Article Tools
Original Contributions
CARE DELIVERY
Article Tools
Race and Ethnicity in the Evidence for Integrating Palliative Care Into Oncology
Abstract
ASCO recommends early involvement of palliative care for patients with advanced cancers on the basis of evidence from 18 randomized trials. We examined racial and ethnic minority representation in these trials and the role of race and ethnicity in the statistical analyses. The goal was to identify specific gaps in the palliative care evidence base for these individuals and potential strategies to address them.
We reviewed the 18 trials cited in the 2012 and 2017 ASCO clinical statements on integrating palliative care into oncology. We extracted data on the reporting and categorization of race and ethnicity, on the enrollment of specific racial and ethnic minority groups, and on how race and ethnicity were addressed in the analyses.
One third of patient trials reported representation of specific racial and ethnic minority groups, one third reported rates of “white” versus “other,” and one third did not report race or ethnicity data. Among the patient trials with race and ethnicity data, 9.9% of participants were Asian, 8.8% Hispanic/Latino, and 5.7% African American. Analyses that used race and ethnicity were primarily baseline comparisons among randomized groups.
Race and ethnicity were inconsistently reported in the trials. Among those that provided race and ethnicity data, representation of specific racial and ethnic minority groups was low. In addition to more research in centers with large minority populations, consistent reporting of race and ethnicity and supplementary data collection from minority patients who participate in trials may be strategies for improvement.
ASCO recommends the early integration of palliative care with oncology care for patients with advanced cancers soon after diagnosis.1,2 In 2012, an ASCO committee reviewed the existing literature on palliative care integrated with oncology care and concluded that there was sufficient evidence to issue this provisional clinical opinion.2 The committee cited seven randomized clinical trials of palliative care interventions as supporting evidence. Five years later in 2017, ASCO published a follow-up statement with even stronger recommendations on the basis of evidence from an additional 11 trials.1 However, the authors of the ASCO recommendations noted that the trials of palliative care interventions were conducted in predominantly non-Hispanic white samples.
Under-representation of racial and ethnic minorities in palliative care trials is particularly problematic because of preexisting differences in factors that are typical outcomes in palliative care trials. A systematic review of disparities in end-of-life (EOL) care for patients with advanced cancer identified multiple differences according to race and ethnicity, including preferences for EOL care, code status documentation, advanced care planning, knowledge of hospice, perceived need for hospice, rates of hospice referrals, use of hospice, duration of hospice admissions, and receipt of preference-concordant care at EOL.3 All these differences at baseline can affect the outcomes of palliative care interventions.
Rather than simply dismissing the sparse data as limitations of these trials, we should recognize that this is the extent of our knowledge about the effects of integrated palliative care in specific racial and ethnic minority groups. A closer examination of these data would reveal the state of the science and identify opportunities for future research on integrated palliative care in racially and ethnically diverse populations. Therefore, we reviewed the details of race and ethnicity in all the palliative care trials cited in the ASCO recommendations. We focused on the numbers of participants from specific racial and ethnic minority groups in each trial, how the trials categorized race and ethnicity, and the role of race and ethnicity in the statistical analyses. To provide context for the attention given to race and ethnicity, we also reviewed the representation of other common sociodemographic variables, such as age and sex, in the trials. The goal was to identify specific gaps in the palliative care evidence base for these individuals and potential strategies to address them.
We reviewed all studies included as evidence in the 2012 and 2017 ASCO statements on palliative care in oncology. The 2012 ASCO committee selected the trials in their statement through a literature search for randomized controlled trials of “interdisciplinary or team-based care to improve the overall cancer experience compared with usual care.” The 2016 ASCO committee conducted an updated systematic review of studies published in English between March 2010 and January 2016 that included patients with cancer and that were phase III randomized controlled trials. The committee also included published secondary analyses of the randomized controlled trials in the 2012 provisional clinical opinion, systematic reviews, and meta-analyses. We extracted data on the racial and ethnic composition of the samples. Specifically, we extracted the total sample size, how race and ethnicity were categorized, and the number of participants in each reported racial and ethnic category. These data were organized into tables and tabulated for presentation in the text. In each trial, we note how race, ethnicity, and other demographic data (age, sex, marital status, education, income, and religion) were used in the analyses. For example, we investigated whether these sociodemographic variables were included in baseline comparisons of trial conditions, as a covariate in the primary analyses, and/or in any subgroup analyses. We also searched PubMed to identify subsequent publications from the trials that included subgroup analyses on the basis of sociodemographic variables. Finally, we examined two trial characteristics: the type of setting where the trial was conducted and language requirements for eligibility.
