JACM

THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE

Volume 00, Number 00, 2019, pp. 1–24
ª Mary Ann Liebert, Inc.
DOI: 10.1089/acm.2018.0046

Acupuncture for Primary Insomnia:

An Updated Systematic Review
of Randomized Controlled Trials
Hui-Juan Cao, PhD,1,* Mei-Li Yu, PhD,2,* Li-Qiong Wang, PhD,1 Yu-Tong Fei, PhD,1
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Hao Xu, PhD, MD,3 and Jian-Ping Liu, PhD, MD1,4

Abstract
Introduction: Acupuncture as one of the alternative therapies for insomnia is widely used in Asia and
increasingly employed in western countries.
Objectives: To provide updated evidence from randomized controlled trials (RCTs) on the effectiveness and
safety of acupuncture for primary insomnia.
Methods: A comprehensive literature search in 11 databases was conducted from January 2008 to October
2017. Two authors independently extracted data and assessed risk of bias independently. Statistical analysis was
performed using RevMan 5.3 software. According to predefined protocol, we combined data in meta-analysis
and performed trial sequential analysis when appropriate. Grading of Recommendations Assessment, Devel-
opment, and Evaluation was also conducted to assess the quality of evidence.
Results: A total of 73 RCTs involving 5533 participants were analyzed. The pooled results showed better
effect from real acupuncture than no treatment (mean difference [MD] -5.58, 95% confidence interval [CI]
-6.85 to -4.31, I2 = 0%, p < 0.00001, 2 trials, fixed effect model, 105 participants) on reducing Pittsburgh Sleep
Quality Index (PSQI) scores with ‘‘very low quality’’ evidence. Acupuncture plus drugs showed better im-
provement than drugs alone on decreasing the PSQI total scores (MD -3.17, 95% CI -4.74 to -1.61, I2 = 72%, 4
trials, random-effects model (REM), p < 0.0001, 253 participants, low quality). Similar benefit favored acu-
puncture compared with no treatment (MD -8.46, 95% CI -9.59 to -7.33, I2 = 0%, p < 0.00001, 2 trials, 65
participants). Acupuncture showed more benefit than estazolam on PSQI (with enough statistical power).
Athens Insomnia Scale (MD -1.64, 95% CI -2.40 to -0.89, I2 = 0%, p < 0.0001, 3 trials, fixed-effects model,
180 participants) or SPIEGEL (MD -2.86, 95% CI -3.54 to -2.18, p < 0.00001, I2 = 0%, 5 trials, fixed-effects
model, 326 participants) with ‘‘very low-quality’’ evidence. Furthermore, low-quality evidence showed less
adverse events from acupuncture than western medications (risk ratio 0.23, 95% CI 0.11–0.48, I2 = 56%,
p < 0.0001, 11 trials, REM, 914 participants). Publication bias was likely present based on the PSQI total scores.
Conclusions: The summary estimates indicate that acupuncture might result in improvement than no treat-
ment on PSQI scores and appears safe. However, the quality of the evidence is varied from very low to low due
to the potential risk of bias and inconsistency among included trials. Further large sample size and rigorously
designed RCTs are still needed.

Keywords: acupuncture, primary insomnia, randomized controlled trials, meta-analysis, systematic review

1
Beijing University of Chinese Medicine, Beijing, China.
2
Beijing First Hospital of Integrated Chinese and Western Medicine, Beijing, China.
3
Cardiovascular Diseases Center, Xi Yuan Hospital, China Academy of Chinese Medicine Sciences, Beijing, China.
4
Center for Evidence-Based Chinese Medicine, Beijing University of Chinese Medicine, Beijing, China.
*These two authors contributed equally.

1
 2 CAO ET AL.
Introduction
I nsomnia is a subjective experience characterized by
impairment of the ability to initiate or maintain sleep and
result in significant impairment in daily function.1,2 Approxi-
mately 35%–50% of adults experience unsatisfactory sleep,3
and 12%–20% of these individuals suffer from insomnia dis-
order.4,5 In China, insomnia affects about 42.5% adults.6
Insomnia treatments include cognitive behavioral therapy
(CBT) and medication treatment.7 CBT, as a multicompo-
nent behavioral intervention, which provides sleep educa-
tion, stimulus control (strengthening associations between
bed and sleep), and therapy for anxiety-provoking beliefs
about sleep, is now commonly recommended as a first-line
treatment for chronic insomnia disorder (Grading of Re-
commendations Assessment, Development, and Evaluation
[GRADE]: strong recommendation, moderate-quality evi-
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
dence), since CBT produces sleep improvements that are
sustained over time.8,9 Pharmacologic interventions are of-
ten prescribed for insomnia disorder, including benzodiaz-
epines and other nonbenzodiazepine hypnotics.
Although CBT and medications may help to alleviate
symptoms, both of them have limitations. Since insomnia is
a chronic condition, CBT is not accepted by patients com-
pletely in the long run and also not readily available in most
clinical settings. Medications may cause undesirable adverse
effects, including morning sedation, anterograde amnesia,
anxiety, or other cognitive and behavioral changes, such as
sleepwalking, sleep-related eating, etc.7,10 Thus, more and
more patients seek for alternative therapies.
Acupuncture treatment for insomnia was widely applied and
reported.11–15 The potential mechanism of acupuncture is dif-
ficult to scientifically understand or measure, such as the in-
visible energy of chi in the body. Preliminary evidence based
on experimental studies suggests that acupuncture works in
many cases, has theorized that the energy meridians are actu-
ally part of the nervous system, and that acupuncture improves
insomnia by cholinergic activation, stimulation of the opioi-
dergic neurons to increase the concentrations of b-endorphin,
and the involvement of the m-opioid receptors.16 Although the
mechanism of acupuncture is still unclear, this treatment con-
tinues to show effectiveness in clinical studies on alleviating
the symptoms of insomnia, and number of researches in acu-
puncture in treating insomnia is currently growing.
However, no confirmative benefit of acupuncture treating
for insomnia could be drawn from our published systematic
review in 2009 (46 trials, 3811 patients, searched until 2008)
yet, mainly due to inadequate sample sizes and poor meth-
odology.17 Since more randomized controlled trials (RCTs)
were published in the past 9 years, we conducted this sys-
tematic review to provide up-to-date evidence-based eval-
uation on the effectiveness and safety of acupuncture as a
monotherapy or in combination with other therapy in
treating primary insomnia.
Materials and Methods
Study registration
The protocol of this systematic review was registered at
PROSPERO (CRD42016039642): YU Mei-Li, Cao Hui-
Juan, Wang Li-Qiong, Fei Yu-Tong, Liu Jian-Ping, and Xu
Hao. Acupuncture for primary insomnia: an updated sys-
tematic review of randomized controlled trials. The fol-
lowing was a brief introduction of methods.
Inclusion criteria
Primary insomnia is defined as chronic and persistent
difficulty in either falling asleep, remaining asleep through
the night, or waking up too early at least three nights a week
for more than a month, and not suffered from any secondary
diseases.18,19 RCTs which compared acupuncture (any forms
of needles invaded into the skin and retaining needle for at
least 15 min) with no treatment, sham acupuncture, conven-
tional therapy, or western medications for primary insomnia
were included. Acupuncture combined with other treatments
compared with the same treatments was also included. The
primary outcome is the overall quality of sleep measured by
internationally recognized sleep quality scales, including
Pittsburgh Sleep Quality Index (PSQI), Athens Insomnia
Scale (AIS) and SPIEGEL. PSQI consists of 19 items in 7
factors and the total scores range from 0 to 21. The total
scores of AIS range from 0 to 24 and SPIEGEL is 0 to 30. A
higher score indicates worse sleep quality. Secondary out-
comes include sleep parameters, which include subjective
sleep quality, sleep onset latency, sleep duration, sleep effi-
ciency, sleep disturbance, and daytime dysfunction, measured
by objective measurements or sleep scores about PSQI sub-
items (each item scored 0 to 3 and a lower score is better),
and quality of life measured by validated scales. Adverse
events were reported as safety outcome. Included trials must
report at least one of the primary outcomes or at least the first
two secondary outcomes or adverse events.
Search strategy
An up-to-date comprehensive search was conducted in
four Chinese databases and six English databases, including
PubMed, EMBASE, Cochrane Central Register of Con-
trolled Trials (CENTRAL), PsycINFO, The Allied and Com-
plementary Medicine Database (AMED), Cumulative Index
to Nursing and Allied Health Literature (CINAHL), Chinese
National Knowledge Infrastructure Databases (CNKI), Chi-
nese Biomedical Literature Database (SinoMed), Chongqing
VIP Chinese Science and Technology Periodical Database
(VIP), and Wanfang Database. Search time was from January
2008 to October 2017. The searching strategy and results are
listed in Appendix Table A1 for details.
Study selection and quality assessment
Two authors selected the eligible studies and extract data
independently and assessed the methodological quality of
included trials using the risk of bias tool.20 The GRADE and
the standards for reporting interventions in controlled trials
of acupuncture (STRICTA) checklist were used to assess the
quality of evidence and completeness and reporting quality
of acupuncture intervention.21
Trial sequential analysis (TSA) was also performed if there
were more than eight included studies in the meta-analysis.
We applied TSA version 0.9.5.5 (Copenhagen: The Copen-
hagen Trial Unit, Center for Clinical Intervention Research,
2016) to calculate the required sample size in a meta-analysis
and to detect the robustness of the result. We used the
diversity-adjusted required information size estimated from a
 SYSTEMATIC REVIEW OF ACUPUNCTURE FOR INSOMNIA
control event proportion of the included trials and a priori
intervention effect of 5%, and the diversity which was esti-
mated in the included trials.
Data analyses
All statistical analyses were performed using RevMan 5.3
software. Data were summarized using risk ratio (RR) with
its 95% confidence interval (CI) for binary outcomes or
mean difference (MD) with 95% CI for continuous out-
comes. Meta-analysis would be conducted with data from
eligible included trials; subgroup analysis and sensitive
analysis were also planned to be conducted when data were
available. Details of the data analysis methods were also
described in the registered protocol.
Results
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
Description of the included trials
A total of 73 RCTs were included in the review. The flow
chart of study searching and screening is shown in Figure 1.
3
The random methods of 24 trials were confirmed by
original authors (contacted by telephone or e-mails) and
the authors for the other 49 trials were failed to contact to
verify the randomization procedure. All of the included
trials were published in Chinese and conducted in China
with a total of 5533 participants (2785 in the intervention
group and 2748 in the control group). The sample sizes of
all included trials varied from 22 to 196 participants (10–
98 participants in each group). There was a wide variation
in the age of subjects (18–60 years) and disease duration
(30 days to 18 years). The treatment duration ranged from
7 to 60 days.
Acupuncture was used alone or combined with medica-
tions in the included trials. Needle types included manual
acupuncture (51 trials),22–72 electroacupuncture (14 trials),73–86
scalp acupuncture (6 trials),87–92 wrist–ankle acupuncture (1
trial),93 and eye acupuncture (1 trial).53 The controls in-
cluded medication (70 trials), sham acupuncture (needles
were located at the same acupoints with accessory device,
but were not inserted into the skin) (1 trial),94 or no treat-
ment (2 trials).31,88
FIG. 1. Flow chart of study search and
selection. Preferred reporting items for
systematic review and meta-analysis
(PRISMA). RCTs, randomized controlled
trials.
 4 CAO ET AL.
Most of the included trials used semistructured prescrip-
tion of the points. The number of the selected acupoints
ranged from 2 to 20, and the top 10 commonly used acupoints
were Baihui (GV20), Sishencong (EX-HN1), Shenmen (HT7),
Sanyinjiao (SP6), Anmian (EX-HN22), Neiguan (PC6), Tai-
chong (LR3), Shenting (GV24), Zhaohai (KI6), and Yintang
(GV29). The top three meridians with the most selected
points were Dumeridian, HandShaoyin meridian, and Hand-
Jueyin meridian.
Sixty-nine trials used the PSQI scores as the outcome
measure. SPIEGEL25,39,70,74,83 and AIS25,70,74 were also
reported. Twenty-six trials reported sleep parameters and
daytime functioning, which are sub-items of PSQI score. Sleep
onset latency was reported in three trials57,84,88 and one trial88
reported quality of life measured by the Medical Outcomes
Study item short form health survey (SF-36). Seventeen
of the included trials25,26,28,31,32,38,42,48,50,66,67,81,83,85,88,89,94
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
mentioned the safety issue. Twelve trials reported data of
adverse events and the remaining five trials32,48,66,67,88
mentioned no adverse events. Five trials26,30,88,90,94 men-
tioned the follow-up duration of post-treatment, which
varied from 7 to 90 days. Characteristics of the 73 included
trials are listed in Appendix Table A2.
Methodology quality assessment
Risk of bias of included trials. Ten24,26,28,30,73,74,87,88,89,94
of the included trials were judged as high quality (at least 4
of 7 items were assessed as low risk of bias, which were
strictly designed with randomization, must be included).
Twenty-four of included trials used proper methods to
generate randomization sequence and this was confirmed by
the original authors through telephone/emails. Sixteen trials
failed to contact but are described in the articles that random
sequence was generated by random number table. Fourteen
trials24,26,28–30,32,34,73–75,87–89,94 described methods of allo-
cation concealment. Only one trial94 used sham acupuncture
as control and was assessed as low risk of bias on blinding to
participants. Six trials25,26,71,73,88,94 used the third party to
evaluate the outcome indicators and were assessed as low
risk of bias on blinding of outcome assessors. Thirty-two
trials26,29,34,38,39–43,45,46,49–51,57,58,60,62–66,68,70,78–80,82,84,85,91,93
were assessed as unclear of bias on the item of incomplete
outcome data, since there were dropouts (less than 5%
of the total) excluded from analyses. Baseline character-
istics were comparable between the two groups, but
the reason of the dropout was not provided and not per-
formed by the intention-to-treat analysis. Other 41 tri-
FIG. 2. Risk of bias graph of 73 included
studies.
als22–25,27,28,30–33,35–37,44,48,52–56,59,61,67,69,71–77,81,83,86–90,92,94,95
reported no dropouts and assessed as low risk of bias. Protocol of
all included trials could not be accessible, but we assessed them
as having low risk of reporting bias, since they all reported the
important outcome measurements for insomnia. Twenty-four
trials26,39,41,44,46,48,49,52,54,55,58,59,61,62,64,67,77,80,81,83,88,90,93,94
were assessed as low risk of other bias by the good baseline
comparability, detailed descriptions on inclusion/exclusion
criteria, and adequate follow-up. However, none of the in-
cluded trials reported sample size calculations, essential for
ensuring adequate statistical power, or described the funding
issue. We evaluated the remaining trials as having an unclear
risk of other bias. Risk of bias graph of 73 included trials is
shown in Figure 2.
STRICTA assessment of 73 included trials. Almost all
of the included trials (98.63%) had detailed descriptions on
acupuncture rationale and treatment regimen. More than
69.49% trials provided a comprehensive description on the
needling details. At least 68 trials (93.15%) have reported a
precise description of the control interventions. However,
for practitioner background, only four of the trials (5.48%)
described, but insufficiently (only mentioned performed by
professional acupuncturist, but their professional affiliation,
years in acupuncture practice, or other relevant experience
were not described in detail). According to predefined in-
clusion criteria, only acupuncture therapy was included and
all of include trials do not need to be reported on the item of
other interventions. The detailed assessment results are
summarized in Table 1.
Estimate Effects of Acupuncture for Primary Insomnia
Comparison 1: acupuncture versus no treatment
Overall quality of sleep. The pooled results of two tri-
als27,34 compared with no treatment showed significant
difference in decreasing the PSQI total scores (MD -5.58,
95% CI -6.85 to -4.31, I2 = 0%, p < 0.00001, 105 partici-
pants, very low-quality) (Appendix Table A3).
Quality of life.One trial27 reported that acupuncture had
an improvement in the physical component scores compared
to no treatment on the SF-36 (MD 197.57, 95% CI 148.40 to
246.74, p < 0.00001, 1 trial, 65 participants).
Sleep onset latency.One trial27 reported sleep onset
latency detected by monitoring devices, and the results
showed acupuncture had at least 40 min shorter sleep onset
 SYSTEMATIC REVIEW OF ACUPUNCTURE FOR INSOMNIA 5

