- Email: [email protected]
Using music to reduce anxiety among older adults in the emergency department: a randomized pilot study
Journal homepage:
www.jcimjournal.com/jim www.elsevier.com/locate/issn/20954964
Available also online at www.sciencedirect.com. Copyright © 2017, Journal of Integrative Medicine Editorial Office. E-edition published by Elsevier (Singapore) Pte Ltd. All rights reserved.
●
Research Article
Using music to reduce anxiety among older adults in the emergency department: a randomized pilot study Laura Belland1, Laura Rivera-Reyes2, Ula Hwang2,3 1. Center for Family Medicine, NewYork-Presbyterian/Columbia University Medical Center, New York, NY 10032, USA 2. Department of Emergency Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 3. Geriatric Research, Education and Clinical Center, James J. Peters VAMC, New York, NY 10029, USA ABSTRACT BACKGROUND: An emergency department (ED) visit may be distressing and anxiety-provoking for older adults (age > 65 years). No studies have specifically evaluated the effect of music listening on anxiety in older adults in the ED. OBJECTIVE: The objective of this pilot study was to evaluate the effect of music listening on anxiety levels in older ED patients. DESIGN, SETTING, PARTICIPANTS AND INTERVENTIONS: This was a randomized pilot study in the geriatric ED of an urban academic tertiary medical center. This was a sample of English-speaking adults (age > 65 years) who were not deaf (n = 35). Subjects consented to participate and were randomized to receive up to 60 min of music listening with routine care, while the control group received routine care with no music. Subjects in the music treatment group received headphones and an electronic tablet with pre-downloaded music, and were allowed to choose from 5 selections. MAIN OUTCOME MEASURES: The primary outcome was change in anxiety levels, measured by the statetrait anxiety inventory (STAI), at enrollment and 1 h later. RESULTS: A total of 35 participants were enrolled: 74% were female, 40% were white, and 40% were black; of these, 32 subjects completed the study protocol. When comparing control (n = 18) against intervention subjects (n = 17), there were no significant differences in enrollment STAI scores (43.00 ± 15.00 vs. 40.30 ± 12.80, P = 0.57). STAI scores 1 hour after enrollment (after the music intervention) were significantly reduced in the intervention subjects compared to the control subjects (with reduction of 10.00 ± 12.29 vs. 1.88 ± 7.97, P = 0.03). CONCLUSION: These pilot results suggest that music listening may be an effective tool for reducing anxiety among older adults in the ED. Keywords: music; anxiety; emergency department; randomized controlled trial Citation: Belland L, Rivera-Reyes L, Hwang U. Using music to reduce anxiety among older adults in the emergency department: a randomized pilot study. J Integr Med. 2017; 15(6): 450–455.
http://dx.doi.org/10.1016/S2095-4964(17)60341-8 Received February 13, 2017; accepted March 23, 2017 Correspondence: Laura Belland; E-mail: [email protected]
November 2017, Vol. 15, No. 6
450
Journal of Integrative Medicine
www.jcimjournal.com/jim
1 Introduction A visit to the emergency department (ED) is anxiety provoking for patients by nature. Contributing factors may include the acuity of the visit,[1] a noisy environment,[2] an unfamiliar staff,[3] fear of painful tests or studies[3] and waiting in anticipation of a serious diagnosis or bad news.[4] Studies suggest that nearly 75% of adult ED patients experience mild to severe anxiety in relation to their ED visit.[5] In fact, anxiety or worry can be the second most common source, behind pain, of self-perceived acute “suffering” among patients visiting the ED.[4] A visit to the ED may be particularly distressing for older adults (age > 65 years), who are more likely than younger adults to have a greater ED length of stay before discharge, [6] receive more diagnostic tests and venipuncture for intravenous access [6] and have less effective pain care.[7–9] Anxiety can have deleterious effects on a patient in the clinical setting. Patients may report excessive pain complaints and manifest the typical signs and symptoms of anxiety (e.g., anorexia, dry mouth, nausea and chest pain), which complicate diagnosis.[10] Previous studies have found that patients awaiting procedures or surgery who have pre-procedural anxiety have lowered pain thresholds,[11] increased analgesic use,[5] greater need for sedation[12] and longer post-procedural recovery.[13] Patients may refuse evaluation or treatment because of anxiety surrounding the procedure or possible outcome.[10] Lastly, patient anxiety can impose barriers to communication with ED staff,[14] hindering successful delivery of important medical information. Music listening as an anxiolytic has been studied across a variety of clinical settings. Four separate Cochrane systematic reviews have reported the beneficial effects of music listening on anxiety in mechanically ventilated patients,[15] in perioperative patients,[16] in patients with coronary heart disease[17] and in oncology patients.[18] Other studies have shown the anxiolytic effect of music listening in patients undergoing cystoscopy, [19] in pulmonary rehabilitation patients [20] and in pediatric ED patients undergoing laceration repair. [21] However, there is a relative paucity of published data on the use of music listening for adult patients in the ED. While some studies have demonstrated mixed findings on the impact of music on ED patients,[22–25] no studies, to our knowledge, have specifically evaluated the effect of music listening on anxiety in older adults in the ED. Our aim was to perform a randomized pilot study comparing music listening plus standard care to standard care, with the goal of decreasing levels of anxiety among older adults as measured by the statetrait anxiety inventory (STAI).[26] The STAI is a validated Journal of Integrative Medicine
tool that measures state anxiety using a four-point forced choice Likert response scale (i.e., not at all, somewhat, moderately so, very much so). STAI scores range from 20 to 80, with 20 indicating mild to no anxiety while 80 indicates severe anxiety. 2 Methods 2.1 Study design and setting This was a randomized pilot study that took place in the geriatric ED of The Mount Sinai Hospital, an urban academic tertiary care medical center, during the months of April–May, 2015. The geriatric ED is a space connected to the main adult ED that is dedicated to providing an environment more conducive to treating patients aged > 65 years (e.g., softer lighting, reduced noise, multiple private rooms). Only those who are triaged with an emergency severity index (ESI) of 2, 3, 4 or 5 (1 to 5, 1 = urgent, 5 = nonurgent) are evaluated in the geriatric ED, while those with an ESI of 1 are seen in the main ED. This study received institutional review board approval from the Icahn School of Medicine at Mount Sinai. This study was not registered as a clinical trial as it did not meet applicable clinical trial guidelines under the Federal Drug and Administration Act. 2.2 Selection of participants Subjects were recruited from 8:00 a.m. to 8:00 p.m. Inclusion criteria included speaking English. Exclusion criteria included being deaf, contact isolation precautions (given the nature of the intervention) and prisoners. Patients unable to participate in the process of informed consent (e.g., due to delirium or severe dementia) were excluded. Also excluded were patients who had been given a disposition of discharge by their ED provider at the time of screening, since these patients were likely to leave the ED before completion of the study. Screening of eligible participants was done through chart review using an electronic ED tracking board accessed through the ED electronic health record, which lists patients alphabetically; all subjects deemed eligible through screening were then approached alphabetically. As this was a pilot study, power analysis/sample size calculation was not required. As many subjects were recruited as possible during the study period (2 months). 