Navigation

Hilliard Family Medicine

The purpose of this study was to evaluate the effects of music therapy on quality of life, length of life in care, physical status, and relationship of death occurrence to the final music therapy interventions of hospice patients diagnosed with terminal cancer. Subjects were adults who were living in their homes, receiving hospice care, and were diagnosed with terminal cancer. A total of 80 subjects participated in the study and were randomly assigned to one of two groups: experimental (routine hospice services and clinical music therapy) and control (routine hospice services only). Groups were matched on the basis of gender and age. Quality of life was measured by the Hospice Quality of Life Index-Revised (HQOLI-R), a self-report measure given every visit. Functional status of the subjects was assessed by the hospice nurse during every visit using the Palliative Performance Scale. All subjects received at least two visits and quality of life and physical status assessments. A repeated measures ANOVA revealed a significant difference between groups on self-report quality of life scores for visits one and two. Quality of life was higher for those subjects receiving music therapy, and their quality of life increased over time as they received more music therapy sessions. Subjects in the control group, however, experienced a lower quality of life than those in the experimental group, and without music, their quality of life decreased over time. There were no significant differences in results by age or gender of subjects in either condition. Furthermore, there were no significant differences between groups on physical functioning, length of life, or time of death in relation to the last scheduled visit by the music therapist or counselor. This study provides an overview of hospice/palliative care, explains the role of music therapy in providing care, and establishes clinical guidelines grounded in research for the use of music therapy in improving the quality of life among the terminally ill.

Introduction and Review of Literature

Although music therapists have been working in palliative care and hospice settings, empirical research evaluating the efficacy of music therapy with the terminally ill is exiguous. Palliative care serves to treat symptoms and provide comfort measures in order to respect quality of life, and the curative approach in promoting length of life is forgone. Symptom management can be complex in palliative care as the whole person is considered. Distress in the form of physical, psychological, financial, or spiritual concerns can impact one's quality of life, and palliative care provides an interdisciplinary team to treat the various forms of distress among the dying. Because music therapy can treat one's mind, body, and spirit, it has been used as an important component within the interdisciplinary team.

The literature is rich with qualitative descriptions reporting the use of music therapy in the treatment of various problems of terminally ill clients. Music therapy programs and techniques are described in the literature, and anecdotal information is provided supporting the role of music therapy in promoting quality of life of people diagnosed with a terminal illness. Theoretical articles encourage the use of music therapy in palliative care as a form of emotional expression, a prompt for life-review and reminiscence, a source of spiritual and psychological support, a form of communication, a treatment for pain, and a source of grief support and expression (Gilbert, 1977; O'Callaghan, 1993, 1996; Porchet-Munro, 1993; Salmon, 2001; West, 1994). Program descriptions illustrate the pragmatic uses of music therapy in hospices and palliative care settings (Hogan, 1996; Mramor, 2001; Nakagami, 1997; Weber, 1996). Through case studies, it has been shown how music therapy has been used to treat a variety of needs exhibited by people with terminal illnesses (Beggs, 1991; Hilliard, 2001; Martin, 1991; Whittall, 1991). The use of Guided Imagery and Music with the terminally ill was explained through case studies (Skaggs, 1997; Wylie & Blom, 1986) and song-writing was used in other cases (O'Callaghan, 1995). Other qualitative approaches including phenomenological research (Forinash, 1990; Hogan, 1999) and modified grounded theory (O'Callaghan, 2001) were used in documenting the use of music therapy in hospice care.

There are few quantitative studies within the palliative care music therapy literature. Curtis (1986) evaluated the effects of music therapy on pain relief and relaxation of nine subjects to find no statistically significant differences among groups. In a pilot study, Whittall (1989) documented decreases in blood pressure and heart rate of eight terminally ill patients following music therapy interventions. Calovini's (1993) and Abbott's (1995) master's theses evaluated the effects of music therapy on the terminally ill, and although movement towards goals was made, significant differences were not found. Krout (2001) studied the effects of single-session music therapy sessions on pain, physical comfort, and relaxation of hospice patients of varying diagnoses to find significant differences on dependent variables. Quantitative research in palliative care music therapy is quite limited, and there is increasing demand in the field for efficacy reports. Palliative care administrators, physicians, and funding sources require empirical evidence in order to support the use of music therapy in hospices and other palliative care settings.

The purpose of this study was to evaluate the use of music therapy on quality of life of people diagnosed with terminal cancer residing in their homes receiving hospice care. Although quality of life is complex, multidimensional, subjective, and encompasses many aspects of one's experiences, it is an important issue at the end of life, especially for those who have selected palliative care over curative treatment (Donnelly & Walsh, 1996). Hospice and palliative care clinicians seek to maintain or increase their clients' quality of life at the end of life, and most clients report quality of life to be an important personal issue.