Table 1. Checklist of 73 Included Trials for Items in STRICTA 2010

No. of trials
eligible to
Item Detail the items Percentage
1. Acupuncture (1a) Style of acupuncture (e.g., Traditional Chinese Medicine, 73 100
rationale Japanese, Korean, Western medical, Five Element, ear
acupuncture, etc)
(1b) Reasoning for treatment provided, based on historical 73 100
context, literature sources, and/or consensus methods,
with references, where appropriate
(1c) Extent to which treatment was varied 69 94.52
2. Details of needling (2a) Number of needle insertions per subject per session 73 100
(mean and range where relevant)
(2b) Names (or location if no standard name) of points used 72 98.63
(uni/bilateral)
(2c) Depth of insertion, based on a specified unit of 50 68.49
measurement, or on a particular tissue level
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

(2d) Response sought (e.g., de qi or muscle twitch response) 57 78.08

(2e) Needle stimulation (e.g., manual, electrical) 73 100
(2f) Needle retention time 72 98.63
(2g) Needle type (diameter, length, and manufacturer 61 83.56
or material)
3. Treatment regimen (3a) Number of treatment sessions 72 98.63
(3b) Frequency and duration of treatment sessions 73 100
4. Other components (4a) Details of other interventions administered to the 73 100
of treatment acupuncture group (e.g., moxibustion, cupping, herbs,
exercises, lifestyle advice)
(4b) Setting and context of treatment, including instructions to 6 8.21
practitioners, and information and explanations to patients
5. Practitioner (5) Description of participating acupuncturists (qualification 4 5.48
background or professional affiliation, years in acupuncture practice,
other relevant experience)
6. Control or (6a) Rationale for the control or comparator in the context of 68 93.15
comparator the research question, with sources that justify this choice
interventions (6b) Precise description of the control or comparator. If sham 72 98.63
acupuncture or any other type of acupuncture-like control
is used, provide details as for Items 1 to 3 above.
STRICTA, Standards for Reporting Interventions in Controlled Trials of Acupuncture.

