effect of lavender

By Jamie Lytle, RN, BSN, Catherine Mwatha, RN, BS, and Karen K. Davis, RN, PhD

Background Sleep deprivation in hospitalized patients is com-mon and can have serious detrimental effects on recovery fromillness. Lavender aromatherapy has improved sleep in a varietyof clinical settings, but the effect has not been tested in theintermediate care unit.Objectives To determine the effect of inhalation of 100%lavender oil on patients vital signs and perceived quality ofsleep in an intermediate care unit.Methods A randomized controlled pilot study was conductedin 50 patients. Control patients received usual care. The treat-ment group had 3 mL of 100% pure lavender oil in a glass jarin place at the bedside from 10 PM until 6 AM. Vital signs wererecorded at intervals throughout the night. At 6 AM all patientscompleted the Richard Campbell Sleep Questionnaire toassess quality of sleep.Results Blood pressure was significantly lower between mid-night and 4 AM in the treatment group than in the control group(P = .03) According to the overall mean change score in bloodpressure between the baseline and 6 AM measurements, thetreatment group had a decrease in blood pressure and thecontrol group had an increase; however, the difference betweenthe 2 groups was not significant (P = .12). Mean overall sleepscore was higher in the intervention group (48.25) than in thecontrol group (40.10), but the difference was not significant.Conclusion Lavender aromatherapy may be an effective way toimprove sleep in an intermediate care unit. (American Journalof Critical Care. 2014;23:24-29)

EFFECT OF LAVENDER AROMATHERAPY ON VITAL SIGNS AND PERCEIVED QUALITY OF SLEEP IN THE INTERMEDIATE CARE UNIT: A PILOT STUDY

2014 American Association of Critical-Care Nursesdoi: http://dx.doi.org/10.4037/ajcc2014958

Innovative Approaches

24 AJCCAMERICAN JOURNAL OF CRITICAL CARE, January 2014, Volume 23, No. 1 www.ajcconline.org

by guest on August 20, 2015ajcc.aacnjournals.orgDownloaded from

A systematic review4 of nursing interventionsindicated that little research is available on use ofaromatherapy to improve sleep in the hospital.Aromatherapy is the use of pure essential oils fromfragrant plants to help relieve health problems andimprove quality of life in general.5 Aromatherapyhas been used in a variety of settings to assist inrelaxation and aid in sleep. In a study of 122 patientsin an intensive care unit by Dunn et al,6 patientsexposed to aromatherapy had significantly greaterimprovements in mood and perceived levels ofanxiety than did patients not exposed. In a random-ized controlled study7 in a hospice, the use of laven-der and massage resulted in improved sleep scores,but the results were not significant because of thesmall number of patients in the sample. Severalstudies have indicated that lavender aromatherapyaffects the autonomic nervous system, reducing anx-iety in patients in different settings, such as inpa-tients and outpatients.8 Sleep deprivation leads tomarkedly impaired glucose tolerance and reduc-tions in acute insulin responses to glucose. Glucosecontrol is an important marker in healing for criti-cally ill patients.9 Lavender aromatherapy can alsoreduce mild insomnia in patients in their home set-ting.10 We chose lavender rather than anotheressential oil because of the studies that supportusing lavender to promote rest and relaxation indifferent settings. Little scientific research has been

done to establish effects of lavender aromatherapyon sleep in an acute care hospital setting.

Investigating ways to promote and provide arestful night of sleep for hospitalized patients isimportant. Vital signs can be a measure of the bodysresponse to stress, illness, and relaxation. A decreasein blood pressure, heart rate, andrespiratory rate correlates with agreater relaxation state.11 Our aimwas to evaluate the effectiveness ofinhalation of 100% lavender oil onthe vital signs and perceived qualityof sleep of IMCU patients. Wehypothesized that patients whoreceived the aromatherapy wouldexperience a decrease in blood pres-sure, heart rate, and respiratory rateduring the night and would report higher quality ofsleep than would patients who did not receive thearomatherapy.

