《气味科学中的健康密码》课程教学资源（阅读材料）The Effects of Aromatherapy on Sleep Improvement：A Systematic Literature Review and Meta-Analysis

THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE Volume 21,Number 2,2015,pp.61-68 Review Articles Mary Ann Liebert,Inc. D0:10.1089/acm.2014.0113 The Effects of Aromatherapy on Sleep Improvement: A Systematic Literature Review and Meta-Analysis Eunhee Hwang,PhD:and Sujin Shin,PhD2 Abstract Objectives:To evaluate the existing data on aromatherapy interventions for improvement of sleep quality Design:Systematic literature review and meta-analysis on the effects of aromatherapy. Study Sources:Electronic databases,including the Korea Education and Research Information Service (KERIS),Korean studies Information Service System(KISS),National Assembly Library,and eight academies within the Korean Society of Nursing Science,were searched to identify studies published between 2000 and August 2013. Study Selection:Randomized controlled and quasi-experimental trials that included aromatherapy for the improvement of sleep quality.Of the 245 publications identified,13 studies met the inclusion and exclusion criteria,and 12 studies were used in the meta-analysis. Results:Meta-analysis of the 12 studies using a random-effects model revealed that the use of aromatherapy was effective in improving sleep quality (95%confidence interval [CI],0.540-1.745;Z=3.716).Subgroup analysis revealed that inhalation aromatherapy (95%CI,0.792-1.541;Z=6.107)was more effective than massage therapy (95%CI,0.128-2.166;Z=2.205)in unhealthy (95%CI,0.248-1.100;Z=3.100)and healthy (95%CI,0.393-5.104;Z=2.287)participants,respectively. Conclusion:Readily available aromatherapy treatments appear to be effective and promote sleep.Thus,it is essential to develop specific guidelines for the efficient use of aromatherapy. Introduction sion during the daytime and habitual drug abuse*have in- creased interest in nonpharmacologic approaches to sleep LEEP AIDS IN RECOVERY FROM physical activities as well disorders.Consequently,several interventional studies have as the maintenance of strength and health,and thus ob- been conducted to verify the effectiveness of behavioral and taining an adequate amount of sleep is essential.Sleep ac- circadian rhythm therapies. counts for one third of an average day,and while sufficient Recently,aromatherapy has also become a common sleep enhances health and well-being,a lack of sleep can nursing intervention because it is economical,has fewer trigger fatigue,drowsiness,nervousness,dizziness,instabil- adverse effects than drug therapies,and improves the sense ity,disorientation,and attention disorders.In fact.a con- of well-being.>Aromatherapy entails the use of oils ex- tinuous lack of sleep may even cause death.In 2006,the tracted from flowers,stems,leaves,roots,and fruits of Korean Sleep Research Society surveyed 5000 adults and various plants absorbed into the body through the skin or discovered that one in four individuals experienced a sleep respiratory system to improve mental and bodily health. disorder caused by stress,aging,substance abuse,or changes The effects of aromatherapy are achieved through inhala- in sleep cycle.Given the large number of people with sleep tion,bathing,or massage.Aromatherapy use is simple, disorders,as well as the incessant changes and increasing does not require special instruments,and results in positive complexity of society,attention to sleep disorders and the effects even when used for only a short time.The scent availability of effective intervention therapies are necessary. particles reach the limbic system through the olfactory Drug therapies,including sleeping pills and antidepres- nerves and can produce sedative and relaxant effects, sants,are widely used as clinical interventions for sleep which in turn influence blood pressure,heart rate,repro- disorders.However,problems such as lethargy and confu- duction,memory,and stress response. 'Wonkwang University,Iksan,Republic of Korea 2Soonchunhyang University,Cheonan,Republic of Korea. 61