Altogether, the 2012 and 2017 ASCO statements on palliative care in oncology included 18 trials. The studies are listed in Table 1. Fifteen studies reported patient outcomes only, three reported caregiver outcomes only, and one reported both patient and caregiver outcomes. The studies included a total of 3,960 patients and 1,262 caregivers.
|
Of the 15 studies that reported patient outcomes, two thirds (n = 10) reported race data, and six reported ethnicity data. All studies that reported ethnicity also reported race. Across all studies, race/ethnicity data were available from 65.9% of the patient participants (2,611 of 3960), which included 43 patients being assigned “missing” as a category in one study. Of note, 10 of the 12 studies conducted in the United States reported race/ethnicity data compared with none of the three conducted solely in other countries (Canada, the Netherlands, and United Kingdom).
Of the four studies that reported caregiver outcomes, three (75%) reported data on both the race and ethnicity of caregivers, including a study that also reported on patient outcomes. Racial and ethnic data were available from 76.4% of the caregiver participants (964 of 1,262), which included four caregivers being assigned missing as a category in one study. The one study that did not report race and ethnicity was conducted outside the United States (Australia).
Of the 10 patient studies that reported any racial data, approximately one half (n = 4) reported race dichotomously as “white” or “other” without differentiating ethnicity. One half of the studies reported the number of participants who were African American or black. In addition, one half of the studies (n = 5) reported the number of participants who were Asian. Three of the studies (30%) included American Indian or Native Alaskan as a category. With respect to ethnicity, six studies reported the number of participants who were Hispanic or Latino, and this category was separated from race in all cases, as is standard for National Institutes of Health enrollment reporting.
Of the three caregiver studies that reported racial data, one (33.3%) reported race as “white" or "other". The other two (66.7%) reported the number of participants who were African American or black, Asian, and American Indian or Native Alaskan. Two (66.7%) of the three studies reported the number of participants who were Hispanic or Latino. Both studies separated the reporting of Hispanic ethnicity from race.
In contrast to race and ethnicity, all studies reported age and sex. After age and sex, education and marital status were the other sociodemographic variables most frequently reported in patient studies, with 10 (66.7%) and nine (60%), respectively. Employment, living situation, and religion were each reported in five patient studies (33.3%). All four of the caregiver studies reported relationship with the patient and employment status. One half (n = 2) of the caregiver studies reported the marital status and religion of caregivers.
Among the patient studies that included racial/ethnicity data, 1,910 participants (73.2%) were listed as white or Caucasian. Among the studies that included African American or black as a racial category, 84 patients (5.7%) were African American or black. Among the studies that included Asian as a racial category, 153 patients (9.9%) were Asian, and among those that reported Hispanic ethnicity, 137 patients (8.8%) were Hispanic/Latino. Of note, no native Hawaiian/Pacific Islander and no patients categorized as multiple races were reported.
Among the caregiver studies that included racial/ethnic data, 733 participants (76.0%) were listed as white or Caucasian. Among the studies that included African American or black as a racial category, 35 caregivers (5.7%) were African American or black. Among the studies that included Asian as a racial category, 72 caregivers (8.6%) were Asian, and among the studies that reported Hispanic ethnicity, 66 caregivers (7.8%) were Hispanic/Latino.
Table 2 lists the role of race and ethnicity in the statistical analyses of the trials. The majority of patient studies (seven of 10) that reported race compared the baseline rates of racial/ethnic categories among study conditions. None of the studies presented subgroup analyses on the basis of race or ethnicity in the primary or subsequent articles. However, one of the 18 subsequent publications that stemmed from the trials included race as a control variable in the main analysis.22
|
The two caregiver studies that reported race compared the rates of racial and ethnic categories among participants assigned to different study conditions. None of the studies presented subgroup analyses on the basis of race or ethnicity in the primary or subsequent articles.