MethodsStudy Design and Sample

This randomized controlled pilot study wasconducted in the IMCU of a large academic teach-ing hospital between August 2, 2011, and Decem-ber 2, 2011. Patients were eligible if they were olderthan 21 years and admitted to the IMCU for at least2 nights. Patients were excluded if they could notspeak English, were confused, had respiratory prob-lems requiring mechanical ventilation or continu-ous positive airway pressure, were receiving oxygenvia mask, had an allergy or sensitivity to oils or fra-grances, or had received a new blood pressure med-ication or a sleeping pill on the night of the study.Potential patients were referred to the study teamby the nursing staff. Any time after the first night in

Sleep is an essential component of health and is related to physical and psychologi-cal well-being. Inadequate quality and quantity of sleep in hospitalized patientsare common problems, particularly in intensive care or intermediate care units(IMCUs) and can have serious detrimental effects on health and recovery from ill-ness.1 The association between the severity of illness and sleep disturbance inpatients in the intensive care unit has been evaluated.2 Sleep disruption was greater in patientswho died and in patients who had a higher disease severity score than in patients who survivedand had lower scores. In addition, sleep deprivation has a adverse effect on the immune sys-tem and is associated with increased morbidity in critically ill patients. Among patients whoreceived an influenza vaccine, patients who were sleep deprived produced less than half thelevel of antibodies produced by patients who had normal sleep times.3 Also, sleep deprivationis one of the most frequent complaints of patients after hospital stays.2

About the AuthorsJamie Lytle and Catherine Mwatha are nurse cliniciansand Karen K. Davis is director of nursing, Department ofMedicine, Johns Hopkins Hospital, Baltimore, Maryland.

Corresponding author: Karen K. Davis, RN, PhD, JohnsHopkins Hospital, 1830 E Monument St, 9th Floor, Room9061, Baltimore, MD 21287 (e-mail: [email protected]).

www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, January 2014, Volume 23, No. 1 25

Patients who dieand those withhigher diseaseseverity scoreshave greatersleep disruption.


Table 2 Mean (SD) for vital signs and change scores

Vital sign Control Intervention

Baseline (10 PM) Change from 10 PM to midnight Change from midnight to 4 AM

Control Intervention P Control Intervention P

Mean arterial pressure,mm Hg

Heart rate, beats perminute

Respiratory rate,breaths per minute

Oxygen saturation, %

87.7 (14.7)

87.8 (16.8)

22.7 (11.2)

96.6 (2.5)

89.9 (17.6)

87.1 (17.1)

22.3 (6.2)

97.4 (2.7)

2.41 (9.1)

3.5 (10.2)

3.1 (8.8)

-0.3 (2.2)

2.9 (12.9)

1.4 (7.5)

2.2 (5.2)

-0.2 (2.0)

.87

.41

.67

.79

-3.4 (10.8)

1.4 (7.8)

0.5 (5.9)

0.4 (2.0)

3.5 (10.4)

2.3 (9.4)

1.7 (3.8)

0.4 (1.2)

.03

.70

.40

.93

vital sign spread sheet, and administer the sleepquestionnaire at 6 AM. The care providers document-ing the vital signs and collecting the perceived sleepdata were aware of each patients group.

MeasurementsThe Richard Campbell Sleep Questionnaire is a

visual analog scale that provides self-reported scoreson depth of sleep, ease of falling asleep, frequencyof awakening, ease in return to sleep, and qualityof sleep.12 The items are marked by the patient on a100-mm line with words describing the poorestpossible sleep (0 mm) to the best possible sleep(100 mm). Patients place a mark on the 100-mmline, and a score from 0 to 100 can be calculatedfor each item by measuring the distance from theright end of the line with a millimeter ruler. Higherscores indicate better sleep. The total sleep score iscalculated by determining the mean of the 5 scores.The tool has been tested in intensive care patientsand has adequate reliability (Cronbach = 0.82).9

Vial signs were obtained from the electronic med-ical record and transferred to a data collection tooldesigned for this study. Mean arterial blood pressureis the mean pressure within an artery over a com-plete cycle of 1 heartbeat. The following equation isused for calculation: mean arterial pressure = ([2 xdiastolic] + systolic)/3. This measure for blood pres-sure was used to compare mean changes in bloodpressure over time.

