self-acupressure for symptom management

Complementary Therapies in Medicine (2015) 23, 68—78 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.elsevierhealth.com/journals/ctim Effect of self-acupressure for symptom management: A systematic review Hyun Jin Song a, Hyun-Ju Seo b,∗, Heeyoung Lee c, Heejeong Son d, Sun Mi Choi e, Sanghun Lee e a School of Pharmacy, Sungkyunkwan University, Suwon, South Korea b Department of Nursing, College of Medicine, Chosun University, 309 Pilmum-daero, Dong-gu, Gwangju 501-759, South Korea c Public Health Medical Service, Seoul National University Bundang Hospital, Seongnam, South Korea d Graduate School of Public Health, Seoul National University, Seoul, South Korea e Korea Institute of Oriental Medicine, Daejeon, South Korea Available online 4 December 2014 KEYWORDS Abstract Acupressure; Objectives: To assess the efficacy and safety of self-administered acupressure to alleviate Self-administration; symptoms of various health problems, including allergic disease, cancer, respiratory disease, Signs and Symptoms dysmenorrhea, perceived stress, insomnia, and sleep disturbances. Methods: We searched core, Korean, Chinese, and Japanese databases, including Ovid-MEDLINE, Ovid-EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), the Cumulative Index to Nursing and Allied Health Literature (CINAHL), six representative electronic Korean medical databases, China National Knowledge Infrastructure (CNKI), and Japan Science and Technology Information Aggregator (J-STAGE). We included randomized controlled trials (RCTs) and quasi-RCTs that examined disease-specific effects or symptom relief, adverse reactions, and quality-of-life (QOL) for self-administered acupressure. Data collection and assessment of the methodological quality of the included studies were conducted by two independent reviewers. Results: Eight RCTs and two quasi-RCTs showed positive effects and safety of self-acupressure therapy in clinically diverse populations. Quality assessment revealed moderate quality for the RCTs, with 50% or more of the trials assessed as presenting a low risk of bias in seven domains. All of the selected 10 studies reported positive effects for primary outcomes of self- acupressure therapy for symptom management, including significant improvements in symptom scores in allergic disease, nausea and vomiting in cancer, symptom scores in respiratory disease, pain symptoms in dysmenorrhea, and stress/fatigue scores and sleep disturbances in healthy people. ∗ Corresponding author. Tel.: +82 62 230 6320; fax: +82 62 230 6329. E-mail address: [email protected] (H.-J. Seo). http://dx.doi.org/10.1016/j.ctim.2014.11.002 0965-2299/© 2014 Elsevier Ltd. All rights reserved.

Effect of self-acupressure for symptom management: A systematic review 69 Conclusions: Our findings suggest that self-administered acupressure shows promise to alleviate the symptoms of various health problems. Therefore, further research with larger samples and methodologically well-designed RCTs is required to establish the efficacy of self-administered acupressure. © 2014 Elsevier Ltd. All rights reserved. Introduction intervention for high sensitivity, and there were no limita- tions on diseases, patients, or language. Acupressure involves applying pressure to specific points on the body using the hands, fingers, thumbs, elbows, feet, Study selection or various devices.1—3 According to the meridian theory, acupressure increases the flow of energy (qi) in the body; The inclusion criterion was self-administered acupressure. therefore, it is effective for managing symptoms of disease.4 The final treatment interventions were carried out by the Acupuncture, which is considered a representative treat- participants after they received training in the interven- ment of alternative and complementary medicine, uses fine tion procedure. All studies related to self-administered needles inserted in specific points on the body. Acupres- acupressure were included, regardless of the acupressure sure utilizes application of pressure to acupoints without techniques applied. The studies included in this systematic penetrating the skin.1,2 Acupressure is non-invasive and typ- review were randomized controlled trials (RCTs) or studies ically painless, and adverse reactions caused by insertion with a quasi-RCT design. Comparisons were sham acupres- of fine needles can be avoided.3 Moreover, self-acupressure sure, standard of care, and no treatment. The participants can be administered in any situation, regardless of time came from the general population or were patients with and place, and self-treatment can be conducted without any type of disease. The outcomes included disease-specific expensive costs. Self-acupressure is acupressure performed effects, symptom relief, adverse reactions, and quality of by trained participants without treatment by practitioners life. Studies with final interventions performed by prac- or healthcare providers. titioners