Music Therapy for Pain in Black and White Cancer Patients: A Retrospective Study

Abstract

Context:

Racial pain disparities present challenges to cancer symptom management. Music therapy has demonstrated benefits for pain and is a promising treatment option for diverse populations due to music’s multicultural presence. However, Black cancer patients are under-represented in music therapy trials.

Purpose:

This study compared pain severity, treatment approaches, and responses to music therapy between Black and white cancer patients. The findings will be used to generate hypotheses for future music therapy research to address racial disparities in pain management.

Methods:

We conducted a retrospective program evaluation of Black and white patients who received music therapy at an NCI-Designated Comprehensive Cancer Center. We used the Edmonton Symptom Assessment Scale (ESAS) to assess pain. We abstracted opioid use, music therapy referral reasons, and treatment approaches from the electronic health record.

Results:

Among 358 patients, 18% were Black, 42% reported moderate-to-severe pain, and 47% received opioids. Black patients reported higher baseline pain than white patients, but similar proportions of Black and white patients received opioids. Greater proportions of Black patients received music therapy referrals for pain (73% vs. 56%, P=0.04) and engaged in active techniques (92% vs. 82%, P=0.04). Black and white patients reported clinically meaningful pain reduction of similar magnitude after music therapy. Black patients discussed spirituality more commonly during music therapy, whereas white patients focused on family bonds.

Conclusions:

Black and white patients reported clinically meaningful pain reduction through varying music therapy approaches. Our findings may help inform cultural adaptations of music therapy to address racial pain disparities in oncology.

INTRODUCTION

Pain affects more than two-thirds of hospitalized cancer patients and is associated with increased risk of opioid overdose, inpatient mortality, longer hospital stay, and higher hospital charges.^1,2 Despite an increased focus on symptom management in oncology, recent studies indicate over a third of cancer patients remain under-treated for pain.^3,4

Racial disparities in oncological pain management are well documented,^5,6 with one study demonstrating that, compared to non-Hispanic white patients, the odds of receiving inadequate analgesic prescriptions was twice as high for patients from other racial groups.⁶ Race is a social construct that has been shown to affect health outcomes through socioeconomic, environmental, and other structural determinants.⁷ Indeed, Black cancer patients have historically experienced barriers to pharmacological pain management.^5,6 Recent studies found that Black patients had 25% lower odds of receiving an opioid prescription and received opioid doses that were 36% lower than those prescribed to white patients.^8,9 Growing research suggests these racial disparities stem from the complex interplay of patient-, provider-, and system-related factors.¹⁰ Interviews with Black cancer patients revealed attitudinal barriers to pain management, including fear of developing tolerance or becoming addicted to opioids, as well as reluctance to discuss pain with their healthcare providers.¹¹ Prior studies also suggested that providers tend to under-estimate the pain severity of Black cancer patients¹² and over-estimate their risk of opioid misuse.¹³ These implicit biases have been shown to influence the opioid prescribing approaches of healthcare providers, resulting in less frequent opioid prescription renewals for Black patients with metastatic cancer pain relative to white patients.¹⁴ In addition to these patient- and provider-related factors, cancer pain management may be hindered by systemic factors, such as limited availability of pain specialists and low pharmacy opioid stocks in neighborhoods with predominantly non-white racial groups.^15–17 These patient-, provider-, and system-related barriers, coupled with the opioid crisis, have sparked growing interest in the role of non-pharmacological interventions to address racial pain disparities in oncology.

