Racial differences in familiarity, interest, and use of integrative medicine among patients with breast cancer

Abstract Purpose Integrative medicine (IM) has received ASCO endorsement for managing cancer treatment-related side effects. Little is known about racial differences in familiarity, interest, and use of IM among breast cancer patients. Methods Breast cancer patients enrolled in the Chicago Multiethnic Epidemiologic Breast Cancer Cohort were surveyed regarding familiarity, interest, and use of IM: acupuncture, massage, meditation, music therapy, and yoga. Familiarity and interest, measured by a 5-point Likert scale, was modeled using proportional odds. Use was self-reported, modeled using binary logistic regression. Results Of 1,300 respondents (71.4% White and 21.9% Black), Black patients were less likely than White patients to be familiar with acupuncture (aOR 0.60, 95% CI: 0.41-0.87). While there was no differences in interest in acupuncture between Black and White patients (aOR 1.12, 95% CI: 0.76-1.65), Black patients were more interested in massage (aOR 1.86, 95% CI: 1.25-2.77), meditation (aOR 2.03, 95% CI: 1.37-3.00), music therapy (aOR 2.68, 95% CI: 1.80-3.99) and yoga (aOR 2.10, 95% CI: 1.41-3.12). Black patients were less likely than White to have used acupuncture (aOR 0.49, 95% CI: 0.29-0.84); but there were no racial differences in use of massage (aOR 0.83, 95% CI: 0.53-1.30), meditation (aOR 0.82, 95% CI: 0.47-1.43), music therapy (aOR 1.65, 95% CI: 0.82-3.32) and yoga (aOR 0.67, 95% CI: 0.37-1.20). Conclusion Black patients expressed more interest in IM than their White counterparts; there were no racial differences in IM use, except lower acupuncture use among Black patients. A breast program focused on equity should provide access to these services for breast cancer patients.


Introduction
Breast cancer is the most common cancer type, with more than 3.6 million people living with breast cancer in the United States (U.S.) [1].Breast cancer and its treatment can be associated with numerous side effects and symptoms, from cancer pain to lymphedema to hot ashes, [2] that can negatively impact patients' treatment adherence and quality of life [3].Reduction of side effects and management of symptoms typically consists of further medications, which carry their own adverse effect pro les.In 2017, the Society for Integrative Oncology (SIO) performed a systemic review focusing on randomized controlled trials from 1990 through 2015 of the use of integrative medicine (IM) modalities during and after breast cancer treatment [4].This resulted in a set of evidence-based practice guidelines on the use of IM in breast cancer, which was subsequently endorsed by the American Society of Clinical Oncology (ASCO) in 2018 [5].IM modalities, including acupuncture, therapeutic massage, meditation therapy, music therapy, and yoga therapy all have received ASCO's endorsement for the treatment or management of various side effects and symptoms, particularly: hot ashes, nausea, anxiety or stress reduction, depression or mood disorders, and improving quality of life [5].
A recent study has documented that more than 50.0% of cancer patients and their caregivers are familiar with acupuncture, yoga therapy, and meditation therapy [6].Multiple previous studies have reported breast and gynecological cancer patients' growing interest in and demand for IM, however, these studies included only White female patients [7][8][9][10].Use of IM has increased in recent decades [11][12][13].A survey conducted in breast cancer survivors found that more than 80.0% of the survivors had either used a complementary and alternative therapy or visited a therapist in the past [11].Black race and lower socioeconomic status (SES) have historically been associated with lower prevalence of IM utilization among U.S. adults, according to a study using the 2002 National Health Interview Survey [12].Therefore, IM services are generally marketed toward White populations and those with higher SES.However, little is known about familiarity, interest, and use of IM among Black or African American breast cancer patients and survivors.
To date, the best study about racial differences in interest and use of IM comes from a 2017 study out of the University of Texas MD Anderson Cancer Center, which surveyed 165 cancer patients, 43% of which were Black or African Americans, at an urban community hospital about interest and use of complementary and alternative therapies [14].The study found that 90.6% of the patients were interested in therapeutic massage, followed by 72.7% in meditation therapy, 69.8% in yoga therapy, and 49.7% in acupuncture [14].However, most were unprofessionally guided use, and both past and current IM use were low.In this study, 13.8% of the Black patients had used yoga therapy as compared to 42.9% of Asian and 25.7% of White patients, with no signi cant differences among these racial groups based on a Pearson's Chi-square test.Moreover, the study did not perform multivariable regression analyses due to the small sample size [14].
Prior studies are small and descriptive, with the majority of cancer patients and survivors being White.
Additionally, there is paucity of data on racial differences in familiarity, interest, and use of IM speci cally among breast cancer patients.To ll these gaps in the literature, we sought to assess racial differences in familiarity, interest, and use of ve ASCO-endorsed IM modalities for breast cancer symptom management: acupuncture, therapeutic massage, meditation therapy, music therapy, and yoga therapy in a large cohort of patients with breast cancer having been treated at the University of Chicago Medicine.