or healthcare providers were excluded. Studies reporting a mixture of interventions were also excluded, Several systematic reviews of acupressure performed by as were those reporting non-human target populations and practitioners have been reported.2,3,5 However, no system- interventions other than acupressure. In addition, there was atic review of self-acupressure with its various advantages, no a priori exclusion criterion based on publication language. such as economical and convenient use, has been published. In order to develop acupressure as a self-administered ther- Data extraction and quality assessment apy, it was necessary to understand the evidence supporting self-acupressure. Thus, this systematic review aimed to esti- Two researchers independently extracted the data using a mate the efficacy and safety of self-acupressure to alleviate predetermined extraction form. For papers where the major symptoms of various health problems. outcomes were presented in graphical form, the corre- sponding author was contacted. The intention-to-treat (ITT) Methods analysis was used if both ITT and per-protocol (PP) analy- ses were presented in the study. Disagreements regarding Data sources and searches extracted data were resolved by consensus with a third person. The extraction form included the study design, We conducted an extensive and comprehensive literature the number and characteristics of participants, follow-up search of core, Korean, Chinese, and Japanese databases. periods, primary and secondary outcomes, country of pub- The core databases were Ovid-MEDLINE, Ovid-EMBASE, the lication, treatment regimens, main acupoints, and methods Cochrane Central Register of Controlled Trials (CENTRAL), of teaching self-acupressure. and the Cumulative Index to Nursing and Allied Health Lit- erature (CINAHL). Korean databases included KoreaMed, The quality of the RCTs was assessed using the Cochrane Korean Medical Database (KMbase), Korean Studies Informa- Risk of Bias (RoB) tool.6 The RoB tool assessed seven tion Service System (KISS), National Discovery for Science dimensions: sequence generation, allocation concealment, Leaders (NDSL), and Korea Institute of Science Technology blinding of participants, personnel and outcomes, incom- Information (KISTI). We also utilized the Oriental Medicine plete outcome data, selective outcome reporting, and other Advanced Searching Integrated System (OASIS). Chinese and sources of bias. In this study, funding sources were con- Japanese databases were included in our search, specifically firmed in order to assess other sources of bias. Quasi-RCTs China National Knowledge Infrastructure (CNKI) and Japan were assessed using the Risk of Bias Assessment tool for Science and Technology Information Aggregator (J-STAGE). Non-randomized Studies (RoBANS), which evaluated selec- The search was conducted from 28 April 2013 to 7 May 2013. tion of participants, confounding variables, measurement The search terms used were ‘‘acupressure,’’ ‘‘self,’’ and of intervention (exposure), blinding for outcome assess- ‘‘acupuncture.’’ Medical Subject Headings (MeSH) terms ment, incomplete outcome data, and selective outcome and text words were used to adjust the search character- reporting.7 istics for each database. The only search limitation was

70 H.J. Song et al. Figure 1 Study flow diagram. The studies included in this review were all designed as head-to-head trials. Eight RCTs and two quasi-RCTs Data synthesis and analysis were selected, with one crossover study design among Because the included studies were not limited to partici- the RCTs. The total number of participants was 659. pants with a specific disease, the estimation of outcomes Two of the trials reported results for two comparative did not possess inherent homogeneity of patients and groups: placebo (sham acupressure or no acupressure) interventions. Therefore, outcomes were not combined and/or usual care (Table 1). Clinical problems included quantitatively, but were presented descriptively in this atopic dermatitis, allergic rhinitis, breast cancer, stom- study. ach cancer, bronchiectasis, chronic obstructive pulmonary disease (COPD), dysmenorrhea, stress, sleep disturbance, Results and fatigue. Follow-up periods ranged from three days to three months. The research countries included three trials The search produced 793 potential articles from all from the United States, two trials from South Korea, and databases. After reviewing the abstracts, 768 studies were one trial each from Taiwan, Iran, Japan, Hong Kong, and excluded, and 25 potential studies were selected. Review Australia. of the full articles resulted in exclusion of an additional 15 studies.3,8—21 Studies were excluded because they did not Selected acupoints varied depending on the disease. report self-acupressure, did not directly compare acupres- Even when the same disease was being treated, different sure and placebo/control, did not include extracted clinical acupoints