Music therapy is a non-pharmacological, evidence-based intervention, in which board-certified music therapists engage patients in personally tailored experiences with music to achieve therapeutic goals.¹⁸ These experiences range from receptive activities (e.g., guided music listening, music-based relaxation) to more active forms of engagement (e.g. singing, playing an instrument, composing lyrics, creating playlists).¹⁸ Music therapy has long been used for pain management in clinical practice,^18–20 and mounting evidence suggests that musical engagement may modulate key neurochemical systems (e.g., mesolimbic) implicated in pain.^21,22 In a recently updated Cochrane review, music therapy was associated with significant pain reduction among cancer patients (standardized mean difference −0.47, 95% CI −0.86 to −0.07, P = 0.02).²³ Due to this growing evidence, the clinical guidelines from the American Society of Clinical Oncology (ASCO),²⁴ the National Comprehensive Cancer Network (NCCN),²⁵ and the Society for Integrative Oncology (SIO)²⁶ recommend music therapy as a treatment option for pain. Music therapy has also become increasingly available to cancer patients. From 2009 to 2016, the number of National Cancer Institute (NCI)-designated Comprehensive Cancer Centers that provided information about music therapy on their websites increased by 31%.²⁷ Currently, music therapy is offered at 75% of NCI-designated Comprehensive Cancer Centers and 55% of community cancer hospitals.²⁸ Ethnographic and phylogenetic research has identified music as a unique characteristic of humankind across various cultures and societies around the world,^29,30 suggesting that music therapy has the potential to resonate culturally with diverse cancer populations.³¹

Despite the multicultural presence of music and the growing evidence base and availability of music therapy, most music therapy trials have not included racially diverse cancer populations. In a recent Cochrane meta-analysis of 81 trials with a total of N=5576 cancer patients, 31 trials failed to report racial demographic information; among the trials that reported this information, only 10% of participants identified as Black.²³ While other trials have examined music therapy for pain management in predominantly Black populations, these studies were conducted in patients without cancer.^32,33 To address this gap in knowledge, we conducted a retrospective study to compare the pain severity, as well as the treatment approaches and responses, of Black and white hospitalized cancer patients who received music therapy as part of their routine inpatient care. The goal of this exploratory study is to generate hypotheses for future music therapy research to address racial pain disparities in oncology.

METHODS

Study Design, Setting, and Participants.

This study was a retrospective program evaluation of single, initial sessions of music therapy delivered between February 2018 and October 2019. Music therapy recipients were self-identified Black or white adult cancer patients hospitalized at Memorial Sloan Kettering Cancer Center (MSK), a tertiary NCI-designated Comprehensive Cancer Center. MSK had 22,792 adult inpatient admissions in 2018 and 24,175 adult inpatient admissions in 2019. Integrative medicine services (i.e. music therapy, massage therapy, acupuncture) were available to hospitalized patients for symptom management. Medical providers (e.g. oncologists, physician assistants, registered nurses, social workers) could request music therapy for their patients by placing referrals in the electronic health record and indicating the referral reason (e.g. pain, anxiety). Although most music therapy referrals were placed in the electronic health record, some providers contacted music therapists who were rotating through the inpatient wards and verbally requested music therapy for their patients without placing an electronic referral. Music therapists documented their treatment sessions in the electronic health record and asked patients to complete assessments via tablet immediately before and after each session. This retrospective study protocol was reviewed and approved by the MSK institutional review board (IRB#18-445).

Interventions.

Music therapy sessions were delivered at bedside by board-certified music therapists with over five years of clinical experience. Treatment sessions lasted approximately 20–30 minutes. Music therapists used a variety of therapeutic tools, such as harmonic instruments (e.g., guitar, keyboard, harp), percussive instruments (e.g., drums, xylophones, bells, shakers, claves), and/or music-based technology (e.g., digital instrument apps). Music therapy interventions included receptive (e.g., listening, music-guided relaxation) and active (e.g., singing, improvising, playing instrument, song-writing) techniques.^18,34 Music therapists selected appropriate techniques based on their clinical judgment, taking into consideration the referral reason, as well as the symptom burden, co-morbidities, medical contraindications, cultural background, and musical preferences of each patient.^18–20 In addition to the type of therapeutic techniques used (i.e., receptive, active), music therapists typically documented treatment goals (e.g., clinical outcomes, skills they were seeking to cultivate), session content (e.g. topics discussed, themes or genres of songs performed), and notable responses and reactions from patients.

Outcomes.