In terms of other sociodemographic variables, nine (60%) of the 15 patient studies and three (75%) of the four caregiver studies compared baseline age between trial conditions. One study of both patients and caregivers reported a subgroup analyses for age in the primary article. Among subsequent publications from the original trials, one article explored the effect of age as a moderating variable in subgroup analyses.23
Approximately three quarters of patient studies (11 [73.3%] of 15) compared sex among participants assigned to different study conditions. In subsequent publications, one article explored the effect of sex in subgroup analyses.23 In addition, two subsequent patient studies included marital status as a covariate or control variable. Religious affiliation also was included as a predictor in two subsequent studies, one of which focused on patients and one on caregivers.
Both the patient and the caregiver trials were conducted in a variety of settings that included academic cancer centers and teaching hospitals, health maintenance organizations, and home-based care. The three trials that reported the highest accrual of minorities were conducted in a health maintenance organization, the general medicine practice of an academic hospital, and the emergency department of an academic hospital.
Seven of the patient trials required participants to speak English, two required English or Spanish, and six did not specify language requirements. The five patient trials that did not report race were among the six that did not specify language requirements. The two trials that required English or Spanish were among the top three trials with the highest reported accrual of minority participants. Three of the four caregiver trials required participants to speak English, and one did not have specific language requirements.
Although the percentage of racial and ethnic minority participants in palliative care trials is small, our examination of the data used in recent ASCO recommendations for integrating palliative care with oncology care revealed several notable findings. First, quantification of the number of specific racial and ethnic groups that have participated in these trials is difficult. Only one third of the patient trials specified whether participants were African American/black, Asian, or Hispanic/Latino, whereas an additional one third only categorized participants as “white” versus “other”. Although this dichotomization may be reported when there are only a small number of patients from racial and ethnic minority groups, this still occurred in one of the studies with the largest “other” group.
The lack of explicitly stating the racial/ethnic enrollment in studies prohibits readers from understanding the generalizability of the study results. In addition, combining different racial and ethnic groups into a single “other” category in the larger studies does not make conceptual sense because there can be, and often are, divergent patterns within the “other” category. For example, compared with white patients, native Hawaiians/Pacific Islanders are more likely to use hospice, and African American patients are less likely.3 In addition, studies have demonstrated that African American patients may perceive greater utility in hospice compared with non-Hispanic white patients, but Hispanic patients perceive less need for hospice services because they often believe it is the family’s role to provide care at the EOL.3 By lumping these groups together, studies may dilute differences and miss the opportunity to identify potential true differences among nonwhite participants. Furthermore, this precludes the pooling of data across samples to conduct subgroup analyses by race or ethnicity.
One third of the trials cited in the ASCO statements did not report race or ethnic groups at all. Although we found this to be surprising, other therapeutic clinical trials have shown similar results. We found that the majority of studies (three of five) that did not report race and ethnicity were conducted outside the United States. This finding may relate to the differential emphasis on race/ethnicity reporting expectations in other countries. However, to put it in context, all studies reported age and sex. The frequency of reporting race/ethnicity was similar to how often the studies reported marital status and education. Furthermore, these studies reported race and ethnicity more often than work/employment status, living situation, and religion. Thus, our findings highlight the prevalence of unreported data and underscores the need for enhanced reporting of these important participant characteristics to improve our understanding of the generalizability of study results.
Although similar findings of inconsistent reporting of race and ethnicity and low accrual of minorities have been reported in reviews of clinical trials of cancer treatments, the current examination of trials is the first to our knowledge of integrated palliative care in oncology.24-26 Compared with clinical trials of cancer treatments, these trials may have greater representation of African Americans/blacks. We found that African Americans/blacks represented 5.7% of the participants across these trials, whereas reports of cancer clinical trials are 3%. However, adequate representation in clinical trials of cancer treatments is determined by the incidence of the specific targeted cancer in different racial and ethnic groups.27 For integrated palliative care trials, representation should be based on the incidence of advanced cancers in each racial and ethnic group. African Americans/blacks are more likely to present at later stages, and African American/black men have the highest incidence of cancer,28 which might contribute to the higher percentage of this racial group in trials of integrated palliative care.
