Review Manager 5

Background Music therapy in end-of-life care aims to improve a person's quality of life by helping relieve symptoms, addressing psychological needs, offering support, facilitating communication, and meeting spiritual needs. In Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 2 addition, music therapists assist family and caregivers with coping, communication, and grief/bereavement. Objectives To examine effects of music therapy with standard care versus standard care alone or standard care combined with other therapies on psychological, physiological, and social responses in end-of-life care. Search methods We searched CENTRAL, MEDLINE, CINAHL, EMBASE, PSYCINFO, LILACS, CancerLit, Science Citation Index, www.musictherapyworld.de, CAIRSS for Music, Proquest Digital Dissertations, ClinicalTrials.gov, Current Controlled Trials, and the National Research Register to September 2009. We handsearched music therapy journals and reference lists, and contacted experts to identify unpublished manuscripts. There was no language restriction. Selection criteria We included all randomized and quasi-randomized controlled trials that compared music interventions and standard care with standard care alone or combined with other therapies in any care setting with a diagnosis of advanced life-limiting illness being treated with palliative intent and with a life expectancy of less than two years. Data collection and analysis Data were extracted, and methodological quality was assessed, independently by review authors. Additional information was sought from study authors when necessary. Results are presented using weighted mean differences for outcomes measured by the same scale and standardized mean differences for outcomes measured by different scales. Posttest scores were used. In cases of statistically significant baseline difference, we used change scores. Results Five studies (175 participants) were included. There is insufficient evidence of high quality to support the effect of music therapy on quality of life of people in end-of-life care. Given the limited number of studies and small sample sizes, more research is needed. No strong evidence was found for the effect of music therapy on pain or anxiety.These results were based on two small studies. There were insufficient data to examine the effect of music therapy on other physical, psychological, or social outcomes. Authors' conclusions A limited number of studies suggest there may be a benefit of music therapy on the quality of life of people in end-of-life care. However, the results stem from studies with a high risk of bias. More research is needed. Plain language summary Music therapy for end-of-life care Music therapy is increasingly used in end-of-life care, with a growing number of music therapists being employed in hospices and hospital-based palliative care programs each year. Music therapy in end-of-life care aims to improve a person's quality of life by helping relieve symptoms, addressing psychological needs, offering support and comfort, facilitating communication, and meeting spiritual needs. In addition, music therapists assist family and caregivers with coping, communication, and grief/bereavement. Music therapy requires the implementation of a music intervention by a trained music therapist, the presence of a therapeutic process, and the use of personally tailored music experiences. These music experiences may include listening to live, therapist-composed, improvised, or pre-recorded music, performing music on an instrument, Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 3 improvising music spontaneously using voice or instruments, composing music, and music combined with other modalities (e.g. movement, imagery, art). Results indicate that music therapy may have a beneficial effect on the quality of life of people in end-of-life care. However, the results stem from a limited number of studies and the quality of the evidence is not strong. More research is needed. No evidence of effect was found for pain or anxiety. This may be due to the fact that only two studies with very small samples examined the effects of music therapy on these outcomes. There were insufficient data to examine the effect of music therapy on other physical, psychological, or social outcomes. More research is needed. Background Music therapy is increasingly used in end-of-life care, with a growing number of music therapists being employed in hospices and hospital-based palliative care programs each year (Hilliard 2005). Data from a survey study of 300 randomly selected hospices in the U.S. indicated that the most popular forms of complementary therapies were massage therapy and music therapy (Demmer 2004). This is also true for use of complementary therapies in Canadian hospices, according to a recently completed survey (Oneschuk 2007). Music therapists in end-of-life care work with a broad range of populations with many types of illnesses including cancer (Hanser 2005; Hilliard 2003; Magill 2001), HIV/AIDS (Lee 1996; Neugebauer 1999), congestive heart failure (Dileo 2005c), dementia (Patrick 2005) and neurodegenerative disorders (Magee 2004 ; Scheiby 2005). The primary aim of music therapy in this context is to improve a person's quality of life by helping relieve symptoms, addressing psychological needs, offering support and comfort, facilitating communication, and meeting spiritual needs. In addition, music therapists assist family and caregivers with coping, communication, and grief/bereavement (Dileo 2005b). Research on the effects of music and music therapy in healthcare has grown rapidly during the past 20 years and has included a variety of outcome measures in a wide range of specialty areas including medical care, geriatric care, and rehabilitation (Dileo 2005a). It is important, however, to make a clear distinction between music interventions administered by medical or healthcare professionals (music medicine) and those implemented by trained music therapists (music therapy). A substantive set of data (Dileo 2005a) indicates that music therapy interventions are more effective than music medicine interventions for improving physiological as well psychological outcomes in medical patients. This difference might be attributed to the fact that music therapists individualize their interventions to meet patients' specific needs, more actively engage the patients in the music making, make use of the therapeutic relationship established with the patient to meet clinical goals and employ a systematic therapeutic process that includes assessment, treatment, and evaluation. As defined by Dileo 1999, interventions are categorized as 'music medicine' when passive listening to pre-recorded music is offered by medical personnel. In contrast, music therapy requires the implementation of a music intervention by a trained music therapist, the presence of a therapeutic process, and the use of personally tailored music experiences. These music experiences include: 1. listening to live, therapist-composed, improvised, or pre-recorded music; 2. performing music on an instrument; 3. improvising music spontaneously using voice or instruments, or both; 4. composing music; and 5. music combined with other modalities (e.g., movement, imagery, art) (Dileo 2007). In end-of-life care, receptive approaches, i.e. listening to live or pre-recorded music, are common due to the physical limitations of many patients. An example of a receptive intervention aimed at providing psychosocial support is song choice in which the patient selects a song according to specific criteria, e.g., how he or she is feeling (Dileo 2005b). For terminally ill patients, verbally expressing their emotions may be too difficult or threatening. In advanced stages of cancer, for example, speech impairments due to brain damage may prevent patients from verbally expressing their emotions, thoughts, and needs. Other patients may be hesitant to openly express their emotions because of the intensity of the feelings or the need to protect their loved ones. These patients may benefit from song choice as it gives them an "alternative, creative, and Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 4 non-threatening medium through which to experience and express their emotions" (Hogan 1999). If the patient is able to engage in music making, active music therapy methods such as songwriting, instrumental improvisation and vocal improvisations are used to improve sense of empowerment, enhance self-esteem, facilitate expression of ideas and emotions, increase socialization, facilitate creativity, and find meaning and hope (O'Callaghan 1997). Music listening, as well as active music making, is also used to help manage physical symptoms such as labored breathing, pain, agitation, and insomnia. Finally, music therapists play an important role in addressing the spiritual needs of patients as music can offer the "creative, lyrical, and symbolic means to address existential and spiritual needs during the process of dying" (Magill 2002, p. 996). Several research studies on the use of music in end-of-life care have reported positive results. For example, positive effects of music on pain, nausea/vomiting, anxiety, depression, mood and sense of well-being were reported in a meta-analysis combining studies conducted with cancer, terminally ill and AIDS patients (Dileo 2005a). However, this meta-analysis did not examine terminally ill patients as a separate group. In addition, differences in factors such as study designs, methods of interventions, and intensity of treatment have led to varying results. A systematic review is needed to more accurately gauge the efficacy of music therapy in end-of-life care as well as to identify variables that may moderate its effects. Objectives 1. To investigate the effectiveness of music therapy in end-of-life care. 2. To compare the effects of music therapy combined with standard care with: a. standard care alone, or b. standard care and other therapies. 