Patients completed the Edmonton Symptom Assessment Scale (ESAS)³⁵ pain item pre- and post-treatment using a secure electronic form on a tablet. The ESAS pain item is a validated single-item instrument that asks patients to use an 11-point numeric rating scale to indicate how they currently feel (0 = no pain, 10 = worst possible pain). A score ≥4 indicates moderate-to-severe pain.³⁵ A one-point difference on an ESAS item represents the minimal clinically important difference; thus, a one-point or greater reduction in the ESAS pain item score is considered clinically meaningful improvement in pain.³⁵ The ESAS pain item has been used in prior research to evaluate the effects of music therapy on pain in an inpatient cancer population.³⁶

Statistical Analysis.

We abstracted socio-demographic characteristics and cancer-related variables from the electronic health record. We also abstracted opioid use, music therapy referral reasons, and music therapy techniques from the electronic health record. Opioid use was categorized as “yes” or “no” based on whether or not patients were administered an opioid in the seven days preceding the music therapy session. Music therapy referral reasons were categorized as “pain” or “non-pain” (i.e., anxiety). Music therapy techniques were categorized as “receptive” (e.g., listening, music-guided relaxation) or “active” (e.g., singing, improvising, playing instrument, song-writing). We used chi-squared tests to examine racial differences in opioid use, music therapy referral reasons, and the type of music therapy techniques used. The analyses of referral reasons were limited to patients who had a referral placed in the electronic health record. We used student’s t-test to examine pre-treatment differences in ESAS pain score by race. We also conducted paired t tests to evaluate pre- and post-treatment change in ESAS pain scores by race. To evaluate the differences in post-treatment ESAS pain scores between Black and white patients, we used linear regression models, adjusting for pre-treatment ESAS pain scores. Statistical significance was set at P<0.05. All analyses were completed in STATA (Windows version 15.0, StataCorp LLC, College Station, TX).

We performed a detailed chart abstraction of clinical notes from music therapists who treated patients with moderate-to-severe pain (ESAS pain score ≥4). Since analyses of ESAS scores were limited to initial music therapy sessions, we also limited the chart abstraction to clinic notes from initial music therapy sessions. The music therapist notes of all Black patients (N=35) with moderate-to-severe pain were selected for chart abstraction. Given that there were substantially more white patients in our study sample, we used STATA to randomly select an equivalent number of white patients (N=35) with moderate-to-severe pain and performed detailed chart abstraction of music therapist notes from this random sample. Two researchers performed chart abstraction independently to ensure consistency and accuracy; chart abstraction was not performed by any of the music therapists who delivered the interventions. We abstracted information from the key sections of the music therapist note (i.e., treatment goals, session content, and patient responses) and compared the findings between Black and white patients.

RESULTS

From February 2018 to October 2019, 358 patients received music therapy; 64 (17.9%) were Black, and 294 (82.1%) were white. Table 1 summarizes the socio-demographic and clinical characteristics of the study population. Among 358 patients, 150 (41.9%) reported moderate-to-severe pain (ESAS pain score ≥4) prior to receiving music therapy. In the seven days preceding the music therapy session, 169 (47.2%) patients received an opioid.

Table 1.

Patient Characteristics (N=358).

Age
Mean (SD), years	62.0 (13.7)
Median (Range), years	63.5 (21.6–93.8)
Race
White, N (%)	294 (82.1)
Black, N (%)	64 (17.9)
Gender
Female, N (%)	244 (68.2)
Male, N (%)	114 (31.8)
Cancer Type
Hematologic, N (%)	114 (31.8)
Digestive, N (%)	75 (21.0)
Gynecological, N (%)	37 (10.3)
Genitourinary, N (%)	24 (6.7)
Breast, N (%)	23 (6.4)
Respiratory, N (%)	23 (6.4)
Soft Tissue, N (%)	21 (5.9)
Skin, N (%)	10 (2.8)
Multiple Primaries, N (%)	5 (1.4)
Other/Unknown, N (%)	26 (7.3)
Cancer Stage
Localized, N (%)	51 (14.3)
Regional, N (%)	45 (12.6)
Metastatic, N (%)	95 (26.5)
Unstaged/Unknown, N (%)	167 (46.7)

Table 2 presents a comparison of Black and white patients with regards to pre-treatment pain severity, opioid use, music therapy referral reasons, and the type of music therapy techniques received. Black patients reported significantly higher pre-treatment pain severity than white patients (mean ESAS pain score, 4.2 vs 3.1, P=0.006). Compared with white patients, a greater proportion of Black patients reported moderate-to-severe pain prior to receiving music therapy (54.7% versus 39.1%, P=0.02). However, in the seven days preceding the music therapy session, the proportion of Black patients who received an opioid was similar to the proportion of white patients who received an opioid (54.7% versus 45.6%, P=0.19).

