQI: MAP: 6 weeks 10 weeks 49 PSQI, AIS MAP > SHE L, L, H, U,
(29) insomnia media and arms (MAP; 10.2/ in PSQI and L, L, L
community SHE) SHE: 10.2 AIS
Family Practice, 2018, Vol. 35, No. 4

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

Table 1. Continued

First author, Country/type of case Source of Mean age, Diagnostic Study Baseline Treatment Follow-up Sample size Major Results Risks of biasa
year participants year/% female criteria design severity, mean duration outcomes reported
score

 Dawson USA/adults with NR 53.6/68% NR Two-parallel ISI: CBT-I: 4 weeks Immed and 87 ISI CBT-I > SHE U, U, U, U,
et al. (30) insomnia arms (CBT-I; 17.0/SHE: 3 months in ISI U, U, U, U
SHE) 19.0
 Falloon New Primary care 53.5/77.3% Insomnia > 3 Two-parallel PSQI: SSR 2 weeks 3 and 97 PSQI, ISI, Sd, SSR+SHE > L, L, L, H, L,
et al. (31) Zealand/16–75 years nights/weeks arms 10.4/SHE: 6 months actig SH in PSQI, L, L
with insomnia (SSR+SHE; 10.3 ISI and
> 6months SHE) act-SE
 Gellis USA/psychology University NR/64.7% ISI ≥ 8 and Two- ISI: NR 1 months 51 ISI CRT+SHE > L, U, H, U,
et al. (32) undergraduates WASO/SOL parallel arms CRT+SHE: SHE in ISI L, L, L
Sleep hygiene education for insomnia

> 30 min ≥ 3 (CRT+SHE; 15.3/SHE:

nights/weeks SHE) 16.8
>1 months
 McCrae USA/ ≥ 65 years Physician 77.2/65% ICSD, Two-parallel Sd-SE: MBT: 2 weeks 2 weeks 20 Sd MBT > SHE U, U, H, U,
et al. (33) referral and DSM-IV, and arms (MBT; 72.5/SHE: in Sd-SOL L, L, L
advertisement SOL/WASO SHE) 76.8 and SE
≥ 31 min ≥ 3
nights/weeks
> 6 months

 Nishinoue Japan/Dayshift IT company 31.3/14.2% PSQI ≥ 6 in Two- PSQI: 1 week 3 months 127 PSQI MBT + SHE U, U, H, U,
et al. (34) office workers 62.2% of parallel arms MBT+SHE:6.9/ > SHE in L, L, L
subjects (MBT+SHE; SHE: 6.3 PSQI
SHE)
 Sun et al. China/ ≥ 65 years Community 69.7/74.7% PSQI >5 Two-parallel PSQI: 4 weeks 3, 6 and 75 PSQI, ESS Rel+SHE > L, U, H, U,
(35) arms Rel+SHE: 9.6/ 12 months SHE in PSQI L, L, L
(Rel+SHE; SHE: 9.5 and ESS
SHE)
 Waters USA/18–59 years Mass media 45.6/79.3% SOL/WASO Two-parallel NR 2 weeks Immed 26 PSG, Sd CBT-I > SHE U, U, H, U,
et al. (36) > 30 min ≥ 4 arms (CBT-I; in Sd-SOL L, L, L
nights/weeks SHE) and WASO
> 1 month
 Wang China/18–65 years Psychology 41.2/54.4% Insomnia and Two-parallel PSQI: CBT-I 4 weeks 4 weeks 79 PSQI, ISI CBT-I > SHE L, L, H, L, L,
et al. (37) clinic hypnotics use arms (CBT-I; 10.4/ in PSQI and L, L
> 1 month SHE) SHE: 10.1 ISI