3. To compare the effects of different types of music therapy interventions (e.g. music listening, songwriting, improvisation). Methods Criteria for considering studies for this review Types of studies We included all randomized controlled trials (RCTs), published or unpublished, in any language. Due to the limited number of studies that used proper methods of randomization, we also included studies with quasi-randomized or systematic methods of treatment allocation (for example, alternate allocation of treatments). Types of participants This review included participants in specialist palliative care or hospice settings (inpatient or community) or participants in any care setting with a diagnosis of advanced life-limiting illness being treated with palliative intent and with a life expectancy of less than two years (Hancock 2007). There were no restrictions as to age, gender, or ethnicity. Types of interventions This review included all studies in which standard care combined with music therapy was compared with: 1. standard care alone, or 2. standard care combined with other therapies. In addition, studies were considered only if: 1. music therapy was delivered by a formally trained music therapist or by trainees in a formal music therapy program; Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 5 2. a therapeutic process was present, and 3. one of the following personally tailored music therapy interventions was used in an individual or group setting: a. listening to live, therapist-composed, patient-composed, therapist and patient-composed, improvised, or pre-recorded music; b. performing music on an instrument; and c. improvising music spontaneously using voice or instruments, or both. Types of outcome measures The following outcome measures were included in this review: 1. symptom relief (e.g. of nausea, fatigue, pain); 2. psychological outcomes (anxiety, depression, fear); 3. physiological outcomes (e.g. respiratory rate, heart rate, IgA levels); 4. relationship and social support (e.g. family support, isolation); 5. communication (e.g. verbalization, facial affect, gestures); 6. quality of life; 7. spirituality; and 8. participant satisfaction. In addition, this review considered the following outcome measures for family members and caregivers: 1. psychological outcomes (e.g. depression, distress, coping, grief); 2. relationship and social support; 3. communication with participant; 4. quality of life. Search methods for identification of studies Electronic searches We searched the Cochrane Cancer Network Register and the Cochrane Pain, Palliative & Supportive Care Register. In addition, we searched the following electronic databases and trials registers: 1. Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, 2009, issue 3); 2. MEDLINE (1950 to September 2009); 3. EMBASE (1980 to October 2009); 4. CINAHL (1982 to September 2009); 5. PsycINFO (1967 to September 2009); 6. LILACS ( 1982 to September 2009); 7. CancerLit (1983 to September 2009); 8. CAIRSS for Music (retrieved on September 23 2009); 9. Proquest Digital Dissertations (1861 to September 2009); 10. ClinicalTrials.gov (www.clinicaltrials.gov) (retrieved on September 23 2009); 11. Current Controlled Trials (www.controlled-trials.com) (retrieved on September 23 2009); 12. National Research Register (www.update-software.com/National) (retrieved on September 23 2009); 13. www.musictherapyworld.de (retrieved on May 8 2009). Please see the Appendices (Appendix 1; Appendix 2; Appendix 3; Appendix 4; Appendix 5; Appendix 6; Appendix 7; Appendix 8; Appendix 9; Appendix 10; Appendix 11; Appendix 12; Appendix 13) for the search Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 6 strategies that were employed for each database. Searching other resources Handsearching We handsearched the following journals from their first available date: 1. Australian Journal of Music Therapy 2. Australian Music Therapy Association Bulletin 3. Canadian Journal of Music Therapy 4. The International Journal of the Arts in Medicine 5. Journal of Music Therapy 6. Musik-,Tanz-, und Kunsttherapie (Journal for Art Therapies in Education, Welfare and Health Care) 7. Musiktherapeutische Umschau 8. Music Therapy 9. Music Therapy Perspectives 10. Nordic Journal of Music Therapy 11. Music Therapy Today (online journal of music therapy) 12. Voices (online international journal of music therapy) 13. New Zealand Journal of Music Therapy 14. The Arts in Psychotherapy 15. British Journal of Music Therapy 16. Journal of Society for Integrative Oncology 17. Evidence Based Complementary and Alternative Medicine 18. Japanese Journal of Music Therapy In an effort to identify further published, unpublished and ongoing trials, we searched the bibliographies of relevant studies and reviews, contacted experts in the field, and searched available proceedings of music therapy conferences. Music therapy association websites were consulted to help identify music therapy practitioners and conference information (e.g., American Music Therapy Association (www.musictherapy.org), The British Society for Music Therapy (www.bsmt.org), The Association of Professional Music Therapists (APMT) (www.apmt.org), Music Therapy World (http://musictherapyworld.de)) . Data collection and analysis Selection of studies One review author (JB) conducted the electronic searches. One review author (JB) and a research assistant scanned the titles and abstracts of each record retrieved from the search. If information in the