Table 2.

Pain Severity, Opioid Use, and Music Therapy Referrals and Techniques by Race.

Variable	Black	White	P-value
Pre-Treatment Pain, ESAS Score, Mean (SD)	4.2 (2.8)	3.1 (2.8)	0.006
Proportion with ESAS Pain Score ≥4, N (%), Total N=64 Black Patients and Total N=294 White Patients	35 (54.7%)	115 (39%)	0.02
Proportion Receiving Opioid in Seven Days Preceding Music Therapy Session, N (%), Total N=64 Black Patients and Total N=294 White Patients	35 (55%)	132 (46%)	0.19
Proportion Referred to Music Therapy for Pain, N (%), Total N=44 Black Patients and Total N=155 White Patients with Electronic Referrals*	32 (73%)	86 (56%)	0.04
Proportion Receiving Active Music Therapy Techniques, N (%), Total N=64 Black patients and Total N=294 White Patients	59 (92%)	240 (82%)	0.04

^*N=159 patients did not have a referral placed in the electronic health record.

Among the 358 music therapy recipients, 118 (33.0%) were referred for pain, 81 (22.6%) were referred for non-pain reasons, and 159 (44.4%) did not have a referral in the electronic health record and were instead referred by other means (i.e., verbal request). Compared with white patients, a greater proportion of Black patients received a music therapy referral for pain (72.7% versus. 55.5%, P=0.04). Among the 358 music therapy recipients, 299 (83.5%) received active techniques. Compared with white patients, a greater proportion of Black patients received active techniques (92% versus 82%, P=0.04).

Patients who received music therapy reported significantly lower post-treatment ESAS pain scores relative to pre-treatment scores (Table 3). The mean ESAS pain score of Black patients decreased from 4.2 (SD 2.9) to 2.9 (SD 2.4), a clinically meaningful reduction of 1.3 points (95% CI 0.8 to 1.8, P<0.001). The mean ESAS pain score of white patients decreased from 3.1 (SD 2.8) to 2.2 (SD 2.4), a statistically significant, although not clinically meaningful, reduction of 0.9 points (95% CI 0.7 to 1.1, p<0.001). After adjusting for pre-treatment score in regression analyses, the mean post-treatment ESAS scores did not differ significantly by race (−0.03, 95% CI −0.4 to 0.3, P=0.89).

Table 3.

Pre- and Post-Treatment Change in ESAS Pain Scores by Race.

Sample	Race	Pre-Treatment, Mean (SD)	Post-Treatment, Mean (SD)	Pre-Post Change (95% CI)	P value
All Patients (N=358)	Black (N=64)	4.2 (2.9)	2.9 (2.4)	1.3 (0.8 to 1.8)	<0.001
	White (N=294)	3.1 (2.8)	2.2 (2.4)	0.9 (0.7 to 1.1)	<0.001
Patients with Moderate-to-Severe Pain (N=150)	Black (N=35)	6.4 (1.9)	4.3 (2.3)	2.1 (1.3 to 2.9)	<0.001
	White (N=115)	6.0 (1.9)	4.3 (2.4)	1.8 (1.4 to 2.1)	<0.001

When analyses were restricted to patients with moderate-to-severe pre-treatment pain, the post-treatment ESAS pain scores were also significantly lower relative to pre-treatment scores (Table 3). The mean ESAS pain score of Black patients decreased from 6.4 (SD 1.9) to 4.3 (SD 2.3), a clinically meaningful reduction of 2.1 points (95% CI 1.3 to 2.9, P<0.001). The mean ESAS pain score of white patients decreased from 6.0 (SD 1.9) to 4.3 (SD 2.4), a clinically meaningful reduction of 1.8 points (95% CI 1.4 to 2.1, P<0.001). After adjusting for pre-treatment score in regression analyses, the mean post-treatment ESAS scores did not differ significantly by race (−0.2, 95% CI −1.0 to 0.5, P=0.52).