>, is significantly more effective than; actig, actigraphy; AIS, Athens insomnia scale; BIS, Bergen insomnia scale; CBT-I, cognitive behavioural therapy for insomnia; CMI, comorbid insomnia and psychiatric disorders; CRT,
cognitive refocusing treatment; DSM, diagnostic and statistical manual of mental disorders; H, high; ICSD, international classification of sleep disorders; immed, immediate; ISI, insomnia severity index; L, unclear; MAP,
mindful awareness practice; MBB, mind body bridging program; MBT, multicomponent behavioural treatment; MM, mindfulness meditation; MOS-SS, medical outcome study sleep scale; NR, not reported; Rel, relaxation
training; PI, primary insomnia; PSG, polysomnography; PSQI, Pittsburgh sleep quality index; Sd, sleep diary; SE, sleep efficiency; SHE, sleep hygiene education; SOL, sleep onset latency; SSR, simplified sleep restriction; TST,
total sleep time; U, unclear; WASO, wake after sleep onset.
a
Risks of bias assessment using Cochrane’s criteria in random sequence generation, allocation concealment, blinding of participant and personnel, blinding of outcome assessors, incomplete outcome data addressed,
selective outcome reporting and other sources of bias.
369

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

370 Family Practice, 2018, Vol. 35, No. 4

complementary and alternative medicine therapy or no treatment. The pre- to post-treatment difference in actigraphy variables was
Subjects were recruited through multiple sources, and the crite- not significant. The leave-one-out sensitivity analysis found that the
ria used for diagnosis of insomnia varied between studies. There significant finding in PSQI and ISI was still present when an outlying
were also great differences in subject characteristics. Four studies study was removed. Funnel plot was not performed due to the small
included only older adults, while two studies were on cancer sur- number of studies.
vivors, one on university students and one on patients with fibro-
myalgia. The most common outcome measure was Pittsburgh Sleep Between-group difference
Quality Index (PSQI), which is a 19-item self-rated questionnaire Pooled analyses showed that CBT-I was significantly more effective
for evaluating subjective sleep quality over the past month (38). than SHE in terms of SOL, WASO, SE, PSQI and ISI, but no signifi-
The PSQI score ranges from 0 to 21 with a score of 5 or above cant difference in actigraphy measures (Table 3). There was moder-
being suggestive of poor sleep and an improvement of three points ate heterogeneity between studies in PSQI, but the significant finding
or more has been used to define treatment response (39). The other was still present when outlying studies were removed. The between-
commonly used outcome measures are sleep diary variables. Sleep group effect size was medium for SOL, WASO and SE (0.48–0.67)
efficiency (SE) is a summary index of sleep diary variables; a SE and medium to large for PSQI and ISI (0.67–0.92). CBT-I was more