latency than no treatment (MD -40.42 min, 95% CI
-50.84 min to -30 min, p < 0.001, 1 trial, 65 participants).
Comparison 2: acupuncture versus sham acupuncture
Overall quality of sleep. One trial94 showed that real
acupuncture reduced more PSQI total scores than sham
acupuncture (MD -8.46, 95% CI -9.59 to -7.33, p < 0.00001,
1 trial, 65 participants).
Sleep parameters. The trial94 reported that real acu-
puncture was more likely to have improvement in PSQI sub-
items, including the subjective sleep quality, sleep onset
latency, sleep duration, sleep efficiency, sleep disturbance,
and daytime dysfunction (MD -0.43 to -1.78, p < 0.0001).
Comparison 3: acupuncture plus estazolam versus
estazolam alone
Overall quality of sleep. The pooled results of four tri-
als29,30,54,62 found a better effect of the intervention group
on PSQI total scores compared with the control group (MD
-3.17, 95% CI -4.74 to -1.61, I2 = 72%, p < 0.0001, four
trials, random-effects model (REM), 253 participants, low
quality) (Appendix Table A3).
One trial30 showed similar result measured by AIS (MD
-1.06, 95% CI -1.75 to -0.37, p = 0.003, 1 trial, 70 partic-
ipants) and Epworth Sleepiness Scale (MD -2.73, 95% CI
-4.56 to -0.90, p = 0.004, 1 trial, 70 participants).
Comparison 4: acupuncture versus western drugs
Overall quality of sleep. According to the different types
of sedatives and hypnotics, we conducted two subgroup
analyses to compare acupuncture with benzodiazepines and
nonbenzodiazepines, respectively.
Subgroup 1: Acupuncture versus benzodiazepines. The
pooled results of 32 trials showed that acupuncture had a
better effect on decreasing the PSQI total scores than esta-
zolam (1 mg once daily) (MD -1.73, 95% CI -2.11 to
-1.35, I2 = 69%, REM, p < 0.00001, 32 trials, 2452 partici-
pants, very low quality). With increasing dose of estazolam,
the differences of curative effect between acupuncture and
estazolam was getting smaller. When 2–4 mg of estazolam
 6 CAO ET AL.
was used, the results of two trials were smaller compared
with the difference between acupuncture and 1 mg estazo-
lam (MD -1.54, 95% CI -1.87 to -1.20, p < 0.00001, 2
trials, 142 participants) (Supplementary Fig. S1).
The pooled results of five trials24,50,56,75,82 showed that
manual acupuncture had a better effect on decreasing PSQI
total scores than estazolam with 2 mg (MD -1.30, 95% CI
-2.31 to -0.30, I2 = 67%, p < 0.01, 5 trials, REM, 328 par-
ticipants) (Supplementary Fig. S1). The results of electro-
acupuncture and scalp acupuncture were not pooled due to
high heterogeneity (I2 > 75%), and the result showed a better
effect compared with estazolam. Other five tri-
als48,65,66,79,87,88 with other benzodiazepines (alprazolam,
chlordiazepoxide, diazepam, clonazepam) reported the
similar result. The results of individual trials are shown in
Appendix Table A3.
The pooled results of three trials25,70,74 showed that
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
acupuncture had a better improvement on the overall quality
of sleep than estazolam measured by AIS (MD -1.64, 95%
CI -2.40 to -0.89, p < 0.0001, 180 participants, low quality)
and SPIEGEL (MD -2.86, 95% CI -3.54 to -2.18,
p < 0.00001, 326 participants, low quality) (Appendix
Table A3).
Subgroup 2: Acupuncture versus nonbenzodiazepines.
Nine trials compared acupuncture with nonbenzodiazepines,
including trazodone, zolpidem, oryzanol, zopiclone, and
melatonin. The results of meta-analysis found a minor effect
of acupuncture in decreasing the PSQI total scores than
nonbenzodiazepines (MD -0.94, 95% CI -1.55 to -0.33,
I2 = 46%, p = 0.003, 9 trials, REM, 643 participants, very
low quality) (Appendix Table A3).
Sleep parameters. The pooled results of 21 trials26,27,33–
35,43,50,51,59,66–68,69,73,76,77,79,81,83,84,89
on sleep parameters
measured with PSQI sub-items reported that acupuncture
had some improvement than benzodiazepines in terms of
subjective sleep quality (MD -0.32, 95% CI -0.41 to -0.23,
I2 = 68%, p < 0.00001, REM, 1812 participants), sleep effi-
ciency (MD -0.25, 95% CI -0.36 to -0.15, p < 0.00001,
I2 = 63%, 19 trials,26,27,33–35,43,50,51,59,66–69,76,77,81,83,84,89
REM, 1618 participants), and sleep duration ( MD -0.18,
95% CI -0.29 to -0.08, p < 0.0004, I2 = 64%, 18 tri-
als,26,27,33,35,43,50,51,59,66–69,76,77,81,83,84,89 REM, 1586 par-
ticipants) (Supplementary Fig. S2). However, pooling analysis
on sleep disturbance and daytime functioning were not used
due to the significant statistical heterogeneity (I2 > 75%).The
result of individual trial showed a better effect when com-
paring acupuncture with benzodiazepines. The results of in-
dividual trials are listed in Appendix Tables A4 and A5.
Sleep onset latency. The results of sleep onset latency
measured by PSQI sub-item were not pooled due to high
heterogeneity (I2 > 75%). The subgroup analysis results of
14 trials26,27,33–35,43,50,51,59,66–69,81 showed that manual acu-
puncture had an improvement in the sleep onset latency than
benzodiazepines (MD -0.35, 95% CI -0.47 to -0.23, I2 = 65%,
p < 0.001, REM, 1213 participants) (Appendix Table A6).
However, the pooled results of two trials43,84 measured by
monitoring devices showed no difference between acu-
puncture and benzodiazepines on this outcome (MD
-0.86 min, 95% CI -3.70 to 5.42, p = 0.71, 110 participants).
Safety
Seventeen trials25,26,28,31,32,38,42,48,50,66,67,81,83,85,88,89,94
reported outcome of adverse events. One trial94 found no
difference between real and sham acupuncture on frequency
of adverse events (RR 1.43, 95% CI 0.29–6.92, 67 partici-
pants). Other five trials32,48,66,67,88 comparing acupuncture
with western medication reported no occurrence of adverse
events. The remaining 16 trials compared acupuncture with
western medication, 11 of which showed lower incidence of
adverse events in acupuncture group than the western
medication group (RR 0.23, 95% CI 0.11–0.48, I2 = 56%,
p < 0.0001, REM, 914 participants). The adverse events of
acupuncture group were fainting, bleeding, dizziness, skin
flushing, and so on. The adverse events in acupuncture
group were fade fasting, which was recovery after removing
the needles. The details of adverse events are listed in
Appendix Table A7.
Assessment for Publication Bias
The funnel plots of 58 trials comparing acupuncture with
western drugs for the outcome of PSQI total score was asym-
metrical, suggesting publication bias (Supplementary Fig. S3).
An asymmetrical funnel plot of 21 trials comparing
acupuncture with benzodiazepines in PSQI sub-items score
is shown (Supplementary Fig. S4).
GRADE: Summary of Findings on Acupuncture for Insomnia
We summarized the main outcomes by using a GRADE
system, and the results from low or very low-quality evi-
dence showed acupuncture appeared more beneficial than
western medications on PSQI reduction and moderate
quality evidence showed acupuncture was superior to no
treatment on this outcome as well. This was mainly due to
the poor methodological quality and high heterogeneity
among relevant trials. The details of GRADE summary of
finding are shown in Table 2.
Trial Sequential Analysis
We conducted TSA with the data from 2 meta-analyses,
in which more than eight trials were included. One was
conducted with data from 25 trials, which compared manual
acupuncture with 1 mg estazolam on PSQI total scores’ re-
duction. TSA illustrated the cumulative Z-curve across the
traditional boundary of 5% significance (horizontal line) and
the monitoring boundaries (inward sloping curves) (in
Fig. 3a). After Liu YY 2014, the significance testing had
been performed each time a new trial was added to the meta-
analysis, which means the sample size achieved the required
244 participants and we had enough power to confirm the
evidence (that the acupuncture may decrease 1.7 more PSQI
total scores than estazolam 1 mg daily) controlling for the
risk of random error.
A similar result was shown in another TSA with data
from 11 trials, which also compared acupuncture with es-
tazolam on incidence rate of adverse events. TSA also il-
lustrated the cumulative Z-curve across the horizontal line
and the inward sloping curves (in Fig. 3b), which means the
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Table 2. GRADE Summary of Finding

Illustrative comparative risksa (95% CI)
Quality of the
Assumed risk Corresponding risk Relative effect No of Participants evidence
Outcomes Control Acupuncture (95% CI) (studies) (GRADE)
PSQI (scale from: 0 to 21)— The mean PSQI (total score) The mean PSQI (total score) — 105 (2 studies) 4. . .
acupuncture versus no treatment was 10.99 was 5.58 lower (6.85–4.31 Very low1,2,3,5
lower)
PSQI (scale from: 0 to 21)— The mean PSQI (total score) The mean PSQI (total score) — 43,233 (32 studies) 4...
acupuncture versus was 10.23 was 2.13 lower (3.28–1.47 Very low1,2,3,5
benzodiazepines lower)
PSQI (Scale from: 0 to 21)— The mean PSQI (total score) The mean PSQI (total score) — 643 (9 studies) 4...
acupuncture versus in the control groups was 0.94 lower (1.55–0.33 Very low1,2,3,4
non-benzodiazepines was 9.95 lower)
SPIEGEL score (scale from: The mean SPIEGEL scores The mean SPIEGEL was 3.09 — 266 (4 studies) 4...
0 to 30)—acupuncture versus was 10.25 lower (3.9–2.28 lower) Very low1,2,3

7
benzodiazepines
AIS score (scale from: 0 to 24)— The mean AIS scores was The mean AIS score was 1.99 — 120 (2 studies) 4...
acupuncture versus 10.29 lower (3.04–0.95 lower) Very low1,2,3,5
benzodiazepines
PSQI (scale from: 0 to 21)— The mean PSQI (total score) The mean PSQI (total score) — 253 (4 studies) 44..
acupuncture plus estazolam was 11.00 was 3.17 lower (4.74–1.61 Low1,2,3,5
versus estazolam lower)
Adverse events—acupuncture 195 per 1000 208 per 1000 RR 0.23 (0.11–0.48) 914 (11 studies) 44..
versus western medication Low1,2,3
a
The basis for the assumed risk (e.g., the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison
group and the relative effect of the intervention (and its 95% CI).
1
Blinding of participants and personnel (performance bias) was high risk of bias.
2
Allocation concealment (selection bias) and blinding of outcome assessment (detection bias) were unclear.
3
High heterogeneity.
4
The medicine of control was different.
5
There is a certain publication bias.
AIS, Athens Insomnia Scale; CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; PSQI, Pittsburgh Sleep Quality Index; RR, risk ratio.
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