Review Articles The Effects of Aromatherapy on Sleep Improvement: A Systematic Literature Review and Meta-Analysis Eunhee Hwang, PhD,1 and Sujin Shin, PhD2 Abstract Objectives: To evaluate the existing data on aromatherapy interventions for improvement of sleep quality. Design: Systematic literature review and meta-analysis on the effects of aromatherapy. Study Sources: Electronic databases, including the Korea Education and Research Information Service (KERIS), Korean studies Information Service System (KISS), National Assembly Library, and eight academies within the Korean Society of Nursing Science, were searched to identify studies published between 2000 and August 2013. Study Selection: Randomized controlled and quasi-experimental trials that included aromatherapy for the improvement of sleep quality. Of the 245 publications identified, 13 studies met the inclusion and exclusion criteria, and 12 studies were used in the meta-analysis. Results: Meta-analysis of the 12 studies using a random-effects model revealed that the use of aromatherapy was effective in improving sleep quality (95% confidence interval [CI], 0.540–1.745; Z = 3.716). Subgroup analysis revealed that inhalation aromatherapy (95% CI, 0.792–1.541; Z = 6.107) was more effective than massage therapy (95% CI, 0.128–2.166; Z = 2.205) in unhealthy (95% CI, 0.248–1.100; Z = 3.100) and healthy (95% CI, 0.393–5.104; Z = 2.287) participants, respectively. Conclusion: Readily available aromatherapy treatments appear to be effective and promote sleep. Thus, it is essential to develop specific guidelines for the efficient use of aromatherapy. Introduction Sleep aids in recovery from physical activities as well as the maintenance of strength and health, and thus ob￾taining an adequate amount of sleep is essential.1 Sleep ac￾counts for one third of an average day, and while sufficient sleep enhances health and well-being, a lack of sleep can trigger fatigue, drowsiness, nervousness, dizziness, instabil￾ity, disorientation, and attention disorders.2 In fact, a con￾tinuous lack of sleep may even cause death. In 2006, the Korean Sleep Research Society surveyed 5000 adults and discovered that one in four individuals experienced a sleep disorder caused by stress, aging, substance abuse, or changes in sleep cycle.3 Given the large number of people with sleep disorders, as well as the incessant changes and increasing complexity of society, attention to sleep disorders and the availability of effective intervention therapies are necessary. Drug therapies, including sleeping pills and antidepres￾sants, are widely used as clinical interventions for sleep disorders. However, problems such as lethargy and confu￾sion during the daytime and habitual drug abuse4 have in￾creased interest in nonpharmacologic approaches to sleep disorders. Consequently, several interventional studies have been conducted to verify the effectiveness of behavioral and circadian rhythm therapies. Recently, aromatherapy has also become a common nursing intervention because it is economical, has fewer adverse effects than drug therapies, and improves the sense of well-being.5 Aromatherapy entails the use of oils ex￾tracted from flowers, stems, leaves, roots, and fruits of various plants absorbed into the body through the skin or respiratory system to improve mental and bodily health. The effects of aromatherapy are achieved through inhala￾tion, bathing, or massage.6 Aromatherapy use is simple, does not require special instruments, and results in positive effects even when used for only a short time. The scent particles reach the limbic system through the olfactory nerves and can produce sedative and relaxant effects, which in turn influence blood pressure, heart rate, repro￾duction, memory, and stress response.7 1 Wonkwang University, Iksan, Republic of Korea. 2 Soonchunhyang University, Cheonan, Republic of Korea. THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE Volume 21, Number 2, 2015, pp. 61–68 ª Mary Ann Liebert, Inc. DOI: 10.1089/acm.2014.0113 61