Questions arise regarding the impact music therapy has on the quality of life among the terminally ill and which dependent measures document this. Specific questions of the current study are: Is there a difference on quality of life scores between hospice subjects who received clinical music therapy and those who did not, specifically Hospice Quality of Life Index-Revised self-report and Palliative Performance Scale nursing assessments? How does quality of life change over the course of end of life with and without multiple treatment sessions of music therapy? Does physical status (continuous physical decline) correlate with self-assessed quality of life? Are there differences in the length of life (hospice admission until time of death) among those who receive music therapy versus those who do not? Is there a relationship between time of death and the last professional visit by discipline, namely, family support counselor or music therapist. Techniques used by music therapists and clinical care plan needs are analyzed. Although people face terminal illnesses without the assistance of hospice or other medical institutions, this study is limited to people who have been diagnosed with terminal cancer and have elected hospice services.

Background

This researcher conducted a preliminary investigation into clinical outcomes of hospice patients receiving music therapy. In the ex post-facto study, the researcher analyzed medical records randomly selected using Hospice Patient Management System software at Big Bend Hospice. There was a total of 40 subjects, 20 of whom had been referred to music therapy, and 20 of whom had not. Subjects in the post-hoc study were not randomly assigned to music therapy, were of varying diagnoses, and lived both in their own homes and in skilled nursing facilities or assisted living facilities. It was found that subjects receiving music therapy lived longer (M= 68 days) on the hospice program than those who did not receive music therapy (M = 25 days). Therefore, subjects' length of stay was evaluated in the current randomized, controlled study in order to ascertain if music therapy plays a role in one's life span on the hospice program. In the same preliminary investigation, it was found that subjects died closer to the last music therapist's visit than to the counselor's visit. The current study evaluates whether or not the same occurs in a randomized, controlled experiment. Subjects' medications used during the study were analyzed and it was found that patients residing in their homes have inaccurate accounts of medications taken. Although prescriptions were accurate in the medical record, documentation of actual medicines taken was not. It was advised that studies evaluating medicine consumption should be conducted in a more controlled environment than in the home setting (i.e., hospice inpatient setting).

The preliminary investigation identified means to better control for confounding variables and develop a sounder research design. Subjects in the preliminary investigation were of varying diagnoses, and this appeared to create a confounding variable as different diagnoses present different reactions to treatment and have varying prognoses even at the end of life. The literature also reflects the need to control for diagnosis, and McMillan and Mahon (1994) encouragE:d future research to control for diagnosis as different diseases have different courses. Therefore, subjects' diagnoses in this current investigation have been limited to terminal cancer. Place of residence can be a confounding variable in evaluating quality of life, and subjects in this current study have been limited to their home environment as is typical in the literature (Eischens, Elliott, & Elliott, 1998; McMillan, 1996a, 1996b; McMillan & Mahon, 1994; McMillan & Weitzner, 1998, 2000). Furthermore, random assignment of subjects to conditions reduces confounding variables and allows for more objective testing of the experimental question.

Method

Subjects

All selected subjects (N= 80) were diagnosed as having terminal cancer by at least two physicians and were given a prognosis of 6 months or less duration for life. Types of cancer included lung, colon, kidney, nasopharynx, prostate, liver, esophagus, breast, pancreatic, brain, oral cavity, ovarian, stomach, endometrial, sinus, larynx, rectal, adeno, leukemia, melanoma, and unspecified cancer. Subjects were all newly admitted patients to hospice, were adults, and resided in their homes. The research design controlled for gender and age, providing for an equal number of males and females and those above and below age 65 randomly assigned to two conditions. The mean age of the control group was 65, and the experimental group was 66. Although race was not controlled for, analysis after randomization showed even distribution with 25% Blacks and 75% Caucasians in each.

This study was conducted with patients of Big Bend Hospice, an organization providing services to terminally ill patients and their families in North Florida. Upon admission into the hospice program subjects were reviewed by the admissions nurse for candidacy into the study. Criteria for subject participation were: (a) a terminal cancer diagnosis was made, (b) the subject was an adult, (c) the subject resided in her or his home, (d) the admissions nurse assessed that the subject would live at least 2 weeks, (e) the subject was able to answer questions regarding perceived quality of life, and (f) the subject consented to participation in the study. If permission to participate was refused, potential subjects remained in the hospice program and received routine hospice care, including access to music therapy. Those refusing participation, however, were no longer contacted regarding the study. Informed consent was obtained by either the family support counselor (control group) or music therapist (experimental group) upon the first visit to the subjects' homes, and subjects or their legal guardians signed the consent form which was approved by the Florida State University Human Subjects Committee and the Management Team of Big Bend Hospice.