The limited number of racial and ethnic minority participants in palliative care trials limits how these data can be analyzed. When race/ethnicity were analyzed, it was to compare intervention to control groups at baseline, and most trials likely were underpowered to detect significant differences. None of the trials used race or ethnicity as a covariate in the primary analyses or conducted subgroup analyses; therefore, compared with age and sex, which were included as covariates in some primary analyses and even in one subgroup analysis, the collection of race and ethnicity data contributed little to the studies beyond simple description.
Characteristics of the trial itself may bias toward lower accrual of minority patients. Because integrated palliative care requires participants to already be receiving care at a particular site, trials of integrated palliative care depend on the population that is receiving care at the trial site, perhaps more than trials of cancer treatments. Because racial and ethnic minority patients are significantly less likely to have their cancers treated at National Cancer Institute–designated cancer centers,29,30 trials conducted at these centers have less potential to enroll minorities. In fact, the top three trials with the highest accrual of minorities were not conducted in academic cancer centers; they were conducted in a health maintenance organization, a general medicine practice, and an emergency department. In addition, the requirement that participants be English speakers also limits a trial’s ability to accrue Hispanic patients and other minorities.
In identifying some of the problems and challenges, this analysis of the existing evidence suggests two potential strategies to increase the data on palliative care in racial and ethnic minorities. First, because integrated palliative care trials depend more on the patients who receive treatment at cancer centers, trials need to be conducted in centers with greater numbers of racial and ethnic minorities, which might not be the academic cancer centers with palliative care and research resources. Therefore, investigators might have to partner with other centers to address these questions. Community-based models of palliative care integrated across various cancer centers is another possibility, although it is more complicated.
The second strategy is to maximize the data, even in samples with low rates of minority participation. Consistent reporting of the racial and ethnic composition of samples by using standardized categories is crucial. Specifically, we recommend categories used by the National Institutes of Health. Use of standard categories may allow for meta-analyses to aggregate data from multiple studies. In addition, postintervention interviews with minority participants could provide valuable information on acceptability as well as potential differences in the delivery of or response to the intervention. These qualitative data could inform possible cultural tailoring necessary for integrated palliative care in certain groups.
As the United States moves toward a population that no longer will have a racial and ethnic majority group by 2044,31 it is critical to ensure that medical care is effective for all. Observational data outside trials raise some questions about the benefits of palliative care across racial and ethnic groups. For example, significant differences in outcomes between non-Hispanic white and racial/ethnic minority patients have been reported in two retrospective observational studies of palliative care consultations in patients with cancer. In one with hospitalized patients with metastatic cancers, Sharma et al32 reported that although non-Hispanic whites and African Americans who had palliative care consults were more likely to be referred to hospice than those without palliative care consults, this was not observed in Hispanics. In the other trial, non-Hispanic white patients with lung cancer in an outpatient palliative care clinic were observed at their first follow-up visit to have significant improvements in pain, depression, and fatigue, whereas Hispanics had improvement in only depression, and blacks had no improvement in any of the three symptoms.33
Although a need for more research clearly exists, the paucity of evidence should not interfere with racial and ethnic minority patients with advanced cancer receiving palliative care integrated with oncology care. Despite cultural differences among people with differing racial and ethnic backgrounds, the patient- and family-centered nature of palliative care may foster a personalized approach that supports each person’s individual values. An important goal of research on integrated palliative care is to ensure that all groups of patients receive care that aligns with their unique needs. Thus, patients with different racial and ethnic backgrounds may benefit from new models of care or adaptations of existing care models. In this way, research that integrates palliative care with oncology care can decrease racial and ethnic disparities in EOL care in oncology.
Conception and design: William F. Pirl, Estefany Saez-Flores
Administrative support: William F. Pirl
Collection and assembly of data: William F. Pirl, Estefany Saez-Flores
Data analysis and interpretation: All authors
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jop/site/ifc/journal-policies.html.