were selected according to the trial’s meth- outcomes, were duplicate studies, or were not original arti- ods (Table 2). Training for participants in the technique cles. At the end of the review process, 10 studies (eight of self-acupressure encompassed a wide variety of teach- RCTs and two quasi-RCTs) out of 793 articles were selected ing methods utilized by the investigators. Investigators in for inclusion in this review (Figure 1). five of 10 trials conducted training using an Acupressure Teaching Booklet. Acupoints were marked using ink or tape Among the eight RCTs, 75% (6/8) of the studies were in three trials. Participants completed training sessions evaluated as having a low risk of bias for random sequence in one trial, and participants practiced self-acupressure generation, blinding of participants, blinding of outcome under direct observation by the practitioner in another assessment, and selective reporting (Figure 2). The per- trial. centage of RCTs assessed as having a low risk of bias for incomplete outcome data and funding source was 50% (4/8). Self-acupressure for allergic disease In the case of allocation concealment, the percentage of low-risk studies was only 25% (2/8). One quasi-RCT was Xue et al. compared real ear acupoints (REAP) with sham evaluated as having a low risk of bias for blinding for out- ear acupoints (SEAP) in 63 patients with allergic rhinitis.22 come assessment, incomplete outcome data, and selective REAP were Shenmen (TF 4), Neibi (TG 4), Fei (CO 14), outcome reporting. All other dimensions evaluated for the Fengxi (SF 1, 2i), and Shenshangxian (TG 2p); SEAP were quasi-RCTs were found to have a high or unclear risk of bias Lun 2 (HX 10), Jian (SF 4, 5), Suogu (SF 6), Zhen (AT (Figure 3). 3), and Ya (LO 1). Outcomes for REAP were significantly better than for SEAP in total nasal symptom score (TNSS) (P = 0.02), global nasal/nonnasal symptom score (P = 0.04), sneezing score (P = 0.01), and regular activities at home and work (P = 0.04). However, there were no significant dif- ferences between groups for results of the standardized Rhinoconjunctivitis Quality of Life Questionnaire (RQLQ) and medication use. Mild adverse reactions were reported by 29.0% of the patients in the REAP group and 21.9% of the patients in SEAP group (P = 0.51). Serious adverse reactions were not reported in either group. Lee et al. studied 15 adult patients with pruritic atopic dermatitis.23 Acupressure plus standard care was compared with standard care only. Acupressure was admin- istered for 3 min to Quchi (LI 11), an acupoint located on the left arm lateral to the antecubital fossa, three times per week for 4 weeks. Standard care included prescription and over-the-counter medications or lotions. Acupressure plus standard care compared with stan- dard care only resulted in significant decreases in the Visual Analogue Scale (VAS) score (P = 0.04), Investigator’s Global Assessment (IGA) score (P = 0.03), and Eczema Area and Severity Index (EASI) lichenification score (P = 0.03). There was no significant difference in the overall EASI score. No adverse reactions were reported in either group.

Table 1 Study characteristics of acupressure trials. First author and Clinical No. of patients Mean age (SD) publication yearproblem NA Acupressure Control (usual care) Randomized controlled trial Xue et al., Allergic rhinitis 31 32 2011 Lee et al., Atopic 8 7 Median 28.5 (range 2012 dermatitis 53 Placebo 53 19—79) Dibble Breast cancer 11 Usual care 49.3 (9.4) et al., 2007 31 54 Placebo 11 59.5 (11.5) Maa et al., Bronchiectasis 88 Standard 2007 19 care 13 67.3 (8.2) 12 31 Maa et al., Chronic 20.0* [Acupressure 1997 obstructive 84 group: 20.0 (1.0) C pulmonary group: 19.9 (0.9)] Bazarganipour disease (COPD) 21 21.8* [Acupressure et al., 2010 Dysmenorrhea group: 22.0 (0.9) C 12 group: 21.6 (0.7)] Wong et al., Dysmenorrhea 28.9 (8.5) 2010 Honda Stress (college et al., 2012 students) Quasi-randomized controlled trial Kim & Youn, Sleep 25 23 26.6* [Acupressure 20 20 group: 26.8 Contro 2004 disturbance, group: 26.3] fatigue (clinical 50.0 [Acupressure 52.8 (10.9) Contro nurses) group: 47.3 (6.8)] Shin et al., Stomach 2004 cancer NA: not available * The mean age of total participants was not reported, so the weighted mean age applied

Female, % Follow-up Country Measures Effect of self-acupressure for symptom management: A systematic review 54.0 8 weeks Australia Symptom score, e 33.3 4 weeks US medication use, quality 21 days US of life 100.0 8 weeks Taiwan Severity of itching and 40.0 6 weeks US atopic dermatitis NA Chemotherapy induced Iran nausea and vomiting e 100.0 5 days Hong Kong (CINV) Control 100.0 Japan Decreased symptoms, 37.5 3 months (3 South Korea improved health-related e days in one quality of life Control month) South Korea Relief of dyspnea and 2 weeks other symptoms e NA 3 days Pain score ol 30.0 5 days Pain level, menstrual group: distress ol Perceived stress Sleep disturbance score, fatigue score Chemotherapy induced nausea and vomiting (CINV) number in each group was calculated. 71