Study design and study population
We conducted a cross-sectional survey among breast cancer patients who were enrolled in the Chicago Multiethnic Epidemiologic Breast Cancer Cohort (ChiMEC).Brie y, ChiMEC is a hospital-based study having been enrolling patients diagnosed with breast cancer since 1993.Detailed information of ChiMEC has been previously published [15].Eligible participants were aged 18 years or older.From July to September 2021, a REDCap survey was sent to 2,788 ChiMEC participants who consented to be followed up for subsequent surveys.All patients provided their written informed consent prior to study participation.The University of Chicago Institutional Review Board reviewed and approved this study.

Measures of key variables
Familiarity was measured by asking participants how familiar they were with these types of IM, using a 5point Likert scale including not familiar at all, not very familiar, neutral, familiar, and very familiar.Interest was measured by asking participants how interested they would be in these IM modalities if offered at the center, using a 5-point Likert scale including not interested at all, not very interested, neutral, interested, and very interested.
We also assessed cancer treatment-related symptoms as facilitators by asking participants how interested they would be in any type of IM if it treated hot ashes, chemotherapy-induced neuropathy, nausea, joint pain, back or other pain, depression or mood change, fatigue or tiredness, and anxiety or stress reduction.Other facilitating factors for interest included recommendation from a provider, cost not being a barrier, being covered by health insurance, and price willing to pay out-of-pocket for a session.
To measure use of IM, we asked patients to "Select all of the therapies that you have had or received in the past."Response options for each item were yes and no.We also assessed major barriers to IM use, including cost/money, lack of access to services, lack of transportation to service-providing facilities, lack of time, lack of interest, unaware of bene ts of these services, low con dence about the bene ts on these services, and lack of trusted information on these services by asking what, in general, prevents participants from using any IM modality.

Covariates
Demographic and behavioral characteristics, including age, race/ethnicity (Asian, Black, Hispanic, and White), highest level of education (High school/GED or less, post high school, trade/technical school, or some college, Associate's degree, Bachelor's degree, and graduate or professional degree), marital status, annual household income, type of health insurance (Medicare, Medicaid, private, and other), and history of tobacco and alcohol consumption (never, current, and past), were collected from the survey.Clinical characteristics such as duration from diagnosis to survey, American Joint Committee on Cancer (AJCC) stage group, hormone receptor (HR) status (HR-positive/ human epidermal growth factor receptor 2 [HER2]-negative, HER2-positive, and triple negative breast cancer [TNBC]), and Charlson comorbidity index (CCI, i.e., 0, 1, and ≥ 2) was obtained through clinical chart abstraction.

Statistical analysis
We calculated means and standard deviations (SD) for continuous data and tabulated frequencies and percentages (%) for categorical data.Demographic, behavioral, and clinical characteristics between racial/ethnic groups were compared using Student's t tests or ANOA for continuous variables and Pearson's chi-square or Fisher's exact tests for categorical variables.Of note, Asian and Hispanic patients were included in the descriptive analysis but were excluded in subsequent analyses due to small group sample size.Multivariable proportional odds were modeled for familiarity with and interest in different IM modalities.Multivariable binary logistic regression was modeled for self-report of having used these modalities in the past.All models were adjusted for age, highest level of education, marital status, annual household income, type of health insurance, CCI, HR status, and AJCC stage group.To assess racial differences in familiarity, interest, and use of IM modalities, we calculated adjusted odds ratios (aOR) and 95% con dence intervals (CI).The level of signi cance was set at 0.05.All statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).