Research Design

This study utilized a randomized experimental-control group design where subjects were randomly assigned to either the control (routine hospice services only) or experimental (routine hospice services and music therapy) group. Because the review of relevant literature indicated that patients over age 65 report less pain than those under 65, subjects were evenly divided in each group by this age category. Furthermore, subjects' gender was evenly divided as well so each group had the same number of males and females and those over and under age 65, as advised by previous studies (Lester, Lefebvre, & Keefe, 1994; McMillan, 1989, 1996; McMillan & Mahon, 1996; Wachter-Shikora & Perez, 1982).

The null hypothesis for the study states: there will be no significant difference between the experimental and control groups on quality of life, length of life, or the relationship to time of death from last counselor or music therapy visit. Furthermore, there will be no significant difference between groups by age or gender on the above variables. The independent variables were routine hospice services and music therapy, and the dependent variables were quality of life as measured by the Hospice Quality of Life Index-- Revised and the Palliative Performance Scale, length of life in days, and the relationship to the time of death in days from last visit by the music therapist or the family support counselor.

The control group consisted of 40 subjects who received only routine hospice services and the experimental group consisted of 40 subjects who received routine hospice services and music therapy. Routine hospice services included medical care by a registered nurse under the direction of a physician, personal care assistance by a home health aide, spiritual support by a hospice chaplain, psychosocial support by a family support counselor, and volunteers as requested. All medical equipment and medications related to the terminal diagnosis were also provided by the hospice for the subjects' home environments.

Measures

Quality of life was measured by the Hospice Quality of Life Index-Revised (HQLI-R) for hospice patients. The HQLI-R is a selfreport questionnaire using a Likert scale of 0 to 10 with a total of 28 items designed for hospice patients with cancer. Questions pertain to various aspects of quality of life including physical, relationship, psychological, spiritual, and financial issues. The questionnaire is in three main categories which form the three subscales of the Index: (a) functional (i.e., daily activities, social life, concentration ability, enjoyable activity); (b) psychophysiological (i.e., anger, pain, nausea, sex life, worry, anxiety); and (c) social/spiritual (i.e., meaning in life, physical contact, family support, relationship with God). Higher scores on the HQLI-R indicate a higher perceived quality of life with a maximum of 280 points. The first subscale, functional well-being, has a maximum of 70 possible points. Psychophysiological well-being, the second subscale, has a possible 130 points, and the final subscale, social/spiritual wellbeing has a possibility of 80 points. The HQLI-R is reported in the literature to have good content validity (R = .83). Subscales are considered reliable and overall reliability is quite high (R = .88) (McMillan, 1996b; McMillan & Weitzner, 1998). Music therapists and family support counselors attended an in-service training at Big Bend Hospice and received instruction via lecture on completing the HQLI-R.

The second measurement tool was the Palliative Performance Scale (PPS) which is a measurement of physical status in palliative care. It serves as a basis for studying treatment effects at various stages of physical decline. The scale is widely used throughout North America in hospice and palliative care settings for prognosis, research, and case management purposes. The PPS is a nursing assessment scale which takes into account conscious level, intake, self-care, ambulation, activity, and evidence of disease. It is scaled in 10% subdivisions from 0% (death) to 100% (full ambulation, normal activity, no evidence of disease, full self-care, normal nutritive intake, and full conscious level). The PPS is considered a valid tool for rating the physical condition of terminally ill cancer patients as it is correlated with other palliative care tools (Morita, Tsunoda, Inoue, & Chihara, 1999). Because the PPS is a routine measurement tool utilized at Big Bend Hospice, all hospice nurses received instruction on its use during their employee orientation.

The third form of measurement was the length of life for each subject on the hospice program. This was measured from the date of admission into hospice until the date of death for a total number of days on the hospice program. Finally, the date of the last scheduled visit by counselor and music therapist until time of subject death was measured in days. These data were taken from the medical records at Big Bend Hospice.

Procedures

Upon admission into the hospice program, candidates for the study were evaluated by the admissions nurse for appropriateness. If candidates were able to meet the six criteria for participation and agreed to volunteer for the study when contacted by the music therapists or family support counselors via telephone, they were randomly assigned to one of the two groups.

When the subject was assigned to the control group, the family support counselor provided a home visit to complete the psychosocial assessment. During this visit the counselor explained the nature of the study, assisted the subject in understanding the informed consent, and answered any questions presented. At the end of the visit, the counselor asked the subject to complete the HQ,LI-R. Based on the psychosocial assessment of the subject's psychosocial support needs, the counselor determined frequency of visits (weekly or bimonthly). At the end of the counselor's routine visits, the HQLI-R was again completed until the subject's death. The counselor's psychosocial assessment also identified care plan needs for the subject and his/her family. Examples of care plan needs included anticipatory grief, funeral planning, advance directives, depression, estate planning, and limited caregiving.