2.3 Intervention Intervention group participants received an electronic tablet with headphones and were instructed to listen to music for up to one hour. Participants could choose from five different genres of music: classical, jazz, new age, Latin guitar, or Chinese traditional. These genres had been selected in an attempt to offer participants a wide variety of music. All of the available selections contained only instrumental music since lyrics may be “distracting” or “emotional.”[27] Tempos of the music ranged between 60
451
November 2017, Vol. 15, No. 6
www.jcimjournal.com/jim
and 80 beats per minute, which have been identified in the literature as soothing and therapeutic tempos.[28] The volume level of the music was pre-set and was adjusted at the bedside according to the participant’s comfort level. Subjects were briefly instructed on how to use the tablet and change genres of music if desired. The tablets were cleaned with alcohol wipes between participants. The headphones were disposable. 2.4 Methods, measurements and outcomes After giving informed consent and prior to randomization, all participants were asked to verbally respond to a set of 20 self-report items from the STAI to establish pre-intervention anxiety scores. After completing the initial STAI assessment, participants were then randomized to either receive standard care plus the music intervention or to receive standard care only. Randomization was done at the bedside using sequentially numbered, concealed opaque envelopes. The randomization was performed with a 1:1 equal allocation ratio. Fifty envelopes had been randomly shuffled and then numbered sequentially at a remote location by the project coordinator who did not participate in patient recruitment. The envelopes were transferred to the ED and the investigator opened a consecutively numbered envelope containing the randomized group assignment for each participant. Following randomization, participants in the control group proceeded with the ED visit as usual, while participants in the intervention group received a tablet with headphones and were asked to listen to music for one hour. As little as 15 min of music listening has been shown to reduce anxiety,[28] however 1 h was chosen given the possibility that participants would be interrupted during the hour, whether by clinicians, staff or family members.[23] In fact, ED providers were not formally made aware of this study; thus, at any point during the 1 h, participants in either group may have interacted with ED clinicians as usual. After 1 h, subjects in both groups were asked to verbally respond to the same STAI questions as before to assess post-intervention anxiety levels. Given research personnel limitations, a single researcher performed the screening and enrollment and also conducted the intervention and surveys. Thus, it was not possible to provide a sham intervention. Patients who were approached were given a thorough explanation of the study and thus were aware of the study goals. Information was collected about patient characteristics including age, gender, race/ethnicity and ESI via retrospective chart review of the electronic health record (Epic ASAP, Epic Systems Corp, Verona, WI, USA). 2.5 Analysis Parametric distributions were assumed for selection of statistical tests of baseline characteristics. Differences November 2017, Vol. 15, No. 6
between study groups in baseline characteristics and clinical outcomes were assessed using Student t-tests for continuous variables and Chi-square tests for categorical variables and independent-samples. Outcomes were checked for normal distributions. All analyses were intention-to-treat and completed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA). 3 Results 3.1 Characteristics of study subjects A total of 1 904 patients were seen in the geriatric ED during the study months. During study hours, a total of 317 patients were screened for the study, 130 of whom were deemed eligible to participate. Of these patients, a total of 35 (27%) agreed to participate. Not feeling well enough to complete the surveys was a main reason for declining participation, as was not liking the choices of music. Of the 35 enrolled participants, 18 were randomized to the control group and 17 to the intervention group. Three subjects, 2 in the control group and 1 in the intervention group, withdrew from the study after the initial STAI survey. Reasons for withdrawal included not feeling well enough to continue participation and dissatisfaction with not being randomized to the intervention group. A total of 16 patients completed the study in each arm (Figure 1). There were no significant differences in baseline characteristics between the control and intervention arms (Table 1). 3.2 Treatment outcomes The intervention subjects had higher mean initial STAI scores (pre-intervention; 43.00 ± 15.00) than the control group (40.30 ± 12.80), but this difference was not statistically significant (P = 0.57). Follow-up mean STAI scores (post-intervention) were not significantly lower in the intervention group (31.60 ± 11.88) than in the control group (37.75 ± 12.61; P = 0.16). Comparison of mean reductions of the STAI scores, however, was statistically significant. The intervention group had a mean reduction of 10.00 ± 12.29, while the control group had a mean reduction of 1.88 ± 7.97 (P = 0.03; Table 2). 4 Discussion The results of this small, randomized pilot study suggest that listening to music may serve to decrease state anxiety in older adults (age > 65 years) seen in the ED. There was a statistically significant reduction of STAI scores in the music group compared to the control group. A caveat (and limitation) to these findings, as is the case with most pilot studies, is the study’s small sample size with low power and whether the results represent a true effect. Based on these results, we calculated that a sample size of at least
452
Journal of Integrative Medicine
www.jcimjournal.com/jim
50 participants, 25 in each arm, would detect a minimum effect size of 0.4 (differences in mean change in anxiety scores between the intervention and control group), with at least 80% power and α at 0.05. Although the clinical significance of our findings was not immediately clear, it is well-established that heightened anxiety can contribute to poor patient outcomes [5,11–13] and