8
FIG. 3. TSA of two meta-analyses compared manual acupuncture with estazolam 1 mg. (a) TSA from 25 trails on PSQI total score’ reduction. (b) TSA from 11 trails on
incidence rate of adverse events. PSQI, Pittsburgh Sleep Quality Index; TSA, trial sequential analysis.
 SYSTEMATIC REVIEW OF ACUPUNCTURE FOR INSOMNIA
sample size achieved the required 502 participants and we had
enough power to confirm the evidence (that the acupuncture
may have 75% fewer adverse events than estazolam).
Discussion
Summary of findings
A total of 73 RCTs were included, and 10 of them were
assessed as high quality according to risk of bias. For com-
parison of acupuncture with no treatment, sham acupuncture,
or acupuncture plus estazolam compared with estazolam
alone, the results showed acupuncture appeared to be more
effective on decreasing PSQI total scores (range of MD from
-3.17 to -8.46, low quality). Acupuncture showed better
effect on decreasing the PSQI total scores than estazolam
(very low quality), and the degree of decline is inversely
proportional to the dose of estazolam. The pooled results also
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
showed that acupuncture could improve patients’ overall
quality of sleep than benzodiazepines on AIS scale (low
quality) and SPIEGEL scale (low-quality evidence).
For the secondary outcomes, the results of 21 trials
showed positive effect of acupuncture on subjective sleep
quality, sleep efficiency, and sleep duration than benzodi-
azepines. However, the scores of PSQI sub-items average
decreased less than 0.5 score (MD -0.25 to -0.32) and
might not be clinically relevant.
The results showed that the incidence of adverse events in
acupuncture group was about a quarter of the western drugs
group (low-quality evidence).
Limitation and strength
We had contacted the original authors to confirm the ran-
domization procedure as far as possible. However, there were
still 47 authors who failed to contact due to their personal
reasons. Therefore, random sequence generation was not con-
firmed and blinding of acupuncturists and patients was hardly
applied. Therefore, the quality of included trials was low.
Compared with the previous reviews, our study conducted
an update comprehensive search and included 73 trials,
which is the best-rounded systematic review of acupuncture
treatment for primary insomnia yet. According to the
STRICTA assessment, we conducted an assessment of
acupuncture report quality of included trials. Although we
draw a consistent conclusion with previous trials in terms of
PSQI total score for comparisons between acupuncture and
no treatment/sham acupuncture, we found a better effect of
acupuncture for primary insomnia compared with estazolam
(1 mg once daily). The incidence of adverse events was only
a quarter than estazolam, which suggests acupuncture may
reduce the dose of estazolam to avoid the side effect. This
was supported by TSA results.
Implications for clinical practice
In this review, we included primary insomnia with dis-
ease courses more than a month. The baseline total scores of
PSQI ranged from 12 to 19 (total scores >5 associated with
poor sleep quality).96 It can be seen from the results that
acupuncture showed better improvement than no treatment
or sham acupuncture on PSQI total score (by reducing 5–8
scores). Some research also suggests that increased risk of
adverse effects of benzodiazepines would occur with high-
9
dose drug.97 So, for the insomniacs with no effect or inef-
ficiency treated by the single drug therapy, the treatment of
acupuncture-combined drug therapy may be a better choice.
The top five acupoints used frequently were GV20, EX-HN1,
HT7, EX-HN22, and GV24. The combinations of GV20 and
EX-HN1, BL62 and KI6 were mostly used. According to the
acupoints distribution, meridians, including Du meridian,
Hand Shaoyin, and Hand Jueyin were selected more. Fur-
thermore, this study suggests that acupuncture may get a
better effect when the treatment course is more than 14 days.
However, due to the inadequate follow-up, the conclusion
needed further researches to be confirmed.
Implications for future research
According to the STRICTA assessment, less description
was mentioned about the qualifications of researchers in the
involved trials. Generally, acupuncturist who could participate
in a clinical study in china should have a formal practicing
certificate already. It is necessary to pay attention to the im-
portant role of researchers in the clinical curative effect of
acupuncture. Meanwhile, researchers are encouraged to reg-
ister their study protocols before trial implementation to en-
sure a high standard of research is maintained so that valid
conclusions may be generated.
Future researches are urged to use more rigorous method-
ology, including appropriate sample sizes and adequate follow-
up with long-term duration. In the outcome assessment, the
objective outcome may be better than personal satisfaction of
improvement, such as recognized sleep scale (PSQI, AIS, etc.)
or other monitoring facility, to confirm the conclusion less
affected by subjective factors. Acupoint selection (the number
or different position) might have an influence on the curative
effect of acupuncture treatment for insomnia. The relationship
between the dose of western medicine and the effect of acu-
puncture needs further investigation.
Conclusion
With the results of fifty more included trials and the TSA
analysis, we would like to suggest that the clinicians may
consider to use acupuncture alone or plus western drugs for
patients with primary insomnia according to the actual sit-
uation of the patients. Of course, further large sample-size
and rigorously designed RCTs are still needed (see PRISMA
2009 Checklist in Appendix Table A8).
Acknowledgments
The authors thank Dr. Xia Yun (affiliated to the Oriental
Hospital, Beijing University of Chinese Medicine) for her
guidance on the TSA methods used in this review. This work
was partially supported by the National Natural Science Foun-
dation of China (81673828) and Beijing Municipal Organiza-
tion Department (2017000020124G292) to Hui-Juan Cao.
Author Disclosure Statement
No competing financial interests exist.
References
1. American Psychiatric Association. Diagnostic and Statis-
tical Manual of Mental Disorders, Revised. 4th ed. Wa-
shington, DC: American Psychiatric Association, 2000.
 10
2. Zhao ZX, Pan JY, Ji JL, et al. Expert consensus panel on
definition, diagnosis and drug therapy for insomnia. Expert
consensus (draft) on definition, diagnosis and drug therapy
for insomnia. Chin J Neurol 2006;39:141–143.
3. Walsh JK, Coulouvrat C, Hajak G, et al. Nighttime in-
somnia symptoms and perceived health in the America
Insomnia Survey (AIS). Sleep 2011;34:997–1011.
4. Roth T, Coulouvrat C, Hajak G, et al. Prevalence and per-
ceived health associated with insomnia based on DSM-IV-
TR; International Statistical Classification of Diseases and
Related Health Problems, Tenth Revision; and Research
Diagnostic Criteria/International Classification of Sleep
Disorders, Second Edition criteria: Results from the America
Insomnia Survey. Biol Psychiatry 2011;69:592–600.
5. Morin CM, LeBlanc M, Bélanger L, et al. Prevalence of
insomnia and its treatment in Canada. Can J Psychiatry
2011;56:540–548.
6. Li YH, Liu Y, Jin T, Gao J. Survey on sleep disorders in
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
middle-aged population. Capital Food Med 2005;12:6–15.
(In Chinese).
7. Buysse DJ. Insomnia. JAMA 2013;309:706–716.
8. Morin CM, Vallières A, Guay B, et al. Cognitive behav-
ioral therapy, singly and combined with medication, for
persistent insomnia: A randomized controlled trial. JAMA
2009;301:2005–2015.
9. Morin CM, Benca R. Chronic insomnia. Lancet 2012;379:
1129–1141.
10. Walsh JK, Roth T. Pharmacologic treatment of insomnia:
Benzodiazepine receptor agonists. In: Buysse DJ, ed.
Principles and Practices of Sleep Medicine. Fifth edition.
St. Louis: Elsevier, 2011:905–915.
11. Guo J, Wang LP, Liu CZ, et al. Efficacy of acupuncture for
primary insomnia: A randomized controlled clinical trial.
Evid Based Complement Alternat Med 2013;2013:163850.
12. O’Brien K, Weber D. Insomnia in Chinese medicine: The
heart of the matter. J Altern Complement Med 2016;22:
684–694.
13. Sun YJ, Yuan JM, Yang ZM. Effectiveness and safety of
moxibustion for primary insomnia: A systematic review
and meta-analysis. BMC Complement Altern Med 2016;
16:217.
14. Lin YF, Liu ZD, Ma W, et al. Hazards of insomnia and the
effects of acupuncture treatment on insomnia. J Integr Med
2016;14:174–186.
15. Bo A, Si L, Wang Y, et al. Clinical trial research on
Mongolian medical warm acupuncture in treating insomnia.
Evid Based Complement Alternat Med 2016;2016:
6190285.
16. Cheng CH, Yi PL, Lin JG, et al. Endogenous opiates in the
nucleus tracts solitarius mediate electroacupuncture-
induced sleep activities in rats. Evid Based Complement
Alternat Med 2011;2011:159209.
17. Cao H, Pan X, Li H, et al. Acupuncture for treatment of
insomnia: A systematic review of randomized controlled
trials. J Altern Complement Med 2009;15:1171–1186.
18. Shamseer L, Moher D, Clarke M, et al. Preferred reporting
items for systematic review and meta-analysis protocols
(PRISMA-P) 2015: Elaboration and explanation. BMJ
2015;349:g7647.
19. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA
statement for reporting systematic reviews and meta-
analyses of studies that evaluate health care interventions:
Explanation and elaboration. Ann Intern Med 2009;151:
W65–W94.
CAO ET AL.
20. Higgins J, Green S. Cochrane Handbook for Systematic
Reviews of Interventions. Version 5.0.2 ed. West Sussex,
United Kingdom: The Cochrane Collaboration, 2009.
21. MacPherson H, Altman DG, Hammerschlag R, et al. Re-
vised Standards for Reporting Interventions in Clinical Trials
of Acupuncture (STRICTA): Extending the CONSORT
statement. J Altern Complement Med 2010;16:ST1–ST14.
22. Lv JC. Clinical observation of acupuncture treatment for
primary sleep disorders [In Chinese]. China Modern Doctor
2010;48:42–43.
23. Wang ZH. The Clinical Research of Using Acupuncture
‘An Mianwuxue’ to Cure Simple Insomnia [In Chinese].
Hebei Shen, China: Chinese Hebei Medical University 2014.
24. Meng FY. A Clinical Study on the Treatment of Insomnia
by Acupuncture with Back-Shu Points and Front-Mu Points
[In Chinese]. Heilongjiang Sheng, China: Heilongjiang
University of Chinese Medicine 2011.
25. Liu Y. Clinical Research of Acupuncture Treatment of
Insomnia with TongduTiaoShen Methods [In Chinese].
Anhui Sheng, China: Anhui College of Traditional Chinese
Medicine 2014.
26. Zhou JC. Randomized Control Study of Treatment with
Scalp Acupuncture for Primary Insomnia [In Chinese].
Chengdu, Sichuan, China: Chengdu University of Tradi-
tional Chinese Medicine 2013.
27. Wang ZY, Liu XG, Zhang W. Clinical observation on
acupuncture treatment of primary insomnia [In Chinese]. J
Tradit Chin Med 2015;33:165–167.
28. Liang LZ. Clinical Study of Insomnia with Electro-
acupuncture on Vasomotor Area [In Chinese]. Guangdong
Sheng, China: Guangzhou University of Chinese Medicine
2009.
29. Sun JQ. Clinical observation on the treatment of primary
insomnia with the method of regulating and supplementing
the Ren and Du methods [In Chinese]. J Tradit Chin Med
2011;39:81–83.
30. Xu SF, Zhuang LX, Yin P, et al. Clinical evaluation of
acupuncture treatment for insomnia of deficiency of heart
and spleen [In Chinese]. J Guangzhou Univ Chin Med
2016;33:31–34.
31. Xiao BB, Luo XY, Shen TY. Efficacy observation on re-
fractory insomnia treated with the balance needing therapy
[In Chinese]. Chin Acupunct Moxibustion 2013;33:101–104.
32. Wu X, Hu H, Xing J, et al. Effect of picking Shenmai
(BL62) and Zhaohai (KI6) on daytime arousal of patients
with chronic insomnia: A randomized controlled trial [In
Chinese]. World J Acupunct Moxibustion 2014;24:1–5.
33. Gu XT. Research on Clinical and Neurobiological Me-
chanisms of Scalp Acupuncture Treatment in Insomnia [In
Chinese]. Guangdong Sheng, China: Guangzhou University
of Chinese Medicine, 2013.
34. Wu X. Effect of Acupuncture on Sleep Quality in Patients
with Primary Insomnia: A Randomized Controlled Trial
[In Chinese]. Chaoyang Qu, China: Beijing University of
Chinese Medicine, 2007.
35. Wang YJ, Zhang LH, Han YX, et al. Efficacy observation on