When the subject was assigned to the experimental group, the music therapist provided a home visit to complete the music therapy assessment. These visits were usually made within the first week of hospice admission and followed the initial assessments by the nurse and counselor. The counselor did not discuss the nature of the study with the subjects in the experimental group. During the first music therapy visit, the music therapist explained the nature of the study, assisted the subject in understanding the informed consent, and answered any questions presented. The music therapist also conducted a music therapy assessment and provided live music therapy. Following the music therapy, subjects were asked to complete the HQLI-R. The frequency of music therapy visits were the same as the frequency of counselor visits for subjects in the experimental group. At the end of each routine music therapy visit, subjects were asked to complete the HQLI-R. The music therapist's assessment also identified care plan needs for each subject. Examples of care plan needs included spiritual support, pain management, depression, and isolation and were analyzed in comparison to the nurses' and counselors' identified care plan needs.

For subjects in both groups, the nurse visited within 48 hours of the subject's hospice admission and conducted the nursing assessment. As part of the nursing assessment, the PPS was completed for all subjects. During virtually all nursing visits for all subjects, the PPS was completed as is the routine procedure for nurses at Big Bend Hospice. Nurses did not discuss the study with the subjects and did not ask the subjects to complete the HQLI-R, although they were aware of subjects assignments to groups. Each discipline documented their visits on the Big Bend Hospice progress notes which reflect how each care plan need was exhibited by the subjects, treated by the staff, and subjects responded to treatment during each visit, and data were taken from progress notes.

Music Therapy

Music therapists at Big Bend Hospice provided music therapy visits to subjects in the experimental group. Music therapists were either Board Certified professionals (three participated in this study) or music therapy interns (two participated in this study under proper clinical supervision). Both the professional music therapists and music therapy interns were trained in hospice music therapy by this researcher. The professional staff were former interns at hospice under this researcher's supervision prior to their employment, and the interns were practica students at hospice prior to their internship under this researcher's supervision. Training consisted of observation, co-facilitation, supervision, and group discussion. Sessions were conducted in the subjects' homes, and visits were planned with subjects and their families when appropriate. Following the initial music therapy assessment visit, the music therapist designed each session to meet the individual identified needs of each subject. Big Bend Hospice utilized the music therapy assessment as designed by Hilliard (1995).

The philosophical foundation of the music therapy program at Big Bend Hospice is a cognitive-behavioral approach where music therapy interventions are designed to treat identified problems and allow for the expression of emotions while respecting the process inherent within the live musical dialogue. All music therapy interventions in this study utilized live music, and all music selected was subject-preferred. Each subject received at least two music therapy sessions. During the course of the study, subjects died at varying intervals. Therefore, some subjects received only two sessions while one subject received thirteen sessions.

Music therapists treated a variety of clinical needs of their subjects during the study and utilized a myriad of techniques. While needs and interventions varied, some themes emerged. Among these, most music therapists utilized one or more of the following music therapy techniques: song choice, music-prompted reminiscence, singing, live music listening, lyric analysis, instrument playing, song parody, singing with accompaniment using the iso-- principle, planning of funerals or memorial services, song gifts, and music-assisted supportive counseling.

Progress notes from last music therapy sessions prior to subjects' deaths were analyzed. The last music therapy sessions may differ from previous sessions by way of care plan needs of the subjects and techniques utilized by the music therapists as subjects were often quite ill in the last sessions and less able to engage actively in sessions. Therefore, care plan needs may have changed from the initial music therapy sessions until the last sessions as subjects' needs were continuously changing with their physical decline. Figures 1 and 2 show an analysis of treatment objectives and music therapy techniques used in the last sessions for experimental subjects prior to their deaths.

Results

Quality of life was measured by the HQLI-R, a self-report tool administered to subjects with the assistance of the counselor (control group) or music therapist (experimental group). All subjects were able to complete the HQLI-R at least twice, therefore, the first two scores were used in statistical analysis. A repeated measures ANOVA was used in data analysis of HQLI-R (QOL). There was a statistically significant difference on the two QOL scores by group and scoring (time), F(1, 72) = 8.437; p .05, or age, F (1, 72) = 3.189; p > .05. Table 1 offers results of the repeated measures ANOVA for QOL.

Because there was significance in Time x Music, Duncan's Test was used for post-hoc analysis. There was a significant difference between the two experimental scores combined and the two control groups combined on the QOL scores. There was also significant difference between the first and second QOL scores in the control group. Results indicated that the quality of life for those receiving music therapy was higher than for those who did; and for those in the control group, quality of life lessened significantly from the first to the second assessment.

Because the analysis by gender approached significance, Duncan's Test was used for post-hoc analysis. There was, however, no significance in the post hoc analysis for gender. Table 2 offers n, means, and standard deviations of QOL. Results indicated that gender and age showed no signs of effecting quality of life.