pose barriers to patient care.[10,14] There are several ways in which music listening may affect anxiety levels among older adult patients in the ED. It may serve as a distraction and help patients regain a sense of control over their environment.[28] Especially when delivered via personal headphones, music can mask direct noise such as a patient talking in the next bed, as well as ambient noise such as
Screened for eligibility (N = 317) Excluded (n = 282) ·Did not meet inclusion criteria (n = 90) ·Met 1 or more exclusion criteria (n = 97) ·Declined to participate (n = 95) Randomized (n = 35)
Music (n = 17)
Control (n = 18)
Lost to follow-up (n = 2) Patient not feeling well (n = 1) Patient dissatisfied with group assignment (n = 1)
Lost to follow-up (n = 1) Patient not feeling well (n = 1)
Completed the study (n = 16)
Completed the study (n = 16)
Figure 1 Flow diagram of participant eligibility, enrollment, randomization and follow-up Table 1 Participant characteristics Characteristic
Control group (n = 18)
Music group (n = 17)
P value
Age (mean ± SD, years)
75 ± 7
73 ± 6
0.21
Female (n (%)) Race/ethnicity (n (%)) White Black Hispanic/other/unknown ESI score (mean ± SD)
13 (72)
13 (76)
8 (44) 9 (50) 1 (6) 2.8 ± 0.5
6 (35) 5 (30) 6 (35) 2.8 ± 0.7
0.77 0.08
0.96
Student t-test was used to analyze the differences of age and ESI score; Chi-square was used to analyze the differences of gender and race/ethnicity. SD: standard deviation; ESI: emergency severity index (1–5, 1 = urgent, 5 = non-urgent).
Table 2 Differences in state anxiety (intention-to-treat analysis) STAI score
Control group (n = 18)
Music group (n = 17)
P value
Baseline (pre-randomization)
40.30 ± 12.80
43.00 ± 15.00
0.57
Follow-up (post-randomization)
37.75 ± 12.61
31.60 ± 11.88
0.16
1.88 ± 7.97
10.00 ± 12.29
0.03
Reduction in STAI
Student t-test was used to analyze differences in state anxiety. Values are expressed as mean ± standard deviation. STAI: state-trait anxiety inventory.
Journal of Integrative Medicine
453
November 2017, Vol. 15, No. 6
www.jcimjournal.com/jim
machine alarms, ringing phones, or medical staff talking at the desk. Additionally, music listening can provide a sense of familiarity and comfort.[29] Based on effective music interventions in the literature,[28] the music used in this study was slow (60–80 beats/min) and non-lyrical. After controlling for these elements, participants were able to choose from a selection of genres since individual preference of, and inherent response to, different styles of music play a large role in whether a listener will find a certain type of music therapeutic.[30] In turn, music genre was not considered a confounder for which to control. The exact number of music-listening minutes also was not controlled for, as we found it difficult to monitor participants without risking a Hawthorne effect or altering standard care. Intervention participants did not report dissatisfaction with or adverse outcomes due to the music intervention. One intervention participant withdrew after completing the first STAI survey but before receiving the intervention, citing not feeling well enough to participate as her reason not to continue. Few obstacles, which were easily overcome, were revealed during the study. Some participants initially expressed hesitation in using a tablet, citing inexperience, but this fear was usually alleviated after the 30-second tutorial. A second obstacle, which some subjects cited as a reason for study decline, was dissatisfaction with the music selection. Although not addressed in this study, adding more genres is a simple solution. There are several limitations to this study, including the small sample size and low power. The sole researcher was not blinded to randomization allocation. This did not affect the approaching and enrollment of subjects nor the administration of the first STAI survey, since randomization was done after these steps, but it increased risk for expectation bias during administration of the second STAI survey. Likewise, with the absence of a sham intervention, subjects were not blinded, and their answers to the second STAI survey could have been subject to attention bias. Future studies on a larger scale could improve upon these problems by employing a sham intervention (e.g., a tablet and headphones without music or with white noise), as well as a second, blinded researcher to conduct the post-intervention STAI. Since the study took place at a single site and excluded all non-English speaking patients, the results are not generalizable. Additionally, the geriatric ED is designed to be less noisy than an average adult ED and, as such, may be more suitable for conducting a music intervention. Future studies should focus on incorporating non-English speaking patients in an average ED. Other considerations for future studies evaluating the effect of music on anxiety in older ED patients include November 2017, Vol. 15, No. 6
controlling for potential confounders such as medications received, tests or procedures performed, whether family or friends provide company at the bedside and whether subjects are seen by a care provider during the study hour. It would also be useful to determine a dose-response relationship between listening time and decreased anxiety levels (both duration and magnitude of the effect). This could be done by re-administering the STAI at regular intervals until anxiety levels return to pre-intervention levels or the patient is discharged from the ED. It could prove difficult to record the exact amount of time a patient listens to the music given the nature of the ED visit. In conclusion, preliminary data show that music listening may serve to reduce anxiety among older adults in the ED. Music listening is a noninvasive, safe, and easy-to-administer intervention to reduce anxiety among patients age > 65 years old being evaluated in the ED. Additional studies are warranted to evaluate this intervention on a larger scale. 5 Acknowledgements All significant contributors to this study have been named as authors. Laura Belland conceived of the study. Laura Belland, Ula Hwang, and Laura Rivera-Reyes designed the pilot. Laura Belland conducted the pilot study and data collection. Laura Belland undertook data management with the assistance of Laura Rivera-Reyes and Ula Hwang undertook statistical analyses. Laura Belland drafted the manuscript, and all authors contributed to its revision. Laura Belland takes responsibility of the paper as a whole. 6 Conflict of interests The authors have no conflicts of interest to declare. REFERENCES 1
2 3 4
5
454
Ekwall A, Gerdtz M, Manias E. Anxiety as a factor influencing satisfaction with emergency department care: perspectives of accompanying persons. J Clin Nurs. 2009; 18(24): 3489–3497. Tijunelis MA, Fitzsullivan E, Henderson SO. Noise in the ED. Am J Emerg Med. 2005; 23(3): 332–335. Byrne G, Heyman R. Patient anxiety in the accident and emergency department. J Clin Nurs. 1997; 6(4): 289–295. Body R, Kaide E, Kendal S, Foex B. Not all suffering is pain: sources of patients’ suffering in the emergency department call for improvements in communication from practitioners. Emerg Med J. 2015; 32(1): 15–20. Craven P, Cinar O, Madsen T. Patient anxiety may influence the efficacy of ED pain management. Am J Emerg Med. 2013; 31(2): 313–318.