Statistical AnalysisDescriptive statistics were computed for all study

variables. Frequencies and percentages were used todescribe categorical variables, and means and stan-dard deviations were used to describe continuousvariables. Group differences in demographics anddependent variables at baseline were assessed byusing 2 analysis and the Fisher exact test for cate-gorical variables and t tests for continuous variables.

the unit, patients were approached by 1 of 2 trainedresearch assistants and asked if they would be willingto participate in the study. Oral consent was obtained.The investigation conformed to the principles out-lined in the Declaration of Helsinki. Approval forthe study was obtained from the appropriate insti-tutional review board.

Demographics, clinical characteristics, and vitalsigns were collected from the electronic record. Unitguidelines require that vital signs be collected at thepatients bedside, so remote monitoring was not pos-sible. Fifty patients were randomized to the controlgroup or the intervention group by using a computer-generated assignment list. Each group had 25 patients.The control group received usual care. Each patient

in the treatment group receivedusual care and had 3 mL of 100%pure therapeutic-grade essential oilof lavender, sold by Edens Garden,in a small glass jar at the bedsideplaced within 1 m of the patientfrom 10 PM until 6 AM. Diffusers orheat administration of the lavenderwas not approved by the infectiousdisease department because of therisk of spreading germs and of heat-

related injuries. Because Dunn et al6 noted that theeffects of lavender were not sustained, placement ofthe oil of lavender was maintained throughout thenight. The oil was premeasured by a study teammember before use in the aromatherapy.

Patients in both groups had vital signs measuredat the start of the intervention (10 PM), then at 4-hour intervals throughout the night (per unit stan-dards), and at the end of the intervention at 6 AM.At 6 AM all patients were asked to fill out the sleepquestionnaire. The nurses on the unit received train-ing to administer the aromatherapy, complete the


The treatmentgroup received

3 mL lavender oilplaced within 3feet from 10 PM

until 6 AM.


Table 1 Baseline demographic characteristics (N = 50)a

54 (15)

8 (32)17 (68)

5 (20)8 (32)8 (32)1 (4)2 (8)0 (0)1 (4)

5 (20)20 (80)

4 (16)21 (84)

50 (20)

9 (36)16 (64)

3 (12)8 (32)3 (12)1 (4)3 (12)4 (16)3 (12)

7 (28) 18 (72)

3 (12)22 (88)

.45

.77

.24

.37

.50

Age, mean (SD), y

Sex Male Female

Diagnosis Cardiac Digestive Endocrine/metabolic Autoimmune Infectious Hematologic Urologic

Oxygen therapy Yes No

Pain medication Yes No

a Unless otherwise indicated, data in the table are expressed as number (percentage).b From 2 analysis, Fisher exact test, or t test.

Intervention group(n = 25)

Control group(n = 25)Characteristic Pb

Change from 4 AM to 6 AM Final at 6 AM Overall mean change 10 PM to 6 AM

Control Intervention PControl InterventionControl Intervention P

-2.3 (10.2)

-1.7 (6.3)

-2.2 (7.0)

-0.8 (2.4)

-3.6 (9.4)

0.9 (8.6)

-1.3 (4.0)

-0.5 (2.0)

.64

.23

.61

.65

91.1 (18.9)

84.8 (15.5)

21.3 (7.6)

97.2 (2.3)

87.0 (14.5)

82.6 (14.9)

19.7 (5.1)

97.6 (2.4)

-3.4 (12.8)

3.1 (10.4)

1.4 (10.3)

-0.6 (2.6)

2.9 (15.2)

4.6 (11.4)

2.6 (4.8)

-0.3 (2.1)

.12

.64

.61

.59


The scores on the Richard Campbell Sleep Ques-tionnaire were calculated by using the establishedstandards.12 The change scores between the inter-vention group and the control group were analyzedby using mean scores and compared by using inde-pendent sample t tests. The value = 0.05 was set asan acceptable level of significance. This investigationwas a pilot study, so power analysis and sample-sizecalculations were not performed.

