Music Therapists� Practice-based Research in Cancer and Palliative Care: Creative Methods and Situated Findings

By Clare OCallaghan & Philippa Barry

Abstract

Although randomized controlled trials are described as the gold standard in medical research in Australia, their superiority is being questioned in palliative care which is focused on addressing individualized needs to maximize life quality. We use creative practice-based research to examine the usefulness of our music therapy work amongst people with life threatening conditions. Examined voices include "collective" (patients, visitors, staff, and music therapist), "their" (patients or caregivers), "our" (a group of music therapists), and "my voice" (one music therapist). Data sources have included clinical journals, semi-structured questionnaires, interview responses, a focus group, reflexive groupwork supervision transcripts, and patients song lyrics. Findings, situated within varied theoretic lenses, substantiate music therapys role in oncology and palliative care settings. Readers are invited to devise creative ways to voice their clients, bystanders, and own wisdom about music therapy to meaningfully extend the knowledge base.

Keywords: practice-based research; palliative care; music therapy; cancer

Introduction

The origin of the quantitative and qualitative research divide [1] is described as appearing at the end of the 1800s, when German philosophers distinguished between: naturwissenschaften, that is scientific ideas of truth seeking which was focused on the physical world; and Geisteswissenschaften, that is, human sciences focused on the meaning of action and the human spirit. The divide is arguably older however, evident in, for example, Vicos (1668-1744) critique that the Descartes process of understanding (based on deduction and reason) avoids acknowledgement of how accumulated human wisdom, alongside the human minds imaginative, inventive, and poetic capacities, generates scientific thinking (Coats, 2003, p. 48). By the early 1900s anthropology was focused on studying "alien cultures", and then city life through fieldwork (Clark, 1997). In the 1960s, comparable research emerged in health care. Field studies, conducted by Glaser and Strauss (1967), occurred in oncology and geriatric hospital wards in California. From this work, they developed an "awareness of dying" theory characterized by how dying patients and their carers interact with each other in a closed or open manner, or through suspicion or a shared mutual pretense. Importantly, Glaser and Strausss work pioneered the grounded theory research paradigm (Corbin & Strauss, 2008; Glaser & Strauss, 1967), one of the most widely used qualitative research approaches in health and palliative care. Grounded theory has been pivotal in our research on music therapy in cancer and palliative care, which Clare (author 1) commenced practicing in 1985, and Philippa (author 2) in 2006.

When Clare began to work as a music therapist she was told that research was essential to expand the profession. Quantitative research, including randomized controlled trials (RCTs), was then, and still are often described as the "gold standard" for producing best evidence to inform health care practice (NHMRC [2], 2000). Some palliative care practitioners, however, question the superiority of RCTs to indiscriminately provide best evidence in palliative care (Aoun & Kristjanson, 2005). This paper will first briefly describe music therapy and palliative care, describe some of the problems associated with using quantitative research methods in this area, and then define "practice-based research", which is aligned with the qualitative research tradition. We will then outline some of the practice-based research methods that we have used to uncover varied views about music therapys efficacy in palliative care and to justify its role. It is hoped that this description will inspire readers to consider creative ways to examine clients, caregivers bystanders, and their own wisdom about music therapy to meaningfully extend knowledge.

Music Therapy and Palliative Care

Music therapy in palliative care can be described as the professionally informed and creative use of music within a therapeutic relationship with people identified as needing biopsychosocial or spiritual help, or who desire further self-awareness, to enable increased life quality (OCallaghan, 2004). Palliative care, which is also sometimes described as hospice care, is:

an approach that improves the quality of life of patients and their families facing the problems associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial and spiritual. (WHO [3], 2008)

Traditionally palliative care was considered appropriate for people dying from cancer, however, its principles are now considered relevant throughout the course of all degenerative and life threatening conditions (Doyle, Hanks, Cherny, & Calman, 2004), and the bereaved (Kissane, 2004). Music therapists work for inpatient and home based palliative care services. We also practice palliative care in cancer wards, nursing homes, neurological, and other settings where prognoses are limited yet uncertain. The aim is to help people have the best quality of life possible when enduring life threatening conditions and end-of-life care, including the patients and their loved ones. Methods used by palliative care music therapists include music supported counseling, improvisation, imagery, relaxation, and song writing. Through music therapy, patients may relive important memories and experience validation of their lives, increased self-awareness, pleasure, or a spiritual connection. Symptoms, including pain, nausea, fatigue, and weakness, may be alleviated through distraction or working through unresolved issues, including regret. Music therapy with patients and families may also positively support their communication and connectedness (OCallaghan, Petering, Thomas, & Crappsley, 2009). This is especially significant when patients are brain-impaired (OCallaghan, 2004), because music and language abilities are apparently a function of separate neural pathways (Levitin, 2006). Increasingly music therapists also describe legacy work which includes creating physical items and memories to validate lives and support adaptive (Coyle, 2006) and creative living (Boog & Tester, 2008) until one dies. This is achieved through helping patients to compose (O'Brien, 2003; OCallaghan, 1996; Salmon, 1995) and record songs (Dileo & Dneaster, 2005) and performances (Beggs, 1991; Dileo & Magill, 2005; Whittall, 1991), leave comforting music therapy memories (Lindenfelser, 2005), create songbooks (Whittall, 1991), and leave music based life reviews (Hogan, 1999; O'Callaghan, 1984).