Governor Vessel-unblocking and mind-calming acupuncture
for insomnia. J Acupunct Tuina Sci 2016;14:274–278.
36. Zhou DP. Clinical observation and analysis on the clinical
effect of acupuncture and moxibustion in the treatment of
insomnia [In Chinese]. Contemp Med 2016;22:24–25.
37. Bi XJ, He XT, Li JN. Clinical observation of acupuncture
treatment for 40 cases insomnia [In Chinese]. Chin J Eth-
nomedicine Ethnopharmacy 2014;68:71.
 SYSTEMATIC REVIEW OF ACUPUNCTURE FOR INSOMNIA
38. Deng XF. Clinical observation of abdominal acupuncture
treat insomnia with syndrome of liver-fire disturbing heart
[In Chinese]. J Sichuan Tradit Chin Med 2014;32:161–163.
39. Fu FC. 43 cases of acupuncture treated insomnia [In Chi-
nese]. Guangming J Chin Med 2012;27:320–321.
40. He FL, Jiang YM. Clinical observation of acupuncture
treats with insomnia by balancing yin and yang [In Chi-
nese]. Chin Med Emerg 2010;19:940, 1006.
41. Ji XD, Wang QS, Zhu WX. Expression of neurotransmitters
in patients of insomnia differentiated as liver stagnation
transforming into fire treated with acupuncture [In Chi-
nese]. Chin Acupunct Moxibustion 2015;35:549–552.
42. Kou JY. Clinical Study of Insomnia with the Treatment of
Acupuncture (Tiaozang an Shen) [In Chinese]. Heilongjiang
Sheng, China: Heilongjiang University of Chinese Medicine,
2003.
43. Li XY, Zhou ZL, Zhang B. Clinical observation of acu-
puncture for treating middle or elderly aged patients with
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
insomnia [In Chinese]. Tianjin J Tradit Chin Med 2010;27:
386–388.
44. Liu YY. Clinical Study of Sishencong Combined Ding-
shenzhen Treatment for Primary Insomnia [In Chinese].
Guangdong Sheng, China: Guangzhou University of Chi-
nese Medicine, 2014.
45. Lai XY. Clinical and Animal Experimental Study on Acu-
puncture Treatment of Insomnia with Insomnia Theory [In
Chinese]. Guangdong Sheng, China: Guangzhou University
of Chinese Medicine, 2014.
46. Ma XW, Deng LJ, Kuang JC, et al. Clinical observation of
acupuncture treat insomnia with Ren meridians [In Chi-
nese]. Guide China Med 2013;11:198–199.
47. Ma YX. Clinical observation of acupuncture treatment of
43 patients with insomnia by ten points [In Chinese]. Chin J
Ethnomedicine Ethnopharmacy 2014;13:67–68.
48. Ren LN, Li X. Clinical control study of acupuncture treated
senile primary insomnia [In Chinese]. J Taishan Med
College 2014;35:660–661.
49. Sun ZY. Clinical effect of acupuncture (Yong quan, Tai xi,
Sanyinjiao) in the treatment of 40 cases of insomnia [In
Chinese]. Shanxi J Chin Med 2010;31:731–732.
50. Su D, Zhao J, Jin H. The clinical efficacy of acupuncture at
yuan-primary point and luo-connecting point of the heart
meridian (HT) for the treatment of deficiency insomnia [In
Chinese]. J Clin Acu & Mox 2011;27:34–36.
51. Song JF. The Clinical Research of Acupuncture for Dis-
harmony Between the Heart and Kidney Insomnia [In
Chinese]. Heilongjiang Sheng, China: Heilongjiang Uni-
versity of Chinese Medicine, 2013.
52. Tao HX, Jin JL. Clinical study of acupuncture with extra
nerve points treatment for 58 cases of insomnia [In Chi-
nese]. Jilin J Tradit Chin Med 2009;29:52–53.
53. Xu H, Wang PQ, He W. Clinical Observation on eye-
acupuncture in treatment of insomnia [In Chinese]. J
Liaoning Univ TCM 2014;16:109–111.
54. Wang JP, Wang JB, Zhang YM. Senile insomnia treated
with integrated acupuncture and medication therapy a
randomized controlled trial [In Chinese]. Chin Acupunct
Moxibustion 2015;35:544–548.
55. Tian JL, Fu Y. Clinical study of eye needle combined with
acupuncture treatment 120 cases of insomnia [In Chinese].
Beijing J Tradit Chin Med 2015;34:563–565.
56. Tian T, Ji P, Zhang RY. Acupuncture treated with 30 cases of
insomnia from fire derived from stagnation of liver-qi type
[In Chinese]. Yunnan J Tradit Chin Med 2015;36:61–62.
11
57. Tang ZY. Clinical effect and experience of acupuncture
treatment of 60 patients with insomnia [In Chinese]. Tradit
Chin Med 2016;21:101–101.
58. Wang HY. Clinical effect of acupuncture in the treatment
of 50 cases of insomnia [In Chinese]. Inner Mongol J Tradit
Chin Med 2015;34:70–71.
59. Lai XS. Clinical Study of Insomnia with the Treatment of
Acupuncture with ‘‘Tong du tiaoshen’’ [In Chinese].
Guangdong Sheng, China: Guangzhou University of Chi-
nese Medicine 2015.
60. Xia XF. Clinical observations on Ren-Dumeridian—
Regulating acupuncture therapy for insomnia of yin defi-
ciency with effulgent fire type [In Chinese]. Shanghai J
Acupunct Moxibustion 2012;31:869–870.
61. Xu HB, Liu YM. Clinical observation of electro acupunc-
ture with Qi xue treatment for primary insomnia [In Chi-
nese]. J New Chin Med 2014;46:173–174.
62. Ye TS, Wang QJ, Xie WX, et al. A randomized control
study of abdominal acupuncture treatment for primary in-
somnia [In Chinese]. Shanghai J Acupunct Moxibustion
2008;27:3–5.
63. Zhang Q, Bai Y, Dong GR. Clinical effect of acupuncture
the point by the principle of adjusting Yin-Yang and Qi-
Blood on insomnia by polysomnography. A seminar of new
technology of acupuncture moxibustion and its clinical
application [In Chinese]. Guide China Med 2009;28:140–
144.
64. Zhang ZQ. Clinical observation on acupuncture on back-
shu points for treatment of insomnia [In Chinese].
Guangming J Chin Med 2011;26:1624–1626.
65. Zhao YY, Shang FM, Zong ZY. Clinical observation of
HuatuoJaji points for treating insomnia [In Chinese].
Changzhou Pract Med 2012;28:308–311.
66. Zhang ZH, Liu ZH. Clinical efficacy and impact on in-
flammatory cytokine levels in patients with insomnia acu-
puncture treatment [In Chinese]. Shanxi J Chin Med 2015;
36:1420–1422.
67. Zhou CH. Intractable insomnia parallel randomized con-
trolled study Du Thirteen combined abdominal acupuncture
needles [In Chinese]. J Prac Tradit Chin Inter Med 2015;29:
144–147.
68. Zhao LH. Analysis of acupuncture on the application of
primary Insomnia [In Chinese]. China Contin Med Educ
2016;8:218–220.
69. Shi F. Clinical Effect Observation on the Treatment of
Superficial Needling for Insomnia with the Type of Yin-
xuhuowang [In Chinese]. Fuzhou, China: Fujian University
of Chinese Medicine 2014.
70. Li PF, Guo M. Clinical curative effect evaluation and
analysis of TongduTiaoshen method in treating insomnia
[In Chinese]. Clin J Tradit Chin Med 2017;29:875–878.
71. Liu Y, Feng H, Liu W, et al. Regulation action and nerve
electrophysiology mechanism of acupuncture on arousal
state in patients of primary insomnia [In Chinese]. Chin
Acupunct Moxibustion 2017;37:19–23. 72. Teng YH, XIE
R, Hu B, et al. Forty-two case with insomnia treated by
puncturing ‘‘sleep-promoting acupoints’’ [In Chinese].
Henan Tradit Chin Med 2017;37:882–884.
73. Luo J. Clinical Research on Treating Insomnia with Con-
fluence of Ren Du Yin Qiao and Yang Qiao Meridians [In
Chinese]. Guangdong Sheng, China: Guangzhou University
of Chinese Medicine, 2012.
74. Yao W. Clinical observation of acupuncture to treat in-
somnia by regulating mind [In Chinese]. Chin Med Clin
 12
Res 2014:4–6. 75. Li F, Ren YJ, An S. Clinical observation
on acupuncture Jiaji therapy for insomnia [In Chinese].
JCAM 2014;33:16–18.
76. Bai WJ, Zhang ZQ. Clinical research of heel vessel acu-
puncture treating in insomnia [In Chinese]. Chin Arch
Tradit Chin Med 2011;29:413–414.
77. Luan YH. Clinical Study of Electro-Nape-Acupuncture
Treatment for Primary Insomnia [In Chinese]. Heilongjiang
Sheng, China: Heilongjiang University of Chinese Medi-
cine 2012.
78. Liu XS, Liu F, Sun ML. Clinical observation of electric
acupuncture treatment of 40 patients with insomnia by
acupoints according to Compendium of Acupuncture and
Moxibustion [In Chinese]. Chin Foreign Health Digest
2013;10:392–393.
79. Liu JY, Wei XT. Clinical observation of electro acupunc-
ture treat insomnia with insufficiency of heart and spleen
[In Chinese]. Shenzhen J Integr Tradit Chin West Med
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.
2013;23:168–170.
80. Li F. Clinical Curative Effect Observation of Jiaji Electro-
acupuncture for Patients with Primary Insomnia [In Chi-
nese]. Heilongjiang Sheng, China: Heilongjiang University
of Chinese Medicine, 2015.
81. Liu F, Wang L, Wei QP, et al. Clinical curative observation
of applying acupuncture at Shen-You time to regulate 196
cases of insomnia [In Chinese]. J Sichuan Tradit Chin Med
2015;33:174–176.
82. Wang Y, Li BJ, Liu L, et al. 36 cases of insomnia treat with
shock wave acupuncture [In Chinese]. Jiang xi J Tradit
Chin Med 2015;46:72–73, 80.
83. Xu YQ. Clinical Research on Treating Insomnia of Hy-
peractivity of Fire Due to Yin Deficiency Type by
Strengthening Brain and Tranquilizing Mind and Guiding
Fire to Origin Acupuncture Moxibustion Treatment [In
Chinese]. Zhuangzuzizhiqu, China: Guang Xi University of
Chinese Medicine, 2014.
84. Zhang CH. Effect of Electro Acupuncture Sishencong
Points on PSG and PSQI in the Insomnia [In Chinese].
Jinan, China: Shandong University of Chinese Medicine,
2006.
85. Zhu LL. Clinical Study of Needing by Five Zang-Organs
Regulating Five Zang-Soul in Insomnia Patients [In Chi-
nese]. Changchung, China: Chang Chun University of
Chinese Medicine, 2014.
86. Zhu YL. Effect Observation on Treatment of Primary In-
somnia by Head of a Pin Electric Stimulation [In Chinese].
China Health Standard Management, 2015;26:130–132.
CAO ET AL.
87. Liu ZB, Niu WM. 160 cases of insomnia treated by con-
tinual needling along the hair Line [In Chinese]. Shanxi
Tradit Chin Med 2007;28:716–717.
88. Feng XF. Clinical Efficacy Study of the Scalp Acupuncture
Treat for Sub-health State of Sleep Latency [In Chinese].
Chengdu, China: Chengdu University of Traditional Chi-
nese Medicine, 2014.
89. Dong HL. Shaanxi Scalp Acupuncture in the Treatment of
Primary Insomnia: A Randomized Controlled Clinical Trial
[In Chinese]. Xianyang, China: Shaanxi University of Chi-
nese Medicine 2010.
90. Lai DP. Frontoparietal Nine-Pin Clinical Effect in the
Treatment of Insomnia [In Chinese]. Beijing, China: Beij-
ing University of Chinese Medicine, 2015.
91. Huang LN, An JM, Dong HL, et al. Clinical study on the
efficacy of scalp acupuncture for primary insomnia [In Chi-
nese]. Shanghai J Acupunct Moxibustion 2011;30:596–597.
92. Zhao QH, Yun SW, Ma YP, et al. Efficacy analysis of scalp
acupuncture therapy for insomnia [In Chinese]. Chin Ma-
nipulation Rehabil Med 2014;5:52–54.
93. Song Y, Li H, Zhao CH, et al. 48 cases of insomnia treat
with ankle needle [In Chinese]. Chin J Gerontol 2014;34:
5233–5234.
94. Zhang LX. Acupuncture for Insomnia Accompanying with
Anxiety: A Randomized Placebo Controlled Trial [In
Chinese]. Chengdu, China: Chengdu University of Tradi-
tional Chinese Medicine, 2014.
95. Cheuk DK, Yeung WF, Chung KF, et al. Acupuncture for
insomnia. Cochrane Database Syst Rev 2012:CD005472.
96. Shergis JL, Ni X, Jackson ML, et al. A systematic review of
acupuncture for sleep quality in people with insomnia.
BMC Complement Ther Med 2016;26:11–20.
97. Liu Q, Wang XJ, Zhao Y, et al. Evaluate efficacy,
intervention-level of acupuncture therapy based on meta-
analysis of ordinal data: Primary insomnia [In Chinese].
Liaoning J Tradit Chin Med 2017;1:134–137.
Address correspondence to:
Jian-Ping Liu, PhD, MD
Centre for Evidence-Based Chinese Medicine
Beijing University of Chinese Medicine
Bei San Huan Dong Lu 11
Chaoyang District
Beijing 100029
China
E-mail: liujp@bucm.edu.cn; jianping_l@hotmail.com
(Appendix follows/)
 Appendix Table A1. Search Strategy and Results
Final
Database Search strategy Outcome Update results
Chinese National #1 (topic) ‘‘Zhenjiu’’ (acupuncture) OR ‘‘Zhenci’’ 2723 115 studies update 2000
Knowledge (acupuncture) OR ‘‘Dianzhen’’ on January 5, 2017
Infrastructure (electroacupuncture) OR ‘‘Tizhen’’ (body 172 updates on
acupuncture) OR ‘‘Fuzhen’’ OR ‘‘Toupizhen’’ October 20, 2017
(scalp acupuncture) OR ‘‘Yanzhen’’ (eye
acupuncture) OR ‘‘Wanhuaizhen’’ (Wrist-Ankle
acupuncture)
#2 (topic) ‘‘Shimian’’ (insomnia) OR ‘‘Bumei’’ 35,902
(insomnia) OR ‘‘Shuimianzhangai’’ (sleep
initiation and maintenance and maintenance
disorder)
#3 #1 and #2 and (Full text) ‘‘Random’’ 1713
VIP database (((((Abstract) ‘‘Zhenjiu’’ (acupuncture) OR 710 72 studies update on 880
‘‘Zhenci’’ (acupuncture) OR ‘‘Dianzhen’’ January 5, 2017
(electroacupuncture) OR ‘‘Tizhen’’ (body 98 studies update
acupuncture) OR ‘‘Fuzhen’’ OR ‘‘Toupizhen’’ October 20, 2017
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