ResultsOf the 50 patients who participated in the study,

25 were allocated to the intervention group and 25to the control group. The sample was predominantlyfemale, with a mean age of 52 years. The majorityof patients were admitted to the IMCU because ofcardiac, digestive, or endocrine conditions. Mostpatients did not receive oxygen therapy or pain med-ication during the night of the study. The 2 groupsdid not differ in any of the baseline demographic orclinical characteristics (Table 1).

Vital SignsBaseline vital signs for both groups were similar.

Mean change scores for the interval 10 PM to mid-night were similar for both groups (Table 2). How-ever, mean change scores for the interval midnightto 4 AM indicated that patients in the interventiongroup had a decrease in blood pressure, whereasthose in the control group had an increase in bloodpressure; this difference between the 2 groups wassignificant (P= .03). For the interval 4 AM to 6 AM,both groups had a decrease in blood pressure. Theoverall mean change score between blood pressureat 10 PM and blood pressure at 6 AM indicated thatpatients in the intervention group had a decrease inblood pressure and patients in the control group hadan increase; however, this difference between the 2groups was not significant (P= .12). Similar trendsoccurred in the changes in heart rate, respiratory

rate, and oxygen saturation, but none of the differ-ences were significant.

Perceived Quality of SleepMean sleep scores for depth of sleep, ease of

falling asleep, ease in return to sleep, and quality ofsleep were higher in the intervention group than inthe control group, but the difference was not signif-icant (Table 3 .) Scores for frequency of awakeningwere similar across both groups. Mean overall sleepscore was higher in the intervention group (48.25)than in the control group (40.10), but this differ-ence was not significant.

DiscussionAlthough aromatherapy has been used in a vari-

ety of settings, to our knowledge, no interventional


Louis and Kowalski17 measured not only vitalsigns but also pain, anxiety, depression, and senseof well-being in a group of 17 cancer patients whoeach received aromatherapy, a humidified-watertreatment, and then no treatment during a 3-dayperiod. The aromatherapy and the treatment withhumidified water lowered blood pressure and pulseand had a positive effect on the other variables, butthe differences were not significant. Perhaps largersample sizes are needed to detect the subtle changesin vital signs that may occur with aromatherapy.Further testing in larger numbers of patients isneeded to determine what strategies will improvesleep and restfulness in hospitalized patients.

Limitations and RecommendationsMost of the patients in our study were women.

Perhaps sleep is affected by sex-related factors,which were not explored. We used the RichardCampbell Sleep Questionnaire, which involvesself-reporting by patients, and so our results aresubjective. Other methods provide a more accuratemeasure of sleep. Polysomnography is the goldstandard for measuring sleep quality and quantity;however, this method is expensive and requirescontinuous electroencephalographic monitoring anda specialized trained technician throughout the study.We did not have financial support for polysomnog-raphy in the pilot phase. Aromatherapy is typicallyconducted by using a heat source or a diffuser, whichour infectious disease department prohibited in ahospital setting. We recognize that oxygen therapycould disrupt aromatherapy, but oxygen deliveredvia nasal cannula was allowed during the studybecause we thought patients would still be able tosense the lavender aroma and because most patientson the IMCU receive some form of oxygen therapy.

Our research was a pilot study; thus the numberof patients in our sample might not have been ade-quate to detect subtle changes in vital signs andperceptions of sleep. Although we did monitor vitalsigns at baseline and throughout the night, we meas-ured sleep only once, at the end of the intervention,because the Richard Campbell Sleep Questionnairehas not yet been validated as a tool for use before andafter treatment. Studies in which a pretest-posttestdesign was used or sleep was measured more than 1night were more likely to have significant results.15,17