Researching Music Therapy in Palliative Care: Issues with Quantitative Approaches

Quantitative based research approaches in health care were originally devised to test pharmacological interventions and gradually emerged as the acceptable method for investigating all other health outcomes (Edwards, 2001), from surgery to quality of life. Such approaches include closed-ended surveys and interviews, and statistical methods applied to score comparisons. In RCTs, research mean scores from one group of similar participants are compared with the mean scores from another similar group of participants who are also subjected to an experimental condition. For example, in a RCT study Clare is currently involved in, the aggregate anxiety measures from a group of people receiving "standard care" radiotherapy are being compared with aggregate anxiety measures from a similar group receiving radiotherapy plus music. The anxiety measurement is based on the State Trait Anxiety Inventory [STAI] (Spielberger, 1983)[4]. Similarly, in Philippas (Barry, 2008a) mixed methods research exploring CD creation with children undergoing radiation therapy treatment, aggregate Kidcope (Spirito, Stark, & Williams, 1988) measures from a group of children with cancer receiving "standard care" radiotherapy were compared with aggregate Kidcope measures from a group of children with cancer receiving radiotherapy plus the music condition. It is assumed that findings from these studies using quantitative methodologies can be used to predict outcomes from health interventions, however, studies with similar methodologies in health care frequently disagree (Edwards, 2001).[5] Mixed methods research may also produce apparently conflicting findings. For example, in Philippas pediatric radiation research, quantitative data reflected that there was a non-significant difference in distress between music therapy and standard care groups, but the qualitative data generated themes suggesting the music therapy helped to reduce distress.

The National Health and Medical Research Council (NHMRC) of Australia states that the best evidence in health care comes from quantitative studies and, most eminently, systematic reviews of all relevant RCTs (NHMRC, 2000) [6]. This is despite the fact that the NHMRC also provides guidelines for conducting qualitative research, stating that it can enable "researchers to gain a better understanding of complex concepts or social processes, and investigate how individuals and communities interpret and make sense of their experiences" (NHMRC, 2007, p. 25). The NHMRC also funds qualitative researchers, including Clares current post doctoral research, and Dr Pam McGraths important research with the International Program for Psycho-Social Health Research at Central Queensland University. In Australian medical care, there is increasing interest in qualitative research, with local researchers publishing guidelines to help promote high standards (Kitto Chesters, & Grbich, 2008), including standards for a hierarchy of evidence for assessing quality (Daly, Willis, Small, et al, 2007). In Clares personal experience, while many multidisciplinary colleagues promote quantitative research, believing that it will be more effectively convince health care funders about musics potential contribution, there are signs that health care leaders are increasingly open to qualitative research, for example, the Medical Journal of Australia recently accepted qualitative research on the effects of a hospital staff choir (OCallaghan, Hornby, Pearson, & Ball, in press). This encourages us to continue using qualitative research methods when we believe that they are most appropriate for our research questions.

While quantitative research in music therapy and palliative care can be useful (and the authors use it for some research questions) their findings are not necessarily superior to those from qualitative research studies on human relationship therapies because quantitative research often cannot be used to make predictive generalizations (OCallaghan, 2009). One can, however, usually make conceptual generalizations (Kitto et al, 2008) from quantitative research, as one can in much qualitative music therapy research (OCallaghan, 2009). Inherent methodological problems in RCTs in palliative care include high attrition rates and limited survival times; patients cannot be blinded to group allocation (informed consent means that they know they are receiving music therapy and this can bias their responses as they may not want to offend); palliative care patients greatly vary in their illnesses, treatments, treatment responses, and backgrounds so that finding two homogenous groups (with evenly distributed variables) for comparison is unlikely, even with randomization; some argue that it is unethical to withhold offering potentially supportive care, like music therapy, from those assigned to control groups (Keeley, 1999) or wait-list controls (at the end of life); and the music therapy process does not allow for a standardized treatment as patients often choose how they wish to engage in therapy, and the therapists approaches vary (so that future patients do not receive exactly the same experience as the experimental groups). Similarly, the "therapist effect" can exist whereby two different music therapists using the same music therapy techniques can produce different outcomes (one significant and one insignificant) because of uncontrolled human relationship factors (Kain, Caldwell-Andrews, & Krivutza, 2004).

Furthermore, a problem associated with any quantitative research using standardized measures is that, while they may provide information important to the researcher, that information does not necessarily reflect what is important for the respondents (McGrath, 2000). Such measures, arguably, contradict palliative care principles of addressing individualized need, patient-and-family-centred care (Kvale & Bondevik, 2008), and patient involvement in service delivery developments (Health Evidence Network, 2004). Accessing funding for quantitative research studies can also be difficult, as one may need to pay for using measurement scales (like the STAI), research assistants, and statisticians. It is appropriate to further consider how creative research methods can be developed to understand music therapys role in palliative care, preferably at minimal or no cost. We will now describe some of our responses.

Practice-based Research

The research well outline here can be described as "practice-based research" which is "the use of research-inspired principles ... (within practice) ... to answer questions that emerge from practice in ways that inform practice" (Epstein, cited in Epstein, 2001, p. 17). This is distinguished from research-based practice which is "the use of research-based concepts ... so that hypotheses concerning cause-effect relationships (in practice) ... may be rigorously tested" (ibid). In our research we have examined two components of our daily music therapy work: (a) patients, visitors, staff carers, and our own [7] perceptions, and (b) therapeutic products, that is, patients song lyrics. Data has been gathered through collecting retrospective perceptions about music therapy through reflexive journals, semi-structured interviews, questionnaires, and focus groups, including reflexive group supervision. It has included clinical data-mining which privileges the metaphorical mine of precious information often contained in therapists routinely kept records, and is potentially transformable into data for research purposes (Epstein, 2001). Therapists clinical memories and artifacts also offered data for prospective analysis which is a component of clinical data-mining research (OCallaghan, 2005). The research findings were then situated through theoretical lenses to extend meaning about music therapys effects. The research well describe now can be categorized further as: "collective voices" (patients, visitors, staff, and mine); "their voice" (patients or staff); "our voice" (other music therapists with mine), and "my voice". An illustration of each type of research follows.