Of the patients who were selected for detailed chart abstraction, two (5.7%) of the 35 Black patients and three (8.6%) of the 35 white patients were missing the clinical note for the initial music therapy session. Therefore, we abstracted data from the clinical notes of 33 Black patients and 32 white patients. Regarding the treatment goals of music therapy sessions, self-expression was more commonly documented among Black patients (54.5% of Black patients vs. 40.6% of white patients), whereas relaxation was more commonly documented among white patients (65.6% of white patients vs. 45.5% of Black patients). With regards to the content of music therapy sessions, spirituality was more commonly documented among Black patients (39.4% of Black patients vs. 3.1% of white patients), whereas family bonds were more commonly documented among white patients (40.6% of white patients vs. 18.2% of Black patients). In terms of patient responses, gratitude (75.8% of Black patients versus 53.1% of white patients) and brighter affect (78.8% of Black patients vs. 46.9% of white patients) were more commonly documented among Black patients, whereas relaxed (37.5% of white patients vs. 12.1% of Black patients) and reflective (46.9% of white patients vs. 33.3% of Black patients) were more commonly documented among white patients (Figure 1).

Figure 1.

Commonly Documented Music Therapy Approaches and Responses Among Black and White Cancer Patients with Moderate-to-Severe Pain.

DISCUSSION

In this retrospective study of 358 hospitalized adults with cancer, both Black and white patients reported a similar magnitude of pain reduction after receiving music therapy. The observed variations in the treatment goals, session content, and responses of Black and white patients suggest that personalization and cultural adaptation of music therapy, rather than a one-size-fits-all approach, may be necessary to achieve equitable pain outcomes in racially diverse cancer populations.

A large body of literature has documented the unmet pain management needs of cancer patients.^2–4 Nearly half of our study population received opioids, but 40% still reported moderate-to-severe pain. These findings highlight that under-treatment of pain remains a challenge in cancer symptom management. Although Black patients reported higher pain severity at baseline relative to white patients, a similar proportion of Black and white patients received opioids, echoing the findings from prior research that Black patients may encounter difficulties obtaining adequate doses of opioids and other analgesics.^8,9

While Black patients have historically experienced patient-, provider-, and healthcare system-related barriers to pharmacologic pain management,¹⁰ our study suggests that music therapy may be viewed as an acceptable non-pharmacological option, not only by Black patients, but also by their healthcare providers. We found that providers referred a higher proportion of Black patients than white patients to music therapy for pain management. While the reasons underlying these referral patterns require further study, our findings suggest that inpatient providers are willing to use music therapy to treat pain. This stands in contrast to prior research that documented reluctance among providers to prescribe opioids and analgesics to Black patients, which oftentimes stems from implicit bias and false perceptions of Black patients.^13,14 Further, a higher proportion of Black patients received active music therapy techniques compared to white patients, indicating a willingness to engage with music beyond receptive listening. In contrast to prior research documenting “pharmacy deserts” in minority-predominant neighborhoods,^15–17 a preliminary study suggests that the availability of music therapy at community cancer hospitals does not differ by the racial demographics of their catchman areas.²⁸ Taken together, these findings support the accessibility of music therapy, as well as its acceptability for pain management among both Black patients and their providers.