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

<85% represents poor sleep and an improvement ≥10% is sug- effective than SHE for improving SOL by 11 min, WASO by 14 min,
gestive of treatment response (39). Both PSQI and sleep diaries are SE by 8%, PSQI by two points and ISI by four points. Forest plots of
well-established assessments of sleep and insomnia (40). Objective SE and PSQI are presented in Figure 2c and d. Pooled analyses also
measures are rarely used. Only three studies used actigraphy and found that mindfulness-based therapy produced greater improve-
one study used polysomnography. Baseline insomnia severity was ment in PSQI than SHE, but only two studies were available for
mild to moderate, as indicated by a mean PSQI score ranging from 6 analysis (Hedges’s g = 1.13, CI = 0.64, 1.62, P < 0.001).
to 15 across studies. Eight of the 15 studies had only one follow-up,
which was arranged at immediate post-treatment or up to 3-month Subgroup analyses
post-treatment. There were no significant differences in subgroup analysis of the
impact of insomnia nature (primary versus comorbid), delivery
Description of SHE modality (in-person versus printed material) and the number of SHE
The number of sessions of SHE ranged from 1 to 6, with a median sessions (1–2 versus ≥ 3) on PSQI and SE (Supplementary Table 2).
of three sessions (Table 2). Six studies used group approach, five
studies used individualized approach, and four studies used printed
material. General knowledge about sleep architecture, substance use, Discussion
regular exercise and bedroom arrangement were commonly covered Our study showed that SHE was associated with sleep improvements,
during SHE, followed by sleep-wake regularity and avoidance of based on significant pre- to post-treatment changes, but it was less
daytime naps in seven programs, and stress management in five pro- effective than CBT-I and mindfulness-based therapy. Within-group
grams. Ten studies mentioned the use of a standardized manual, 10 improvements and between-group differences were shown only in
studies provided therapist training, 8 studies had therapist supervi- subjective measures. Subgroup analyses could not detect any impact
sion and 5 studies had treatment fidelity monitoring. of comorbid insomnia, delivery modality and the number of sessions
on outcomes. The overall finding seems to suggest that CBT-I is more
Assessment by the Cochrane’s risk of bias effective than SHE for the treatment of insomnia. However, there are
uncertainties in the finding due to methodological problems in stud-
assessment
ies comparing SHE and CBT-I and practical and cost-effectiveness
Results are shown in Table 1. Blinding of participants and person-
issues regarding the implementation of CBT-I. A recommendation to
nel was most difficult, with 11 of the 15 studies having a high risk
abandon using SHE in primary care cannot be made with certainty.
of bias. Allocation concealment was also unclear in 11 of the 15
A systematic review found that psychological placebo in the form
studies, while blinding of outcome assessors was unclear in 9 of the
of sham procedure had small pre- to post-treatment effect sizes on
15 studies. The risk of bias due to incomplete or selective outcome
sleep diary measures (0.12 to 0.36) and a moderate effect size on
reporting and other sources of bias were low in all studies, except the
subjective sleep quality (0.52) (41). Our study showed that the pre-
study by Dawson et al. (30).
to post-treatment effect sizes of SHE were quite similar to psycho-
logical placebo. If treatment response was defined as an improvement
Efficacy assessment in PSQI by 3 points or SE by 10% (39), the pre- to post-treatment
Within-group difference improvement following SHE was not up to the level.
Table 3 presents the within-group meta-analyses on subjective and Compared to CBT-I, SHE was shown to be significantly less effi-
objective measures. Forest plots on sleep-diary-derived SE and cacious. The difference in effect size was medium to large, depending
PSQI are shown in Figure 2a and b. Supplementary Figures S1–S17 on the outcome measures. In terms of native units, CBT-I outper-
present the forest plots of other variables. Other than PSQI and formed SHE in SE by 8% and PSQI by two points. Although most
Insomnia Severity Index (ISI), there was no significant heterogen- of the included studies used standardized manuals and had therap-
eity between studies. There were significant pre- to post-treatment ist training and supervision, only five studies had treatment fidelity
improvements in sleep-diary-derived sleep onset latency (SOL), wake monitoring. It remains unclear whether the efficacy of SHE can be
after sleep onset (WASO), total sleep time (TST) and SE, PSQI and enhanced by treatment fidelity monitoring and a more comprehen-
ISI. The within-group effect size was small for sleep diary variables sive coverage of sleep hygiene recommendations.
(0.23–0.35) and medium for PSQI and ISI (0.51–0.67). In their If SHE was introduced as an entry-step treatment for insomnia
native units, SOL was improved by 5 min, WASO by 12 min, TST in primary care, a standardized and comprehensive SHE package
by 25 min, SE by 5%, PSQI by two points and ISI by three points. should be developed, instead of information leaflets alone. Due to
Table 2. Summary of sleep hygiene education program of a systematic review and meta-analysis of sleep hygiene education as a treatment of insomnia (up to May 2017)

First author, No. of Group (Gp)/ Therapist Duration of Avoid Avoid Avoid Regular Manage Reduce Sleep time Avoid General sleep
year sessions individual sessions (min) caffeine nicotine alcohol exercise stress bedroom regularity daytime knowledge
(Ind) noise naps

Comorbid insomnia
 Edinger 4 Ind Clinical 30–60 √ NR √ √ NR √ NR NR √
et al. (23) psychologist
 Epstein 4 Gp Psychiatric nurse 60–120 √ √ √ √ √ √ √ √ √
Sleep hygiene education for insomnia