Collective Voices

Music Therapys Relevance in Oncology

Almost two years after commencing work as a music therapist at Peter MacCallum Cancer Centre (Peter Mac), in 1998, Clare examined the programs relevance, that is, what was it doing and was it helping, through a PhD. Peter Mac is Australias sole comprehensive cancer centre, and currently includes 94 inpatient beds, day treatment, and outpatient radiotherapy (OCallaghan, 2001a). The constructivist research paradigm inspired this research because it acknowledges that varying people can have multiple views about a research phenomenon: results emerge from the comparisons of appropriately selected respondents "constructions" (Guba & Lincoln, 1994). The research consisted of five studies which gathered five sources of data through criterion sampling, that is, all cases that met a selected criteria were considered (Rice & Ezzy, 1999). Over a three month period, all people affected by music therapy sessions conducted, and able to participate in the research, were invited to write anonymous thoughts on semi-structured questionnaires about the music therapys relevance. Music therapy (by Clare) was offered for 16 hours-a-week in five inpatient wards. Patients were either referred or offered music therapy as Clare visited them in their rooms which consisted of 1-to-4 beds. Methods mostly included patient and family selected live and familiar music (played by the therapist, and sometimes patients) on an electric keyboard, music relaxation, music based life review, or music supported counseling. Questionnaires were returned in strategically placed feedback boxes from 128 patients who participated in music therapy, 27 patients who overheard or witnessed music therapy, 41 visitors, and 61 staff. Clares interpretations about music therapys relevance were typed in a reflexive clinical journal. A thematic analysis, informed by grounded theory, was performed on each of the five data groups with the support of qualitative data management software (ATLAS.V.5.0, 2004). This type of thematic analysis is an inductive, comparative, and iterative process comprising of these phases: descriptive labels called codes were created to represent segments of text; comparable codes were grouped into categories; and comparable categories were grouped into themes. Summaries of research findings, reported elsewhere, follow:

(a) Findings from the five data groups were separately presented (O'Callaghan, 2001b; OCallaghan & McDermott, 2004) and then contrasted and compared into a final statement, an abridged form of which follows:

Music from peoples lifetimes often elicited beneficial affective experiences and altered multisensorial imaginings .... "The relevance of music therapy was evident through its capacity to enable, in many individuals, a renewed or heightened sense of personal historical significance and ongoing presence" (OCallaghan & McDermott, 2004, p. 178). "Many patients, visitors, and staff members affective, contemplative, and imagined moments in music therapy affirmed their "aliveness", resonating with an expanded consciousness, in a context where lifes vulnerability is constantly apparent" (ibid, p. 151).

(b) An interpretative subgroup analysis" was applied to the 128 oncologic patient responses. This is a method which adapts principles of subgroup analysis in quantitative research to textual data analysis. Males were much more likely to return questionnaires (84% males compared to 64% females)[8] but much less likely to participate in music therapy (57% of males and 42.5% of females declined). Other examples of subgroup differences included that, unlike females, some males described positive emotional responses when experiencing both good and sad memories, and the middle-age patients were more reflective about how music therapy brought people together (OCallaghan & Hiscock, 2007).

(c) A discourse analysis was inspired by comments by the main supervisor of this research, who questioned why the findings were so positive. A discourse analysis is a framework for examining the production of social reality, and follows the premise that multiple understandings about the world arise from the different dialogues we have throughout our individual life histories (Gee, 1999). This question and the consequent discourse analysis drew Clare to literature questioning, for example, whether positive findings were a "fair" representation of how the cancer patients really felt about their care because research participants with cancer tend to respond in socially desirable ways (OCallaghan & McDermott, 2007; Wilkinson & Kitzinger, 2000).

(d) Experienced researchers practice wisdom, that is, the tacit knowledge that informs practice, is now regarded as an important data font for developing professional knowledge (Schön, 1991; Scott, 1990) through research upon it (Mulder & Gregory, 2000). This led to an article illustrating the research of Clares practice wisdom, as evident in the analysis of her own reflections kept in a reflexive journal written throughout the research period. This culminated in a theme and clarifying statement. The theme follows:

In a cancer hospital, music therapy offers patients who engage in or overhear sessions, as well as visitors and staff, opportunities to encounter dynamic music spaces enabling altered intra-awareness and transient community participation, although music therapy was not appropriate for everyone. (OCallaghan, 2005, p. 225)

These findings were situated within varied literature, including psychodynamic theory, cultural studies, medical anthropology, medical ethnomusicology, and palliative care, to extend understanding about the positive effects of music and music therapy in cancer inpatient wards.

Their Stories

Effect of Music Therapy on Staff Bystanders

In the aforementioned research, 38 of the 61 staff who anonymously wrote their beliefs about music therapys relevance unexpectedly described personal benefits from being bystanders during sessions with patients and their families. (Clare had expected that they would discuss what they thought music therapy did for patients). After this, Lucanne Magill invited 62 staff to participate in research interviews examining music therapys effect on them as they conducted their work at Memorial Sloan Kettering Cancer Center, New York. Staff were theoretically sampled, meaning that staff expected to report both positive and negative feedback were interviewed. Notes on these interviews were thematically analyzed, as described earlier, and then compared with findings from Clares staff study in her PhD. Data saturation was discovered, negative cases uncovered (that is, findings that contradicted the emerging findings) and findings were repetitive, leading to Lucanne and Clares claim that the findings could inform a substantive grounded theory [9]. An abridged version of the theory was:

Staff witnessing music therapy can experience personally helpful emotions, moods, self-awarenesses, teamwork, and thus perceive improved patient care. Intrusive effects are uncommon (they included initial suspicion, audibility, and relaxation causing slowing of work pace). Music therapys benefits for staff are attributed to the presence of live music, the human presence of the music therapist, and the observed positive effects in patients and families.