Each HQLI-R contained three subscales: functional, psychophysiological, social/spiritual well-being. Subscale scores were analyzed using a Two-Way Repeated Measures ANOVA. Results indicated that on the first (functional well-being) subscale, there was no significant difference between groups, F(1, 78) = 2.034; p > .05. There was a significance difference between groups on the second (psychophysiological well-being) subscale, F(1, 78) = 3.995; p .05. Table 3 shows the results of the Two-Way Repeated Measures ANOVA for each subscale. Table 4 shows means of QOL subscale scores. Because there was significance on the second subscale, Duncan's Test was used for post-hoc analysis, and a significant difference was found between the combined experimental group scores and the combined control group scores. Results indicated that the music therapy group scores were significantly higher on quality of life on the psychophysiological subscale.

Subjects' physical status was measured by the hospice nurses using the Palliative Performance Scale (PPS). Nurses were able to assess the PPS on all subjects at least twice, therefore, the first two scores were used in statistical analysis. A Repeated Measures ANOVA was used in data analysis of the PPS. There were no significant differences by group, F(1, 72) = .758; p > .05, gender, F(1, 72) = .212; P > .05, or age, F(1, 72) = .077; p > .05. Table 5 shows the results of the Repeated Measures ANOVA for PPS scores. Table 6 shows mean PPS scores. Results indicate that the treatment of music therapy has no significant effect on subjects' physical functioning. Furthermore, subjects' gender and age have no significant effect either.

Pearson's Correlations Test was used to recognize possible correlations between QOL and PPS scores. Analysis of both scorings of the QOL and PPS showed weak correlation and no significant difference (p > .05). Results indicated that the two measures are independent of one another, therefore, subjects' quality of life scores were not dependent on the physical decline as assessed by the PPS. Table 7 shows the Pearson's Correlations.

To further evaluate the relationship between QOL and PPS scores over time between the two groups, graphic analysis was conducted. Figures 3 and 4 illustrate means of QOL and PPS scores throughout the duration of the treatment. There are varying numbers of subjects after the second assessment/treatment due to subject mortality over time. As PPS scores declined over time, quality of life scores increased in the music therapy group. This indicates that the population was becoming increasingly more ill, yet their quality of life was becoming better. This was not the case in the control group; quality of life scores lowered as physical health declined (lower PPS scores). Therefore, music therapy can positively influence quality of life, even though the patients' disease processes continue and death becomes imminent.

To determine differences between groups on subjects' length of life in the hospice program, an independent samples test was conducted. Table 8 offers n, mean, and standard deviation for length of life. Levine's Test for Equality of Variances was conducted to test for equal variance among variables. Equal variance (p > .05) was found for length of life, therefore, conducting independent samples t-test was appropriate. There was no statistically significant difference found between groups on subjects' length of life (t = .980, df = 78, p > .05). Therefore, whether or not the subjects received music therapy played no significant role on how long they lived during their hospice experience.

An independent samples Rest was conducted to determine differences between the subjects' time of death from the last music therapist and counselor's visit (in days) within the experimental group. Levine's Test for Equality of Variances was conducted to test for equal variance among variables. Equal variance (p > .05) was found, therefore, conducting independent samples Rest was appropriate. There was no statistically significant difference found between visits made by disciplines and time of death (t = .852, df = 74, p > .05). Results indicated no relationship between time of death and the last visit made by counselor or music therapist. Table 9 shows means for the last scheduled visits by discipline.

Care plan needs for subjects as assessed by the nurse, counselor, and music therapist were analyzed. Music therapists were more likely to address quality of life issues (30 times more than the counselor and 28 times more than the nurse) of subjects. Other than the chaplain, the music therapist was the only discipline providing spiritual support (16 times more than nurse and counselor). Music therapists were more likely to treat nausea and vomiting, pain, comfort measures, confusion, lethargy, changes in locus of control, and semi-coma than were counselors. Nurses, however, routinely addressed these issues. Counselors and music therapists addressed the following issues on a more equal basis: safety and grief for imminent death, anxiety, fear, depression, anger, anticipatory grief, isolation, denial, irritability, and shock. Counselors were much more likely to address funeral arrangements, advanced directives, legal and financial issues, limited caregiving, and role changes than music therapists and nurses. Nurses addressed medical issues of sensory impairment, nausea/vomiting, nutritional needs, bowel and bladder issues, pain, cardio-pulmonary distress, mobility and activity impairments, and skin integrity at high frequencies.

The null hypothesis regarding no significant difference between the experimental and control groups on quality of life scores was rejected. The null hypothesis relating to no significant difference between groups for length of life, the relationship to time of death from last counselor or music therapy visit, or by age or gender failed to be rejected.

