Skip to main content

Effect of Music Therapy on Pain Perception, Anxiety, and Opioid Use During Dressing Change Among Patients With Burns in India: A Quasi-experimental, Cross-over Pilot Study

Empirical Studies

Effect of Music Therapy on Pain Perception, Anxiety, and Opioid Use During Dressing Change Among Patients With Burns in India: A Quasi-experimental, Cross-over Pilot Study

Index: Ostomy Wound Management 2018;64(10):40-46. doi: 10.25270/owm.2018.10.4046


The effectiveness of music therapy for relieving pain and anxiety during burn dressing changes has not been reported from India. Purpose: This study was conducted to assess the effect of music therapy on pain, anxiety, opioid use, and hemodynamic variables during burn dressing change.Methods: Patients in a tertiary care burn unit who were >10 years old, conscious, able to respond, and oriented to time, place, and person participated in a 2-month, quasi-experimental, cross-over pilot study. Each served as his/her own control. Dressings were changed every other day alternating between the control (standard pain management) and experimental (control plus patient-selected music) intervention. Pain was assessed using a numerical rating scale, anxiety was scored using the State Trait Anxiety Test (higher scores indicated more pain and anxiety), and hemodynamic parameters and analgesics were recorded. Wilcoxon Test and chi-squared tests were utilized for statistical analysis. Results: Median pain scores (5, interquartile range [IQR] IQR: 3-7; and 6, IQR: 5-8) and median anxiety scores (12, IQR: 8-17; and 14, IQR: 10-19) were significantly lower during the experimental than during the standard dressing change, respectively (P <.001), and opioids were used significantly less frequently during the experimental change (P = .002). Conclusion: Music therapy helps reduce anxiety, pain, and opioid use during burn dressing change. 


Study location. The study was conducted in an 8-bed burn unit of a tertiary level hospital (Post Graduate Institute of Medical Education and Research, Chandigarh, North India) during July and August 2015. 

Study design. A quasi-experimental, cross-over (repeated measures) design was employed. Every burn dressing change was considered a study data point. In order to ensure complete homogeneity of the control and experiment group, each patient undergoing dressing changes served as his/her own control. All the patients irrespective of their post burn day were identified and screened for eligibility using the inclusion and exclusion criteria. The enrolled patients had a control dressing change on the first day of assessment. Following the cross-over design, the same patients went through an experimental dressing change on the third day (ie, during their next dressing change), and this cycle was repeated until discharge. Each patient was studied for approximately 10 days. The washout period was an important consideration in the cross-over design; dressing changes were performed at least 48 hours apart to reduce recall bias among the patients. 

Participants. Using the total enumeration technique, all the dressing changes performed in the burn unit during the study period were assessed. Inclusion criteria stipulated study participants must be burn patients >10 years old, conscious, able to respond, and oriented to time, place, and person. Burn patients on ventilator support, hearing impaired, and/or declared hemodynamically unstable by the attending plastic surgeon were excluded. In addition, patients had the choice of opting out of the study at any time if they were not comfortable with the intervention. Had this occurred, they would have been reported as drop outs.

Conceptual framework. The conceptual framework for this research plan was based on Melzack and Wall’s Gate Control Theory of Pain.21 This theory proposes that pain can be reduced by stimulating sensory neurons, which are larger and faster than pain neurons and effectively close the gate-carrying pain stimuli. Cognitive processes such as distraction reduce pain perception by consuming attention and stimulating the sensory neurons.12

Study tools. The primary outcome variables were pain, anxiety, and opioid use during the burn dressing change. Secondary outcome variables were hemodynamic parameters. An interview/assessment schedule was prepared to obtain patient sociodemographic (age, gender, education, occupation, and marital status) and burn profile (etiology and degree of burn, total burn surface area) data; these data were validated by experts from nursing education and clinicians from the department of plastic surgery of the authors’ institute. Using a NRS,22 a standardized ordinal level scale was used to measure pain perception. Patients rated their pain from 0 to 10, where 0 means no pain and 10 means extremely intolerable pain. The Indian version of the State-Trait Anxiety Test (STAT), developed by Vohra,23 of the STAI for adults developed by Speilberger et al,12 was used to assess patient anxiety levels 30 minutes before dressing changes. The STAT is a standardized tool with established reliability and validity.23 Forty (40) items (20 measuring state anxiety and 20 measuring trait anxiety) are used to calculate the score. State anxiety is defined as an emotional state that exists at a given moment in time (ie, how the respondent feels “right now, at this moment,” in this case immediately before the dressing change). Trait anxiety is described as the likelihood of the person to experience anxiety when perceiving a stressful situation as dangeous or threatening.12 Total scores for state and trait sections separately range from 20 to 80, with higher scores denoting higher levels of anxiety.23