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
No relationship to disclose
| 1. | Ferrell BR, Temel JS, Temin S, et al: Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol 35:96-112, 2017 Link, Google Scholar |
| 2. | Smith TJ, Temin S, Alesi ER, et al: American Society of Clinical Oncology provisional clinical opinion: The integration of palliative care into standard oncology care. J Clin Oncol 30:880-887, 2012 Link, Google Scholar |
| 3. | LoPresti MA, Dement F, Gold HT: End-of-life care for people with cancer from ethnic minority groups: A systematic review. Am J Hosp Palliat Care 33:291-305, 2016 Crossref, Medline, Google Scholar |
| 4. | Bakitas M, Lyons KD, Hegel MT, et al: Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: The Project ENABLE II randomized controlled trial. JAMA 302:741-749, 2009 Crossref, Medline, Google Scholar |
| 5. | Brumley R, Enguidanos S, Jamison P, et al: Increased satisfaction with care and lower costs: Results of a randomized trial of in-home palliative care. J Am Geriatr Soc 55:993-1000, 2007 Crossref, Medline, Google Scholar |
| 6. | Gade G, Venohr I, Conner D, et al: Impact of an inpatient palliative care team: A randomized control trial. J Palliat Med 11:180-190, 2008 Crossref, Medline, Google Scholar |
| 7. | Pantilat SZ, O’Riordan DL, Dibble SL, et al: Hospital-based palliative medicine consultation: A randomized controlled trial. Arch Intern Med 170:2038-2040, 2010 Crossref, Medline, Google Scholar |
| 8. | Rabow MW, Dibble SL, Pantilat SZ, et al: The comprehensive care team: A controlled trial of outpatient palliative medicine consultation. Arch Intern Med 164:83-91, 2004 Crossref, Medline, Google Scholar |
| 9. | Temel JS, Greer JA, Muzikansky A, et al: Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med 363:733-742, 2010 Crossref, Medline, Google Scholar |
| 10. | Meyers FJ, Carducci M, Loscalzo MJ, et al: Effects of a problem-solving intervention (COPE) on quality of life for patients with advanced cancer on clinical trials and their caregivers: Simultaneous care educational intervention (SCEI): Linking palliation and clinical trials. J Palliat Med 14:465-473, 2011 Crossref, Medline, Google Scholar |
| 11. | Chochinov HM, Kristjanson LJ, Breitbart W, et al: Effect of dignity therapy on distress and end-of-life experience in terminally ill patients: A randomised controlled trial. Lancet Oncol 12:753-762, 2011 Crossref, Medline, Google Scholar |
| 12. | Dyar S, Lesperance M, Shannon R, et al: A nurse practitioner directed intervention improves the quality of life of patients with metastatic cancer: Results of a randomized pilot study. J Palliat Med 15:890-895, 2012 Crossref, Medline, Google Scholar |
| 13. | Ferrell B, Sun V, Hurria A, et al: Interdisciplinary palliative care for patients with lung cancer. J Pain Symptom Manage 50:758-767, 2015 Crossref, Medline, Google Scholar |
| 14. | Grudzen CR, Richardson LD, Johnson PN, et al: Emergency department–initiated palliative care in advanced cancer: A randomized clinical trial. JAMA Oncol 2:591-598, 2016 Crossref, Medline, Google Scholar |
| 15. | Higginson IJ, Bausewein C, Reilly CC, et al: An integrated palliative and respiratory care service for patients with advanced disease and refractory breathlessness: A randomised controlled trial. Lancet Respir Med 2:979-987, 2014 Crossref, Medline, Google Scholar |
| 16. | Uitdehaag MJ, van Putten PG, van Eijck CH, et al: Nurse-led follow-up at home vs. conventional medical outpatient clinic follow-up in patients with incurable upper gastrointestinal cancer: A randomized study. J Pain Symptom Manage 47:518-530, 2014 Crossref, Medline, Google Scholar |
| 17. | Zimmermann C, Swami N, Krzyzanowska M, et al: Early palliative care for patients with advanced cancer: A cluster-randomised controlled trial. Lancet 383:1721-1730, 2014 Crossref, Medline, Google Scholar |