abstract clearly indicated that the study did not meet the inclusion criteria, we rejected the study. When a title or abstract could not be rejected with certainty, the full article was retrieved and the two review authors independently inspected the article. Both review authors used an inclusion criteria form to assess the study's eligibility for inclusion. If a study was excluded, we kept a record of both the article and the reason for its exclusion. Data extraction and management Both review authors independently extracted data from the selected studies using a standard coding form. We discussed any differences in the data extraction. The data that was extracted from the included studies is outlined in Additional Table 1. Where data was unavailable from the studies identified, we contacted the study author for clarification. Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 7 Assessment of risk of bias in included studies Both review authors, blinded to each other's assessment, assessed all included studies for quality, using the following criteria. 1. Method of randomization: was the study reported as randomized? Yes or no; was the method of randomization appropriate? Yes, no, or unclear. We rated randomization as appropriate if every participant had an equal chance to be selected for either intervention. We regarded the use of date of birth, date of admission, or alternation as inappropriate. 2. We used allocation concealment ratings of: A (adequate), B (unclear), and C (inadequate) in accordance with section 6.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2006). A: adequate, where methods to conceal allocation included 1. central randomization; 2. serially-numbered, opaque, sealed envelopes; or 3. other descriptions with convincing concealment. B: unclear, where authors did not adequately report on method of concealment. C: inadequate, where allocation was not adequately concealed (e.g., alternation methods were used). 3. Blinding: yes, no, or unclear With music therapy studies it is not possible to blind participants and those providing the music therapy interventions. However, outcome assessors can be blinded. In studies that used self-report measures, blinded outcome assessment was, of course, not possible. In this review, we marked blinding as yes, no, or unclear as it pertains to blinding of outcome assessors for objective outcomes. 4. Incomplete outcome data addressed: adequate, inadequate, unclear We gave a rating of adequate when numbers of dropouts and reasons for drop-out were reported or if we were able to obtain this information from the study author. If there were no withdrawals and this was indicated in the study, the study received a rating of adequate. The above four criteria were used to give each article an overall quality rating (based on section 6.7.1 of the Cochrane Handbook for Systematic Reviews of Interventions) (Higgins 2006). A. Low risk of bias: all four criteria met. B. Moderate risk of bias: one or more of the criteria only partly met. C. High risk of bias: one or more criteria not met. We did not exclude studies based on a low quality score. Dealing with missing data Data were analyzed on an endpoint basis, including only participants for whom final data point measurement was obtained (available case analysis). It was not assumed that participants who dropped out after randomization had a negative outcome. Assessment of heterogeneity Heterogeneity was investigated using the I-squared test with I 2 > 50% indicating significant heterogeneity. Assessment of reporting biases We had planned to examine publication bias visually in the form of funnel plots. However, this was not possible because of the limited number of studies per outcome. It needs to be noted though that three out of the five Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 8 studies are unpublished studies. Data synthesis JB entered the data of included studies into Review Manager (RevMan 2008). CD checked data entry for errors. We presented the main outcomes in this review as continuous variables. Where studies used different instruments to measure the same conceptual phenomenon (for example, quality of life) we reported the standardized mean difference (SMD) with 95% confidence intervals (CI). When there were sufficient data available from various studies using the same measurement instrument (for example, The Hospice Quality of Life Index-Revised) we computed a weighted mean difference (MD) with 95% CI. We calculated pooled estimates using the fixed-effect model. In case of significant heterogeneity (I-squared value > 50%), we used the random-effects model. We determined the levels of heterogeneity by I-squared (I 2 ) (Higgins 2002). The following treatment comparisons were made: 1. standard care and music therapy versus standard care alone; 2. standard care and music therapy versus standard care combined with other treatment. Subgroup analysis and investigation of heterogeneity The following sub-analyses were planned a priori as described by Deeks et al (Deeks 2001) and as recommended in section 8.8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2006), but could not be carried