Care protocol. According to routine protocol of the unit, dressings are changed on patients in the burn unit every other day. Patients are bathed with water and a betadine scrub, and the dressing change is performed by the attending plastic surgeon. The wounds are dried using sterile dressing pads (gauze dressing wrapped over cotton). Antibiotic ointment (silver sulfadiazine or colloidal silver) then is applied over the wounds. The wound areas then are covered with a double layer of sterile dressing pads (12 inch x 12 inch) and the dressing is secured using sterile (6-inch) bandages. The routine pain management in the unit included pharmacological interventions prescribed by the physician and administered to the patients as needed. 

The burn unit’s analgesic routine was not manipulated during the study. The drug name, dosage, and route of the analgesics administered to the patients by the burn unit team during the dressing change were recorded per se. A clinical assessment proforma was developed to collect data about hemodynamic parameters (blood pressure, pulse rate, respiration, and temperature) using the cardiac monitors of the unit. All of the above study tools were paper/pencil instruments, and the questions were asked by the investigator who completed the forms. 

Music therapy protocol. A protocol for music therapy during burn dressing change was prepared on the basis of a review of literature and suggestions from medical experts. As advocated in literature, nonlyrical instrumental music pieces8,17 were included in the therapy under 4 classes: Spiritual/religious (Bhajans and Sikhism shabad), Western instrumental (guitar, piano), Classical instrumental (tabla, flute, sitar), and Bollywood instrumental (ajeeb dastan hai ye, khamoshiyan). The 15 music pieces were assessed by 20 nursing research experts. The experts were asked individually to rate each music piece by giving it a score out of 10 for suitability to be used in music therapy in the patient population (1 = unsuitable and 10 = most suitable). The scores for each music piece were totalled and the music pieces with the lowest scores were eliminated, leaving 10 instrumental music pieces. The patient was given an opportunity to listen to all the music pieces on the first day of therapy; the music selection made by the patient was used every time s/he received music therapy during the experimental dressing change. 

The music was played using MP3 players and earphones at the bedside of the patient 30 minutes before and for 30 minutes after each dressing change. Ear phones were used to help the patient focus on the music and to remove distracting sounds of the burn unit (eg, cardiac monitors or people talking). The MP3 players and earphones were kept separate for each patient and were cleaned with alcohol swabs before and after use to prevent infections. Ear phones were used at bedside, and loudspeakers were used where dressings were changed. During control dressing changes, patients received standard care as per the unit’s protocol.

Ethical considerations. The research protocol was approved by the Ethics Review Committee of the institute. Informed written consent was obtained from all participants after explaining the research, objectives, and duration of the study. All the participants were informed about their right to refuse to participate or withdraw from the study at any time. Patient anonymity and confidentiality were maintained while collecting data and reporting the study by restricting data access to the core investigators. Nothing interfered with the routine treatment of the participants. Care was taken not to cause any harm or discomfort to them. The STAT was purchased with legal rights of use. The study was registered with the Clinical Trials Registry of India (Registration number: CTRI/2016/09/007281).

Data collection and analysis. The sociodemographic (gender, age, employment status, marital status, level of education), burn profile (type of burn, percent of body affected; ie, total burn surface area [TBSA]), and hemodynamic variable (blood pressure, heart rate, respiration, and temperature) data were collected by interview and clinical assessment. Using the NRS, patient pain scores were recorded 3 times: 30 minutes before, during, and 30 minutes after the dressing change. To obtain the pain score during dressing change, patients were asked to rate their overall pain during the procedure after the dressing change was over; this score was analyzed using median and interquartile range (IQR) data. Degree of anxiety was determined 30 minutes before burn dressing change as per the total score obtained on STAT. Analgesic use was recorded and classified into 2 categories for purposes of analysis: opioids (eg, tramadol, morphine) and nonopioids (eg, paracetamol and diclofenac sodium). The hemodynamic parameters were assessed at 3 instances: 30 minutes before, during, and 30 minutes after dressing changes. 