Research on aromatherapy has several method-ological challenges. In this pilot study, the data col-lectors, care providers, and participants knew thegroup assignment of each patient because of thelavender odor at the bedside. This situation couldintroduce both interviewer and participant bias. In

studies of the outcomes have been conducted inpatients in an IMCU. Our investigation is the firstpilot study on the effects of aromatherapy on bloodpressure, heart rate, respiratory rate, and perceivedquality of sleep in the IMCU. The primary finding isthat aromatherapy with 100% essential oil of laven-der resulted in lower blood pressure after 6 hours oftherapy. Other vital signs did not seem to be affected.Other researchers have had similar results. Accord-ing to a systematic literature review,13 aromatherapyyielded positive improvement in physiological stressin 1 study, as indicated by heart rate, systolic bloodpressure, and respiratory rate; caused no change inphysiological stress in 2 studies; and had no effecton sleep quality in 1 study. Chien et al14 found sim-ilar results in vital signs and sleep improvement. Intheir study in 67 middle-aged women on the effectof lavender aromatherapy on the autonomic nervoussystem, the 34 women in the aromatherapy grouphad a significant decrease in mean heart rate. Chienet al also found that lavender aromatherapy led to asignificant improvement in sleep quality when thewomen had a 20-minute exposure twice a weekduring a 12-week period. Using the St Marys Hos-pital Sleep Questionnaire to measure sleep in a sam-ple of 64 patients in 2 cardiac care units in Iran,Moeini et al15 found significant improvement inmean sleep quality scores after lavender aromather-apy. In their study,15 sleep was measured before andafter therapy for 3 nights of lavender aromatherapy.

In our study, patients who had aromatherapyreported higher quality of sleep than did patients inthe control group; however, the difference betweenthe 2 groups was not significant. This finding maybe related to the small number of patients in thestudy sample. The size of the study sample is acommon limitation in the studies on lavender aro-matherapy; most of the samples consist of a smallnumber of patients and have flaws in the methodsused. Thus, results should be interpreted with somedegree of caution.16

Table 3 Mean (SD) scores for Richard Campbell Sleep Questionnaire

41.44 (32.50)

36.92 (30.83)

46.36 (34.61)

36.20 (33.22)

39.56 (32.52)

40.10 (23.42)

52.60 (34.09)

47.76 (34.41)

46.24 (35.47)

49.48 (37.22)

45.16 (38.99)

48.25 (32.09)

.24

.25

.99

.19

.58

.31

Deep/light sleep

Ease of falling asleep

Awakenings

Ease of return to sleep

Quality of sleep

Overall sleep score

Intervention (n = 25)

Control (n = 25)Item P



addition, determining the purity of the oil and howmuch a patient actually sensed or perceived thearoma is difficult and could introduce interventionbias. Despite these challenges, patients in the treat-ment group had higher sleep scores than did thecontrol group for most questions, suggesting thatlavender aromatherapy may be a promising inter-vention to enhance perceived sleep. On the basisof our results, we suggest that future studies be con-ducted in larger numbers of patients, with a pretest-posttest design and use of methods to reduce bias.Because of the adverse effect of poor sleep qualityin critically ill patients, research on ways to improvesleep in acute care settings should continue.

ConclusionDespite limitations, the results of our pilot

study on alternative therapy to enhance sleep haveimportant implications. Sleep is essential to healing,and finding ways to offer patients more restful sleepwhile they are hospitalized is critical, particularly inmore acute care settings such as the IMCU. Wedetected a decrease in blood pressure after the 6 hoursof treatment and higher satisfaction with sleep afterthe use of lavender aromatherapy. We think thatconducting a randomized controlled trial of aro-matherapy in an IMCU is feasible. Research usinglarger numbers of patients is required to understandthe effects of lavender aromatherapy on sleep in thehospital and whether a combination of alternativetherapies, such as massage or music, would havegreater effects than aromatherapy alone.

ACKNOWLEDGMENTSWe acknowledge Maddy Biggs, Kathy Wagner-Kosmakos,and the nurses on the medical progressive care unit.

FINANCIAL DISCLOSURESThis work was supported through the Crickett JuliusMemorial Scholarship Fund.

REFERENCES1. Dogan O, Ertekin S, Dogan S. Sleep quality in hospital-

ized patients. J Clin Nurs. 2005;14(1):107-113.2. Patel M, Chipman J, Carlin BW, Shade D. Sleep in the

intensive care unit setting. Crit Care Nurs Q. 2008;31(4):309-318.

3. Spiegel K, Sheridan JF, Van Cauter E. Effect of sleepdeprivation on response to immunization. JAMA. 2002;288(12):1471-1472.