Both Black and white cancer patients with moderate-to-severe pain reported statistically significant, clinically meaningful reductions in pain severity after music therapy. The magnitude of pain reduction was similar to the results observed in other studies of hospitalized cancer patients who received music therapy.³⁶ Notably, Black patients reported similar improvements in their pain symptoms as white patients. These findings stand in contrast to prior research of pain interventions, which demonstrated minimal or fewer benefits for Black patients compared with white patients.^37,38 Given that music is found in various societies around the world,^29,30 music therapy has unique potential as a culturally affirming practice for diverse cancer patients.³¹ Indeed, certain music therapy approaches, particularly resource-oriented frameworks, focus on nurturing individuals’ strengths, capabilities, and cultural identities, rather than correcting weaknesses or imposing predetermined health views and practices.³⁹

Consistent with this resource-oriented framework, most music therapy sessions of Black patients in our study focused on facilitating self-expression. This approach echoes the historical role of music within Black communities. Work songs and spirituals, coined by W.E.B. DuBois as “Sorrow Songs,” often functioned as safe containers for airing grievances and voicing shared hopes for the “ultimate justice of things.”⁴⁰ Others have also argued that jazz and the blues have roots in the creative expressions of freedom or longing for freedom among Black artists.^41,42 These musical outlets provided an escape from oppressive systems that discourage the sharing of Black perspectives, voices, and lived experiences. Systemic oppression may foster stoicism, causing Black patients to hide, ignore, or accept pain, rather than discuss it openly with loved ones or healthcare providers.^43,44 Our findings suggests that music therapy can tap into the cultural resources of Black communities and provide a safer, more accessible medium for processing and discussing Black experiences with pain.³¹

Spirituality was another common topic of discussion during music therapy sessions with Black patients. This finding builds on prior research demonstrating that Black patients rely on faith to cope with cancer pain.^11,44 A recent study found that spirituality was linked to decreased cancer pain and symptom burden among Black patients.⁴⁵ Others have identified music and spirituality as the key mechanisms behind the healing qualities of Black churches.⁴⁶ Since spirituality and religion are integral parts of culture,⁴⁷ music therapy may offer a culturally resonant approach to addressing pain among Black patients. As pain management shifts towards a team-based, multi-disciplinary model, our findings highlight the potential for collaboration between music therapists and chaplains to reduce disparities and improve pain outcomes among Black cancer patients.^48,49 Future research is needed on the personalization and cultural adaptation of music therapy for diverse cancer populations.⁵⁰

Finally, our abstraction of music therapist charts revealed a wide range of patient responses, including gratitude and brighter affect among Black patients and relaxation and reflection among white patients. These preliminary findings are consistent with the growing evidence that music therapy affects other physical or psychological outcomes besides pain.²³ Since pain often co-occurs with other symptoms and is influenced by emotional and thinking processes,^51,52 future research should investigate the capacity of music therapy to target pain through mood, cognition, or other pathways. Understanding how pain is experienced differently by people of diverse backgrounds will also help tailor music therapy to individuals and their unique coping styles, cognitive tendencies, and emotional dispositions.^31,50

Given that this study was a retrospective program evaluation, the findings should be interpreted as associations, rather than causal relationships, and the results may have been affected by confounders that are not captured by our evaluation methods. Additionally, we used a validated, single-item ESAS score, rather than a more comprehensive instrument, to assess pain. Since patients were referred by clinicians, selection bias may have influenced the findings. It is also possible that the personal biases of music therapists may have influenced their treatment approaches, although the incorporation of patient feedback and preferences could theoretically help to mitigate this. Further, we only examined the immediate pre- and post-treatment change after a single, initial session; thus, the durability of treatment effects and the cumulative impact of multiple treatments remains unknown. We are also unable to assess how the treatment approaches may evolve over time as the therapist-patient relationship develops. Finally, this study was conducted at a tertiary academic cancer center, so the generalizability of findings may be limited.

Despite these limitations, our study represents an important step towards understanding the diverse range of music therapy experiences among Black and white cancer patients with pain. Music therapy is recommended for pain management in the clinical guidelines of leading cancer organizations.^24–26 However, Black patients are often drastically under-represented in the trials that form the evidence base of the clinical guidelines.²³ Our study may help inform future research on the cultural adaptation of music therapy to improve pain outcomes and reduce disparities in racially diverse cancer populations.

KEY MESSAGE:

In this retrospective study of hospitalized cancer patients, music therapy was associated with clinically meaningful pain reduction among Black and white patients. Music therapists more commonly incorporated spirituality with Black patients compared to white patients, suggesting that culturally attuned treatments are needed to address racial pain disparities in oncology.