The data collected were directly entered into the Statistical Package for Social Sciences (SPSS for Windows, version 16.0, Chicago, IL) for statistical analysis. Descriptive analysis was used for the sociodemographic and burn profile data to assess frequency and percentage. Because the data did not follow normal distribution, nonparametric tests were used, with a level of significance of 0.05. The median values of pain, anxiety, and hemodynamic parameters of the control and experimental groups were compared using Wilcoxon test. The frequency of opioid use between the control and experimental dressing change was compared using chi-squared test. Post hoc analysis of the power of the study was performed using the “Gpower” software, version 3.0.10 (Universität, Dusseldorf, Germany). With an effect size of 0.5, with the given sample size of 52, a power of 0.95 was achieved.


Demographic and burn characteristics. During the data collection period, 25 patients were treated in the authors’ unit. Of these, 10 were eligible for the study according to the inclusion and exclusion criteria. Participants included 8 men, 2 women, mean age 28.5 ± 13.7 (range 14–51) years; 6 were educated up to the secondary level, 3 up to primary level, and 1 was illiterate. Five (5) were unemployed, 6 were married, 7 had electric burns, and 3 had second-degree flame burn injury. The mean TBSA was 27% ± 19.0% (range 5%–60%). 

A total of 104 dressing changes (52 control, 52 experimental) were completed as study data points. The mean number of dressing changes of each participant was 10 ± 6.09 (range 2–18), 5 as part of the experiment (with music therapy) and 5 as control (without music therapy). The majority of the participants (7) chose spiritual music (gayatri mantra and shabad) for the therapy.

Procedural pain and anxiety outcomes. The overall pain score (median, IQR) for the experimental dressing changes (3, range 1–5) was significantly lower than for the control dressing changes (4.5, range 2–6) (P <.001) (see Table 1). Median state anxiety scores before the experimental dressing changes (12, IQR: 8-17) were significantly lower as compared to the control dressing changes (14, IQR: 10-19) (P <.001). The trait anxiety scores were not significantly different during the control and experimental dressing changes (see Table 2). 

Analgesic use outcomes. Most patients required analgesics during the dressing change. Of the total 104 dressing changes, an analgesic was provided 38 times in control group and 24 times in the experimental group. The frequency of opioid use was significantly lower during the experimental dressing change (9.6%) as compared to the control dressing change (34.6%), (P = .002) (see Table 3). Interestingly, the overall frequency of analgesic use was not reduced, but the type of medication changed. Patients were given weaker analgesics such as paracetamol more often during the experimental dressing change pertinent to the lower levels of pain perception. In all the patients, morphine was charted as 3 mg intravenously as needed. Thus, the objective to reduce the opioid use (as-needed use of morphine) was successfully achieved. 

Hemodynamic parameters outcomes. None of the hemodynamic parameters (blood pressure, heart rate, respiratory rate, temperature) was significantly different between the control and experiment groups (see Table 4). 


Since time immemorial, music has been widely used to enhance well-being, decrease emotional difficulties, and distract patients from symptoms such as pain.24 Despite this fact, no studies have been conducted in India that explore the use of music therapy to relieve patient pain and anxiety during burn wound dressing changes. In the current study, every patient served as his or her own control. This helped ensure the control and experimental dressing changes had total homogeneity in terms of participant characteristics. As previously described, Fratianne et al14 employed a similar repeated measures design with patients serving as their own control to test the efficacy of music-based imagery and musical engagement to manage pain and anxiety during burn debridement. A quasi-experimental pilot study was conducted by Kahar et al25 in Singapore to assess whether music helps reduce the pain experienced by burn patients during dressing changes. Using convenience sampling, 30 dressing changes were alternatively assigned to control (without listening to music) and experimental (listening to music) groups. The authors used the NRS, a pain behavioral tool, and physiological monitoring. The music was soothing and distracting, but its effects were not found to be statistically significant. The authors concluded that further high-quality studies are needed to evaluate the impact of music. Also, they did not allow patients to select the type of music; they acknowledged that allowing the patient to select the music may have enhanced the study. Thus, the current study was built on the work of Kahar et al25 by using patient-selected music. Self-selected music also was used in the Son and Kim11 and Wang et al19 studies. In the systematic review by Nilsson,18 29 out of 42 studies used self-selected prerecorded music. However, researchers also have used live music13,17 and participatory music interventions such as playing instruments or singing,14,16 showing music therapy has a wide scope in terms of use and practice. 