72 H.J. Song et al. Table 2 Study protocols for acupressure trials. First author and Main acupoints Placebo/control Treatment Teaching methods publication year Acupressure regimen of self-acupressure Randomized controlled trial Sham ear acupoints: Three times daily, both Participants were Xue et al., 2011 Real ear acupoints: Lun 2 (HX 10), Jian (SF ears, alternating instructed to press 4, 5), Suogu (SF 6), weekly; adhesive stainless steel pellets Shenmen (TF 4), Neibi Zhen (AT 3), and Ya (LO stainless steel until they felt heat and (TG 4), Fei (CO 14), 1) press-pellets slight soreness at the Fengxi (SF 1, 2i), and (Acuneeds, Australia) acupoints Shenshangxian (TG 2p) No acupressure were pressed until participants felt heat Participants were Lee et al., 2012 Quchi (LI 11), located Placebo group: sham and slight soreness at instructed by principal on the left arm lateral point (Houxi, SI the points investigators to to the antecubital fossa 3)/Usual care only 3 min, 3 times per administer acupressure group week, for 4 weeks; with a titanium Dibble et al., 2007 Neiguan (PC 6), located acupressure was acupellet on both forearms Sham acupressure: performed using a near real 1.2 mm titanium Participants were Maa et al., 2007 Zhongfu (LU 1), Chize acupoints/Standard acupellet (Lhasa OMS, instructed to perform (LU 5), Yuji (LU 10), care Weymouth, acupressure to PC 6 Fenglong (ST 40) and Massachusetts, USA) points every morning. Zusanli (ST 36). Each Sham acupressure: Each morning; When participants point is located non-acupoints acupressure was asked, ink marks were bilaterally. applied using the provided to make thumb of the opposite locating the points Maa et al., 1997 Zhongfu LU 1, Yunmen hand easier. LU 2, Dazhui GV 14, A skilled instructor Yuji LU 10, Laogong PC First one side of the directly demonstrated 8, Zusanli ST 36, Hegu body and then the acupressure for the LI 4 other side, for 30 s to participants. After 2 min; acupoints were that, participants were stimulated using the asked to perform fingers to apply gentle acupressure by but firm pressure themselves to make sure of their Real acupoints for 6 acquisition of the weeks, followed by technique. An sham acupressure for 6 instructional booklet weeks; participants explaining how to applied gentle but firm locate and stimulate pressure to the acupoints was also acupoint at provided. approximately a 90◦ A skilled instructor angle to the skin explained the surface, using one or acupressure technique two fingers using the instruction booklet. Acupressure was performed for each participant by the instructor for teaching purposes. When necessary, acupoints were marked on the skin.

Effect of self-acupressure for symptom management: A systematic review 73 Table 2 (Continued) First author and Main acupoints Placebo/control Treatment Teaching methods publication year Acupressure Simple pressure to the regimen of self-acupressure placebo point Bazarganipour Simple pressure to the (between the third and 20 min for 3 to 7 days Participants were et al., 2010 Taichong (LR 3) point fourth toe) before the menstrual instructed to apply the (between the first and No acupressure cycle, in a sitting acupressure protocol second toe) position, applying on the acupoints No intervention pressure using the predetermined by the Wong et al., 2010 Sanyinjiao (SP 6) thumb investigator. acupressure 20 min during the first After an acupressure 3 days of their next 3 intervention by a Honda et al., 2012 Three acupressure menstrual cycles; practitioner, points (Wangu GB 12, pressure was applied participants were Tianrong SI 17, Futu LI using the participant’s taught to practice 18) on the left and thumb acupressure at home. right sides of the neck An Acupressure 5 s, five sessions during Teaching Booklet was the morning, midday, provided. and night Participants were asked to complete five self-acupressure sessions Quasi-randomized controlled trial Kim and Youn, Wangu (GB 12), No intervention 3 days, 4 times per The investigator Usual care day; acupressure was demonstrated 2004 Shenmen (HT 7), and administered using two acupressure using an thumbs Acupressure Teaching Sanyinjiao (SP 6) Booklet. After that, 5 min before the participants acupoints chemotherapy, performed acupressure mealtimes, and by themselves to make Shin et al., 2004 Neiguan (PC 6) anytime sensations of sure they understood acupressure point nausea were felt; the technique. participants applied Acupoints were first acupressure using their marked with tape. finger After it was removed, the participants located acupoints following written instructions. Participants and their family in the ward were taught how to locate the acupressure point by investigators. Acupressure Teaching Booklets were distributed to the ward. Participants were asked to perform the technique on the investigators to assure they could locate the points and apply enough pressure.