62 HWANG AND SHIN Various studies have evaluated the effects of aromather- matherapy,scent therapy;sleep disorder,insomnia,quality apy,and results have indicated that aromatherapy de- of sleep,and degree of satisfaction with sleep. creased many psychological factors,including depression, anxiety,and stress;had positive effects on physiologic Literature selection and exclusion criteria factors,such as fatigue,blood pressure,pulse,and sleep:9.10 and relieved pain.The prior studies analyzed diverse uses Reports of studies published in Korea after 2000 that of aromatherapy,as well as the use of aromatherapy in included aromatherapy used by adults as a main intervention various populations,including night shift nurses,elderly in- and reported at least one of the intervention results(quality dividuals,patients with chronic disease,inpatients,middle- of sleep,degree of satisfaction with sleep,sleep efficiency, aged women,and college students..10.12.13 Prior studies on or sleep improvement)were selected.College research pa- the use of aromatherapy for sleep improvement used such pers,theses,exhibitions,presentations,review articles, methods as inhalation,massage,and skin application and overlapping publications,and research involving animal experiments were excluded. mixtures of such oils as lavender,geranium,mandarin, bergamot,and marjoram.The results of these studies like wise indicated that the effects of aromatherapy on sleep Literature search results improvement were significant.14.15 Two researchers,using three different processes,selected Although investigations into the effects of aromatherapy the publications included in the current study (Fig.1).The on improving sleep quality have continued,the application words chosen for the literature search were mixed and then methods,areas to which aromatherapy oils are applied,the used to query the RISS,KISS,National Assembly Library, duration of application,the types of aromatherapy oils used, and the academies under the Korean Society of Nursing and the mixture ratios of the oils have differed across Science databases.A total of 202 publications,excluding studies.Therefore,a systematic analysis is necessary to overlapping studies,were selected from the 245 studies in- verify the reported effects and to present scientific stan- cluded in the databases.A total of 176 studies were excluded dards for the conduct of aromatherapy studies.Given the on the basis of title and abstract,and an additional 13 studies current emphasis on proof-based practice,it is critical to were excluded after an evaluation of the full text.Thirteen provide conclusive evidence of the effectiveness of aro- studies were selected for final inclusion in the current study. matherapy application methods as a nursing intervention for sleep improvement. Evaluation of study quality The objectives of the current study were to conduct a systematic literature review and meta-analysis to evaluate The quality of the studies was evaluated using the the reported effects of aromatherapy on sleep improvement quantitative research evaluation framework presented by the and to present evidence and scientific standards for the ap- McMaster University Evidence-Based Practice Research plication of aromatherapy. Group.The framework comprised questions that evaluated eight areas:the purpose of the study,background literature, Methods study design,sampling,results,interventions,conclusions, and meanings.Each question was judged by the responses Study design "yes,”“no,”or“not applicable.”Judgments regarding literature that did not satisfy the quality standards were This systematic literature review and meta-analysis ana- made following a discussion between two researchers. lyzed the reported effects of aromatherapy used as a treat- ment for sleep improvement. Data analysis Literature search strategies Systematic literature review.Results from the 13 papers The key parameters of the literature searches conducted selected following the systematic literature review were an- were as follows.The participants were adults aged 20 years alyzed,encoded,and arranged into forms.The forms included or older.The intervention consisted of all types of ar- sections on research design,number of participants,standards omatherapies (inhalation,massage,and skin application). for participant selection and exclusion.measurement vari- The comparisons included no intervention or other inter ables and tools,result variables,and statistical values.To vention (massage without application of aroma oil).Out- maintain accuracy during the process,two researchers inde- comes were measures of sleep improvement(sleeping time, pendently conducted evaluations,and the results were cross- sleep efficiency,the quality of sleep,and the degree of checked and