Discussion

Results of this investigation clearly support the use of music therapy in hospice and palliative care for improving the quality of life of people diagnosed with terminal cancer. For subjects in the music therapy group, quality of life scores after the initial music therapy session were significantly higher than those who received only routine hospice services. Not only were scores significantly higher after the first music therapy session, but scores increased further following the second music therapy session whereas scores in the control group lowered after the second counselor visit. For example, when PPS scores declined on the third assessment in the experimental group to a mere 22, quality of life remained stable at 210. Throughout the duration of treatment, quality of life was consistently higher in the experimental group than in the control group, even when physical function declined. These data are important because they illustrate that music therapy treatment served to increase quality of life initially and increased quality of life even more over time. These data support the use of music therapy treatment for patients upon admission into hospice and its provision with high frequency throughout their remaining lifetime.

When comparing mean scores of the HQOLI-R of the current study to those in the literature, it is evident that music therapy positively influences quality of life. Two studies in the literature measured quality of life using the HQOLI-R with adult hospice patients diagnosed with terminal cancer who resided in their home and found mean scores similar to those in the control group of the current investigation. The mean scores from the studies were 168 and 191 which is similar to the mean score of the control group of this investigation (183). The mean score for those receiving music therapy, however, was 211 (McMillan, 1996a, b; McMillan & Weitzner, 2000).

In another study, 100 adult hospice patients with terminal cancer residing in their homes were given the HQOLI-R and results were compared to the same number of healthy adults given the test. The mean score in the healthy adult population was 233 whereas 191 was the mean score of the hospice population (McMillan & Weitzner, 1998). Subjects in the current study who were terminally ill and receiving music therapy scored closer to the healthy adult population than to any other terminally ill population on the HQOLI-R. A comparison of results throughout the literature further indicates that those receiving music therapy scored higher on quality of life scores than those who do not.

The subscales analysis offers several implications for music therapy practice in palliative care and hones in on specific target areas treated by music therapists. Psychophysiological well-being is successfully treated by music therapists, and needs in this area which are often identified by the music therapy assessment include: worry, loneliness, anger, pain, nausea, and shortness of breath (anxiety). Care plan needs analysis indicated that music therapists treated anxiety, pain, comfort measures, and nausea with higher frequencies than did the counselors. Therefore, music therapists working in palliative care may need to pay special attention to the needs of their patients in this area. Furthermore, referral sources to music therapy should recognize the unique impact music therapy offers patients in these domains. Because patients in palliative care often suffer from psychophysiological problems, palliative care professionals need to recognize that a music therapy referral may be warranted for treatment.

The functional well-being and social/spiritual well-being subscales showed no significance between groups. There were fewer items on each of these subscales, and it appears that music therapists were least likely to address needs in these areas. Issues such as nutritive intake, concentration, level of independence, and fatigue were included on the functional well-being subscale. Many of these items appear to be disease-driven, and therefore music therapy would have minimal effects. The spiritual/social well-being subscale evaluated issues including physical care, support from family and friends, physical contact with loved ones, and satisfaction with the emotional and spiritual support provided by the hospice team. While spiritual support was identified with many subjects as a care plan need, the subscale showed no significance between groups. This may be because the subscale asks questions related to spiritual concerns that are different from the way the music therapists viewed the treatment of spiritual needs. Perhaps phrases including spiritual practices (rituals, singing of hymns, etc.) would better capture spiritual issues related to one's quality of life.

Analysis of care plan needs indicates the need for music therapists to be a routine part of the interdisciplinary care team in palliative care. They serve as adjuncts to the care plan team. Music therapists provide services that others on the team do not, and they complement on-going care successfully. The provision of palliative care is inherently creative and patient-centered. Within this framework, it makes sense that music therapy is offered as an integral part of the palliative care team of professionals.

The research design controlled for age group and gender so an equal number were represented in both the experimental and control groups. Statistical analysis evaluated the effects of gender and age on the QOL scores. There were no differences regarding subjects' gender. This indicates that both women and men can benefit equally from music therapy to promote quality of life in hospice and palliative care. While meta-analyses in the medical literature show that women respond to music therapy better than men (Standley, 2000), it was not the case for this investigation. This may be because each subject received music therapy services based on their own unique needs, and the music therapist assessed subjects' needs, strengths, preferences, and musical motivation using the music therapy assessment. Furthermore, while hospice services are generally considered within the medical model, the hospice philosophy embraces the treatment of the whole person (physically, psychologically, socially, and spiritually). The meta-analysis viewed music research in medical settings where music therapy treated more medical than emotional or spiritual needs. Therefore, the difference in treatment goals may help explain the differing gender responses.

Subjects were divided into two age groups: those age 65 and older and those under age 65. Age group played no role in subjects' perceived quality of life as measured by the HQLI-R. Subjects' physical decline was not affected by age group or by gender. Physical decline was also consistent between groups, and those receiving music therapy experienced the same physical decline as those in the control group. It appeared that physical decline was driven primarily by the disease process, and subjects' age, gender, and assigned group (experimental or control) played no role. Treatment for patients, then, should focus more on quality of life issues than on physical issues which is consistent within the hospice philosophy.