These findings substantiated Lucanne and Clares claim that music therapy is valuable in oncology settings to support staff (OCallaghan & Magill, 2009). This is especially important given that staff appear to experience greater stress than their palliative care colleagues (Pierce, Dougherty, Panzarella, et al., 2007; Vachon, 2004) and that cancer center managers are requested to explore ways to enhance oncology staff support and reduce their work stress (Kash, Holland, Breitbart, et al., 2000).

Adolescent and Young Adult Music Therapy Group

In 2008 the onTrac Victorian Adolescent and Young Adult Cancer Service developed a multidisciplinary clinical research project aimed at providing young people with the opportunity to connect through a six-week song creation group. The group was co-facilitated by a music therapist and social worker. The aim was to examine the groups effect on participants self esteem and satisfaction, and their interpretations about its usefulness. Eight participants aged 19 to 25 years (4 male, 4 female) from five different treatment centers across Melbourne, Australia, attended the group. There was a median number of 6 participants per group. The group interventions included song sharing with discussion and group song writing. Data available for the thematic analysis included: (a) transcribed responses from the participants focus group "evaluation of the music therapy program" session (n = 7); (b) open-ended responses from a participant post-group questionnaire (n = 3); and (c) spontaneous email correspondence sent to the facilitators from the participants (n = 2). Thematic analysis revealed benefits, including: (a) connecting and expressing their self with a group who understands the cancer experience; (b) reducing loneliness, isolation, and risk of depression; and (c) song writing and the music allowed it to be a group not focusing on cancer. The focus group was a helpful way of capturing the young peoples voices and included the following comments:

Before this group I felt really alone, like just having gone through cancer and everything and all my friends, none of them obviously had anything... they could compare to it. Its just good to meet all of you guys, to, you know, get rid of that loneliness you know, being alone with cancer.

Yeah I think through having the music... it means that its not like a group where were all talking about how were sick (Barry, 2008b)

Adolescents need opportunities for authentic self-expression and supportive social interaction (McFerran-Skewes, 2003). A cancer diagnosis and required treatment, however, reduces time spent with peers which can have deleterious consequences for self esteem and identity development (Lewis. 1996). The use of music in therapeutic group work, with adolescents and young adults living with cancer, can enable these young people to comfortably express their experiences of cancer and connect with supportive peers who are undergoing comparable treatments.

Analysis of Therapeutic Products: Cancer Inpatients Song Lyrics for their Children

Another way of understanding music therapy participants experiences is through analyzing therapeutic products. Research into musical elements in clients improvisations with therapists is one approach (Aldridge & Aldridge, 2008). Another is analysing song lyrics emergent from therapeutic song writing. To examine potential communicative and therapeutic effects through helping cancer inpatients write songs for their children, Clare and three colleagues comparatively analyzed two song lyric groups (based on grounded theory). One group included 19 songs written by 12 patients in sessions with Clare, and the other group included 16 songs written by 15 patients with three other music therapists which were previously published or recorded for the public. These two song groups were analysed separately and then compared. An abridged final statement, representing the lyrics contained in the songs, composed by 20 mothers and 7 fathers for at least 46 offspring, follows:

Parents ... may convey their felt and enduring love and hopes for their children, and their metaphysical presence in the childrens lives. While many grieve with their children, some also look forward to, or yearn for, a shared life together. ... Parents may associate their children with miracles and lifes wonder, and describe, encourage, and compliment their childrens wholistic qualities and lifespan experiences. Parents can also describe how they find their childrens many qualities helpful .... While some parents apologize or request their children's optimism and forgiveness, many convey their personal reflections, including their existential and metaphysical beliefs ... (such as) being present now and after life. Parents sometimes offer their children supportive strategies in the songs. For young children, parents may write "playsongs" (which) can affirm, support, and encourage the youngsters, and include parents remembered and imagined depictions of their children. (OCallaghan, O'Brien, Magill, & Ballinger, 2009)

This statement, framed through developmental, attachment, grief, and neurobiological theories, substantiated the authors claims that parents song lyric messages may support and promote parent-child connectedness, the children during their parents illnesses, and the childrens developmental transitions and possible bereavement. Song writing may also help parents as they know that their children will hear what they want known, and that, through the songs, they may have a continuing attachment with them after death (OCallaghan, O'Brien, et al., 2009). This is especially important given service gaps in supporting parent-child communication during the parents cancer hospitalizations (Turner, Yates, Hargraves, & Hausmann, 2007) and, "woefully", during hospice care (Saldinger, Cain, Porterfield, & Lohnes, 2004).

Our Voices

Unfinished Legacy Work in Palliative Care

Earlier, we mentioned that an important knowledge font exists in the practice wisdom that exists in experienced therapists that can be researched to advance our profession. This font of knowledge can be examined both in individual therapists (described later) and across groups of music therapists (our voices) as will now be described.