compared for inter-rater agreement. satisfaction with sleep). Meta-analysis.Statistical analysis of the effect size was Databases used for literature search conducted by using Comprehensive Meta-Analysis soft- and words searched ware,version 2.2.064 (Biostat,Englewood,NJ).Out of the 13 studies,one lacked appropriate statistical analysis and The databases searched included the Korea Education and was excluded.Thus,12 studies were included in the effect Research Information Service (RISS),the Korean studies size analysis.Studies used for the meta-analysis had dif- Information Service System (KISS),and the National As- ferent scales and therefore integration of the measured sembly Library.Academic journals from eight academies values into one unit was necessary.Accordingly,a stan- within the Korean Society of Nursing Science were also dardized mean difference was calculated and used for searched.The words searched included the following:aro- transformation into effect size (Z)

Various studies have evaluated the effects of aromather￾apy, and results have indicated that aromatherapy de￾creased many psychological factors, including depression, anxiety, and stress;8 had positive effects on physiologic factors, such as fatigue, blood pressure, pulse, and sleep;9,10 and relieved pain.11 The prior studies analyzed diverse uses of aromatherapy, as well as the use of aromatherapy in various populations, including night shift nurses, elderly in￾dividuals, patients with chronic disease, inpatients, middle￾aged women, and college students.9,10,12,13 Prior studies on the use of aromatherapy for sleep improvement used such methods as inhalation, massage, and skin application and mixtures of such oils as lavender, geranium, mandarin, bergamot, and marjoram. The results of these studies like￾wise indicated that the effects of aromatherapy on sleep improvement were significant.14,15 Although investigations into the effects of aromatherapy on improving sleep quality have continued, the application methods, areas to which aromatherapy oils are applied, the duration of application, the types of aromatherapy oils used, and the mixture ratios of the oils have differed across studies. Therefore, a systematic analysis is necessary to verify the reported effects and to present scientific stan￾dards for the conduct of aromatherapy studies. Given the current emphasis on proof-based practice, it is critical to provide conclusive evidence of the effectiveness of aro￾matherapy application methods as a nursing intervention for sleep improvement. The objectives of the current study were to conduct a systematic literature review and meta-analysis to evaluate the reported effects of aromatherapy on sleep improvement and to present evidence and scientific standards for the ap￾plication of aromatherapy. Methods Study design This systematic literature review and meta-analysis ana￾lyzed the reported effects of aromatherapy used as a treat￾ment for sleep improvement. Literature search strategies The key parameters of the literature searches conducted were as follows. The participants were adults aged 20 years or older. The intervention consisted of all types of ar￾omatherapies (inhalation, massage, and skin application). The comparisons included no intervention or other inter￾vention (massage without application of aroma oil). Out￾comes were measures of sleep improvement (sleeping time, sleep efficiency, the quality of sleep, and the degree of satisfaction with sleep). Databases used for literature search and words searched The databases searched included the Korea Education and Research Information Service (RISS), the Korean studies Information Service System (KISS), and the National As￾sembly Library. Academic journals from eight academies within the Korean Society of Nursing Science were also searched. The words searched included the following: aro￾matherapy, scent therapy; sleep disorder, insomnia, quality of sleep, and degree of satisfaction with sleep. Literature selection and exclusion criteria Reports of studies published in Korea after 2000 that included aromatherapy used by adults as a main intervention and reported at least one of the intervention results (quality of sleep, degree of satisfaction with sleep, sleep efficiency, or sleep improvement) were