Familiarity with integrative medicine
Overall, 59.8% of the patients were familiar or very familiar with therapeutic massage, followed by 47.7% acupuncture, 47.6% meditation therapy, 47.0% yoga therapy, and 35.4% music therapy.Compared with White patients, Black patients had a higher percentage of being familiar or very familiar with music therapy (44.2% vs. 32.7%).However, a higher percentage of White patients reported being familiar or very familiar with therapeutic massage (60.9% vs. 56.4%),acupuncture (49.9% vs. 40.5%),or yoga therapy (47.8% vs. 44.3%)than Black patients (Table 2).After adjusting for covariates, Black patients were less likely than their White counterparts to be familiar with acupuncture (aOR 0.60, 95% CI: 0.41-0.87).We did not observe differences between Black and White patients in familiarity with music therapy, meditation therapy, therapeutic, or yoga therapy (Table 3).

Interest in integrative medicine and facilitators
Overall, 62.9% of the patients reported being interested or very interested in therapeutic massage, followed by 49.4% yoga therapy, 47.1% meditation therapy, 43.9% acupuncture, and 40.4% music therapy.
Participants reported being interested or very interested when asked for interest in any of the IM modalities to address speci c symptoms: joint paints due to aromatase inhibitors (Black 71.6% vs. White 66.2%), back pain or other pain (Black 71.0% vs. White 65.9%), fatigue (Black 63.8% vs. White 64.0%), anxiety or stress reduction (Black 61.3% vs. White 63.25), depression or mood changes (Black 50.4% vs. White 52.5%), hot ashes (Black 49.2% vs. White 44.1%), chemotherapy-induced neuropathy (Black 43.2% vs. White 38.0%), and nausea (Black 32.0% vs. White 30.2%).However, there were no signi cant differences in symptoms as facilitators for interest in IM between the races (Table 5).Generally, Black respondents expressed more interest in IM modalities if they were recommended by their doctors or nurses and were covered by health insurance.Black respondents typically were less willing to pay more than $0-$19 out of pocket for a session of any IM services (Table 6).

Use of integrative medicine and barriers
Overall, 41.6% of the patients had used therapeutic massage, followed by 26.1% acupuncture, 19.0% yoga therapy, 18.5% meditation therapy, and only 7.7% music therapy.Higher percentages of White patients reported having used therapeutic massage (42.9% vs. 37.5%), acupuncture (29.5% vs. 14.7%),yoga therapy (20.8% vs. 13.0%), and meditation therapy (18.8% vs. 17.5%)than Black patients.However, Black patients had a higher percentage of prior use of music therapy than did White patients (11.2% vs. 6.6%)(Table 7).In the adjusted logistic regression model, there were no differences between Black and White patients in self-reported use of the IM modalities surveyed, with the exception of acupuncture as Black patients were less likely than their White counterparts to have used acupuncture (aOR 0.49, 95% CI: 0.29-0.84)(Table 3).
When participants were asked about barriers to use of any IM modality, higher percentages of Black patients reported cost (55.1% vs. 31.4%),lack of awareness of bene ts of IM services (35.8% vs. 23.9%),lack of access to services (24.6% vs. 19.4%),and lack of transportation to service-providing facilities (13.0% vs. 9.3%) than White patients.White patients had a higher percentage of lack of time as a barrier to use than their Black counterparts (32.2% vs. 19.3%).Of note, con dence in the bene ts of IM was high in both races (White 91% vs. Black 90.5%) (Table 8).