74 H.J. Song et al. Figure 2 Risk of bias graph and summary of all included randomized controlled trials using Cochrane’s RoB assessment tool. (a) RoB graph (b) RoB summary. RoB: Risk of Bias; (+): low risk of bias; (−): high risk of bias; (?): unclear risk of bias. Self-acupressure for cancer of vomiting and the intensity of nausea over time, was signif- icantly decreased in the experimental group compared with Dibble et al. conducted a RCT with 160 breast cancer both the placebo group (P = 0.002) and the control group patients who were receiving chemotherapy.24 The experi- (P < 0.0001). However, acute nausea or emesis was not sig- mental group used their thumbs to press the Neiguan (PC nificantly different among the three groups (P = 0.71). 6) acupoint located bilaterally on the forearm, while the placebo group pressed the Houxi (SI 3) acupoint located on Shin et al. conducted a quasi-RCT of 40 stomach can- the ulnar side of the hand. The control group was treated cer patients receiving chemotherapy.25 Nausea and vomiting with only usual care. Patients in all groups received usual in the intervention and control groups was the pri- care consisting of antiemetic therapy prescribed by their mary outcome. All patients received antiemetic therapy physicians. Delayed nausea or emesis, as well as the amount (metoclopramide and ondansetron) as prescribed by their physicians as usual care. Patients in the intervention group

Effect of self-acupressure for symptom management: A systematic review 75 LU 10, Laogong PC 8, Zusanli ST 36, and Hegu LI 4) for 6 weeks, followed by sham acupressure for an additional 6 weeks. The other group performed sham acupressure for the first 6 weeks and real acupressure for the following 6 weeks. Sham acupressure was applied with the same tech- nique to seven points that were not true acupoints. The real acupressure group had less dyspnea as measured by a VAS scale and lower decathexis scores compared with the sham acupressure group (P = 0.009 and P = 0.044, both one- tailed). However, sham acupressure was more effective in reducing peripheral sensory symptoms than real acupressure (P = 0.002, two-tailed). Self-acupressure for dysmenorrhea Figure 3 Risk of bias graph and summary of all included quasi- A study of 172 dysmenorrhea patients by Bazarganipour randomized controlled trials using the RoBA. NS tool. (a) RoBANS et al. compared simple pressure to the Taichong (LR 3) graph (b) RoBANS summary. RoBANS: Risk of Bias Assessment for acupoint (between the first and second toe) with simple Non-randomized Studies; (+): low risk of bias; (−): high risk of pressure to a placebo point.28 The Taichong acupressure bias; (?): unclear risk of bias. group showed a significant decrease in dysmenorrhea sever- ity (P < 0.01), and, by the fourth menstrual cycle, there was also performed acupressure treatment at the Neiguan (PC a significant difference in dysmenorrhea severity (P < 0.02) 6) acupoint for 5 min before chemotherapy, at mealtimes, between the Taichong and placebo groups. and any time they experienced nausea. According to the Rhodes Index of Nausea, Vomiting, and Retching (INVR), out- Wong et al. studied acupressure in 40 dysmenorrhea comes in the intervention group were significantly better patients divided into acupressure intervention and control for severity of nausea and vomiting (P < 0.01), duration of groups.29 The intervention group performed acupressure at nausea (P < 0.01), and frequency of vomiting (P < 0.01) when Sanyinjiao (SP 6) for 20 min upon waking and at bedtime compared with the control group. during the first three days of three successive menstrual cycles. The Pain Visual Analogue Scale (PVAS) and Short- Self-acupressure for respiratory disease Form MacGill Pain Questionnaire (SF-MPQ), administered 20 min after acupressure, showed significant reductions in In 2007, Maa et al. compared acupressure plus standard pain scores in the acupressure group compared with the care, sham acupressure plus standard care, and standard control group (P = 0.008 and P = 0.012). Moreover, scores on care only in 35 patients with bronchiectasis.26 The acupres- the PVAS, SF-MPQ, and Short-Form Menstrual Distress Ques- sure points were Zhongfu (LU 1), Chize (LU 5), Yuji (LU 10), tionnaire (SF-MDQ) administered to the acupressure group Fenglong (ST 40) and Zusanli (ST 36), with each point located were significantly different when compared with the con- bilaterally. Sham acupressure was performed on nearby real trol group at three months after the intervention (P = 0.008, acupoints, and standard care included oral medication and P = 0.012, and P = 0.024, respectively). chest physiotherapy. Outcomes for the Saint George Respira- tory Questionnaire (SGRQ) in the acupressure plus standard Self-acupressure for healthy people care group were significantly better when compared with the standard care only group (P = 0.01). However, there was Honda et al. studied the difference between acupressure no significant difference in outcomes between the sham acu- intervention