selected. College research pa￾pers, theses, exhibitions, presentations, review articles, overlapping publications, and research involving animal experiments were excluded. Literature search results Two researchers, using three different processes, selected the publications included in the current study (Fig. 1). The words chosen for the literature search were mixed and then used to query the RISS, KISS, National Assembly Library, and the academies under the Korean Society of Nursing Science databases. A total of 202 publications, excluding overlapping studies, were selected from the 245 studies in￾cluded in the databases. A total of 176 studies were excluded on the basis of title and abstract, and an additional 13 studies were excluded after an evaluation of the full text. Thirteen studies were selected for final inclusion in the current study. Evaluation of study quality The quality of the studies was evaluated using the quantitative research evaluation framework presented by the McMaster University Evidence-Based Practice Research Group.16 The framework comprised questions that evaluated eight areas: the purpose of the study, background literature, study design, sampling, results, interventions, conclusions, and meanings. Each question was judged by the responses ‘‘yes,’’ ‘‘no,’’ or ‘‘not applicable.’’ Judgments regarding literature that did not satisfy the quality standards were made following a discussion between two researchers. Data analysis Systematic literature review. Results from the 13 papers selected following the systematic literature review were an￾alyzed, encoded, and arranged into forms. The forms included sections on research design, number of participants, standards for participant selection and exclusion, measurement vari￾ables and tools, result variables, and statistical values. To maintain accuracy during the process, two researchers inde￾pendently conducted evaluations, and the results were cross￾checked and compared for inter-rater agreement. Meta-analysis. Statistical analysis of the effect size was conducted by using Comprehensive Meta-Analysis soft￾ware, version 2.2.064 (Biostat, Englewood, NJ). Out of the 13 studies, one lacked appropriate statistical analysis and was excluded. Thus, 12 studies were included in the effect size analysis. Studies used for the meta-analysis had dif￾ferent scales and therefore integration of the measured values into one unit was necessary. Accordingly, a stan￾dardized mean difference was calculated and used for transformation into effect size (Z). 62 HWANG AND SHIN

EFFECTS OF AROMATHERAPY ON SLEEP IMPROVEMENT 63 The Higgins 12 test for homogeneity was conducted to (23.1%).With regards to study design,4 studies had a summarize the statistical results of each study.When the randomized control group (30.9%),I study was a cohort calculated 12 value was lower than 25%,statistical hetero- study (7.7%),and 8 studies were pre/post experimental geneity was considered low,and when the value ex- group studies that included a control group (61.5%). ceeded 50%.statistical heterogeneity was assumed. According to the analysis results,the hypothesis of homo- Aromatherapy variables geneity was rejected,and the effect size calculated using a random effect model and 95%confidence interval [CI]were Results of the analysis of variables in the papers related presented.In addition,subgroup analysis by each cate- to aromatherapy indicated that lavender and bergamot aroma oils were used most often.Two studies used lav- gorical variable of major research characteristics was con- ducted to explore mediator variables.A funnel plot and ender oil (15.4%),and two each used a mixture of lav- Egger linear regression asymmetry test were used to test ender and bergamot;lavender and peppermint;or publication bias and sensitivity of the research papers used lavender,bergamot,and chamomile (15.4%).The re- in the current study.1 maining studies involved clary sage,rosewood,lemon, marjoram,ylang,eucalyptus,and rosemary.Among the Results aroma application methods,seven studies involved mas- sage (53.7%).three studies entailed direct inhalation Research trends on the effects of aromatherapy on sleep methods after spraying the aroma on a gauze or a hand- Research on the effects of aromatherapies included in this