(scalp acupuncture) OR ‘‘Yanzhen’’ (eye

acupuncture) OR ‘‘Wanhuaizhen’’ (Wrist-Ankle
acupuncture)) AND ‘‘Shimian’’ (insomnia) OR
‘‘Bumei’’ (insomnia) OR
‘‘Shuimianzhangai’’(sleep initiation and
maintenance and maintenance disorder)) AND
(Full text)‘‘Random’’)) AND 2008–2016 (Date))
Chinese Biomedical #1 (Common filed) ‘‘Zhenjiu’’ (acupuncture) OR 19,688 90 studies update on 859
Literature Service ‘‘Zhenci’’ (acupuncture) OR ‘‘Dianzhen’’ January 5, 2017
System (electroacupuncture) OR ‘‘Tizhen’’ (body 88 studies update on
acupuncture) OR ‘‘Fuzhen’’ OR ‘‘Toupizhen’’ October 20, 2017
(scalp acupuncture) OR ‘‘Yanzhen’’ (eye
acupuncture) OR ‘‘Wanhuaizhen’’ (Wrist-Ankle
acupuncture)
#2 (Common filed) ‘‘Shimian’’ (insomnia) OR 9567
‘‘Bumei’’ (insomnia) OR ‘‘Shuimianzhangai’’
(sleep initiation and maintenance and
maintenance disorder)
#3 (Full text)‘‘Random’’ AND Human 616,886
(Characteristic word) AND 2008–2016 (Date)
#4 #1 and #2 and #3 681
Wanfang database (((((Abstract) ‘‘Zhenjiu’’ (acupuncture) OR 1149 128 studies update 1389
‘‘Zhenci’’ (acupuncture) OR ‘‘Dianzhen’’ on January 5, 2017
(electroacupuncture) OR ‘‘Tizhen’’ (body 112 studies update
acupuncture) OR ‘‘Fuzhen’’ OR ‘‘Toupizhen’’ on October 20,
(scalp acupuncture) OR ‘‘Yanzhen’’ (eye 2017
acupuncture) OR ‘‘Wanhuaizhen’’ (Wrist-Ankle
acupuncture)) AND ‘‘Shimian’’ (insomnia) OR
‘‘Bumei’’ (insomnia) OR
‘‘Shuimianzhangai’’(sleep initiation and
maintenance and maintenance disorder)) AND
(Full text)‘‘Random’’)) AND 2008–2016 (Date))
PubMed #1 (((((((((((((Insomnia[Title/Abstract]) OR (sleep 34,836 5 studies update on 84
initiation and maintenance disorders [Mesh October 20, 2017
Terms]))) OR sleepless[Title/Abstract])) OR
wakefulness[Title/Abstract])) OR
dyssomnia[Title/Abstract])) OR Sleep
deprivation[Title/Abstract])) OR Sleep Initiation
Dysfunction[Title/Abstract])) OR sleep
stages[Title/Abstract]
#2 (((needling[Title/Abstract]) OR 20,319
acupunct*[Title/Abstract])) OR
((electroacupunct*) OR electro-acupunct*)
#3 #1AND #2 240
#4 selected article types for clinical trial 79
(continued)

13
 Appendix Table AT1. (Continued)
Final
Database Search strategy Outcome Update results
Cochrane Library #1 acupunct* or electroacupunct* or ‘‘electro- 167 0 study update on 167
acupunct*’’ or needling October 20, 2017
#2 Insomnia or sleepless or wakefulness or
dyssomnia or [mh ‘‘Sleep Initiation and
Maintenance Disorders’’] or ‘‘Sleep Initiation
Dysfunction’’ or ‘‘sleep stages’’ or ‘‘Sleep
deprivation’’
#3 #1 and #2
#4 selected article types for clinical trial.
Embase #1 ’acupunct*’ OR ’electroacupunct*’ OR 4483 0 study update on 86
’electro-acupunct*’ OR ’needling’ AND October 20, 2017
[randomized controlled trial]/lim AND
[humans]/lim AND ([embase]/lim OR [medline]/
lim)
#2 ’insomnia’: ab OR ’sleepless’: ab OR 2138
’wakefulness’: ab OR ’dyssomnia’: ab OR ’sleep
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

initiation and maintenance disorders’: ab OR

’sleep deprivation’: ab AND [randomized
controlled trial]/lim AND [humans]/lim
#3 #1 AND #2 86
MEDLINE #1 (acupunct* or electroacupunct* or electro- 13,323 3 studies update on 171
acupunct* or needling). ab. October 20, 2017
#2 limit 1 to humans 10,913
#3 (insomnia or sleepless or wakefulness or 19,380
dyssomnia). ab.
#4 (sleep initiation and maintenance disorders). ab. 6
#5 sleep deprivation. ab. 4981
#6 3 or 4 or 5 23,547
#7 1 and 6 171
#8 2 and 7 168
PsycINFO #1 (acupunct* or electroacupunct* or electro- 1796 0 study update on 37
acupunct* or needling) October 20, 2017
#2 (insomnia or sleepless or wakefulness or 13,302
dyssomnia)
#3 sleep deprivation 3687
#4 (sleep initiation and maintenance disorders) 6
#5 #2 or #3 or #4 16,407
#6 #1 and #5 37
AMED #1 (acupunct* or electroacupunct* or electro - 4248 0 study update on 58
acupunct* or needling). ab. October 20, 2017
#2 (insomnia or sleepless or wakefulness or 373
dyssomnia). ab.
#3 sleep deprivation. ab. 20
#4 (sleep initiation and maintenance disorders). 2
mp. [mp = abstract, heading words, title]
#5 #2 or #3 or #4 390
#6 #1 and #5 58
CINAHL S1: AB acupunct* OR AB electroacupunct* OR 4262 0 study update on 59
AB electro - acupunct* OR AB needling Oct. 20, 2017
S2: AB insomnia OR AB sleepless OR AB 4519
wakefulness OR AB dyssomnia OR AB sleep
deprivation OR AB ((sleep initiation and
maintenance disorders))
S3: (AB insomnia OR AB sleepless OR AB 59
wakefulness OR AB dyssomnia OR AB sleep
deprivation OR AB ((sleep initiation and
maintenance disorders))) AND (S1 AND S2)
We totally searched four Chinese databases and six English databases on May 1, 2016 and update on January 5, 2017 and October 20,
2017, separately. Update years were covered from 2008 to 2017, and there was no language restriction. All these searching tasks were run
by Yu Meili and exported to Note Express and EndNote. The strategy searches for complementary were as term and topic, and its alternate
names as term. Acupuncture and insomnia context were included to narrow search results. To further narrow the search results, results were
then combined with a search set for the term clinical trial. Details are listed above.

14
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Appendix Table A2. Characteristics of 73 Included Trials

Interventions
Treatment
Sample Acupuncture group duration
Study ID size (T/C) Age – SD (y) Disease duration (no. of points used) Control group (days) Outcomes

Bai WJ 2011 30/30 T: 38.36 – 13.52 T: 46.89 – 16.42 (d) EA (7 + x), once daily, last for Estazolam 1 mg, once per night for 30 1, 2
C: 39.23 – 13.14 C: 43.35 – 15.94 (d) 30 days 30 days
Bi XJ 2014 40/40 T: 50.5 – 13.1 T: 37.53 – 9.02 (m) MA (6), 20 min per session, once Estazolam 1 mg once per night 30 1, 2
C: 52.7 – 11.5 C: 35.75 – 7.8 (m) daily
Dong HL 2010 92/92 T: 42.27 – 11.67 T: 39.12 – 25.68 (m) SA (5), 30 min per session once Estazolam 2 mg, once per night for 20 1, 2, 6
C: 41.51 – 11.25 C: 39.05 – 26.69 (m) daily, 5 times a week 20 days
Deng XF 2014 30/30 T: 25.64 – 4.85 T: 11.0 – 2.3 (m) MA (7 + x), 30 min per session, once Alprazolam 0.4 mg once per night 20 1, 2, 6
C: 26.17 – 4.64 C: 12.0 – 1.7 (m) daily
Fu FC 2012 43/43 T: 48.5 – 13.6 T: 98.5 – 53.6 (m) MA (6 + x), 30–60 min per session Estazolam 0.4 mg · 2 tablets, once 30 2, 4
C: 46.5 – 14.6 C: 108.5 – 43.6 (m) once daily per night
Feng XF 2014 33/33/33 18–60 2–6 (m) SA (3), 30 min per session once Medical group: Chlordiazepoxide 14 1, 2, 6, 7, 9
daily for 5 days 1 week, last 2 5 mg, once per night for 2 weeks;
weeks Black group: no treatment
Gu XT 2013 30/30 T: 52.33 – 9.998 None MA (4), 30 min per session, once Estazolam 1 mg, once per night for 14 1, 2, 3
C: 52.43 – 8.661 daily for 2 weeks 2 weeks
He FL 2010 30/38 Unknown Unknown MA (4 + x), 30–40 min per session, Estazolam once per night 28 1, 5