In palliative care, legacy creation is encouraged to help patients cope with loss and find meaning, and the legacies may support the bereaved. There has been, however, no consideration about how staff deal with patients incomplete legacies when patients relocate, deteriorate, or die. Hence, four music therapists, with combined 53 years palliative care experience, questioned: What are four music therapists experiences of palliative care patients incomplete tangible music therapy legacies in their practices? A "reflexive group supervision research" method was developed which included a discussion guide. Two shared discussions were recorded and transcribed, which were aimed to promote each others examination of their actions, beliefs, and theories related to their work with incomplete legacies. Analysis was again informed by grounded theory and culminated in five themes about incomplete legacies, including: descriptions, reasons, feelings, suggestions for the "resolution of", and imaginings about legacy work. The research also led to further questions and inquiry into the ethical ramifications of incomplete legacies, including ownership and stewardship issues. Hence, through a seeming straightforward question arising from a music therapists quandary related to a clients unfinished legacy, research findings emerged to, hopefully, help all palliative care workers consider the ramifications of dealing with unfinished legacies in their work (OCallaghan, Peterson, Thomas, & Crappsley, 2009).

My Voice

Lullament

As experienced and professional health workers, we embody a site of data which can be harvested for music therapy knowledge, which is just as valid as knowledge emergent from other researchers examining our practice wisdom, or experimental findings offered to inform our future work. We can reflexively examine our own knowledge. There are many precedents for this in health professions, as outlined elsewhere (Barry & O'Callaghan, 2008; OCallaghan, 2008). One of Clares examples follows.

After more than 20 years of working as a palliative care music therapist, Clare pondered about the roles of lullabies and laments in her work. To examine this further, she developed two research questions: (a) what elicited her conception that lullaby and lament qualities were evident in her work; and (b) what were her beliefs about the relevance of these qualities for patients and their significant others (i.e., what was happening and was it helpful?). Data sources included: (a) a reflexive clinical journal of 12 months practice in a cancer hospital (300 000 words), created for previous research (OCallaghan, 2001a); (b) her publications, and two post graduate theses; and (c) further clinical vignettes, memories, and session artifacts. Informed by grounded theory, data was condensed into groups (called "supercategories") including lullabies and laments evidence; helpfulness; evidence and helpfulness; and ambivalence. A final statement ultimately emerged depicting lullaby and laments relevance in her work as a palliative care (inpatient) music therapist:

[abridged] Lullabies and laments are evident in music contextualized expressions of attachment and detachment; sadness/tears and happiness/laughter; privilege and loss; nurturance and grief; and deterioration, stasis, and moving forward. These seemingly paradoxical phenomena are often witnessed as juxtaposed and transitional and are evident in the "lullament", that is, when ones personal, socio-historical, and possibly archetypal relationship with lullabies and laments is actualized ... enabling resilience when vulnerable. A music therapist can enable the lullament experience through providing opportunities for music contextualized moments of "restorative resounding", articulated psychobiologically, verbally, musically, and metaphorically (but) ... may also need to be cautious of intensifying ones alienation and loss experience through negating peoples control over musics presence in their space (OCallaghan, 2008, pp. 97-8).

These findings were situated in literature discussing how ritualized music usage for health has declined with the emergence of hypothetically driven medicine (Biesele & Davis-Floyd, 1996) and informed the suggestion that palliative care music therapists may be reawakening and re-enabling musics helpful qualities to support people through uncertainty, loss, and mortal death transition (OCallaghan, 2008). This research experience also supported other authors statements that deeply reflecting upon ones clinical work in a focused manner can reinvigorate and, most likely, improve ones practice (Mulder & Gregory, 2000).

Researcher-therapist Partnerships: Supervised Voices

Until now we have examined how therapists can research their own practice, however therapists, especially those with minimal or no research experience, may choose to research their own practice in partnership with another more experienced therapist or academic. Such partnerships can be particularly useful between students and their clinical supervisors. For example, Carolyn Ayson (2008) who was supervised by Sarah Hoskyns during music training at the New Zealand School of Music, examined the role of a single music therapy session for supporting three hospitalized children and their parents, one of whom had the often life limiting condition, Friedereichs ataxia. Data included clinical notes, semi-structured interviews with parents and staff, and the students reflective journal. Thematic analysis indicated that sessions supported and promoted childrens normalization, and improved parents moods and their learning/parenting.

Students and supervisees can also use the method in the examination of personal thoughts about their placement and work experiences. Their voices, "mirrored" and extended through supervision, can valuably contribute to our knowledge. Philippa conducted a retrospective analysis on her student clinical reflexive journal, written during her cancer hospital music therapy placement with Clares supervision. This process revealed five key benefits of the reflexive journal writing process: (a) understanding contextual influences on practice; (b) connecting theory and practice; (c) self-evaluation and supervision; (d) practice development; and (e) understanding the usefulness of music therapy. Journal writing was illustrated as an educational tool for extending self-critique, integrating new insights into practice, and promoting reflexive evaluation (Barry & O'Callaghan, 2008). The findings substantiated Gentleman Byers & Forinashs (2004) claim that reflexive practice can inspire music therapists to confront challenging personal and professional experiences, deepen their understandings, and return to the field with transformed perceptions.

Discussion and Conclusion

One of the criticisms from those lauding quantitative research in palliative care is that qualitative researchers can be biased (Trauer, 2009) when examining their practices because we have a vested interest in uncovering positive findings. Qualitative researchers do not purport to be objective but, indeed, clarify that our work is situated, contextualized (Kuper, Reeves, & Levinson, 2008), and reflexive, that is, we constantly examine what we know and how we know it (Hertz, 1997, p. iix). Qualitative researchers continually self-analyse through every aspect of our research, and acknowledge potential personal and contextual effects on findings. We dont purport to uncover truth, but offer our interpretations of experiences, sometimes personal, and sometimes reflecting our interpretations of collective voices: including our patients, their families, and staff carers.