kerchief(30.8%),three studies involved indirect inhala- study are shown in Table 1.According to the analysis re- tion methods after placing the aroma in a necklace or a sults,9 of the 13 studies used validated tools to measure pillow (23.1%),and one study applied the aroma to a joint effects.The number of individuals in the studies was 36-72 area in a poultice (7.7%). Trends in research conducted on the effects of aroma- therapy on sleep over the past 10 years were examined. Measurement variables and results on the effects Among the 13 studies identified,most were conducted of aromatherapy on sleep within the past 5 years;5 studies(38.5%)were conducted in Effect variables used in the studies were the quality of 2011,2 in 2009,and 2 in 2012.As for study participants,3 sleep,degree of sleep disorder,degree of satisfaction with studies included adults and elderly individuals (23.1%),7 sleep,and sleep patterns.The four studies measured the studies included patients with chronic or other specific dis- quality of sleep as an effect variable,of which three reported eases(53.8%),and 3 studies included hospitalized patients that aromatherapy was effective.Eight studies measured the potentially relevant 245 records .the Korea Education and Research identified through database Information Service (RISS);58 searching 。the Korean studies Information Service System (KISS);25 Identification ↓ the National Assembly Library;153 202 records after duplicates ·academic joumals from eight academies removed within the Korean Society of Nursing Science;9 176 records excluded Screening 202 records screened by title and abstract 26 full text articles assessed for Eligibility eligibility 13 full text articles excluded as follows: 13 articles included in qualitative ·not aroma study n=l analysis one group design n=8 ↓ no results presented n=2 Included others n=2 12 articles included in quantitative synthesis (meta-analysis) FIG.1.Flowchart of study selection

The Higgins I 2 test for homogeneity was conducted to summarize the statistical results of each study. When the calculated I 2 value was lower than 25%, statistical hetero￾geneity was considered low, and when the I 2 value ex￾ceeded 50%, statistical heterogeneity was assumed.17 According to the analysis results, the hypothesis of homo￾geneity was rejected, and the effect size calculated using a random effect model and 95% confidence interval [CI] were presented.18 In addition, subgroup analysis by each cate￾gorical variable of major research characteristics was con￾ducted to explore mediator variables. A funnel plot and Egger linear regression asymmetry test were used to test publication bias and sensitivity of the research papers used in the current study.19 Results Research trends on the effects of aromatherapy on sleep Research on the effects of aromatherapies included in this study are shown in Table 1. According to the analysis re￾sults, 9 of the 13 studies used validated tools to measure effects. The number of individuals in the studies was 36–72. Trends in research conducted on the effects of aroma￾therapy on sleep over the past 10 years were examined. Among the 13 studies identified, most were conducted within the past 5 years; 5 studies (38.5%) were conducted in 2011, 2 in 2009, and 2 in 2012. As for study participants, 3 studies included adults and elderly individuals (23.1%), 7 studies included patients with chronic or other specific dis￾eases (53.8%), and 3 studies included hospitalized patients (23.1%). With regards to study design, 4 studies had a randomized control group (30.9%), 1 study was a cohort study (7.7%), and 8 studies were pre/post experimental group studies that included a control group (61.5%). Aromatherapy variables Results of the analysis of variables in the papers related to aromatherapy indicated that lavender and bergamot aroma oils were used most often. Two studies used lav￾ender oil (15.4%), and two each used a mixture of lav￾ender and bergamot; lavender and peppermint; or lavender, bergamot, and chamomile (15.4%). The re￾maining studies involved clary sage, rosewood, lemon, marjoram, ylang, eucalyptus, and rosemary. Among the aroma application methods, seven studies involved mas￾sage (53.7%), three studies entailed direct inhalation methods after spraying the