15
once daily
Huang LN 2011 30/30 T: 40 T: 17 (m) SA (5 + x), 30 min per session daily, Estazolam 2 mg once per night 20 1, 2
C: 38 C: 19.9 (m) 5 times weekly
Ji XD 2015 35/35 T: 37 – 11 T: 1.8 – 1.2 (y) MA (5), 20 min per session, once Trazodone 50–100 mg once per 28 1, 6
C: 36 – 13 C: 1.7 – 1.4 (y) daily night
Kou JY 2003 30/30 T: 40.25 – 14.9 T: 4.02 – 2.56 (y) MA (17), 20 min per session once Zolpidem tartrate (Stilnox) 10 mg 28 1, 2, 6
C: 41.17 – 15.38 C: 3.86 – 2.77 (y) daily once per night (or 5 mg for those
who are >65 years)
Liu ZB 2007 80/80 T: 48.06 – 6.12 T: 18.05 – 2.8 (m) SA (5), 6 h per session once daily Clonazepam 4 mg once per night 28 1, 2
C: 49.1 – 2.24 C: 16.99 – 8.12 (m) for 5 days a week
Liang LZ 2009 30/30 T: 36.09 – 16.27 T: 64.86 – 12.03 (m) MA (4 + x), 30 min per session, once Zolpidem 10 mg, once per night, for 30 1, 2, 6
C: 36.04 – 15.88 C: 70.36 – 12.60 (m) daily 30 days
Lv JC 2010 30/30 62.3 – 9.5 16.5 – 5.3 (y) MA (7 + x), 30 min per session once Conventional treatmenta once daily 20 1
daily plus control treatment for 3 weeks
Li XY 2010 30/30 T: 59.5 – 8.11 T: 9.52 – 4.66 (m) MA (5 + x), 30 min per session, once Estazolam 1 mg once per night 28 1, 2, 7
C: 58.2 – 7.83 C: 8.93 – 4.73 (m) daily
Luo J 2012 52/52 T: 52.03 – 10.02 T: 4.34 – 1.15 (m) EA (10 + x), 30 min per session, Estazolam 2 mg, once per night, last 60 1, 2
C: 50.60 – 9.24 C: 4.32 – 1.44 (m) once daily 8 weeks
Luan YH 2012 30/30 21–60 T: 27.77 – 23.84 (m) EA (2 + x), 30 min per session, once Estazolam 1 mg once per night 28 1, 2
C: 30.03 – 24.35 (m) daily
(continued)
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Appendix Table A2. (Continued)

Interventions
Treatment
Sample Acupuncture group duration
Study ID size (T/C) Age – SD (y) Disease duration (no. of points used) Control group (days) Outcomes

Liu XS 2013 40/40 T: 43.51 – 3.52 T: 8.41 – 2.33 (m) EA (6), 30 min per session, 3 times Estazolam 0.4–0.8 mg once per 30 1
C: 42.02 – 3.60 C: 8.35 – 2.65 (m) weekly night
Liu JY 2013 45/45 T: 36.74 – 9.31 T: 25.3 – 7.6 (m) EA (7), 30 min per session once Diazepam 5 mg once per night 10 1, 2
C: 35.66 – 8.99 C: 25.9 – 10.3 (m) daily
Liu Y 2014 30/30 T: 46.31 – 18.26 T: 3.78 – 1.98 (m) MA (7), once daily, 6 days a week, Estazolam 1 mg, once per night, last 28 1, 2, 4, 5, 6
C: 41.29 – 17.64 C: 3.54 – 1.52 (m) last 4 weeks 4 weeks
Liu YY 2014 35/35 T: 45.56 – 13.16 unknown MA (2), 30 min per session, once Estazolam 1 mg once per night 14 1
C: 41.36 – 12.50 daily
Lai XY 2014 35/33 T: 41.53 – 13.07 T: 9.10 – 5.33 (y) MA (

16
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

17
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

18
 Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

19
 Appendix Table A3. Estimate Effect of Overall Quality of Sleep
Outcome or subgroup title No. of studies No. of participants Effect estimate, MD (95% CI) pa
1. Total score on the PSQI
1.1. Acupuncture versus no treatment
Feng 2014 1 65 -5.68 (-7.27 to -4.09) <0.01
Xiao 2013 1 40 -5.40 (-7.52 to -3.28) <0.01
Subtotal 2 105 -5.58 (-6.85 to -4.31) <0.01
1.2. Acupuncture versus sham acupuncture
Zhang LX 2014 1 65 -8.64 (-9.59 to -7.33) <0.01
1.3. Acupuncture plus estazolam versus estazolam alone
Sun JQ 2011 1 96 -4.78 (-5.67 to -3.89) <0.01
Xu SF 2016 1 70 -2.86 (-4.90 to -0.82) <0.01
Sun JQ 2011 1 96 -4.78 (-5.67 to -3.89) <0.01
Wang JP 2015 1 65 -2.42 (-4.56 to -0.28) 0.03
Subtotal 4 253 -3.17 (-4.74 to -1.61) <0.01
1.4. Acupuncture versus western medication
Subgroup 1: acupuncture versus benzodiazepine (not-pooled trials)
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Lai XY 2014 1 68 -0.34 (-1.67, 0.99) 0.62

Wang ZH 2014 1 80 -0.52 (-1.91, 0.87) 0.46
Xia XF 2012 1 82 -4.95 (-5.89, -4.01) <0.01
Wang Y 2015 1 72 -1.83 (-2.49, -1.17) <0.01
Xu YQ 2014 1 60 -4.04 (-5.36, -2.72) <0.01
Dong HL 2010 1 175 -5.65 (-6.82, -4.48) <0.01
Huang LN 2011 1 60 -2.57 (-4.44, -0.70) <0.01
Zhang ZH 2015 1 119 -2.92 (-3.25, -2.59) <0.01
Feng XF 2014 1 65 -1.63 (-3.10, -0.16) 0.03
Liu JY 2013 90 -2.22 (-2.95, -1.49) <0.01
Zhao YY 2012 1 50 -1.75 (-4.12, 0.62) 0.15
Liu ZB 2007 1 160 -0.91 (-1.55, -0.27) <0.01
Subgroup 2: Acupuncture versus nonbenzodiazepine
Deng XF 2014 1 60 0.10 (-1.00 to 1.20) 0.86
Liang ZL 2009 1 57 -1.15 (-2.87 to 0.57) 0.19
Ji XD 2015 1 70 -2.60 (-4.33 to -0.87) <0.01
Kou JY 2003 1 60 -0.51 (-2.16 to 1.14) 0.54
Song Y 2014 1 103 -1.56 (-2.68 to -0.44) <0.01
Wang JP 2015 1 63 -1.96 (-3.97 to 0.05) 0.06
Zhang ZQ 2011 1 60 -1.71 (-3.20 to 0.22) 0.02
Zhao QH 2014 1 102 -0.65 (-1.49 to 0.19) 0.13
Ren LN 2014 1 68 0.32 (-0.91, 1.55) 0.61
Subtotal 9 643 -0.94 (-1.55, -0.33) <0.01
2. Total score on the AIS comparing acupuncture with estazolam
Liu 2014 1 30 -2.63 (-4.54 to -0.72) <0.01
Yao 2014 1 60 -1.72 (-2.97 to -0.47) <0.01
Li PF 2017 1 60 -1.26 (-2.36 to -0.16) 0.04
Subtotal 3 180 -1.64 (-2.40 to -0.89) <0.01
3. Total score on the SPIEGEL compared acupuncture with estazolam
Liu Y 2014 1 30 -2.05 (-4.03 to -0.07) 0.04
Li PF 2017 1 60 -2.30 (-3.55 to -1.05) <0.01
Yao W 2014 1 60 -4.00 (-5.69 to -2.31) <0.01
Fu FC 2012 1 86 -2.87 (-4.18 to -1.56) <0.01
Xu YQ 2014 1 60 -3.30 (-5.02 to -1.58) <0.01
Subtotal 5 326 -2.86 (-3.54 to -2.18) <0.01
a
Keep two decimal places.
AIS, Athens Insomnia Scale; CI, confidence interval; MD, mean difference; PSQI, Pittsburgh Sleep Quality Index.

20
 Appendix Table A4. The Result of Single Study on Sleep Disorders Compared Acupuncture
with Benzodiazepines
Study ID No. of studies No. of participants Effect estimate, MD (95% CI) p
Bai WJ 2011 1 60 -0.45 (-0.88 to -0.02) 0.04
Dong HL 2010 1 175 -0.50 (-0.75 to -0.25) <0.01
Gu XT 2013 1 60 0.37 (0.12 to 0.62) <0.01
Lai XS 2015 1 60 -0.37 (-0.64 to -0.10) <0.01
Li XY 2010 1 60 -0.51 (-0.77 to -0.25) <0.01
Liu F 2015 1 196 -0.38 (-0.55 to -0.21) <0.01
Luan YH 2012 1 60 0.03 (-0.22 to 0.28) 0.82
Luo J 2012 1 104 -0.44 (-0.61 to -0.27) <0.01
Shi F 2014 1 60 -0.04 (-0.27 to 0.19) 0.73
Su D 2011 1 76 0.10 (-0.22 to 0.42) 0.54
Wang YJ 2016 1 68 -0.41 (-0.74 to -0.08) 0.01
Wang ZY 2015 1 60 0.05 (-0.28 to 0.38) 0.76
Wu X 2007 1 62 -0.38 (-0.89 to 0.13) <0.01
Xu YQ 2014 1 60 -0.06 (-0.45 to 0.33) 0.76
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Zhang CH 2006 1 50 0.08 (-0.33 to 0.49) 0.70

Zhang ZH 2015 1 119 -0.31 (-0.44 to -0.18) <0.01
Zhao LH 2016 1 106 -1.20 (-1.43 to -0.97) <0.01
Zhou CH 2015 1 92 -0.19 (-0.35 to -0.03) 0.02
Zhou JC 2013 1 70 -0.51 (-0.96 to -0.07) 0.02
Total 19 1598 — —
CI, confidence interval; MD, mean difference.