4. Hellstrom A, Fagerstrm C, Wilman A. Promoting sleepby nursing interventions in health care settings: a sys-tematic review. Worldviews Evid Based Nurs. 2011;8(3):128-142.

5. Buckle J. Literature review: should nursing take aro-matherapy more seriously? Br J Nurs. 2007;16(2):116-120.

6. Dunn C, Sleep J, Collett D. Sensing an improvement:an experimental study to evaluate the use of aromather-apy, massage, and periods of rest in an intensive careunit. J Adv Nurs. 1995;21:34-40.

7. Soden K, Vincent K, Craske S, Lucas C, Ashley S. Arandomized controlled trial of aromatherapy massagein a hospice setting. Palliat Med. 2004;18(2):87-92.

8. Van der Watt G, Janca A. Aromatherapy in nursing andmental health care. Contemp Nurse. 2008;30(1):69-75.

9. Van Cauter E, Spiegel K, Tasali E, Leproult R. Metabolicconsequences of sleep and sleep loss. Sleep Med.2008;9(suppl 1):S23-S28.

10. Lewith GT, Godfrey AD, Prescott P. A single-blinded,randomized pilot study evaluating the aroma of Lavan-dula augustifolia as a treatment for mild insomnia. JAltern Complement Med. 2005;11(4):631-637.

11. Penzel T, Kantelhardt J, Lo C, Voigt K, Vogelmeier C.Dynamics of heart rate and sleep stages in normalsand patients with sleep apnea. Neuropsychopharma-cology. 2003;28(suppl 1):S48-S53.

12. Richards KC, OSullivan PS, Phillips RL. Measurementof sleep in critically ill patients. J Nurs Meas. 2000;8(2):131-144.

13. Halm M. Essential oils for management of symptomsin critically ill patients. Am J Crit Care. 2008;17;160-163.

14. Chien LW, Cheng SL, Liu FC. The effects of lavenderaromatherapy on autonomic nervous system in midlifewomen with insomnia. Evid Based ComplementaryAltern Med. 2012;2012:740813.

15. Moeini M, Khadibi M, Bekhradi R, Mahmoudian SA,Nazari F. Effect of aromatherapy on the quality of sleepin ischemic heart disease patients hospitalized in inten-sive care units of heart hospitals of the Isfahan Universityof Medical Sciences. Iran J Nurs Midwifery Res. 2010;15(4):234-239.

16. Cooke B, Ernst E. Aromatherapy: a systematic review.Br J Gen Pract. 2000;50(455):493-496.

17. Louis M, Kowalski SD. Use of aromatherapy with hospicepatients to decrease pain, anxiety, and depression andto promote an increased sense of well-being. Am JHosp Palliat Care. 2002;19(6):381-386.

To purchase electronic or print reprints, contact theAmerican Association of Critical-Care Nurses, 101Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail,[email protected].

eLettersNow that youve read the article, create or contribute to anonline discussion on this topic. Visit www.ajcconline.organd click Responses in the second column of either thefull-text or PDF view of the article.



http://ajcc.aacnjournals.org/cgi/external_ref?link_type=PERMISSIONDIRECTPersonal use only. For copyright permission information:

Published online http://www.ajcconline.org 2014 American Association of Critical-Care Nurses

doi: 10.4037/ajcc2014958 2014;23:24-29Am J Crit Care

Jamie Lytle, Catherine Mwatha and Karen K. DavisSleep in the Intermediate Care Unit: A Pilot StudyEffect of Lavender Aromatherapy on Vital Signs and Perceived Quality of

http://ajcc.aacnjournals.org/subscriptions/Subscription Information

http://ajcc.aacnjournals.org/misc/ifora.xhtmlInformation for authors

http://www.editorialmanager.com/ajccSubmit a manuscript

http://ajcc.aacnjournals.org/subscriptions/etoc.xhtmlEmail alerts

by AACN. All rights reserved. 2014 CopyrightTelephone: (800) 899-1712, (949) 362-2050, ext. 532. Fax: (949) 362-2049. bimonthly by The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656.journal of the American Association of Critical-Care Nurses (AACN), published AJCC, the American Journal of Critical Care, is the official peer-reviewed research


effect of lavender

Documents