out because of an insufficient number of studies. These sub-analyses would have compared: 1. different types of music therapy interventions; 2. different duration and frequency of music therapy; 3. different diagnoses. Sensitivity analysis We had planned to examine the influence of study quality using a sensitivity analysis where the results including and excluding lower-quality studies are compared. Because all studies received a high risk of bias quality rating, we could not conduct the planned sensitivity analysis for impact of high risk studies. Results Description of studies Results of the search The database searches and handsearching of conference proceedings and journals resulted in 2964 citations. One review author (JB) and a research assistant examined the titles and abstracts, and 23 references were retrieved for possible inclusion. These were then independently screened by the two review authors resulting in five studies that met all the inclusion criteria. Where necessary we contacted chief investigators to obtain additional information on study details and data. Included studies Five studies with a total of a 175 participants (experimental and control) were included. One study provided music therapy in in-home hospice care (Hilliard 2003) and four studies conducted the music therapy sessions in an inpatient hospice setting (Horne-Thompson 2008; Lee 2005; Nguyen 2003; Wlodarczyk 2007). All study participants were adults with an average age of 68 years. The gender distribution in these studies was balanced with 51% female and 49% male participants. For three studies, ethnicity of the participants was not reported (Horne-Thompson 2008; Lee 2005; Nguyen 2003). For those studies that did report on ethnicity Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 9 (Hilliard 2003; Wlodarczyk 2007), the majority of the participants were Caucasian (average of 82.5%). Trial sample size ranged from 10 to 80 participants (see 'Characteristics of included studies' table for sample size of each study). One study provided services exclusively to terminal cancer patients in in-home hospice care (Hilliard 2003) whereas the other studies offered music therapy sessions to inpatient participants with various diagnoses including cancer, congestive heart failure, renal failure, AIDS, amyotrophic lateral sclerosis (ALS). Four studies used parallel group designs, whereas one study (Wlodarczyk 2007) used a cross-over design. Various music therapy interventions were used to meet the individual needs of the participants during the music therapy sessions: song choice, music-prompted reminiscence, singing, listening to live music, lyric analysis, instrument playing, song parody, singing with accompaniment using the Iso-principle, planning of funerals or memorial services, song gifts, music-assisted supportive counseling, music and relaxation, music and imagery, improvisation, songwriting, life review, sing-alongs with family and friends, and music for prayer. One study exclusively used live music based on the Iso-principle (Lee 2005). To establish a musical iso, music that matches the patient's current mood is played, after which the music is gradually changed in the therapeutic direction (e.g., a gradual change from music that has a lot of harmonic tension to music that sounds relaxing and peaceful). The music therapy interventions were aimed at developing a rapport with the patient or family, facilitating family interaction, providing support, enabling reminiscence, providing opportunities for spiritual exploration and validation, addressing feelings of anticipatory mourning and grief, and reducing anxiety and pain. The studies offered the following control conditions: standard care (Hilliard 2003; Nguyen 2003), a visit by a volunteer who engaged the participant in a conversation, read to the participant, or provided emotional support (Horne-Thompson 2008), a visit by the researcher to engage the participant in a conversation about a patient-preferred topic (Wlodarczyk 2007), and a music session where the participant listened to pre-recorded music without a music therapy process or interaction with a music therapist (Lee 2005). Two studies offered one music therapy session (Horne-Thompson 2008; Lee 2005), two studies offered two music therapy sessions (Nguyen 2003; Wlodarczyk 2007). In the Hilliard 2003 study, participants received a minimum of two sessions with some participants receiving up to 13 sessions. However, Hilliard 2003 only included data of the second session in his data analysis. Not all studies measured all outcomes identified for this review. The studies were conducted in two different countries: USA (Hilliard 2003; Lee 2005; Nguyen 2003; Wlodarczyk 2007) and Australia (Horne-Thompson 2008). Further details of the studies included in the review are shown in the table of 'Characteristics of included studies'. Excluded studies We identified 18 additional experimental research studies. However, these were excluded because of the following reasons: (a) no control group or control condition (Brown 2006; Calovini 1993; Kitawaki 2007; Krout 2001; Nakayama 2009; Segall 2007; Whittall 1989; Whittsitt 2006), (b) no random or pseudo-random (i.e. alternate group allocation) assignment to groups or conditions (Abbott 1995; Okamoto 2005; Patrick 2004), (c) ex post facto research without randomization (Gallagher 2006), and not an end-of-life care study (Martinez 2007; Mihara 2006). Four studies were excluded because the intervention was not a music therapy intervention (as defined by the authors in the background section) even though the study was conducted by a music therapist (Curtis 1986) or music therapy student (Choi 2007; Kerr 2004; Kim 2006). In these studies, the participants listen to pre-recorded music without the implementation of a therapeutic process. Details of the excluded trials are listed in the 'Characteristics of excluded studies' table. Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 10 Risk of bias in included studies We included studies that used appropriate methods of randomization (e.g. computer-generated table of random numbers, draw of lots, flip of coins) (Horne-Thompson 2008; Nguyen 2003) as well as studies that used alternate group assignment as allocation method (Lee 2005; Wlodarczyk 2007). One study stated that random assignment was used, but the randomization method was not specified. Attempts to receive this information from the author were unsuccessful (Hilliard 2003). Only one study used allocation concealment (Horne-Thompson 2008). All studies used self-report measures. Blinding of outcome assessors for these measures was, therefore, not possible. Two studies included additional physiological and physical outcomes, but blinding of outcome assessors did not take place (Lee 2005) or was unclear (Hilliard 2003). This inevitably introduced potential for biased assessment and, subsequently, overestimation of the effect size. Blinding of intervention allocation is not possible in music therapy interventions, adding another layer of possible bias. For two studies, the drop-out rate was unclear (Hilliard 2003; Nguyen 2003). Two studies reported a drop-out rate of less than 20% (Horne-Thompson 2008; Lee 2005). One study had a very high drop out rate (51%), but this was likely due to the fact that the data collection was spread over seven days. The authors reported that within that time frame, many of the participants passed away or became non-responsive. High drop-out rates are to be expected in end-of-life studies. As a result, all studies were rated to have a high risk of bias. Risk of bias is detailed for each study in the risk of bias tables included in the 'Characteristics of included studies' table. As all studies were rated as high risk, sensitivity analysis on the basis of overall quality rating was not carried out. Effects of interventions Primary outcomes Symptom relief Two studies (Horne-Thompson 2008; Nguyen 2003) examined the effect of music therapy on pain in hospice patients. Their pooled estimate indicated no strong evidence of effect of music therapy (SMD = -0.33, 95% CI -0.92 to 0.26, P = 0.27) (Analysis 1.1). Lee 2005 compared the effects of live music therapy with listening to pre-recorded music on patients' self-reported pain (using a 10-point graphic rating scale) and found that live music therapy based on the Iso-principle (posttest M = 3.02, SD = 1.86) was more effective (P = 0.025) in reducing pain than the use of pre-recorded music (posttest M = 4.63, SD = 2.34). Horne-Thompson and Grocke (Horne-Thompson 2008) investigated the effects of music therapy on other symptomatic issues common for patients in end-of-life care. Participants were asked to rate the severity of their symptoms using the Edmonton Symptom Assessment System (ESAS) (0 to 10 scale for each symptom). The authors reported that music therapy was effective in reducing tiredness (P = 0.024) and drowsiness (P = 0.018), but not in improving nausea (P = 0.2), appetite (P = 0.09), and shortness of breath (P = 0.07). One study (Nguyen 2003) found a statistically significant difference (P = 0.006) between post-intervention levels of discomfort (as measured on a 100 mm visual analogue scale (VAS)) in hospice patients receiving standard care and music therapy (n = 10, M = 11.1, SD = 14.34) and hospice patients receiving standard care alone (n = 10, M = 52.1, SD = 41.82). Finally, Hilliard 2003 included physical status as measured by the Palliative Performance Scale (PPS), but found no statistically significant difference between the music therapy group (M = 35.8, SD = 13.7) and the standard care group (M = 32.5, SD = 16.9). The PPS is a nursing assessment scale which includes measurement of the patient's conscious level, nutritive intake, self-care, ambulation, activity, and evidence of disease, with 0% indicating death and 100% indicating full functioning and no evidence of disease Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 11 (Hilliard 2003). Psychological outcomes Two studies included anxiety as an outcome measure and had a homogeneous SMD of -0.31 (95% CI -0.90 to 0.28, P = 0.30) (Analysis 1.2). However, this effect was not statistically significant. Two studies (Horne-Thompson 2008; Nguyen 2003) reported on the effect of music therapy on depression or sadness in participants who