and no intervention in 24 college students.30 pressure plus standard care and standard care only groups. Acupressure was administered to three points (Wangu GB 12, Tianrong SI 17, Futu LI 18) on the left and right sides of Maa et al. studied the effects of acupressure in 31 the neck for 5 s. The stress score for the Japanese language patients with COPD in a RCT with a crossover design.27 version of the Rhode Island Stress and Coping Inventory (J- One group performed acupressure at seven bilateral acu- RISCI) was significantly decreased in the acupressure group points (Zhongfu LU 1, Yunmen LU 2, Dazhui GV 14, Yuji compared with the control group after 2 weeks (P < 0.05). Kim and Youn studied sleep disturbance and fatigue among 48 clinical nurses divided into acupressure and no- intervention groups.31 The Wangu (GB 12), Shenmen (HT 7), and Sanyinjiao (SP 6) acupoints were pressed for three days at a frequency of four times per day. The sleep disturbance and fatigue scores were significantly lower in the acupres- sure group compared with the no-intervention group after three days (P < 0.01 and P < 0.01). Discussion This systematic review was performed to investigate the efficacy and safety of self-administered acupressure therapy

76 H.J. Song et al. to relieve symptoms of various health problems, including acupressure groups in comparison with the control allergic disease, cancer, respiratory disease, dysmenorrhea, groups.26,27 The results of the present review were similar perceived stress, insomnia, and sleep disturbances. All of to those of Robinson et al., who conducted a systematic the selected 10 studies reported positive effects for pri- review of Shiatsu or acupressure, including self-acupressure mary outcomes of self-acupressure therapy for symptom and acupressure performed by practitioners.16 This review management. That is, there were significant improvements included three studies of COPD and two studies of bronchiec- in outcomes for allergic disease, chemotherapy-induced tasis. They concluded that there was limited evidence for nausea and vomiting in cancer patients, respiratory dis- chronic respiratory conditions. ease, dysmenorrhea, and stress/fatigue scores and sleep disturbances in healthy people. However, some secondary Dysmenorrhea outcomes such as QOL, medication use in allergic disease, and acute nausea or emesis induced by chemotherapy did Pain scores for dysmenorrhea patients performing self- not show a significant difference. There were no studies acupressure were significantly decreased compared with reporting severe adverse reactions to self-acupressure. scores for patients in the control groups.28,29 Lee and Frazier reported three RCTs showing pain scores that significantly The results of our quality assessment showed a relatively decreased in the acupressure group in comparison with the moderate risk of bias in the RCTs, while the risk of bias in the control group.3 Jiang et al. conducted a meta-analysis of quasi-RCTs was at least 50% higher for most of the evaluated acupressure performed by participants or by practitioners domains. and reported that the VAS, SF-MPQ, and SF-MDQ scores in the acupressure group were significantly better than in the The results of the present review were similar to those control group.5 of previous reviews.2,3,5,16,32 All of the previous systematic reviews included studies of acupressure performed by both Healthy people participants and practitioners. One RCT showed that self-administered acupressure in Allergic disease healthy college students significantly lowered stress lev- els, and a quasi-RCT reported that self-acupressure reduced Two RCTs for allergic disease were included in this review. sleep disturbances and fatigue in healthy nurses, both Xue et al. found that the symptom score for rhinitis in a in comparison with a control group.30,31 Lee and Frazier REAP group was significantly improved compared with a found one RCT that reported that acupressure for insomnia SEAP group.22 In the second trial, the atopic dermatitis was significantly better than control for symptom improve- score for patients receiving acupressure plus standard care ment, while another RCT reported no significant difference was significantly improved in comparison with standard between acupressure and control.3 Lee and Frazier also care only.23 The results of a review of acupuncture and identified three RCTs reporting that acupressure reduced acupressure for allergic rhinitis by Zhang et al. were similar fatigue. Yeung et al. conducted a systematic review of acu- to those of the present study.32 Inclusion criteria were not pressure for insomnia and included studies of acupressure limited to self-administered acupressure or treatment by performed by both participants and practitioners.2 Results practitioner. Among the five included studies, three studies from six RCTs showed that acupressure was significantly bet- compared auricular acupressure with antihistamines or ter than sham acupressure for insomnia. In outcomes for Chinese herbal