et al. (24)
 Martinez 6 Gp Clinical 90 √ √ √ √ NR √ √ √ √
et al. (25) psychologist
 Nakamura 3 Gp Social worker 90 √ √ √ √ √ √ √ √ NR
et al. (26)
Insomnia disorder
 Alessi 5 Gp Master’s degree 60 NR NR NR √ NR √ √ √ √
et al. (27) non-clinician
 Bjorvatn 1 Printed NA NA √ √ √ √ NR √ NR NR √
et al. (28) material
 Black 6 Gp Master in public 120 √ √ √ √ √ √ √ √ √
et al. (29) health
 Dawson 4 Printed NA NA NR NR NR NR NR NR NR NR NR
et al. (30) material
 Falloon 2 Ind General NR √ NR NR NR √ NR NR NR NR
et al. (31) practitioner
 Gellis 1 Ind Clinical 15 √ √ √ √ NR √ NR √ NR
et al. (32) psychologist
 McCrae 2 Ind Counsellor and 50 √ √ √ √ NR NR NR NR √
et al. (33) social worker
 Nishinoue 1 Gp Physician 40 √ √ √ √ √ √ √ √ √
et al. (34)
 Sun 1 Printed NA NA NR NR NR NR NR NR NR NR NR
et al. (35) material
 Waters 2 Printed NA NA √ √ √ √ NR √ √ √ NR
et al. (36) material
 Wang 4 Ind Clinical 15–60 √ NR √ √ NR √ NR NR NR
et al. (37) psychologist

NA, not applicable; NR, not reported.

371

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

 372

Table 3. Summary of within-group and between-group meta-analyses of a systematic review and meta-analysis of sleep hygiene education as a treatment of insomnia (up to May 2017)

No. of Mean differencea CI Q I2 Hedges’s Mean differencea CI Q I2 Hedges’s g Quality of

datasets the evidence
(GRADE)

Within-group meta-analyses Between-group meta-analyses: SHE versus CBT-I

Sleep diary and questionnaires

 Sleep onset 8 5.41** 1.78, 9.03 8.06 13% 0.23** 11.44*** 5.55, 17.34 8.67 19% 0.66** Moderateb
latency, min
 Wake after sleep onset, 8 12.16*** 6.07, 18.25 5.91 0% 0.32*** 14.01*** 6.13, 21.89 2.90 0% 0.48*** Moderateb
min
Total sleep time, min 5 25.06*** 11.86, 38.26 3.21 0% 0.28* 9.14 −10.18, 28.46 0.07 0% 0.13 Lowb,c
Sleep efficiency, % 6 4.72*** 2.66, 6.78 1.16 0% 0.35*** 7.66*** 4.68, 10.64 5.29 5% 0.67*** Moderateb
 Pittsburgh sleep 9 1.75*** 1.05, 2.45 22.76 65% 0.51*** 2.26*** 1.46, 3.05 14.58 45% 0.67*** Lowb,d
quality index score
 Insomnia severity 5 3.00** 0.88, 5.13 23.92 83% 0.67** 3.55*** 2.69, 4.40 3.04 0% 0.92*** Moderateb
index score
Actigraphy
 Sleep onset 3 1.30 −2.38, 4.99 0.23 0% 0.09 3.49 −1.72, 8.71 0.22 0% 0.24 Lowb,c
latency, min
 Wake after sleep 3 0.41 −5.33, 6.15 0.08 0% 0.03 3.88 −3.71, 11.47 1.30 0% 0.22 Lowb,c
onset, min
Total sleep time, min 3 3.97 −8.30, 16.25 0.31 0% 0.10 5.40 −11.72, 22.52 0.01 0% 0.10 Lowb,c
Sleep efficiency, % 4 0.40 −0.89, 1.70 0.06 0% 0.06 1.49 −0.23, 3.22 0.70 0% 0.22 Lowb,c

CBT-I, cognitive-behavioural therapy for insomnia; GRADE, Grading of Recommendations Assessment, Development, and Evaluation; SHE, sleep hygiene education. GRADE Working Group grades of evidence High
quality: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the
effect, but there is a possibility that it is substantially different. Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low quality: We
have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.
a
Positive values indicate pre-post improvements and greater effectiveness of CBT-I.
b
Methods of sequence generation/allocation concealment were unclear, and blinding of participants was not achieved in most of the studies (downgraded by 1 due to limitation of study).
c
The confidence intervals did not exclude no difference so it is difficult to tell whether effects CBT_I and SHE was different (downgraded by 1 due to imprecision).
d
There was important variation between the study results (I2 = 45%, P = 0.07) (downgraded by 1 due to imprecision due to inconsistency).
*P <0.05, **P < 0.01, ***P < 0.001.
Family Practice, 2018, Vol. 35, No. 4