Journal of Integrative Medicine
www.jcimjournal.com/jim 6 7 8
9
10 11
12
13 14
15 16 17 18
Baum SA, Rubenstein LZ. Old people in the emergency room: age-related differences in emergency department use and care. J Am Geriatr Soc. 1987; 35(5): 398–404. Hwang U, Richardson LD, Harris B, Morrison RS. The quality of emergency department pain care for older adult patients. J Am Geriatr Soc. 2010; 58(11): 2122–2128. Platts-Mills TF, Esserman DA, Brown DL, Bortsov AV, Sloane PD, McLean SA. Older US emergency department patients are less likely to receive pain medication than younger patients: results from a national survey. Ann Emerg Med. 2012; 60(2): 199–206. Hwang U, Belland LK, Handel DA, Yadav K, Heard K, Rivera-Reyes L, Eisenberg A, Noble MJ, Mekala S, Valley M, Winkel G, Todd KH, Morrison RS. Is all pain is treated equally? A multicenter evaluation of acute pain care by age. Pain. 2014; 155(12): 2568–2574. Stern TA, Fricchione GL, Rosenbaum JF. Massachusetts general hospital handbook of general hospital psychiatry. Amsterdam: Elsevier Health Sciences. 2010. Wong PS. Anxiety, signal anxiety, and unconscious anticipation: neuroscientific evidence for an unconscious signal function in humans. J Am Psychoanal Assoc. 1999; 47(3): 817–841. Schupp CJ, Berbaum K, Berbaum M, Lang EV. Pain and anxiety during interventional radiologic procedures: effect of patients’ state anxiety at baseline and modulation by nonpharmacologic analgesia adjuncts. J Vasc Interv Radiol. 2005; 16(12): 1585–1592. Jamison RN, Parris WC, Maxson WS. Psychological factors influencing recovery from outpatient surgery. Behav Res Ther. 1987; 25(1): 31–37. Musso MW, Perret JN, Sanders T, Daray R, Anderon K, Lancaster M, Lim D, Jones GN. Patients’ comprehension of their emergency department encounter: a pilot study using physician observers. Ann Emerg Med. 2015; 65(2): 151–155. Bradt J, Dileo C. Music interventions for mechanically ventilated patients. Cochrane Database Syst Rev. 2014; (12): CD006902. Bradt J, Dileo C, Shim M. Music interventions for preoperative anxiety. Cochrane Database Syst Rev. 2013; (6): CD006908. Bradt J, Dileo C, Potvin N. Music for stress and anxiety reduction in coronary heart disease patients (review). Cochrane Database Syst Rev. 2013; (12): CD006577. Bradt J, Dileo C, Magill L, Teague A. Music interventions for improving psychological and physical outcomes in cancer patients. Cochrane Database Syst Rev. 2016; (8):
Journal of Integrative Medicine
CD006911. 19 Raheem OA, Mirheydar HS, Lee HJ, Patel ND, Godebu E, Sakamoto K. Does listening to music during officebased flexible cystoscopy decrease anxiety in patients: a prospective randomized trial. J Endourol. 2015; 29(7): 791–796. 20 Reychler G, Mottart F, Boland M, Wasterlain E, Pieters T, Caty G, Liistro G. Influence of ambient music on perceived exertion during a pulmonary rehabilitation session: a randomized crossover study. Respir Care. 2015; 60(5): 711–717. 21 Sinha M, Christopher NC, Fenn R, Reeves L. Evaluation of nonpharmacologic methods of pain and anxiety management for laceration repair in the pediatric emergency department. Pediatr. 2006; 117(4): 1162–1168. 22 Parlar Kilic S, Karadag G, Oyucu S, Kale O, Zengin S, Ozdemir E, Korhan EA. Effect of music on pain, anxiety, and patient satisfaction in patients who present to the emergency department in Turkey. Jpn J Nurs Sci. 2015; 12(1): 44–53. 23 Short A, Ahern N. Evaluation of a systematic development process: relaxing music for the emergency department. Aust NZ J Med. 2009; 20: 1–27. 24 Weiland TJ, Jellnek GA, Macarow KE, Samartzis P, Brown DM, Grierson EM, Winter C. Original sound compositions reduce anxiety in emergency department patients: a randomised controlled trial. Med J Aust. 2011; 195(11–12): 694–698. 25 Menegazzi JJ, Paris PM, Kersteen CH, Flynn B, Trautman DE. A randomized, controlled trial of the use of music during laceration repair. Ann Emerg Med. 1991; 20(4): 348–350. 26 Spielberger CD, Gorsuch RL. Manual for the state-trait anxiety inventory (form Y): “self-evaluation questionnaire”. Washington, DC: Consulting Psychologists Press. 1983. 27 Krout RE. Music listening to facilitate relaxation and promote wellness: integrated aspects of our neurophysiological responses to music. Art Psychother. 2007; 34(2): 134–141. 28 Nilsson U. The anxiety- and pain-reducing effects of music interventions: a systematic review. AORN J. 2008; 87(4): 780–807. 29 Mok E, Wong KY. Effects of music on patient anxiety. AORN J. 2004; 77(2): 396–397, 401–406, 409–410. 30 Clair AA, Memmott J. Therapeutic uses of music with older adults. Silver Spring: American Music Therapy Association. 2008.