Pain intensity was measured using a NRS because it frequently has been reported to be a useful tool.17,18,22,26 The current study is in consonance with many previous studies in documenting that music has a significant role in reducing the intensity of pain among burn patients during dressing changes.10,11,14,15 However, Whitehead-Pleaux et al13 reported that music therapy for pediatric patients during donor site dressing changes generated mixed results; therefore, the study was statistically inconclusive, suggesting more studies in pediatric burn patient populations are needed. The current study confirms that music therapy is effective in reducing pain, with results similar to studies exploring music in other patient populations besides burns.17,18 

The STAT was used because it is a widely accepted, reliable, and valid tool.23 A significant reduction was noted in the median state anxiety scores before the experimental dressing changes as compared to the control dressing changes (P <.001). These findings are in line with the Hsu et al10 and Son and Kim11 studies that found significant reduction in anxiety scores of burn patients when music-related interventions were employed. Similar to the current findings, studies involving other patient populations also have found music to be useful in relieving anxiety.16,19 However, the Whitehead-Pleaux et al13 and Fratianne et al14 studies differ; they both reported an improvement in the self-report of anxiety scores of the burn patients, but the difference was not significant. The Wang et al19 study utilized physiological measures of anxiety (eg, heart rate, blood pressure, electrodermal activity, serum cortisol levels) along with STAI. Further research using these physiological measures in burn patients is warranted.

In the present study, the frequency of opioid use was found to be significantly lower in the experiment group (P = .002). Because literature is scant regarding the effect of music therapy on opioid use among burn patients, further exploration to generate satisfactory evidence is warranted. None of the physiological parameters such as blood pressure, pulse, respiratory rate, and temperature — assessed in this study as secondary outcomes because they have been reported to be indirect measures of anxiety and pain19 — was significantly different between the control and experiment groups. Wang et al19 had similar findings, with no significant difference found in heart rate and blood pressure values (P = .5) between persons who did/did not have music therapy. However, Almerud and Petersson27 reported a significant decrease in both systolic and diastolic blood pressure during music therapy sessions among intensive care unit patients temporarily on a respirator.


Burn injury results from excessive exposure of tissue to thermal, chemical, electrical, or radioactive agents.1,2 According to a relevant textbook,3 pain is integral to survival. Although no research describes the proportion of burn patients reporting inadequate pain relief, pain relief measures have been anecdotally reported to be insufficient in both adult4-6 and pediatric7 burn populations, mainly due to fear of side effects and addiction to opioids, lack of routine pain evaluations, and standardized analgesic protocols.4 In their review article, Patterson et al5 recommended a structured approach to burn analgesia that includes performing routine pain assessment and incorporating both drugs and individualized alternative therapies based on psychological, cognitive behavioral, and operant learning techniques.8 A variety of psychological techniques such as hypnosis, relaxation, and music therapy may be helpful in managing burn pain by providing multimodal distraction.

An exploratory study9 showed the effect of music therapy depends upon not only the type of music, but also upon the associations and memories of the music. Various studies performed in countries other than India have documented a positive effect of listening to music on burn patients’ pain and anxiety during dressing change. Hsu et al10 conducted a prospective, randomized controlled trial (RCT) (N = 70) to assess the impact of music on burn patients’ pain and anxiety at the time of dressing change. Patients were randomly assigned to 2 groups; the control group was provided standard interventions (ie, routine dressing change) and the experimental group listened to music during the routine dressing change. Pain and anxiety were assessed before, during, and after dressing change. Morphine use in both groups was recorded. A significant reduction in pain and anxiety was reported by the fourth day in the experimental group (P <.05), although morphine dosage remained the same in both groups. Similarly, in a quasi-experimental pretest-posttest design study by Son and Kim,11 32 patients were assigned into control (routine burn dressing changes) and experimental (listening to self-selected music during dressing changes) groups for 3 days. Pain and anxiety scores were self-reported using the State-Trait Anxiety Inventory (STAI)12; significant reductions were noted in both anxiety (P <.017) and pain scores (P <.012) before and after the dressing changes when patients listened to music in the experimental group as compared to the control group. 