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

Sleep hygiene education for insomnia 373

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

Figure 2. (a) Within-group comparison on sleep-diary-derived sleep efficiency (SE), in %; (b) Within-group comparison on Pittsburgh Sleep Quality Index (PSQI),
in total score; (c) Comparison of sleep hygiene education (SHE) versus cognitive-behavioural therapy for insomnia (CBT-I) on sleep-diary-derived sleep efficiency
(SE), in %; (d) Comparison of sleep hygiene education (SHE) versus cognitive-behavioural therapy for insomnia (CBT-I) on Pittsburgh Sleep Quality Index (PSQI),
in total score.
374 Family Practice, 2018, Vol. 35, No. 4

differences in therapists’ expertise and training requirement, studies 2. Morin CM, LeBlanc M, Daley M, Gregoire JP, Mérette C. Epidemiology
should compare the implementation, acceptability and cost-effect- of insomnia: prevalence, self-help treatments, consultations, and determi-
iveness issues between SHE and CBT-I. A qualitative study suggested nants of help-seeking behaviors. Sleep Med 2006; 7: 123–30.
3. Morin CM, Gaulier B, Barry T, Kowatch RA. Patients’ acceptance of psy-
that although general practitioners know CBT-I, they seldom refer
chological and pharmacological therapies for insomnia. Sleep 1992; 15:
patients for treatment (42). More work is needed to educate general
302–5.
practitioners on the health risks of insomnia and the availability of
4. Sivertsen B, Nordhus IH, Bjorvatn B, Pallesen S. Sleep problems in general
psycho-behavioural treatments. As SHE may be able to resolve the practice: a national survey of assessment and treatment routines of general
patient’s problem and has little or no risk of adverse effects; in places practitioners in Norway. J Sleep Res 2010; 19(1 Pt 1): 36–41.
where CBT-I is unavailable or too costly, SHE may be considered as 5. Hauri PJ. Sleep hygiene. In: Hauri PJ (ed). Current Concepts: The Sleep
a first-step treatment. Disorders. Kalamazoo, MI: The Upjohn Company, 1977, pp. 22–35.
Cross-sectional studies have revealed that daytime napping, 6. Hauri PJ. Sleep hygiene, relaxation therapy, and cognitive interventions.
smoking, alcohol use and uncomfortable sleeping environment are In: Hauri PJ (ed). Case Studies in Insomnia. New York: Plenum, 1991, pp.
more common in individuals with insomnia, compared to good 65–84.
7. Stepanski EJ, Wyatt JK. Use of sleep hygiene in the treatment of insomnia.
sleepers; however, the frequencies of these behaviours are not high
Sleep Med Rev 2003; 7: 215–25.