455
November 2017, Vol. 15, No. 6
www.jcimjournal.com/jim www.elsevier.com/locate/issn/20954964
Available also online at www.sciencedirect.com. Copyright © 2017, Journal of Integrative Medicine Editorial Office. E-edition published by Elsevier (Singapore) Pte Ltd. All rights reserved.
●
Research Article
Using music to reduce anxiety among older adults in the emergency department: a randomized pilot study Laura Belland1, Laura Rivera-Reyes2, Ula Hwang2,3 1. Center for Family Medicine, NewYork-Presbyterian/Columbia University Medical Center, New York, NY 10032, USA 2. Department of Emergency Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA 3. Geriatric Research, Education and Clinical Center, James J. Peters VAMC, New York, NY 10029, USA ABSTRACT BACKGROUND: An emergency department (ED) visit may be distressing and anxiety-provoking for older adults (age > 65 years). No studies have specifically evaluated the effect of music listening on anxiety in older adults in the ED. OBJECTIVE: The objective of this pilot study was to evaluate the effect of music listening on anxiety levels in older ED patients. DESIGN, SETTING, PARTICIPANTS AND INTERVENTIONS: This was a randomized pilot study in the geriatric ED of an urban academic tertiary medical center. This was a sample of English-speaking adults (age > 65 years) who were not deaf (n = 35). Subjects consented to participate and were randomized to receive up to 60 min of music listening with routine care, while the control group received routine care with no music. Subjects in the music treatment group received headphones and an electronic tablet with pre-downloaded music, and were allowed to choose from 5 selections. MAIN OUTCOME MEASURES: The primary outcome was change in anxiety levels, measured by the statetrait anxiety inventory (STAI), at enrollment and 1 h later. RESULTS: A total of 35 participants were enrolled: 74% were female, 40% were white, and 40% were black; of these, 32 subjects completed the study protocol. When comparing control (n = 18) against intervention subjects (n = 17), there were no significant differences in enrollment STAI scores (43.00 ± 15.00 vs. 40.30 ± 12.80, P = 0.57). STAI scores 1 hour after enrollment (after the music intervention) were significantly reduced in the intervention subjects compared to the control subjects (with reduction of 10.00 ± 12.29 vs. 1.88 ± 7.97, P = 0.03). CONCLUSION: These pilot results suggest that music listening may be an effective tool for reducing anxiety among older adults in the ED. Keywords: music; anxiety; emergency department; randomized controlled trial Citation: Belland L, Rivera-Reyes L, Hwang U. Using music to reduce anxiety among older adults in the emergency department: a randomized pilot study. J Integr Med. 2017; 15(6): 450–455.
http://dx.doi.org/10.1016/S2095-4964(17)60341-8 Received February 13, 2017; accepted March 23, 2017 Correspondence: Laura Belland; E-mail: [email protected]
November 2017, Vol. 15, No. 6
450
Journal of Integrative Medicine
www.jcimjournal.com/jim
1 Introduction A visit to the emergency department (ED) is anxiety provoking for patients by nature. Contributing factors may include the acuity of the visit,[1] a noisy environment,[2] an unfamiliar staff,[3] fear of painful tests or studies[3] and waiting in anticipation of a serious diagnosis or bad news.[4] Studies suggest that nearly 75% of adult ED patients experience mild to severe anxiety in relation to their ED visit.[5] In fact, anxiety or worry can be the second most common source, behind pain, of self-perceived acute “suffering” among patients visiting the ED.[4] A visit to the ED may be particularly distressing for older adults (age > 65 years), who are more likely than younger adults to have a greater ED length of stay before discharge, [6] receive more diagnostic tests and venipuncture for intravenous access [6] and have less effective pain care.[7–9] Anxiety can have deleterious effects on a patient in the clinical setting. Patients may report excessive pain complaints and manifest the typical signs and symptoms of anxiety (e.g., anorexia, dry mouth, nausea and chest pain), which complicate diagnosis.[10] Previous studies have found that patients awaiting procedures or surgery who have pre-procedural anxiety have lowered pain thresholds,[11] increased analgesic use,[5] greater need for sedation[12] and longer post-procedural recovery.[13] Patients may refuse evaluation or treatment because of anxiety surrounding the procedure or possible outcome.[10] Lastly, patient anxiety can impose barriers to communication with ED staff,[14] hindering successful delivery of important medical information. Music listening as an anxiolytic has been studied across a variety of clinical settings. Four separate Cochrane systematic reviews have reported the beneficial effects of music listening on anxiety in mechanically ventilated patients,[15] in perioperative patients,[16] in patients with coronary heart disease[17] and in oncology patients.[18] Other studies have shown the anxiolytic effect of music listening in patients undergoing cystoscopy, [19] in pulmonary rehabilitation patients [20] and in pediatric ED patients undergoing laceration repair. [21] However, there is a relative paucity of published data on the use of music listening for adult patients in the ED. While some studies have demonstrated mixed findings on the impact of music on ED patients,[22–25] no studies, to our knowledge, have specifically evaluated the effect of music listening on anxiety in older adults in the ED. Our aim was to perform a randomized pilot study comparing music listening plus standard care to standard care, with the goal of decreasing levels of anxiety among older adults as measured by the statetrait anxiety inventory (STAI).[26] The STAI is a validated Journal of Integrative Medicine
tool that measures state anxiety using a four-point forced choice Likert response scale (i.e., not at all, somewhat, moderately so, very much so). STAI scores range from 20 to 80, with 20 indicating mild to no anxiety while 80 indicates severe anxiety. 2 Methods 2.1 Study design and setting This was a randomized pilot study that took place in the geriatric ED of The Mount Sinai Hospital, an urban academic tertiary care medical center, during the months of April–May, 2015. The geriatric ED is a space connected to the main adult ED that is dedicated to providing an environment more conducive to treating patients aged > 65 years (e.g., softer lighting, reduced noise, multiple private rooms). Only those who are triaged with an emergency severity index (ESI) of 2, 3, 4 or 5 (1 to 5, 1 = urgent, 5 = nonurgent) are evaluated in the geriatric ED, while those with an ESI of 1 are seen in the main ED. This study received institutional review board approval from the Icahn School of Medicine at Mount Sinai. This study was not registered as a clinical trial as it did not meet applicable clinical trial guidelines under the Federal Drug and Administration Act. 2.2 Selection of participants Subjects were recruited from 8:00 a.m. to 8:00 p.m. Inclusion criteria included speaking English. Exclusion criteria included being deaf, contact isolation precautions (given the nature of the intervention) and prisoners. Patients unable to participate in the process of informed consent (e.g., due to delirium or severe dementia) were excluded. Also excluded were patients who had been given a disposition of discharge by their ED provider at the time of screening, since these patients were likely to leave the ED before completion of the study. Screening of eligible participants was done through chart review using an electronic ED tracking board accessed through the ED electronic health record, which lists patients alphabetically; all subjects deemed eligible through screening were then approached alphabetically. As this was a pilot study, power analysis/sample size calculation was not required. As many subjects were recruited as possible during the study period (2 months). 2.3 Intervention Intervention group participants received an electronic tablet with headphones and were instructed to listen to music for up to one hour. Participants could choose from five different genres of music: classical, jazz, new age, Latin guitar, or Chinese traditional. These genres had been selected in an attempt to offer participants a wide variety of music. All of the available selections contained only instrumental music since lyrics may be “distracting” or “emotional.”[27] Tempos of the music ranged between 60