| 18. | Bakitas MA, Tosteson TD, Li Z, et al: Early versus delayed initiation of concurrent palliative oncology care: Patient outcomes in the ENABLE III randomized controlled trial. J Clin Oncol 33:1438-1445, 2015 Link, Google Scholar |
| 19. | Hudson P, Trauer T, Kelly B, et al: Reducing the psychological distress of family caregivers of home based palliative care patients: Longer term effects from a randomised controlled trial. Psychooncology 24:19-24, 2015 Crossref, Medline, Google Scholar |
| 20. | Sun V, Grant M, Koczywas M, et al: Effectiveness of an interdisciplinary palliative care intervention for family caregivers in lung cancer. Cancer 121:3737-3745, 2015 Crossref, Medline, Google Scholar |
| 21. | Dionne-Odom JN, Azuero A, Lyons KD, et al: Benefits of early versus delayed palliative care to informal family caregivers of patients with advanced cancer: Outcomes from the ENABLE III randomized controlled trial. J Clin Oncol 33:1446-1452, 2015 Link, Google Scholar |
| 22. | Pirl WF, Greer JA, Traeger L, et al: Depression and survival in metastatic non-small-cell lung cancer: Effects of early palliative care. J Clin Oncol 30:1310-1315, 2012 Link, Google Scholar |
| 23. | Nipp RD, Greer JA, El-Jawahri A, et al: Age and gender moderate the impact of early palliative care in metastatic non-small cell lung cancer. Oncologist 21:119-126, 2016 Crossref, Medline, Google Scholar |
| 24. | Murthy VH, Krumholz HM, Gross CP: Participation in cancer clinical trials: Race-, sex-, and age-based disparities. JAMA 291:2720-2726, 2004 Crossref, Medline, Google Scholar |
| 25. | Stewart JH, Bertoni AG, Staten JL, et al: Participation in surgical oncology clinical trials: Gender-, race/ethnicity-, and age-based disparities. Ann Surg Oncol 14:3328-3334, 2007 Crossref, Medline, Google Scholar |
| 26. | Chen MS, Jr., Lara PN, Dang JH, et al: Twenty years post-NIH Revitalization Act: Enhancing Minority Participation in Clinical Trials (EMPaCT): Laying the groundwork for improving minority clinical trial accrual: Renewing the case for enhancing minority participation in cancer clinical trials. Cancer 120(suppl 7):1091-1096, 2014 Crossref, Medline, Google Scholar |
| 27. | Freedman LS, Simon R, Foulkes MA, et al: Inclusion of women and minorities in clinical trials and the NIH Revitalization Act of 1993--the perspective of NIH clinical trialists. Control Clin Trials 16:277-285, 1995 Crossref, Medline, Google Scholar |
| 28. | Siegel RL, Miller KD, Jemal A: Cancer Statistics, 2017. CA Cancer J Clin 67:7-30, 2017 Crossref, Medline, Google Scholar |
| 29. | Huang LC, Yifei M, Ngo JV, et al: What factors influence minority use of National Cancer Institute-designated cancer centers? Cancer 120:399-407, 2014 Crossref, Medline, Google Scholar |
| 30. | Sultan DH, Gishe J, Hanciles A, et al: Minority use of a National Cancer Institute-designated comprehensive cancer center and non-specialty hospitals in two Florida regions. J Racial Ethn Health Disparities 2:373-384, 2015 Crossref, Medline, Google Scholar |
| 31. | Colby SL, Ortman JM: Projections of the Size and Composition of the U.S. Population: 2014 to 2060, 2015. http://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf Google Scholar |
| 32. | Sharma RK, Cameron KA, Chmiel JS, et al: Racial/ethnic differences in inpatient palliative care consultation for patients with advanced cancer. J Clin Oncol 33:3802-3808, 2015 Link, Google Scholar |
| 33. | Reyes-Gibby CC, Anderson KO, Shete S, et al: Early referral to supportive care specialists for symptom burden in lung cancer patients: A comparison of non-Hispanic whites, Hispanics, and non-Hispanic blacks. Cancer 118:856-863, 2012 Crossref, Medline, Google Scholar |