medicine, and the symptom score for the medication use, seven of 10 RCTs showed a significant reduc- auricular acupressure group was significantly improved in tion in the acupressure group compared with the control comparison with the control group. group, while three of 10 RCTs found no significant difference between the acupressure and control groups. Cancer This study is the first systematic review estimating the We reviewed one RCT and one quasi-RCT showing effects of self-acupressure. Systematic reviews reporting that chemotherapy-induced nausea and vomiting in the both self-acupressure and practitioner-administered acu- self-acupressure groups was significantly decreased in com- pressure have been published; however, no systematic parison with the control groups.24,25 Lee and Frazier review for self-acupressure alone has been reported. conducted a systematic review of acupressure, including self-acupressure and acupressure performed by practition- Furthermore, this study identifies a range of diseases ers, and found four RCTs reporting significantly less nausea that can be treated by self-acupressure. The search strat- and vomiting in the acupressure groups compared with the egy included only the intervention (with no limitation for control groups.3 At the initiatory stage, symptom improve- disease) so that the search could be conducted with a ment was not significant in either of the two groups. The highly sensitive search strategy. The present study included results of our study were similar to those of Lee and Frazier. RCTs and quasi-RCTs for various conditions, including cancer, respiratory disease, dysmenorrhea, and symptom manage- Respiratory disease ment in healthy persons. Two RCTs for respiratory disease were included in this In addition, we conducted an extensive search of systematic review. The symptom scores for bronchiectasis databases, including Eastern databases. Many studies of and dyspnea in COPD were significantly improved in the alternative and complementary treatments have been con- ducted in Eastern countries. Therefore, we expanded our search strategy to include Korean and Japanese databases

Effect of self-acupressure for symptom management: A systematic review 77 as well as core and Chinese databases. Previous systematic 6. Higgins JPT, Green S. Cochrane handbook for systematic reviews reviews only searched core and Chinese databases.2,3,5 of interventions. Version 5.1.0. The Cochrane Collaboration. 2011. http://handbook.cochrane.org/. [accessed 21.11.13]. Interpretation of the results of this study should take into account its limitations. First, we could not perform 7. Kim SY, Park JE, Lee YJ, Seo HJ, Sheen SS, Hahn S, et al. Testing a meta-analysis of the selected studies because of diver- a tool for assessing the risk of bias for nonrandomized stud- sity in participants, intervention methods, and medical ies showed moderate reliability and promising validity. J Clin results. Even when the same disease was studied in different Epidemiol 2013;66:408—14. trials, the outcome measures were not standardized. Conse- quently, we opted to present a narrative description of the 8. Chao HL, Miao SJ, Liu PF, Lee HH, Chen YM, Yao CT, et al. The efficacy and safety of self-acupressure. Second, we found beneficial effect of ST-36 (Zusanli) acupressure on postopera- it difficult to judge the evidence for self-acupressure. The tive gastrointestinal function in patients with colorectal cancer. results of the quality assessment of risk of bias in the tri- Oncol Nurs Forum 2013;40. E61-68. als were not good, and efficacy was not clearly estimated. Moreover, the teaching methods for the self-administered 9. Chen HM, Chen CH. Effects of acupressure at the Sanyinjiao intervention were not standardized, suggesting a high point on primary dysmenorrhoea. J Adv Nurs 2004;48:380—7. probability of inconsistency in the administration of self- acupressure. 10. Fritz DJ, Carney RM, Steinmeyer B, Ditson G, Hill N, Zee-Cheng J. The efficacy of auriculotherapy for smoking cessation: a Conclusions randomized, placebo-controlled trial. J Am Board Fam Med 2013;26:61—70. Even though most of the trials of self-acupressure therapy showed some effect, it was difficult to draw conclu- 11. Lee JS, Lee MS, Min K, Lew JH, Lee BJ. Acupressure for treat- sions about specific intervention methods or their effects ing neurological disorders: a systematic review. Int J Neurosci because the interventions and effect indicators in each study 2011;121:409—14. were different. Furthermore, the evidence level for self- acupressure was difficult to evaluate due to the moderate 12. Lu MJ, Lin ST, Chen KM, Tsang HY, Su SF. Acupressure quality of the studies and the disparities in teaching meth- improves sleep quality of psychogeriatric inpatients. Nurs Res ods for the self-administered intervention. High-quality RCTs 2013;62:130—7. with larger samples are necessary to establish the evidence for self-acupressure as an effective intervention. 