were terminally ill. Their data resulted in a homogeneous SMD of -.51 (95% CI -1.11 to 0.10, P = 0.10), but this was not statistically significant (Analysis 1.3). Physiological outcomes Only two studies included a physiological outcome, namely pulse rate. However, because one study used a volunteer visit as the control condition (Horne-Thompson 2008), and the other study used listening to prerecorded music as the control condition (Lee 2005), their estimates could not be pooled. Neither of the studies found a statistically significant effect of music therapy on pulse rate. Relationship and social support None of the studies included relationship and social support outcomes such as family support or isolation. Two studies used the Hospice Quality of Life Index Revised (HQLI-R) which includes a social/spiritual well-being subscale, but social well-being was not considered separate from spiritual well-being. Results of these two studies are discussed below under the quality of life outcome. Communication We did not find any studies that examined the effect of music therapy on communication variables such as verbalization, facial affect, and gestures. Quality of life The pooled effect of two studies (Hilliard 2003; Nguyen 2003) indicated that music therapy has a beneficial effect (MD = 37.08, 95% CI 22.45 to 51.72, P < 0.00001) on quality of life in patients in end-of-life care and results were consistent between the two studies (I2 = 0%). Both studies used the Hospice Quality of Life Index-Revised (HQLI-R). Horne-Thompson and Grocke (Horne-Thompson 2008) used the Edmonton Symptom Assessment Scale (ESAS) to examine the effects of music therapy on several symptomatic issues in palliative care patients, including well-being (measured on a 0-10 scale); they did not find statistically significant differences between the music therapy group and the control group. Pooling the results of their study with the Hilliard 2003 and Nguyen 2003 study resulted in a moderate but heterogeneous effect size (SMD = 0.69, 95% CI 0.11 to 1.27, P = 0.02; I2 = 51%) (Analysis 1.4). The HQLI-R is a self-report questionnaire using a Likert scale of 0 to 10 with a total of 28 items designed for hospice patients with cancer. It includes three subscales: functional well-being (i.e. daily activities, concentration ability) (total possible score of 70), psychophysiological well-being (i.e. anger, pain, nausea) (total possible score of 130), and social/spiritual well-being (i.e. meaning in life, family support, relationship with God) (total possible score of 80). The results of two studies (Hilliard 2003; Nguyen 2003) suggest that music therapy has a positive effect on psychophysiological well-being (MD = 17.41, 95% CI 9.10 to 25.72, P < 0.0001; I2 = 0%) (Analysis 1.6), functional well-being (MD = 13.40, 95% 7.25 to 19.54, P < 0.0001; I2 = 0%) (Analysis 1.5), and social/spiritual well-being (MD = 6.02, 95% 1.67 to 10.37, P = 0.007; I2 = 0%) (Analysis 1.7). In one study (Hilliard 2003), it was found that even though the physical status of participants receiving music therapy declined over time, as indicated by their scores on the Palliative Performance Scale, their quality of life scores increased. In the standard care group, both physical status and quality of life declined over time. Hilliard 2003 also included length of life as an outcome variable. The average length of life (in days) for the music Music therapy for end-of-life care 02-Nov-2009 Review Manager 5 12 therapy participants was 69.5 (SD = 60.5) days and 57.8 (SD = 45.5) days for the control group. More studies are needed to examine the impact of music therapy interventions on length of life. Spirituality One study (Wlodarczyk 2007) considered spirituality as an outcome. Even though two other studies reported results on social/spiritual well-being on the HQLI-R, this subscale did not separate spiritual well-being from social well-being and, therefore, their results could not be pooled with the Wlodarczyk study. Wlodarczyk 2007 reported that music therapy was significantly more effective (P = 0.01) than a non-music visit in enhancing spiritual well-being (M = 78.5, SD = 20.93 and M = 73.95, SD = 20.76, respectively) in hospice patients. Participant satisfaction None of the studies compared satisfaction of experimental group participants with satisfaction of control group participants. One study included a family satisfaction questionnaire for the music therapy participants. In this questionnaire, participants were asked how beneficial music therapy was for them and their loved ones. The following ratings (10-point scale) were obtained: (a) The use of music therapy within end-of-life celebration is beneficial for me: M = 9.4, SD = 0.97, (b) The use of music therapy within end-of-life celebration is beneficial for my loved one: M = 9.7, SD = 0.48, (c) I enjoy the use of music therapy within the hospital setting: M = 9.7, SD = 4.2, (d) I believe that music therapy with an end-of-life celebration has or will bring closure for me: M = 8.9, SD = 1.55, and (e) I believe that music therapy with an end-of-life celebration has or will bring closure for my loved ones: M = 8.6, SD = 1.7.