Qualitative researchers also ensure that their findings are rigorously derived and trustworthy. Methods to promote "interpretative rigour" (which differ to traditional notions of achieving validity and reliability in quantitative research) include inter-rater reliability, triangulation, member checking (Kitto et al, 2008), "thick description" (Geertz, 1973), and audit trails (Olesen, 2005). Audit trails were kept in all of the research described in this paper. Similarly, in all of the research except the lullament, qualitative inter-rater reliability was integrated. This means that textual data interpretations were shared with other "experts" (usually, the other authors in research mentioned here), validated or discussed further, and reworked to extend their complexity until all analysts were satisfied with the findings representations (Kitto et al, 2008). Member checking was integrated in the incomplete legacy research whereby all therapists re-read transcripts to verify them, and they also added relevant thoughts which were consequently integrated into the analysis (OCallaghan, Peterson, et al, 2009). Triangulated data sources (questionnaires and interviews) across two data collection sites were integrated into the bystander research (OCallaghan & Magill, 2009). Thick description was also integrated into the lullament (OCallaghan, 2008) and other (OCallaghan & Magill, 2009 [10]; OCallaghan, OBrien, et al, 2009) research, so that readers could imagine the contexts from which the findings emerged, and consider potential "conceptual generalizations" (Kitto et al, 2008) to their own contexts.

It is hoped that this description of practice-based research, from two experienced music therapists reflexive analyses of their work, will inspire others to consider practice-based research, on their own, or shared with other therapists, or in therapist-researcher partnerships. Finally, we conclude with three personal beliefs:

Knowledge harvested from "data fonts" existing within experienced therapists is just as valid as experimental findings anticipated to inform it.

Knowledge emergent from therapists who research their own work is just as valid as knowledge emergent from other researchers who examine the therapists work.

Many (albeit not all) cancer and palliative care journal reviewers are listening to our practice-based research work, and publishing it.

Hence, we urge music therapists to honor the wisdom contained within their clients, caregiver bystanders, and themselves, and to continue exploring creative ways to ethically and meaningfully sound these collective voices into our professional knowledge base.

Acknowledgement

The paper was enabled through Dr Clare OCallaghans Australian NHMRC Post Doctoral Fellowship in Palliative Care (2008-9).

Notes

[1]It is necessary to explain how quantitative and qualitative research will be described in this paper. In 2008, the British Medical Journal published a series of papers discussing and comparing the quantitative and qualitative research approaches (Kuper, Reeves, & Levinson, 2008). It was highlighted that quantitative research is now described as the "objectivist" approach, and qualitative research as the "constructivist" approach. Objectivist approaches focus on truth, replication, and predictability, while constructivist approaches endeavor to understand research participants subjective experiences and how varying perspectives about reality are informed by varied backgrounds. As "quantitative" and "qualitative" are arguably more familiar terms in the music therapy field, these terms will be used here.

[2]National Health and Medical Research Council of Australia.

[3]World Health Organization.

[4]This is a self-evaluation questionnaire that is comprised of scales for measuring anxiety (trait and state). Each scale consists of 20 self-descriptive statements and four point frequency scales, which the respondent circles to denote the best description of his or her state.

[5]A summary of cancer and palliative care music therapy research studies, from 1983 to 2009, is found in OCallaghan (2009).

[6]The National Health and Medical Research Council of Australias criteria for rating levels of objectivist evidence is based on the US Preventative Services Task Force Levels (Edwards, 2001) and include the following: Level 1, a systematic review of all relevant RCTs; Level 2, at least one properly designed RCT; Level 3, other comparative studies; and Level 4, case series with either post or pre-post testing (NHMRC, 2000).

[7]This kind of research on ones own clinical practice contains many components of ethnographic research, including reflexive journals and participant observation, which has precedence in other health and sociological fields.

[8]Qualitative, i.e., constructivist research can include quantitative findings to expand understandings, but the emphasis in on subjective experiences (Kuper et al, 2008).

[9]A grounded theory denotes a set of well developed codes, categories and themes that are "systematically interrelated through statements of relationship ... that explains some phenomenon .... even though ... theory may become outdated as new knowledge comes to light" (Corbin & Strauss, 2008, p. 55). Grounded theory encompasses three levels of theory development: substantive, middle range, and formal, and these theoretical levels represent decreasing levels of specificity to a group and/ or place. So, Lucanne Magill and Clare believed that their theory was substantive because it is based on two music therapists similar styles of practice (Lucanne Magill was Clares clinical student supervisor back in 1985). The cancer settings where they did the research have some cultural differences but are similar in many ways. To develop more middle range and formal grounded theories one would need to comparatively analyse more oncology music therapy programs throughout the world including different styles of music therapists practices and different contextual settings.

[10]Including references with more detailed descriptions of the authors music therapy practices.

References

Aldridge, G., & Aldridge, D. (2008). Melody in music therapy: A therapeutic narrative analysis. London: Jessica Kingsley Publishers.

Aoun, S. M., & Kristjanson, L. M. (2005). Challenging the framework for evidence in palliative care research. Palliat Med, 19, 461-465.

ATLAS.V.5.0. (2004). ATLAS.ti Scientific Software Development.Berlin.

Ayson, C. (2008). Child-Parent Wellbeing in a Paediatric Ward: The Role of Music Therapy in Supporting Children and Their Parents Facing the Challenge of Hospitalisation. Voices: A World Forum for Music Therapy. Retrieved August 1, 2009, from http://www.voices.no/mainissues/mi40008000261.php

Barry, P., & O'Callaghan, C. (2008). Reflexive journal writing: A tool for music therapy student clinical practice development. Nordic Journal of Music Therapy, 17(1), 55-66.

Barry, P. (2008a). Effects of music therapy CD creation on paediatric patients first radiotherapy treatment. Unpublished MMus Thesis. The University of Melbourne, Melbourne.

Barry, P. (2008b). Developing a music therapy service for adolescent and young adult cancer patients: Age, gender, and cultural issues. Paper presented at the Australian Music Therapy Association National Conference, Brisbane, September 20th.