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

(43). The findings may explain why SHE may not be a sufficient
8. Irish LA, Kline CE, Gunn HE, Buysse DJ, Hall MH. The role of sleep hy-
treatment, while having poor sleep hygiene may be a prerequisite
giene in promoting public health: A review of empirical evidence. Sleep
for using SHE. Med Rev 2015; 22: 23–36.
Despite an extensive literature search, the major limitation of our 9. Maness DL, Khan M. Nonpharmacologic management of chronic in-
review was the small number of included studies. We did not know somnia. Am Fam Physician 2015; 92: 1058–64.
whether there were missing papers that used non-standard spelling or 10. Morgenthaler T, Kramer M, Alessi C et al.; American Academy of Sleep
non-standard terms to define the intervention. Future studies should Medicine. Practice parameters for the psychological and behavioral treat-
search using a combination of Medical Subject Heading (MeSH) ment of insomnia: an update. An American academy of sleep medicine
and free-text terms. Suggestions on using filters and highly sensi- report. Sleep 2006; 29: 1415–9.
11. Schutte-Rodin S, Broch L, Buysse D, Dorsey C, Sateia M. Clinical guideline
tive strategies for identifying randomized trials are available in the
for the evaluation and management of chronic insomnia in adults. J Clin
Cochrane Handbook for Systematic Reviews of Interventions (44).
Sleep Med 2008; 4: 487–504.
Participants’ characteristics, recruitment source and baseline severity
12. Brasure M, Fuchs E, MacDonald R et al. Psychological and behavioral
varied widely across studies; hence generalization to a specific setting interventions for managing insomnia disorder: an evidence report for a
and patient group was not possible. Methodological quality of the Clinical Practice Guideline by the American College of Physicians. Ann
included studies was fair. Due to the nature of intervention, blinding Intern Med 2016; 165: 113–24.
of participants and personnel and allocation concealment were diffi- 13. Ree M, Junge M, Cunnington D. Australasian Sleep Association position
cult in most studies; however, publication bias was unlikely because statement regarding the use of psychological/behavioral treatments in the
SHE was often used as a control intervention and the results were management of insomnia in adults. Sleep Med 2017; 36 (Suppl 1): 43–7.
mostly consistent across studies. 14. Siriwardena AN, Apekey T, Tilling M, et al. General practitioners’ pref-
erences for managing insomnia and opportunities for reducing hypnotic
In conclusion, SHE resulted in pre- to post-treatment improve-
prescribing. J Eval Clin Pract 2010; 16: 731–7.
ments in sleep; however, it fared worse than CBT-I and mindfulness-
15. Everitt H, McDermott L, Leydon G, et al. GPs’ management strategies for
based therapy for the treatment of insomnia. Although CBT-I was
patients with insomnia: a survey and qualitative interview study. Br J Gen
shown to be more effective than SHE, the difference in sleep-diary- Pract 2014; 64: e112–9.
derived SE was 8%; for PSQI, it was two points’ difference. More 16. Espie CA. “Stepped care”: a health technology solution for delivering cog-
studies are needed to examine whether SHE is better than CBT-I in nitive behavioral therapy as a first line insomnia treatment. Sleep 2009;
terms of acceptability, adherence, understanding, cost-effectiveness 32: 1549–58.
and ease of implementation. To understand the effectiveness of SHE, 17. Ho FY, Chung KF, Yeung WF et al. Self-help cognitive-behavioral therapy
future studies should ensure treatment fidelity and a comprehensive for insomnia: a meta-analysis of randomized controlled trials. Sleep Med
coverage of sleep hygiene recommendations. In addition, studies Rev 2015; 19: 17–28.
18. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred
comparing SHE with placebo or no treatment are needed.
reporting items for systematic reviews and meta-analyses: the PRISMA
statement. Ann Intern Med 2009; 151: 264–9, W64.
19. Higgins JP, Altman DG, Gøtzsche PC et al.; Cochrane Bias Methods Group;
Supplementary material Cochrane Statistical Methods Group. The Cochrane Collaboration’s tool
Supplementary material is available at Family Practice online. for assessing risk of bias in randomised trials. BMJ 2011; 343: d5928.
20. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
Trials 1986; 7: 177–88.
Declaration 21. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency
in meta-analyses. BMJ 2003; 327: 557–60.
Funding: this research did not receive any specific grant from funding agencies 22. Cooper H, Hedges LV, Valentine JC. The Handbook of Research Synthesis
in the public, commercial or not-for-profit sectors. and Meta-analysis. New York: Russell Sage Foundation, 2009.
Ethical approval: none. 23. Edinger JD, Olsen MK, Stechuchak KM et al. Cognitive behavioral therapy
Conflict of interest: none. for patients with primary insomnia or insomnia associated predominantly
with mixed psychiatric disorders: a randomized clinical trial. Sleep 2009;
32: 499–510.
References 24. Epstein DR, Dirksen SR. Randomized trial of a cognitive-behavioral inter-
1. Taylor DJ, Lichstein KL, Durrence HH. Insomnia as a health risk factor. vention for insomnia in breast cancer survivors. Oncol Nurs Forum 2007;
Behav Sleep Med 2003; 1: 227–47. 34: E51–9.
Sleep hygiene education for insomnia 375