451
November 2017, Vol. 15, No. 6
www.jcimjournal.com/jim
and 80 beats per minute, which have been identified in the literature as soothing and therapeutic tempos.[28] The volume level of the music was pre-set and was adjusted at the bedside according to the participant’s comfort level. Subjects were briefly instructed on how to use the tablet and change genres of music if desired. The tablets were cleaned with alcohol wipes between participants. The headphones were disposable. 2.4 Methods, measurements and outcomes After giving informed consent and prior to randomization, all participants were asked to verbally respond to a set of 20 self-report items from the STAI to establish pre-intervention anxiety scores. After completing the initial STAI assessment, participants were then randomized to either receive standard care plus the music intervention or to receive standard care only. Randomization was done at the bedside using sequentially numbered, concealed opaque envelopes. The randomization was performed with a 1:1 equal allocation ratio. Fifty envelopes had been randomly shuffled and then numbered sequentially at a remote location by the project coordinator who did not participate in patient recruitment. The envelopes were transferred to the ED and the investigator opened a consecutively numbered envelope containing the randomized group assignment for each participant. Following randomization, participants in the control group proceeded with the ED visit as usual, while participants in the intervention group received a tablet with headphones and were asked to listen to music for one hour. As little as 15 min of music listening has been shown to reduce anxiety,[28] however 1 h was chosen given the possibility that participants would be interrupted during the hour, whether by clinicians, staff or family members.[23] In fact, ED providers were not formally made aware of this study; thus, at any point during the 1 h, participants in either group may have interacted with ED clinicians as usual. After 1 h, subjects in both groups were asked to verbally respond to the same STAI questions as before to assess post-intervention anxiety levels. Given research personnel limitations, a single researcher performed the screening and enrollment and also conducted the intervention and surveys. Thus, it was not possible to provide a sham intervention. Patients who were approached were given a thorough explanation of the study and thus were aware of the study goals. Information was collected about patient characteristics including age, gender, race/ethnicity and ESI via retrospective chart review of the electronic health record (Epic ASAP, Epic Systems Corp, Verona, WI, USA). 2.5 Analysis Parametric distributions were assumed for selection of statistical tests of baseline characteristics. Differences November 2017, Vol. 15, No. 6
between study groups in baseline characteristics and clinical outcomes were assessed using Student t-tests for continuous variables and Chi-square tests for categorical variables and independent-samples. Outcomes were checked for normal distributions. All analyses were intention-to-treat and completed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA). 3 Results 3.1 Characteristics of study subjects A total of 1 904 patients were seen in the geriatric ED during the study months. During study hours, a total of 317 patients were screened for the study, 130 of whom were deemed eligible to participate. Of these patients, a total of 35 (27%) agreed to participate. Not feeling well enough to complete the surveys was a main reason for declining participation, as was not liking the choices of music. Of the 35 enrolled participants, 18 were randomized to the control group and 17 to the intervention group. Three subjects, 2 in the control group and 1 in the intervention group, withdrew from the study after the initial STAI survey. Reasons for withdrawal included not feeling well enough to continue participation and dissatisfaction with not being randomized to the intervention group. A total of 16 patients completed the study in each arm (Figure 1). There were no significant differences in baseline characteristics between the control and intervention arms (Table 1). 3.2 Treatment outcomes The intervention subjects had higher mean initial STAI scores (pre-intervention; 43.00 ± 15.00) than the control group (40.30 ± 12.80), but this difference was not statistically significant (P = 0.57). Follow-up mean STAI scores (post-intervention) were not significantly lower in the intervention group (31.60 ± 11.88) than in the control group (37.75 ± 12.61; P = 0.16). Comparison of mean reductions of the STAI scores, however, was statistically significant. The intervention group had a mean reduction of 10.00 ± 12.29, while the control group had a mean reduction of 1.88 ± 7.97 (P = 0.03; Table 2). 4 Discussion The results of this small, randomized pilot study suggest that listening to music may serve to decrease state anxiety in older adults (age > 65 years) seen in the ED. There was a statistically significant reduction of STAI scores in the music group compared to the control group. A caveat (and limitation) to these findings, as is the case with most pilot studies, is the study’s small sample size with low power and whether the results represent a true effect. Based on these results, we calculated that a sample size of at least
452
Journal of Integrative Medicine
www.jcimjournal.com/jim
50 participants, 25 in each arm, would detect a minimum effect size of 0.4 (differences in mean change in anxiety scores between the intervention and control group), with at least 80% power and α at 0.05. Although the clinical significance of our findings was not immediately clear, it is well-established that heightened anxiety can contribute to poor patient outcomes [5,11–13] and
pose barriers to patient care.[10,14] There are several ways in which music listening may affect anxiety levels among older adult patients in the ED. It may serve as a distraction and help patients regain a sense of control over their environment.[28] Especially when delivered via personal headphones, music can mask direct noise such as a patient talking in the next bed, as well as ambient noise such as
Screened for eligibility (N = 317) Excluded (n = 282) ·Did not meet inclusion criteria (n = 90) ·Met 1 or more exclusion criteria (n = 97) ·Declined to participate (n = 95) Randomized (n = 35)
Music (n = 17)
Control (n = 18)
Lost to follow-up (n = 2) Patient not feeling well (n = 1) Patient dissatisfied with group assignment (n = 1)
Lost to follow-up (n = 1) Patient not feeling well (n = 1)
Completed the study (n = 16)
Completed the study (n = 16)
Figure 1 Flow diagram of participant eligibility, enrollment, randomization and follow-up Table 1 Participant characteristics Characteristic
Control group (n = 18)
Music group (n = 17)
P value
Age (mean ± SD, years)
75 ± 7
73 ± 6
0.21
Female (n (%)) Race/ethnicity (n (%)) White Black Hispanic/other/unknown ESI score (mean ± SD)
13 (72)
13 (76)
8 (44) 9 (50) 1 (6) 2.8 ± 0.5
6 (35) 5 (30) 6 (35) 2.8 ± 0.7
0.77 0.08
0.96
Student t-test was used to analyze the differences of age and ESI score; Chi-square was used to analyze the differences of gender and race/ethnicity. SD: standard deviation; ESI: emergency severity index (1–5, 1 = urgent, 5 = non-urgent).