Appendix Table A5. The Result of Single Study on Daytime Dysfunction Compared Acupuncture
with Benzodiazepines
Study ID No. of studies No. of participants Effect estimate, MD (95% CI) p
Bai WJ 2011 1 60 -0.62 (-1.06 to -0.18) <0.01
Dong HL 2010 1 175 -0.79 (-1. 05 to -0.53) <0.01
Gu XT 2013 1 60 -1.60 (-1.92 to -1.28) <0.01
Lai XS 2015 1 60 -0.13 (-0.43 to 0.17) 0.40
Li XY 2010 1 60 -0.96 (-1.24 to -0.68) <0.01
Liu F 2015 1 196 -0.69 (-0.90 to -0.48) <0.01
Liu JY 2013 1 90 -0.15 (0.45 to 0.15) 0.33
Luan YH 2012 1 60 0.06 (-0.19 to 0.31) 0.64
Luo J 2012 1 104 -0.67 (-0.93 to 0.41) <0.01
Shi F 2014 1 60 -0.22 (-0.50 to 0.06) 0.12
Song JF 2013 1 70 -0.35 (-0.73 to 0.03) 0.07
Su D 2011 1 76 -0.51 (-0.82 to -0.20) <0.01
Wang YJ 2016 1 68 -0.50 (-0.87 to -0.13) <0.01
Wang ZY 2015 1 60 -0.12 (-0.34 to 0.10) 0.27
Wu X 2007 1 62 -1.72 (-2.11 to -1.33) <0.01
Xu YQ 2014 1 60 -0.97 (-1.29 to -0.65) <0.01
Zhang CH 2006 1 50 -0.44 (-0.86 to -0.02) 0.04
Zhang ZH 2015 1 119 -0.21 (-0.32 to -0.10) <0.01
Zhao LH 2016 1 106 -1.60 (-1.83 to -1.37) <0.01
Zhou CH 2015 1 92 -0.78 (-1.14 to -0.42) <0.01
Zhou JC 2013 1 70 -0.52 (-0.96 to -0.07) 0.02
Total 21 1812 — —
CI, confidence interval; MD, mean difference.

21
 Appendix Table A6. Subgroup Analysis of Sleep Onset Latency Compared Acupuncture
with Benzodiazepines
Subgroup title or study ID No. of studies No. of participants Effect estimate, MD (95% CI) p
1. Electroacupuncture versus benzodiazepines
Bai WJ 2011 1 60 -0.43 (-0.84 to -0.02) 0.04
Liu JY 2013 1 90 -0.79 (-1.06 to -0.52) <0.01
Luan YH 2012 1 60 0.07 (-0.23 to 0.37) 0.65
Xu YQ 2014 1 60 0.46 (0.10 to 0.82) 0.01
Zhang CH 2006 1 50 0.08 (-0.22 to 0.38) 0.60
2. Scalp acupuncture versus benzodiazepines
Dong HL 2010 1 175 -0.15 (-0.40 to 0.10) 0.23
3. Manual acupuncture versus benzodiazepines
Gu XT 2013 1 60 -0.53 (-0.75 to -0.32) <0.01
Lai XS 2015 1 60 -0.23 (-0.51 to 0.05) 0.11
Li XY 2010 1 60 -0.12 (-0.54 to 0.30) 0.57
Liu F 2015 1 196 -0.33 (-0.57 to -0.09) <0.01
Shi F 2014 1 60 -0.59 (-1.00 to -0.18) <0.01
Song JF 2013 1 70 -0.57 (-0.89 to -0.25) <0.01
Su D 2011 1 76 -0.54 (-0.89 to -0.19) <0.01
Wang YJ 2016 1 68 -0.17 (-0.40 to 0.06) 0.15
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Wang ZY 2015 1 60 -0.49 (-0.70 to -0.28) <0.01

Wu X 2007 1 62 -0.38 (-0.89 to 0.13) 0.14
Zhang ZH 2015 1 119 -0.32 (-0.46 to -0.18) <0.01
Zhao LH 2016 1 106 -0.40 (-0.73 to -0.07) 0.02
Zhou CH 2015 1 92 0.03 (-0.12 to 0.18) 0.69
Zhou JC 2013 1 70 -0.63 (-1.29 to 0.04) 0.06
Subgroup analysis 14 1213 -0.35 (-0.47 to -0.23) <0.01
CI, confidence interval; MD, mean difference.

Appendix Table A7. Safety and Adverse Events Reported

No. of adverse events
Acupuncture Control Effect estimate,
Study ID group group RR (95% CI) p Adverse reactions
1. Acupuncture versus western medication
Liu Y 2014 1/30 7/30 0.11 (0.01–0.99) 0.05 AG: bent needle, stuck needle, fainting; CG:
drowsiness, dizziness, dry mouth, fatigue
Zhou JC 2013 13/35 26/35 0.20 (0.07–0.57) 0.02 AG: drowsiness (2), bleeding (8), hematoma (3);
CG: drowsiness (5), dizziness and headache (10),
dry mouth (2), fatigue (6), loss of appetite (3)
Liang LZ 2009 0/29 6/28 0.06 (0.00–1.10) 0.06 CG: dizziness (4), vomiting (2);
Xiao BB 2013 1/20 0/20 3.15 (0.12–82.16) 0.49 AG: fainting during acupuncture treatment (1);
Dong HL 2010 0/88 4/87 0.10 (0.01–1.98) 0.13 CG: dizziness, fatigue, flustered, bloating,
unresponsive, or nausea
Su D 2011 1/39 10/37 0.09 (0.01–0.71) 0.02 AG: fainting during acupuncture treatment (1)
CG: headache (3), drowsiness (2), dizziness (2),
fatigue (2), nausea (1)
Kou JY 2003 1/30 9/30 0.11 (0.01–0.82) 0.03 AG: fainting during acupuncture treatment (1)
CG: headache (2), dizziness (4), fatigue (1), dry
mouth (1), excitement (1)
Xu YQ 2014 7/30 18/30 0.39 (0.19–0.79) 0.09 AG: subcutaneous hematoma (7); CG: daytime
sleepiness (18)
Zhu LL 2014 1/30 15/30 0.07 (0.01–0.47) 0.07 AG: fainting during acupuncture treatment (1)
CG: headache (3), dizziness (7), daytime sleepiness
(3), dry mouth (2)
Deng SF 2014 1/30 20/30 0.05 (0.01–0.35) 0.03 AG: skin flushing (1); Control group: daytime
sleepiness (20)
Liu F 2015 2/98 0/98 5.00 (0.24–102.82) 0.30 AG: hypoglycemia (1), fainting during acupuncture
treatment (1)
Subtotal 28/459 115/455 0.26 (0.18–0.36) <0.01 —
2. Acupuncture versus sham acupuncture
Zhang LX 4/33 3/34 1.43 (0.29–6.92) 0.66 Acupuncture group: bleeding (3), local pain (1);
2014 Sham acupuncture group: bleeding (3)
Another five trials reported no adverse event, including: Wu X 2014, Feng XF 2014, Ren LN 2014, Zhang ZH 2015, Zhou CH 2015.
AG, acupuncture group; CG, control group; CI, confidence interval; RR, risk ratio.

22
 Appendix Table A8. PRISMA 2009 Checklist
Section/topic # Checklist item Reported on page #
TITLE
Title 1 Identify the report as a systematic review, meta-analysis, or 1
both.
ABSTRACT
Structured summary 2 Provide a structured summary, including, as applicable: 1–2
background; objectives; data sources; study eligibility
criteria, participants, and interventions; study appraisal and
synthesis methods; results; limitations; conclusions and
implications of key findings; and systematic review
registration number.
INTRODUCTION
Rationale 3 Describe the rationale for the review in the context of what is 2
already known.
Objectives 4 Provide an explicit statement of questions being addressed with 2
reference to participants, interventions, comparisons,
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

outcomes, and study design (PICOS).

METHODS
Protocol and 5 Indicate if a review protocol exists, if and where it can be 2
registration accessed (e.g., Web address), and, if available, provide
registration information, including registration number.
Eligibility criteria 6 Specify study characteristics (e.g., PICOS, length of follow-up) 2
and report characteristics (e.g., years considered, language,
publication status) used as criteria for eligibility, giving
rationale.
Information 7 Describe all information sources (e.g., databases with dates of 3
sources coverage, contact with study authors to identify additional
studies) in the search and date last searched.
Search 8 Present full electronic search strategy for at least one database, 3, Appendix 2
including any limits used, such that it could be repeated.
Study selection 9 State the process for selecting studies (i.e., screening, 3
eligibility, included in systematic review, and, if applicable,
included in the meta-analysis).
Data collection 10 Describe method of data extraction from reports (e.g., piloted 3
process forms, independently, in duplicate) and any processes for
obtaining and confirming data from investigators.
Data items 11 List and define all variables for which data were sought (e.g., 3
PICOS, funding sources) and any assumptions and
simplifications made.
Risk of bias in 12 Describe methods used for assessing risk of bias of individual 3
individual studies studies (including specification of whether this was done at
the study or outcome level), and how this information is to be
used in any data synthesis.
Summary measures 13 State the principal summary measures (e.g., risk ratio, 3–4
difference in means).
Synthesis of results 14 Describe the methods of handling data and combining results of 3–4
studies, if done, including measures of consistency (e.g., I2)
for each meta-analysis.
Risk of bias across 15 Specify any assessment of risk of bias that may affect the 3
studies cumulative evidence (e.g., publication bias, selective
reporting within studies).
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or 3–4
subgroup analyses, meta-regression), if done, indicating
which were prespecified.
RESULTS
Study selection 17 Give numbers of studies screened, assessed for eligibility, and 4, Figure1
included in the review, with reasons for exclusions at each
stage, ideally with a flow diagram.
Study characteristics 18 For each study, present characteristics for which data were 4, Table 1
extracted (e.g., study size, PICOS, follow-up period) and
provide the citations.
Risk of bias within 19 Present data on risk of bias of each study and, if available, any 4–5, Table 2, Figure2
studies outcome level assessment (see item 12).
(continued)

23
 Appendix Table A8. (Continued)
Section/topic # Checklist item Reported on page #
Results of individual 20 For all outcomes considered (benefits or harms), present, for 5–6, table 3–7
studies each study: (a) simple summary data for each intervention
group (b) effect estimates and confidence intervals, ideally
with a forest plot.
Synthesis of results 21 Present results of each meta-analysis done, including 5–7, table 3–7
confidence intervals and measures of consistency.
Risk of bias across 22 Present results of any assessment of risk of bias across studies 4
studies (see Item 15).
Additional analysis 23 Give results of additional analyses, if done (e.g., sensitivity or 5–7
subgroup analyses, meta-regression [see Item 16]).
DISCUSSION
Summary of evidence 24 Summarize the main findings, including the strength of 7–8
evidence for each main outcome; consider their relevance to
key groups (e.g., healthcare providers, users, and policy
makers).
Downloaded by IDAHO STATE UNIV LIBR from www.liebertpub.com at 04/24/19. For personal use only.

Limitations 25 Discuss limitations at study and outcome level (e.g., risk of 8

bias), and at review-level (e.g., incomplete retrieval of
identified research, reporting bias).
Conclusions 26 Provide a general interpretation of the results in the context of 9
other evidence, and implications for future research.
FUNDING
Funding 27 Describe sources of funding for the systematic review and other 9
support (e.g., supply of data); role of funders for the
systematic review.
From: Moher D, Liberati A, Tetzlaff J, Altman DG; The PRISMA Group. Preferred reporting items for systematic reviews and meta-
analyses: The PRISMA statement. PLoS Med 2009;6:e1000097.
For more information, visit: www.prisma-statement.org.

24