Discussion
To our knowledge, this is of the rst and the largest study to examine racial differences in familiarity, interest, and use of ve ASCO-endorsed IM modalities and to assess speci c symptom-related facilitating factors for interest in, and key barriers to use of, these modalities among U.S. patients with breast cancer.
In this diverse cohort of patients with breast cancer, familiarity of IM modalities was prevalent among Black and White patients.However, higher proportions, ranging from 26.9% to 48.6%, of the patients across the racial groups were still not familiar with these modalities.Black patients were less familiar with acupuncture than their White counterparts, but there were no differences between them in familiarity with therapeutic massage, meditation therapy, music therapy, and yoga therapy.These ndings are somewhat consistent with a recent study of familiarity and interest in IM among cancer patients and their caregivers that non-White patients are less familiar with therapeutic massage than White patients, while levels of familiarity with acupuncture, meditation therapy, music therapy, and yoga therapy are similar between White and non-White patients [6].Our ndings suggest that patient education on IM and its associated bene ts may be needed among breast cancer patients and survivors in order to increase patients' knowledge and awareness of IM.Future research may be needed to explore reasons related to level of familiarity and how the ndings could help inform and tailor IM education campaigns and programs speci cally toward breast cancer patients and survivors.
We found that most patients across the two racial groups were interested in the use of IM modalities, and the percentages of interest of IM use also increased when the patients were asked if any of these modalities were treated for a particular symptom such as joint pains, back pain, fatigue, anxiety or stress reduction, hot ashes, and chemotherapy-induced neuropathy.Our nding in increased interest of IM use aligns with a published study that patients who experience back, joint, or other pain are more likely to use acupuncture and therapeutic massage, though racial differences were not assessed [16].Further, Black patients were twice as likely as their White counterparts to have expressed interest in music therapy, yoga therapy, meditation therapy, and therapeutic massage, which is contrary to prior ndings in the general population that conclude Black patients are less interested in IM services than White patients [12].Breast cancer patients may have unique needs, as a population-based research has indicated that cancer patients and survivors are more likely than the general population to have discussed IM use with a provider and have used these modalities in the past 12 months [17].Our result also contradicts a recent study nding that levels of interest in IM are similar between non-White and White patients [6].However, this study sample was relatively small, probably lacking statistical power.In addition, it included patients with different cancer types and their caregivers, and thus, the ndings may not be comparable to our cohort of breast cancer patients [6].It is worth noting that approximately 21.6%-40.6% of the patients across the racial groups reported being not very interested or not interested at all in using these IM modalities, even when they were asked if these modalities addressed speci c common cancer treatmentassociated symptoms.It is also important to note that most patients expressed greater interest in IM use if recommended by their providers and were willing to pay no more than $19 out of pocket for a session.
Our ndings indicate that Black patients may be in greater need for IM and that both provision and coverage of these modalities should be integrated as part of standard cancer care and services at comprehensive cancer centers in the U.S.
The percentages of past use of IM modalities were low among Black and White breast cancer patients; the majority of the patients, between 57.1% and 93.4%, had not used these modalities in the past.These results are in line with a previous study that both past and current use of acupuncture, meditation therapy, yoga therapy, and therapeutic massage were low, ranging from only 5.6% to 46.3% [14].Compared with White patients, a higher proportion of Black patients reported cost, lack of access to services, unaware of the bene ts of these services as major barriers to IM use.Less than 10.0% of both Black and White patients reported low con dence in the bene ts of the ve surveyed IM modalities.After adjusting for key demographic and clinical characteristics, Black patients in our cohort were signi cantly less likely than their White counterparts to have used acupuncture, while there were no differences between the racial groups in past use of therapeutic massage, meditation therapy, music therapy, or yoga therapy.Our nding is consistent with the previous study partially that use of yoga therapy was similar between Black and White patients, however, multivariable regression modeling was not performed due to the small sample size [14].It is important to point out that we did not ask whether the patients had used these IM modalities before, during, and/or after their cancer therapies, which is worth doing in future research to evaluate whether there are racial and ethnic differences in IM utilization over time and how these differences would impact breast cancer patients' treatment adherence and quality of life.Our ndings also suggest that providers at cancer centers should be promoting these IM services as recommended by ASCO guidelines to all patients which, as shown, may be likely to increase interest in breast cancer patients and survivors Furthermore, Black patients with breast cancer more frequently report nonadherence to endocrine therapy than their White counterparts, with side effect pro le being one of the main causes of discontinuation [18,19].Therefore, the use of IM modalities to reduce side effects from breast cancer treatment and to manage symptoms may lead to greater adherence to endocrine therapy in Black patients.As we have shown Black patients are just as interested, if not more interested, in IM as their White counterparts, and there are likely unmet needs of IM among breast cancer patients, we should ensure equity to access these services for all our breast cancer patients, regardless of race.
Several limitations of this study need to be noted.First, the data collected through the survey were selfreported, which were subject to recall error or social desirability bias.However, we expect such bias to be minimal since our research staff had little interaction with the patients.Therefore, their responses were unlikely in uenced.Second, because we did not ask the participants whether they had use these services before their cancer diagnoses, during, and/or after their cancer treatment, with an approximately 47.0% response rate, the percentages of use of these IM modalities may have been either overestimated or underestimated.Third, we were not able to assess unmeasured characteristics, e.g., cultural background/in uence, employment status, patient-provider discussion of IM, which might affect or help better explain the observed racial differences in familiarity, interest, and past use of IM.Thus, additional cultural and behavioral factors should be taken into consideration in future research.Lastly, participants in the ChiMEC may not be representative of all U.S. patients with breast cancer or other patient populations, and therefore, limiting the generalizability of our study ndings.
Despite the above limitations, this study has several strengths.Our study is the largest to date examining racial differences in familiarity, interest, and use of the ve ASCO-endorsed IM modalities among patients with breast cancer.Another strength of this study was the inclusion of a racially diverse cohort of patients with breast cancer and key clinical characteristics that previous research was not able to assess.
In conclusion, both Black and White breast cancer patients were familiar with the ve ASCO-endorsed IM modalities, but Black patients expressed greater interest in the use of these modalities.There were no racial differences in prior use of IM, except an increased use of acupuncture among White patients.However, Black patients reported more health care and services access-related barriers than did their White counterparts.Promoting bene ts of IM among breast cancer patients and facilitating patientprovider discussion of IM use may be needed.Furthermore, breast programs focused on health equity should provide access to these services for all patients.Note: Multivariable proportional odds were modeled for familiarity with and interest in using different integrative medicine (IM) modalities, while multivariable binomial logistic regression was modeled for self-report of having used these IM modalities in the past.
a Familiarity was assessed by asking participants how familiar they were with these IM modalities.
b Interest was assessed by asking participants how interested they would be in these IM modalities if offered at the center.
c Participants were asked about these IM modalities they have used in the past.
d Adjusted for age, highest level of education, marital status, annual household income, type of health insurance, Charlson comorbidity index, hormone receptor status/subtype, and stage group.
Acupuncture Not interested at all ( Not very interested Very interested ( 5  a p-values were calculated using Chi-square tests.
b Recommendation from your doctor or nurse was assessed by asking participants how interested they would be to have any integrative medicine modality if their doctor or nurse recommended it.
c Cost not being a barrier was assessed by asking participants how interested they would be to have any integrative medicine modality if cost was not a barrier.
d Being covered by health insurance was assessed by participants how interested they would be to have any integrative medicine modality if their health insurance paid for it.a p-values were calculated using Chi-square tests.
b Participants were asked about these integrative medicine modalities they have used in the past.