Table 2 Differences in state anxiety (intention-to-treat analysis) STAI score
Control group (n = 18)
Music group (n = 17)
P value
Baseline (pre-randomization)
40.30 ± 12.80
43.00 ± 15.00
0.57
Follow-up (post-randomization)
37.75 ± 12.61
31.60 ± 11.88
0.16
1.88 ± 7.97
10.00 ± 12.29
0.03
Reduction in STAI
Student t-test was used to analyze differences in state anxiety. Values are expressed as mean ± standard deviation. STAI: state-trait anxiety inventory.
Journal of Integrative Medicine
453
November 2017, Vol. 15, No. 6
www.jcimjournal.com/jim
machine alarms, ringing phones, or medical staff talking at the desk. Additionally, music listening can provide a sense of familiarity and comfort.[29] Based on effective music interventions in the literature,[28] the music used in this study was slow (60–80 beats/min) and non-lyrical. After controlling for these elements, participants were able to choose from a selection of genres since individual preference of, and inherent response to, different styles of music play a large role in whether a listener will find a certain type of music therapeutic.[30] In turn, music genre was not considered a confounder for which to control. The exact number of music-listening minutes also was not controlled for, as we found it difficult to monitor participants without risking a Hawthorne effect or altering standard care. Intervention participants did not report dissatisfaction with or adverse outcomes due to the music intervention. One intervention participant withdrew after completing the first STAI survey but before receiving the intervention, citing not feeling well enough to participate as her reason not to continue. Few obstacles, which were easily overcome, were revealed during the study. Some participants initially expressed hesitation in using a tablet, citing inexperience, but this fear was usually alleviated after the 30-second tutorial. A second obstacle, which some subjects cited as a reason for study decline, was dissatisfaction with the music selection. Although not addressed in this study, adding more genres is a simple solution. There are several limitations to this study, including the small sample size and low power. The sole researcher was not blinded to randomization allocation. This did not affect the approaching and enrollment of subjects nor the administration of the first STAI survey, since randomization was done after these steps, but it increased risk for expectation bias during administration of the second STAI survey. Likewise, with the absence of a sham intervention, subjects were not blinded, and their answers to the second STAI survey could have been subject to attention bias. Future studies on a larger scale could improve upon these problems by employing a sham intervention (e.g., a tablet and headphones without music or with white noise), as well as a second, blinded researcher to conduct the post-intervention STAI. Since the study took place at a single site and excluded all non-English speaking patients, the results are not generalizable. Additionally, the geriatric ED is designed to be less noisy than an average adult ED and, as such, may be more suitable for conducting a music intervention. Future studies should focus on incorporating non-English speaking patients in an average ED. Other considerations for future studies evaluating the effect of music on anxiety in older ED patients include November 2017, Vol. 15, No. 6
controlling for potential confounders such as medications received, tests or procedures performed, whether family or friends provide company at the bedside and whether subjects are seen by a care provider during the study hour. It would also be useful to determine a dose-response relationship between listening time and decreased anxiety levels (both duration and magnitude of the effect). This could be done by re-administering the STAI at regular intervals until anxiety levels return to pre-intervention levels or the patient is discharged from the ED. It could prove difficult to record the exact amount of time a patient listens to the music given the nature of the ED visit. In conclusion, preliminary data show that music listening may serve to reduce anxiety among older adults in the ED. Music listening is a noninvasive, safe, and easy-to-administer intervention to reduce anxiety among patients age > 65 years old being evaluated in the ED. Additional studies are warranted to evaluate this intervention on a larger scale. 5 Acknowledgements All significant contributors to this study have been named as authors. Laura Belland conceived of the study. Laura Belland, Ula Hwang, and Laura Rivera-Reyes designed the pilot. Laura Belland conducted the pilot study and data collection. Laura Belland undertook data management with the assistance of Laura Rivera-Reyes and Ula Hwang undertook statistical analyses. Laura Belland drafted the manuscript, and all authors contributed to its revision. Laura Belland takes responsibility of the paper as a whole. 6 Conflict of interests The authors have no conflicts of interest to declare. REFERENCES 1
2 3 4
5
454
Ekwall A, Gerdtz M, Manias E. Anxiety as a factor influencing satisfaction with emergency department care: perspectives of accompanying persons. J Clin Nurs. 2009; 18(24): 3489–3497. Tijunelis MA, Fitzsullivan E, Henderson SO. Noise in the ED. Am J Emerg Med. 2005; 23(3): 332–335. Byrne G, Heyman R. Patient anxiety in the accident and emergency department. J Clin Nurs. 1997; 6(4): 289–295. Body R, Kaide E, Kendal S, Foex B. Not all suffering is pain: sources of patients’ suffering in the emergency department call for improvements in communication from practitioners. Emerg Med J. 2015; 32(1): 15–20. Craven P, Cinar O, Madsen T. Patient anxiety may influence the efficacy of ED pain management. Am J Emerg Med. 2013; 31(2): 313–318.