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.

Biesele, M., & Davis-Floyd, R. (1996). Dying as a medical performance: The oncologist as charon. In C. Laderman & M. Roseman (Eds.), The performance of healing. (pp. 291-322). New York: Routledge.

Boog, K. M., & Tester, C. Y. (2008). Palliative care: A practical guide for the health professional. Edinburgh: Churchill Livingstone.

Coats, W. J. Jr. (2003). Political theory and practice: eight essays on a theme. Cranbury, New Jersey: Associated University Press.

Clark, D. (1997). What is qualitative research and what can it contribute to palliative care? Palliative Medicine, 11, 159-166.

Corbin, J., & Strauss, A. (2008). Basics of qualitative research 3e: Techniques and procedures for developing grounded theory. Thousand Oaks, California: Sage.

Coyle, N. (2006). The hard work of living in the face of death. J Pain Symptom Manage, 32(3), 266-274.

Daly, J., Willis, K., Small, R., Green, J., Welch, N., Kealy, M., & Hughes, E. (2007). A heirarchy of evidence for assessing qualitative health research. Journal of Clinical Epidemiology, 60, 43-49.

Dileo, C., & Dneaster, D. (2005). Introduction: State of the art. In C. Dileo & J. Loewy (Eds.), Music therapy at the end of life (pp. xix-xxvii). Cherry Hill, NJ: Jeffrey Books.

Dileo, C., & Magill, L. (2005). Songwriting with oncology and hospice adult patients from a multicultural perspective. In F. Baker & T. Wigram (Eds.), Songwriting: Methods and clinical applications for music therapy clinicians, educators and students (pp. 226-245). London: Jessica Kingsley Publishers.

Doyle, D., Hanks, G., Cherny, N., & Calman, K. (2004). Introduction. In D. Doyle, G. Hanks, N. Cherny & K. Calman (Eds.), Oxford textbook of palliative medicine (3rd ed., pp. 1-4). Oxford: Oxford University Press.

Edwards, J. (2001). Evidence based medicine and music therapy. Network: A forum for discussion amongst registered music therapists, June, 2-3.

Epstein, I. (2001). Using available clinical information in practice-based research: Mining for silver while dreaming of gold. In I. Epstein & S. Blumenfield (Eds.), Clinical data-mining in practice-based research. (pp. 15-32). Binghampton, N. Y.: Hamworth Social Work Practice Press.

Gee, P. J. (1999). An introduction to discourse analysis: Theory and method. London: Routledge.

Geertz, C. (1973). The interpretation of cultures. New York: Basic Books.

Gentleman Byers, J. J., & Forinash, M. (2004). Educators, therapists, and artists on reflective practice. New York: Peter Lang Publishing.

Glaser, B., & Strauss, A. (1967). The discovery of grounded theory. New York: Aldine.

Guba, E. G., & Lincoln, Y. S. (1994). Competing paradigms in qualitative research. In N. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 105-117). Newbury Park: Sage.

Health Evidence Network (2004). What are palliative care needs of older people and how might they be met? World Health Organisation. Retrieved 18th February, 2009, from http://www.euro.who.int/HEN/Syntheses/palliative/20040723_1

Hertz, R. (1997). Introduction: Reflexivity and voice. In R. Hertz (Ed.), Reflexivity and voice. (pp. vii-xviii). Thousand Oaks, CA: Sage.

Hogan, B. (1999). 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 Pub Ltd.

Kain, Z., Caldwell-Andrews, A., & Krivutza D. M. (2004). Interactive music therapy as treatment for preoperative anxiety in children: a randomized control trial. Anesth Anal 98, 1260-1266.

Kash, K., Holland, J., Breitbart, W., Berenson, S., Doughtery, J., Quellette-Kobasa, S., et al. (2000). Stress and burnout in oncology. Oncol Nurs Forum, 14(11), 1621-1633.

Keeley, D. (1999). Rigorous assessment of palliative care revisited. Wisdom and compassion are needed when evidence is lacking. BMJ, 319(1447-48).

Kissane, D. (2004). Bereavement. In D. Doyle, G. Hanks, N. Cherny & C. K (Eds.), Oxford Textbook of Palliative Medicine (3rd ed., pp. 1135-1151). Oxford: Oxford University Press.

Kitto, S. C., Chesters, J., & Grbich, C. (2008). Criteria for authors in the submission and assessment of qualitative research articles for the Medical Journal of Australia, 188(4), 243-246.

Kuper, A., Reeves, S., & Levinson, W. (2008). Qualitative research: An introduction to reading and appraising qualitative research. BMJ, 337, 404-407.

Kvale, K., & Bondevik, M. (2008). What is important for patient centred care? A qualitative study about the perceptions of patients with cancer. Scand J Caring Sci, 22, 582-589.

Levitin, D. (2006). This is your brain on music. London: Atlantic.

Lindenfelser, K. (2005). Parents' voices supporting music therapy within pediatric palliative care. Voices: A World Forum for Music Therapy. Retrieved October 27, 2008, from http://www.voices.no/mainissues/mi40005000194.html.

McFerran-Skewes, K. (2003). Contemplating the nature of adolescent group improvisations. Voices: A World Forum for Music Therapy. Retrieved October 28, 2009, from https://normt.uib.no/index.php/voices/article/view/133/109

McGrath, P. (2000). Confronting Icarus: A psychosocial perspective on haematological malignancies. Aldershot, UK: Ashgate.

Mulder, J., & Gregory, D. (2000). Transforming experience into wisdom: Healing amidst suffering. J Palliat Care, 16(2), 25-29.