While there was no significant difference on length of life of subjects between the experimental and control groups, those in the experimental group lived an average of 12 days longer than those not receiving music therapy. For people with a terminal illness and their families, additional life may be viewed as a miracle, especially when quality of life issues remain high. The means of the groups indicate that those receiving music therapy lived longer with a significantly higher degree of quality of life. Even though significance was not found statistically, the length of life might be an area for future study in order to determine if a pattern can be established. Informed choices about treatment could then be made by patients and their families.

In the post-hoc preliminary investigation for this current study, subjects died closer to a music therapist's visit than to the counselor's visit. Because these data were remarkable, this issue was further investigated within the current study which has better controls and randomized sampling. Statistical analysis showed no significant differences between subjects' time of death and the last visit provided by the counselor or music therapist. Results indicate that subjects were 18% more likely to die closer to a counselor's visit than to a music therapist's visit. This percentage is relatively low and may be due to chance, however, future studies should further investigate this issue, especially since several qualitative articles describe the use of music therapy at the death vigil or just prior to a patient's death, and the data here do not support this.

Results of this study are promising for music therapy in palliative care, however, it is not without limitations. The mortality rate interfered with analysis of sessions beyond the first two for many subjects, therefore, only the first two quality of life indexes were utilized in data analysis. This leads to the question of what happens to quality of life over many sessions until the time of death? Further, the majority of these subjects were Caucasian, and they were limited to the same geographic region with the same social norms and similar music preferences. What would a multiple-site study find evaluating the effects of music therapy among hospice patients of varying cultures, races, and musical preferences? Finally, the implementation of music therapy included many varying types of techniques, and the only aspect which was controlled for was that all the interventions utilized live music. Future studies may benefit from a comparison of specific music therapy techniques utilized at the end of life.

Music therapy has been finding its place in hospices and palliative care settings over the last few decades. There are many voices contributing to the literature, however, there is a dearth of quantitative studies. Such studies are important as they offer more opportunities for generalizing, and they better guide clinicians in their practice. Furthermore, they speak to a different audience than qualitative studies. Because there are increasing demands for quantitative research in palliative care from external sources (insurance providers, physicians, hospice administrators, etc.), music therapists will benefit from data supporting the use of music therapy as they will be better able to meet the demand. Moreover, people seeking palliative care may benefit from research as more opportunities for accessing palliative care music therapy may be afforded.

References

Abbott, C. M. (1995). The efects of music therapy on the perceived quality of lift of patients with terminal illness in a hospice setting. Unpublished master's thesis, Western Michigan University; Kalamazoo, MI.

Beggs, C. (1991). Life review with a palliative care patient. In K. Bruscia (Ed.), Case studies in music therapy (pp. 611-616). Phoenixville, PA: Barcelona Publishers.

Calovini, B. S. (1993). The effect of participation in one music therapy session on state anxiety in hospice patients. Unpublished master's thesis, Case Western Reserve University.

Curtis, S. L. (1986). The effect of music on pain relief and relaxation of the terminally ill. Journal of Music Therapy, 23, 10-24.

Donnelly, S., & Walsh, D. (1996). Quality of life assessment in advanced cancer. Palliative Medicine, 10, 275-283.

Eischens, M.J., Elliott, B. A., & Elliott, T. E. (1998). Two hospice quality of life surveys: A comparison. The American Journal of Hospice and Palliative Care, May/june, 143-148.

Forinash, M. (1990). A phenomenology of music therapy with the terminally ill. Unpublished doctoral dissertation, New York University, Ann Arbor: UMI Dissertation Information Service.

Gilbert, J. P. (1977). Music therapy perspectives on death and dying. Journal of Music Therapy, 14, 165-171.

Hilliard, R. E. (1995). Hospice and music therapy: Creating jobs and fulfilling needs. Presented at the annual conference of the National Association for Music Therapy, Houston, TX.

Hilliard, R. E. (2001). The use of music therapy in meeting the multidimensional needs of hospice patients and families. Journal of Palliative Care, 17(3), 161-166.

Hogan, B. (1996). Music therapy at the end of life: Searching for the rite of passage. In D. Aldridge (Ed.), Music therapy in palliative care: New voices, (pp. 68-81). London: Jessica Kingsley Publishers.

Hogan, B. (1999). The experience of music therapy for terminally ill patients: A phenomenological research project. In R. R. Pratt & D. E. Grocke (Eds.), Musicmedicine 3: Musicmedicine and music therapy: Expanding horizons (pp. 242-254). Melbourne, Australia: University of Melbourne.

Krout, R. E. (2001). The effects of single-session music therapy interventions on the observed and self-reported levels of pain control, physical comfort, and relaxation of hospice patients. American, journal of Hospice and Palliative Care, 18(6), 383-390.