Table 1 .
Declarations 19.Wheeler SB, Spencer J, Pinheiro LC, Murphy CC, Earp JA, Carey L, Reeder-Hayes KE (2019) Endocrine Therapy Nonadherence and Discontinuation in Black and White Women.J Natl Cancer Inst 111(5):498-508.https://doi.org/10.1093/jnci/djy136Tables Demographic, behavioral, and clinical characteristics of breast cancer patients, overall and by race/ethnicity Abbreviations: SD, standard deviation; GED, general educational development; HR, hormone receptors; HER2, human epidermal growth factor receptor 2; TNBC, triple negative breast cancer; BCS, breast conserving surgery.Note: Percentages may not be 100 due to rounding.ap-valueswere calculated using ANOVA for continuous data and Chi-square or Fisher's exact tests for categorical data.bAge was measured at the time of survey.cMaritalstatus was documented at the time of diagnosis.dOtherincluded 1 uninsured/self-pay, 17 insurance not otherwise speci ed, 1 TRICARE, and 1 Military.eTobaccoproducts assessed included cigarette, cigar, pipe, snuff, chew, smokeless, or mixed use with more than one type.fDurationfrom diagnosis to survey was de ned as the duration from the time of cancer diagnosis to the time of survey.gStage group was de ned based on the American Joint Committee on Cancer's cancer staging.Table 2. Familiarity with integrative medicine modalities among breast cancer patients, overall and by Note: Percentages may not be 100 due to rounding.a Familiarity was assessed by asking participants how familiar they were with integrative medicine modalities.b p-values were calculated using Student's t tests for continuous data and Chi-square tests for categorical data.

Table 4 .
Interest in using integrative medicine modalities among breast cancer patients, overall and by race

Table 5 .
Symptoms as facilitators for interest in using any integrative medicine modality among breast cancer patients, overall and by race Symptoms as facilitators were assessed by asking participants how interested they would be in any integrative medicine modality if speci cally treated the symptoms above.p-values were calculated using Chi-square tests.
Note: Percentages may not be 100 due to rounding.a Interest was assessed by asking participants how interested they would be in these integrative medicine modalities if offered at the center.b p-values were calculated using Student's t tests for continuous data and Chi-square tests for categorical data.a b

Table 6 .
Facilitators for interest in using integrative medicine modalities among breast cancer patients,

Table 7 .
Self-reported use of integrative medicine modalities among breast cancer patients, overall and by

Table 8 .
Barriers to using any integrative medicine modality among breast cancer patients, overall and by Note: Percentages may not be 100 due to rounding.