25. Martínez MP, Miró E, Sánchez AI et al. Cognitive-behavioral therapy for 34. Nishinoue N, Takano T, Kaku A et al. Effects of sleep hygiene education
insomnia and sleep hygiene in fibromyalgia: a randomized controlled trial. and behavioral therapy on sleep quality of white-collar workers: a rand-
J Behav Med 2014; 37: 683–97. omized controlled trial. Ind Health 2012; 50: 123–31.
26. Nakamura Y, Lipschitz DL, Kuhn R, Kinney AY, Donaldson GW. 35. Sun J, Kang J, Wang P, Zeng H. Self-relaxation training can improve sleep
Investigating efficacy of two brief mind-body intervention programs for quality and cognitive functions in the older: a one-year randomised con-
managing sleep disturbance in cancer survivors: a pilot randomized con- trolled trial. J Clin Nurs 2013; 22: 1270–80.
trolled trial. J Cancer Surviv 2013; 7: 165–82. 36. Waters WF, Hurry MJ, Binks PG et al. Behavioral and hypnotic treatments
27. Alessi C, Martin JL, Fiorentino L et al. Cognitive behavioral therapy for for insomnia subtypes. Behav Sleep Med 2003; 1: 81–101.
insomnia in older veterans using nonclinician sleep coaches: randomized 37. Wang J, Wei Q, Wu X, Zhong Z, Li G. Brief behavioral treatment for
controlled trial. J Am Geriatr Soc 2016; 64: 1830–8. patients with treatment-resistant insomnia. Neuropsychiatr Dis Treat
28. Bjorvatn B, Fiske E, Pallesen S. A self-help book is better than sleep hy- 2016; 12: 1967–75.
giene advice for insomnia: a randomized controlled comparative study. 38. Buysse DJ, Reynolds CF 3rd, Monk TH, Berman SR, Kupfer DJ. The
Scand J Psychol 2011; 52: 580–5. Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice
29. Black DS, O’Reilly GA, Olmstead R, Breen EC, Irwin MR. Mindfulness and research. Psychiatry Res 1989; 28: 193–213.
meditation and improvement in sleep quality and daytime impairment 39. Buysse DJ, Germain A, Moul DE et al. Efficacy of brief behavioral treat-
among older adults with sleep disturbances: a randomized clinical trial. ment for chronic insomnia in older adults. Arch Intern Med 2011; 171:

Downloaded from https://academic.oup.com/fampra/article/35/4/365/4671078 by guest on 21 August 2022

JAMA Intern Med 2015; 175: 494–501. 887–95.
30. Dawson SC, Pillon A, Cousins J et al. Cognitive behavioral therapy for 40. Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, Morin CM.
insomnia reduces night to night variability of insomnia symptoms. Sleep Recommendations for a standard research assessment of insomnia. Sleep
2014;37:A171–2. 2006; 29: 1155–73.
31. Falloon K, Elley CR, Fernando A 3rd, Lee AC, Arroll B. Simplified sleep 41. Bélanger L, Vallières A, Ivers H et al. Meta-analysis of sleep changes in
restriction for insomnia in general practice: a randomised controlled trial. control groups of insomnia treatment trials. J Sleep Res 2007; 16: 77–84.
Br J Gen Pract 2015; 65: e508–15. 42. Davy Z, Middlemass J, Siriwardena AN. Patients’ and clinicians’ experi-
32. Gellis LA, Arigo D, Elliott JC. Cognitive refocusing treatment for in- ences and perceptions of the primary care management of insomnia: quali-
somnia: a randomized controlled trial in university students. Behav Ther tative study. Health Expect 2015; 18: 1371–83.
2013; 44: 100–10. 43. Lacks P, Rotert M. Knowledge and practice of sleep hygiene techniques in
33. McCrae CS, McGovern R, Lukefahr R, Stripling AM. Research Evaluating insomniacs and good sleepers. Behav Res Ther 1986; 24: 365–8.
Brief Behavioral Sleep Treatments for Rural Elderly (RESTORE): a pre- 44. Higgins JP, Green S (ed). Cochrane Handbook for Systematic Reviews of
liminary examination of effectiveness. Am J Geriatr Psychiatry 2007; 15: Interventions. John Wiley & Sons Ltd, 2008.
979–82.