Whitehead-Pleaux et al13 conducted a similar RCT study on pediatric patients. Fourteen (14) patients were randomly assigned to control (verbal interaction) and experimental groups (listening to live music). The participants of the experimental group anecdotally reported positive effects of music on pain and anxiety, although the difference in the pain and anxiety scores was not statistically significant between the experimental and control groups. Fratianne et al14 conducted a repeated measures study among 25 patients (age 7 years and older) serving as their own control to test the efficacy of music-based imagery and musical engagement (ie, activities such as listening to music, identifying or singing familiar songs, practicing deep breathing according to music rhythm, and playing simple musical instruments) in relieving pain and anxiety among burn patients during the debridement process. A significant reduction in self-reported pain was noted among persons who received music therapy versus those who did not (P <.03). Anxiety scores also improved, but the difference of the anxiety scores between the 2 groups was not statistically significant. 

A systematic review and meta-analysis15 of 17 RCTs (804 patients) on the effects of music intervention on burn patients found a significant positive effect of music intervention on pain alleviation (standard mean differences [SMD] = -1.26; 95% confidence interval [CI] -1.83 to -0.68), anxiety relief (SMD = -1.22; 95% CI -1.75 to -0.69), and heart rate reduction (SMD = -0.60; 95% CI [-0.84 to -0.36]). In a systematic review of 26 studies,6 17 showed positive outcomes when nonpharmacological interventions were used to manage procedural pain among adult burn patients. 

Music therapy also has been documented to have a positive effect on pain and anxiety in patient populations other than burns. A repeated measures, pretest–posttest design study (N = 36 men) performed by Hwang and Oh16 compared 3 different types of music therapy (singing, listening to music, and playing instruments for 30 minutes, twice a week for 6 weeks) on levels of depression, anxiety, anger, and stress among alcohol-dependent clients. The authors reported participant anxiety scores were significantly reduced with music therapy (P <.05). Gutgsell et al17 conducted a RCT (N = 200) to assess the efficacy of a single music therapy session to reduce pain in palliative care patients. A significant decrease (P <.0001) was noted in the numeric rating scale (NRS) pain scores in the music therapy group.

Nilsson18 conducted a systematic review of 42 RCTs published from 1995 to 2007 on the effects of music therapy on pain and anxiety in the perioperative period among patients undergoing elective surgery. The type of music employed was soothing (ie, not more than 60 to 80 beats per minute); in 29 studies, a patient self-selected music. In 19 studies, the STAI12 was used to assess anxiety. In the 22 studies that evaluated the effect of music on pain, a visual analogue scale was most commonly used (n = 12); other instruments used to measure pain included a NRS, the McGill Pain Questionnaire, and a verbal rating scale. The results showed (as demonstrated in approximately ~50% of the outcomes) music intervention had a positive effect on reducing patient anxiety and pain in the perioperative setting. 

A RCT by Wang et al19 evaluated the effect of music therapy on preoperative anxiety. The experiment group (n = 48) listened to 30 minutes of patient-selected music 30 minutes before the surgery and self-reported anxiety using the STAI; physiological measures of anxiety included heart rate and blood pressure. Results showed music therapy reduced anxiety levels of the patients in the experimental group by 16% compared with their pre-interventional level. The reduction in the anxiety scores in the experimental group as compared to the control group was also statistically significant (P =.001). In a review of the literature by Henry,20 music therapy was found effective in decreasing pain and anxiety related to acute illness, injury, and painful procedures among critical care patients.

The current authors noted a lack of clinical evidence regarding use of music therapy among burn patients in India and conducted a study to assess the effect of music therapy on pain, anxiety, and opioid use among patients admitted in a tertiary care hospital in northern India.


Lack of blinding (observer and patient) is a major concern. A larger number of patients is needed to increase the power of the study and determine if music truly has an effect on pain and anxiety. The small tertiary care unit provided services to patients of all ages (including neonates) from all over northern India; the majority of these patients have severe burn injuries and are either on inotropic or ventilator support. These patients were not included in this study because their anxiety and pain perception is not easily measurable, limiting the sample size of this pilot study. Further research in this dimension is under consideration, including a multicenter study.


Pain management in burn patients, which has both physiological and psychological outcomes,<sup>28</sup> is an ever-growing challenge faced by the burn care team. Music therapy can help reduce the level of pain, anxiety, and opioid use during dressing change among burn patients. Similar studies can be conducted in different research settings and different patient populations to include a larger sample size, and the effect of instrumental and lyrical music types can be compared. Future research would benefit from blinded observer assessment of pain and anxiety. Also, self-report of pain and anxiety can be compared with objective assessment by an observer. Generating further evidence with a larger sample size to support current and historic findings and facilitate the development of an evidence-based standard music therapy protocol is recommended.