NHMRC. (2000). How to use the evidence: assessment and application of scientific evidence. Retrieved 20th July, 2009, from http://www.nhmrc.gov.au/publications/synopses/_files/cp69.pdf

NHMRC. (2007). National statement on ethical Conduct in research involving humans. Retrieved 19th February, 2009, from http://www.nhmrc.gov.au/publications/humans/part4.htm

O'Brien, E. (2003). Living soul [CD]. Melbourne: Crystal Mastering.

O'Callaghan, C. (1984). Musical profiles of dying patients. Australian Music Therapy Association Bulletin, 7(2), 5-11.

O'Callaghan, C. (1996). Lyrical themes in songs written by palliative care patients. J Music Ther, 33(2), 74-92.

O'Callaghan, C. (2001a). Music therapy's relevance in a cancer hospital researched through a constructivist lens. PhD Thesis, The University of Melbourne, Melbourne. http://eprints.infodiv.unimelb.edu.au/archive/00003827

O'Callaghan, C. (2001b). Bringing music to life: Music therapy and palliative care experiences in a cancer hospital. J Palliat Care, 13(3), 155-160.

OCallaghan, C. (2004). Music therapy in palliative care. In G. Hanks, N. MacDonald, N. Cherny & K. Calman (Eds.), The Oxford textbook of palliative medicine (3rd ed., pp. 1041-1046). Oxford: OUP.

O'Callaghan, C. (2005). Qualitative data-mining through reflexive journal analysis: Implications for music therapy practice development. Journal of Social Work Research and Evaluation, 6(2), 219-231.

O'Callaghan, C. (2008). Lullament: Lullaby and lament therapeutic qualities actualized through music therapy. Am J Hosp Palliat Care, 25(2), 93-99.

O'Callaghan, C. (2009). Objectivist and constructivist music therapy research in oncology and palliative care: An overview and reflection. Music and Medicine, 1(1), 41-60.

O'Callaghan, C., & Hiscock, R. (2007). Subgroup analysis extends modified grounded theory research findings in oncologic music therapy. J Music Ther, 64(3), 256-281.

O'Callaghan, C., & Magill, L. (2009). Effect of music therapy on oncologic staff bystanders: A substantive grounded theory. Palliative and Supportive Care, 7(2), 219-28.

OCallaghan, C., & McDermott, F. (2004). Music therapys relevance in a cancer hospital researched through a constructivist lens. J Music Ther, 41(2), 151-185.

O'Callaghan, C., & McDermott, F. (2007). Discourse analysis reframes oncologic music therapy research findings. Arts Psychother, 34, 398-408.

O'Callaghan, C., O'Brien, E., Magill, L., & Ballinger, E. (2009). Resounding attachment: Cancer inpatients' song lyrics for their children in music therapy. Support Care Cancer, 17(9), 1149-57.

O'Callaghan, C., Petering, H., Thomas, A., & Crappsley, R. (2009). Dealing with palliative care patients incomplete music therapy legacies: Reflexive group supervision research. J Palliat Care, 25(3), 197-205.

O'Callaghan, C., Hornby, C., Pearson, E., & Ball, D. (in press). "The moment is all we have": patients and visitors reflect on a staff Christmas choir. Med J Australia.

Olesen, V. (2005). Early millennial feminist qualitative research: Challenges and contours. In N. K. Denzin & Y. S. Lincoln (Eds). The Sage handbook of qualitative research (3rd ed., pp. 235-78). Thousand Oaks, California: Sage Pub. Inc.

Pierce, B., Dougherty, E., Panzarella, T., Le, L., Rodin, G., & Zimmerman, C. (2007). Staff stress, work satisfaction, and death attitudes on an oncology palliative care unit, and on a medical and radiation oncology inpatient unit. J Palliat Care, 23(1), 32-39.

Rice, P. L., & Ezzy, D. (1999). Qualitative research methods. Oxford: Oxford University Press.

Saldinger, A., Cain, A., Porterfield, K., & Lohnes, K. (2004). Facilitating attachment between school-aged children and a dying parent. Death Studies, 28, 915-940.

Salmon, D. (1995). Music and emotion in palliative care: Assessing inner resources. In C. A. Lee (Ed.), Lonely waters: Proceedings of the International Conference, Music Therapy in Palliative Care (pp. 71-84). Oxford: Sobell House.

Schön, D. A. (1991). The reflective turn: Case studies in and on educational practice (2nd ed.). New York: Teachers College Press.

Scott, D. (1990). Practice wisdom: The neglected source of practice research. Social Work, 35(6), 564-568.

Spielberger, C. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists.

Spirito, A., Stark, L. J., & Williams, C. (1988). Development of a brief coping checklist for use with pediatric populations. J Pediatr Psychol, 13(4), 555-574.

Trauer, T. (2009, July 17th). Overview of principles of quantitative research. Paper presented at the Centre for Palliative Care Research and Education Seminar: Evidence based practice in palliative care. The University of Melbourne.

Turner, J., Yates, P., Hargraves, M., & Hausmann, S. (2007). Development of a resource for parents with advanced cancer: What do parents want? Palliative and Supportive Care, 5, 135-145.

Vachon, M. (2004). The stress of professional caregivers. In G. Hanks, N. MacDonald, N. Cherny & K. Calman (Eds.), Oxford textbook of palliative medicine (pp. 992-1004). Oxford: OUP.

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

WHO. (2008). WHO definition of palliative care [Electronic Version]. Retrieved 17th November, 2008 from http://www.who.int/cancer/palliative/definition/en/

Wilkinson, S., & Kitzinger, C. (2000). Thinking differently about thinking positive: A discursive approach to cancer patients' talk. Soc Sci Med, 50, 797-811.