aroma on a gauze or a hand￾kerchief (30.8%), three studies involved indirect inhala￾tion methods after placing the aroma in a necklace or a pillow (23.1%), and one study applied the aroma to a joint area in a poultice (7.7%). Measurement variables and results on the effects of aromatherapy on sleep Effect variables used in the studies were the quality of sleep, degree of sleep disorder, degree of satisfaction with sleep, and sleep patterns. The four studies measured the quality of sleep as an effect variable, of which three reported that aromatherapy was effective. Eight studies measured the FIG. 1. Flowchart of study selection. EFFECTS OF AROMATHERAPY ON SLEEP IMPROVEMENT 63

uouDpiDa 蓝 2 蓝 蓝 兰 2 蓝 2 蓝 sajqDima awooino (+daals (+入(( (+)daals jo uianed (+).入tgtn ( (+)aoueqImis!p daaIS (-daais jo fieno 童 品 8 识 8 母 6 6 8 6 品 u8isad iaije pue aiojag 三 三 三 Jaije pue auojag uoneleyur 15and uone equr uonereyur 15anpu pyiuaw+%I pyofusnSup vinpuDD1 I SiSuaAID sisuauis snin 'pyofisnsun 'T jo puanq y L.cl00n1话1 I:I jo onel e le D.iopoaDso. IM %0't painl!p sem yorM lo pyofusnSuD T %001 jo doup v to pyofusnSun (ADax)Apnis 11600200x Ee(IIOZ) aey pue uoM 110) uoaf pue Bunf 24(4006) 64

Table 1. Summary of Included Studies Study (year) Aroma Application Design Sample Size (n) Outcome variables (effect) Instrument validation? Kim (2009)8 Lavandula angustifolia 1% + Mentha arvensis 1% Massage RCT 72 Sleep disturbance ( + ) Yes Kim et al. (2007)9 A blend of L. angustifolia, Citrus sinensis at a ratio of 2:1 Direct inhalation Before and after 50 Pattern of sleep ( + ), satisfaction with sleep ( + ) Yes No Park et al. (2010)31 A blend of L. angustifolia, C. bergamia, Ocimum basilicum at a ratio of 2:2:1 Direct/indirect inhalation Before and after 44 Pattern of sleep ( + ) Yes Park et al. (2012)6 A blend of L. angustifolia, Aniba rosaeodora at a ratio of 1:1 Indirect inhalation RCT 60 Quality of sleep ( + ) Yes Seo and Chang (2009)32 A blend of L. angustifolia, C. bergamia, Anthemis nobilis at a ratio of 1:1:1, which was diluted 2.0% with Simmondsia chinensis carrier oil 20 mL Massage Before and after 56 Pattern of sleep( + ) Yes Yang et al. (2011)33 A blend of L. angustifolia, M. arvensis at a ratio of 1:1, which was diluted with S. chinensis carrier oil 20 mL Massage Before and after 62 Quality of sleep ( + ) Yes Won and Chae (2011)11 A blend of M. arvensis, Eucalyptus radiata, A. rosaeodora at a ratio of 5:2:3 Massage RCT 42 Sleep disturbance ( - ) Yes Lee (2008)14 A blend of L. angustifolia, C. bergamia, 20 drops each, diluted with S. chinensis carrier oil 100 mL Massage Cohort 67 Sleep disturbance ( + ) (sleep efficiency) No Lee and Hwang (2011)34 A drop of 100% L. angustifolia oil Direct inhalation Before and after 67 Quality of sleep ( + ) Yes Lee et al. (2011)35 A blend of C. bergamia, L. angustifolia, Anthemis nobilis at a ratio of 2:1:1, diluted with S. chinensis carrier oil 20 mL Massage Before and after 66 Sleep ( + ) Yes Jung and Jeon (2004)36 A blend of L. angustifolia, C. bergamia, Anthemis nobilis, Salvia sclarea at a ratio of 3:2:1 Massage Before and after 37 Sleep disturbance ( + ) No Choi and Lee (2012)10 A blend of L. angustifolia, Origanum majorana, Cananga odorata at a ratio of 4:3:3 Direct inhalation RCT 36 Sleep disturbance ( + ) Yes Hwang et al. (2011)37 L. angustifolia oil Warm, wet pack Before and after 45 Quality of sleep ( - ) Yes RCT, randomized controlled trial. 64

EFFECTS OF AROMATHERAPY ON SLEEP IMPROVEMENT 65 study name Statistics for each study Std diff in means and 95%Cl Std diff Standard Lower Upper in means error Variance limit limit Z-Value p-Value Km(2009) 6.155 0.565 0.3195.049 726110.903 0.000 Kim,Kim Park(2007) 0.684 0.291 0.0850.114 1.255 2.352 0.019 Park et al(2010) 1.269 0.330 0.1090.621 1.916 3.839 0.000 Park et al(2012) 1.344 0.286 0.0820783 1.904 4.701 0.000 Seo Chang(2009) 0.927 0.281 0.0790.376 1.479 3.294 0.001 Yang,Kang,Kim(2011) 0.549 0.259 0.0670.042 1.056 2.121 0.034 Won Chae(2011) -0.563 0.315 0.099-1.180 0.053 -1.790 0.073 Lee(2008) 1.185 0.317 0.1000.564 1.806 3.739 0.000 Lee Hwang(2011) 0.874 0.256 0.065 0.3731.376 3.419 0.001 Lee,Lee,Kim(2011) 0.324 0.248 0.061-0.1620.810 1.308 0.191 Jung Jeon(2004) 0.000 0.329 0.108-0.6450.645 0.000 1.000 Choi Lee(2012) 1.847 0.360 0.1301.141 2.552 5.128 