Journal of Integrative Medicine
www.jcimjournal.com/jim 6 7 8
9
10 11
12
13 14
15 16 17 18
Baum SA, Rubenstein LZ. Old people in the emergency room: age-related differences in emergency department use and care. J Am Geriatr Soc. 1987; 35(5): 398–404. Hwang U, Richardson LD, Harris B, Morrison RS. The quality of emergency department pain care for older adult patients. J Am Geriatr Soc. 2010; 58(11): 2122–2128. Platts-Mills TF, Esserman DA, Brown DL, Bortsov AV, Sloane PD, McLean SA. Older US emergency department patients are less likely to receive pain medication than younger patients: results from a national survey. Ann Emerg Med. 2012; 60(2): 199–206. Hwang U, Belland LK, Handel DA, Yadav K, Heard K, Rivera-Reyes L, Eisenberg A, Noble MJ, Mekala S, Valley M, Winkel G, Todd KH, Morrison RS. Is all pain is treated equally? A multicenter evaluation of acute pain care by age. Pain. 2014; 155(12): 2568–2574. Stern TA, Fricchione GL, Rosenbaum JF. Massachusetts general hospital handbook of general hospital psychiatry. Amsterdam: Elsevier Health Sciences. 2010. Wong PS. Anxiety, signal anxiety, and unconscious anticipation: neuroscientific evidence for an unconscious signal function in humans. J Am Psychoanal Assoc. 1999; 47(3): 817–841. Schupp CJ, Berbaum K, Berbaum M, Lang EV. Pain and anxiety during interventional radiologic procedures: effect of patients’ state anxiety at baseline and modulation by nonpharmacologic analgesia adjuncts. J Vasc Interv Radiol. 2005; 16(12): 1585–1592. Jamison RN, Parris WC, Maxson WS. Psychological factors influencing recovery from outpatient surgery. Behav Res Ther. 1987; 25(1): 31–37. Musso MW, Perret JN, Sanders T, Daray R, Anderon K, Lancaster M, Lim D, Jones GN. Patients’ comprehension of their emergency department encounter: a pilot study using physician observers. Ann Emerg Med. 2015; 65(2): 151–155. Bradt J, Dileo C. Music interventions for mechanically ventilated patients. Cochrane Database Syst Rev. 2014; (12): CD006902. Bradt J, Dileo C, Shim M. Music interventions for preoperative anxiety. Cochrane Database Syst Rev. 2013; (6): CD006908. Bradt J, Dileo C, Potvin N. Music for stress and anxiety reduction in coronary heart disease patients (review). Cochrane Database Syst Rev. 2013; (12): CD006577. Bradt J, Dileo C, Magill L, Teague A. Music interventions for improving psychological and physical outcomes in cancer patients. Cochrane Database Syst Rev. 2016; (8):
Journal of Integrative Medicine
CD006911. 19 Raheem OA, Mirheydar HS, Lee HJ, Patel ND, Godebu E, Sakamoto K. Does listening to music during officebased flexible cystoscopy decrease anxiety in patients: a prospective randomized trial. J Endourol. 2015; 29(7): 791–796. 20 Reychler G, Mottart F, Boland M, Wasterlain E, Pieters T, Caty G, Liistro G. Influence of ambient music on perceived exertion during a pulmonary rehabilitation session: a randomized crossover study. Respir Care. 2015; 60(5): 711–717. 21 Sinha M, Christopher NC, Fenn R, Reeves L. Evaluation of nonpharmacologic methods of pain and anxiety management for laceration repair in the pediatric emergency department. Pediatr. 2006; 117(4): 1162–1168. 22 Parlar Kilic S, Karadag G, Oyucu S, Kale O, Zengin S, Ozdemir E, Korhan EA. Effect of music on pain, anxiety, and patient satisfaction in patients who present to the emergency department in Turkey. Jpn J Nurs Sci. 2015; 12(1): 44–53. 23 Short A, Ahern N. Evaluation of a systematic development process: relaxing music for the emergency department. Aust NZ J Med. 2009; 20: 1–27. 24 Weiland TJ, Jellnek GA, Macarow KE, Samartzis P, Brown DM, Grierson EM, Winter C. Original sound compositions reduce anxiety in emergency department patients: a randomised controlled trial. Med J Aust. 2011; 195(11–12): 694–698. 25 Menegazzi JJ, Paris PM, Kersteen CH, Flynn B, Trautman DE. A randomized, controlled trial of the use of music during laceration repair. Ann Emerg Med. 1991; 20(4): 348–350. 26 Spielberger CD, Gorsuch RL. Manual for the state-trait anxiety inventory (form Y): “self-evaluation questionnaire”. Washington, DC: Consulting Psychologists Press. 1983. 27 Krout RE. Music listening to facilitate relaxation and promote wellness: integrated aspects of our neurophysiological responses to music. Art Psychother. 2007; 34(2): 134–141. 28 Nilsson U. The anxiety- and pain-reducing effects of music interventions: a systematic review. AORN J. 2008; 87(4): 780–807. 29 Mok E, Wong KY. Effects of music on patient anxiety. AORN J. 2004; 77(2): 396–397, 401–406, 409–410. 30 Clair AA, Memmott J. Therapeutic uses of music with older adults. Silver Spring: American Music Therapy Association. 2008.
455
November 2017, Vol. 15, No. 6