Lester, N., Lefebvre,J. C., & Keefe, F.J. (1994). Pain in young adults: A relationship to gender and family pain history. Clinical fournal of Pain, 10, 282-289.

Martin, J. A. (1991). Music therapy at the end of life. In R Bruscia (Ed.), Case studies in music therapy (pp. 617-632). Phoenixville, PA: Barcelona Publishers.

McMillan, S. C. (1989). The relationship between age and intensity of cancerrelated symptoms. Oncology Nursing Forum, 16, 237-241.

McMillan, S. C. (1996a). Pain and pain relief experienced by hospice patients with cancer. Cancer Nursing, 19(4), 298-307.

McMillan, S. C. (1996b). The quality of life of patients with cancer receiving hospice care. Oncology Nursing Forum, 23(8), 1221-1228.

McMillan, S. C., & Mahon, M. (1994). A study of quality of life of hospice patients on admission and at week 3. Cancer Nursing, 17(1), 52-60.

McMillan, S. C., & Weitzner, M. (1998). Quality of life in cancer patients. Cancer Practice, 6(5), 282-288.

McMillan, S. C., & Weitzner, M. (2000). How problematic are various aspects of quality of life in patients with cancer at the end of life? Oncology Nursing Forum, 27(5), 817-823.

Morita, T., Tsunoda,j, Inoue, S., & Chihara, S. (1999). Validity of the palliative performance scale from a survival perspective. Journal of Pain and Symptom Management, 18(1), 1-2.

Mramor, K. M. (2001). Music therapy with persons who are indigent and terminally ill. Journal of Palliative Care, 17(3), 182-187.

Nakagami, Y (1997). Hospice program and palliative medicine. Gan To Kagaku Ryoho, 24(7), 792-799.

O'Callaghan, C C. (1993). Communicating with brain-impaired palliative care patients through music therapy. Journal of Palliative Care, 9(4), 53-56.

O'Callaghan, C. C. (1995). Songs written by palliative care patients in music therapy. In C. A. Lee (Ed.), Lovely waters (pp. 31-40). Oxford: Sobell Publications.

O'Callaghan, C. C. (1996). Pain, music creativity and music therapy in palliative care. The American, journal of Hospice and Palliative Care, March/April, 43-49.

O'Callaghan, C. C. (2001). Bringing music to life: A study of music therapy and palliative care experiences in a cancer hospital. Journal of Palliative Care, 17(3), 155-160.

Porchet-Munro, S. (1993). Music therapy. In D. Doyle, G. W. C. Hanks, & N. MacDonald (Eds.), Oxford textbook of palliative medicine (pp. 555-559). Oxford: Oxford University Press.

Salmon, D. (2001). Music therapy as psychospiritual process in palliative care. Journal of Palliative Care 17(3), 142-146.

Skaggs, R. (1997). The bonny method of guided imagery and music in the treatment of terminal illness: A private practice setting, Music Therapy Perspectives, 15, 39-44.

Standley, J. M. (2000). Music research in medical treatment. In D. Smith (Ed.), Effectiveness of music therapy procedures: Documentation of research and clinical practice (3rd ed.) (pp. 1-64). Silver Spring, MD: American Music Therapy Association, Inc.

Wachter-Shikora, N., & Perez, S. (1982). Unmasking pain. Geriatric Nursing, 5, 392-394.

Weber, S. (1996). Music: A means of comfort. In D. Aldridge (Ed.), Music therapy in palliative care: New voices (pp. 95-104). London, Jessica Kingsley Publishers.

West, T, M. (1994). Psychological issues in hospice music therapy. Music Therapy Perspectives, 12, 117-124.

Whittall, J. (1989). The impact of music therapy in palliative care: A quantitative pilot study. In J. A. Martin (Ed.), The next step foward: Music therapy with the terminally ill (pp. 69-72). Bronx, NY: Calvary Hospital.

Whittall,J. (1991). Songs in palliative care: A spouse's last gift. In K. Bruscia (Ed.), Case studies in music therapy (pp. 603-610). Phoenixville, PA: Barcelona Publishers.

Wylie, M. E., & Blom, R. C. (1986). Guided imagery and music with hospice patients. Music Therapy Perspectives, 3, 25-28.

Russell E. Hilliard, PhD, LCSW, MT-BC

Florida State University

Big Bend Hospice

This study was conducted in partial requirements for the degree of Doctor of Philosophy in Music Education and Therapy at the Florida State University. Russell Hilliard is now Assistant Professor of Music Therapy at the State University of New York at New Paltz.

The author wishes to acknowledge the music therapists and social workers at Big Bend Hospice and Jayne Standley for assisting with this study.

Copyright American Music Therapy Association Summer 2003
Provided by ProQuest Information and Learning Company. All rights Reserved