0.000 1.142 0.307 0.0940.540 1.745 3.716 0.000 -1.00 -0.50 0.00 0.50 1.00 Favors control Favors treatment FIG.2.Effect size of aromatherapy related to sleep.Subgroup inhalation:heterogeneity:tau2=0.09,O=8.04,df=4 (p=0.090),1=50.23%.CI,confidence interval;df,degree of freedom. degree of sleep disorder,of which seven reported that aro- (95%CI,0.54-1.75:p<0.001).which indicated that aro- matherapy was effective.One study measured the degree of matherapy significantly improved sleep(Fig.2). satisfaction with sleep and one study included sleep efficiency, A meta-analysis was conducted based on the intervention both of which reported that aromatherapy was effective. factors of aromatherapy application method and the partic- ipants'health condition.Inhalation therapy (95%CI,0.79- Effect size of aromatherapy 1.54;Z=6.11:p<0.001)was more effective than massage therapy(95%CL,0.13-2.17:Z=2.21;p=0.027),and the Twelve of the selected papers were used to analyze effect effect size for participants with a disease (95%CI 0.25- size of aromatherapy on sleep.Heterogeneity of the 12 pa- 1.10:Z=3.10;p=0.002)was larger than that for healthy pers was verified (P2=91.84%;p<0.001),and a random- participants(95%CL,0.39-5.10;Z=2.29;p=0.022) effects model was used.The resulting effect size was 3.72 (Figs.3 and 4). study name Statistics for each study Std diff in means and 95%Cl Std diff Standard Lower Upper in means error Variance limit limit Z-Value p-Value Kim(2009) 6.155 0.565 0.3195049 726110.903 0.000 Kim,Kim Park(2007) 0.684 0.291 0.0850.1141.255 2.352 0.019 Park et al(2010) 1.269 0.330 0.1090.621 1.916 3.839 0.000 Park et al(2012) 1.344 0.286 0.0820.7831.904 4.701 0.000 Seo Chang(2009) 0.927 0.281 0.079 03761.479 3.294 0.001 Yang,Kang,Kim(2011) 0.549 0.259 0.067 0.0421.056 2.121 0.034 Won Chae(2011) -0.563 0.315 0.099 -1.1800.053 -1.790 0.073 Lee(2008) 1.185 0.317 0.1000.564 1.806 3.739 0.000 Lee Hwang(2011) 0.874 0.256 0.065 0.373 1.376 3.419 0.001 Lee,Lee,Kim(2011) 0.324 0.248 0.061-0.1620.810 1.308 0.191 Jung Jeon(2004) 0.000 0.329 0.108-0.6450.645 0.000 1.000 Choi Lee(2012) 1.847 0.360 0.1301.1412.552 5.128 0.000 1.142 0.307 0.0940.5401.745 3.716 0.000 -1.00 -0.50 0.00 0.50 1.00 Favors control Favors treatment FIG.3.Effect size of aromatherapy according to the method applied.Test for overall effect size:Z=6.107,p<0.001. Subgroup massage:heterogeneity:tau2=1.78,O=119.55,df=6 (p<0.001).P=94.98%.Test for overall effect size: Z=2.21,p=0.027.Total heterogeneity:tau2=1.03,O=134.77,df=11 (p<0.001),P=91.84%.Test for overall effect size: Z=3.72,p<0.001

degree of sleep disorder, of which seven reported that aro￾matherapy was effective. One study measured the degree of satisfaction with sleep and one study included sleep efficiency, both of which reported that aromatherapy was effective. Effect size of aromatherapy Twelve of the selected papers were used to analyze effect size of aromatherapy on sleep. Heterogeneity of the 12 pa￾pers was verified (I 2 = 91.84%; p < 0.001), and a random￾effects model was used. The resulting effect size was 3.72 (95% CI, 0.54–1.75; p < 0.001), which indicated that aro￾matherapy significantly improved sleep (Fig. 2). A meta-analysis was conducted based on the intervention factors of aromatherapy application method and the partic￾ipants’ health condition. Inhalation therapy (95% CI, 0.79– 1.54; Z = 6.11; p < 0.001) was more effective than massage therapy (95% CI, 0.13–2.17; Z = 2.21; p = 0.027), and the effect size for participants with a disease (95% CI, 0.25– 1.10; Z = 3.10; p = 0.002) was larger than that for healthy participants (95% CI, 0.39–5.10; Z = 2.29; p = 0.022) (Figs. 3 and 4). FIG. 2. Effect size of aromatherapy related to sleep. Subgroup inhalation: heterogeneity: tau2 = 0.09, Q = 8.04, df = 4 ( p = 0.090), I 2 = 50.23%. CI, confidence interval; df, degree of freedom. FIG. 3. Effect size of aromatherapy according to the method applied. Test for overall effect size: Z = 6.107, p < 0.001. Subgroup massage: heterogeneity: tau2 = 1.78, Q = 119.55, df = 6 ( p < 0.001), I 2 = 94.98%. Test for overall effect size: Z = 2.21, p = 0.027. Total heterogeneity: tau2 = 1.03, Q = 134.77, df = 11 ( p < 0.001), I 2 = 91.84%. Test for overall effect size: Z = 3.72, p < 0.001. EFFECTS OF AROMATHERAPY ON SLEEP IMPROVEMENT 65