13. Maa SH, Sun MF, Hsu KH, Hung TJ, Chen HC, Yu CT, et al. Effect of acupuncture or acupressure on quality of life of patients with Conflict of interest chronic obstructive asthma: a pilot study. J Altern Complement Med 2003;9:659—70. The authors have no conflicts of interest that are directly relevant to the content of this article. 14. Noroozinia H, Mahoori A, Hasani E, Gerami-Fahim M, Sepehrvand N. The effect of acupressure on nausea and vomi- Acknowledgments ting after cesarean section under spinal anesthesia. Acta Med Iran 2013;51:163—7. This study was performed as part of the Infrastructure Devel- opment Project for Traditional Knowledge-based Remedy 15. Rerksuppaphol L. Efficacy of auricular acupressure combined (K12210) funded by the Korea Institute of Oriental Medicine with transcutaneous electrical acupoint stimulation for weight (KIOM) in South Korea. The results of this project were sub- reduction in obese women. J Med Assoc Thai 2012;95(Suppl. jected to the KIOM appraisal process. No other sources of 12):S32—9. funding were used to assist in the preparation of this article. 16. Robinson N, Lorenc A, Liao X. The evidence for Shiatsu: a sys- References tematic review of Shiatsu and acupressure. BMC Complement Altern Med 2011;11:88. 1. Beal WM. Acupuncture and acupressure application to women’s reproductive health care. J Nursemidwifery 1999;44:217—30. 17. Smith CA, Collins CT, Crowther CA, Levett KM. Acupuncture or acupressure for pain management in labour. Cochrane Database 2. Yeung WF, Chung KF, Poon MM, Ho FY, Zhang SP, et al. Acu- Syst Rev 2011;7:CD009232. pressure, reflexology, and auricular acupressure for insomnia: a systematic review of randomized controlled trials. Sleep Med 18. Wu HS, Wu SC, Lin JG, Lin LC. Effectiveness of acupressure in 2012;13:971—84. improving dyspnoea in chronic obstructive pulmonary disease. J Adv Nurs 2004;45:252—9. 3. Lee EJ, Frazier SK. The efficacy of acupressure for symptom management: a systematic review. J Pain Symptom Manage 19. Yang MH, Wu SC, Lin JG, Lin LC. The efficacy of acupressure for 2011;42:589—603. decreasing agitated behaviour in dementia: a pilot study. J Clin Nurs 2007;16:308—15. 4. National Cancer Institute. Acupuncture. http://www.cancer. gov/cancertopics/pdq/cam/acupuncture/HealthProfessional. ; 20. Yeh CH, Chien LC, Balaban D, Sponberg R, Primavera J, Morone 2007 [accessed 21.11.13]. NE, et al. A randomized clinical trial of auricular point acupres- sure for chronic low back pain: a feasibility study. Evid Based 5. Jiang HR, Ni S, Li JL, Liu MM, Li J, Cui XJ, et al. Systematic Complement Alternat Med 2013:196978. review of randomized clinical trials of acupressure therapy for primary dysmenorrhea. Evid Based Complement Alternat Med 21. Yip YB, Tse SH. The effectiveness of relaxation acupoint stimu- 2013;2013:169692 [Epub 2013 Aug 18]. lation and acupressure with aromatic lavender essential oil for non-specific low back pain in Hong Kong: a randomised con- trolled trial. Complement Ther Med 2004;12:28—37. 22. Xue CC, Zhang CS, Yang AW, Zhang AL, Li Y, Xia J, et al. Semi-self-administered ear acupressure for persistent aller- gic rhinitis: a randomised sham-controlled trial. Ann Allergy Asthma Immunol 2011;106:168—70. 23. Lee KC, Keyes A, Hensley JR, Gordon JR, Kwasny MJ, West DP, et al. Effectiveness of acupressure on pruritus and lichenifica- tion associated with atopic dermatitis: a pilot trial. Acupunct Med 2012;30:8—11. 24. Dibble SL, Luce J, Cooper BA, Israel J, Cohen M, Nussey B, et al. Acupressure for chemotherapy-induced nausea and vomiting: a randomized clinical trial. Oncol Nurs Forum 2007;34:813—20. 25. Shin YH, Kim TI, Shin MS, Juon HS. Effect of acupres- sure on nausea and vomiting during chemotherapy cycle for

78 H.J. Song et al. Korean postoperative stomach cancer patients. Cancer Nurs 29. Wong CL, Lai KY, Tse HM. Effects of SP6 acupressure on pain 2004;27:267—74. and menstrual distress in young women with dysmenorrhea. 26. Maa SH, Tsou TS, Wang KY, Wang CH, Lin HC, Huang YH. Self- Complement Ther Clin Pract 2010;16:64—9. administered acupressure reduces the symptoms that limit daily activities in bronchiectasis patients: pilot study findings. J Clin 30. Honda Y, Tsuda A, Horiuchi S. Effect of a four-week self- Nurs 2007;16:794—804. administered acupressure intervention on perceived stress over 27. Maa SH, Gauthier D, Turner M. Acupressure as an adjunct the past month. Open J Med Psychol 2012;1:20—4. to a pulmonary rehabilitation program. J Cardiopulm Rehabil 1997;17:268—76. 31. Kim WI, Youn HM. The effect of self-acupressure of Qan- 28. Bazarganipour F, Lamyian M, Heshmat R, Abadi MA, Taghavi A. gol (G12), Shinmun (H7), Samumgyo (Sp6) on clinical nurses’ A randomized clinical trial of the efficacy of applying a simple sleep pattern disturbance and fatigue. Kor J Meridian Acupoint acupressure protocol to the Taichong point in relieving dysmen- 2004;21:81—93. orrhea. Int J Gynaecol Obstet 2010;111:105—9. 32. Zhang CS, Yang AW, Zhang AL, Fu WB, Thien FU, Lewith G, et al. Ear-acupressure for allergic rhinitis: a systematic review. Clin Otolaryngol 2010;35:6—12.