Distress and Supportive Care Needs of Ethnically Diverse Older Adults With Advanced or Recurrent Cancer

The population of the United States is becoming more diverse, and the number of Americans aged older than 65 years is increasing. For example, from 1982–2003, the number of adults aged 65 years or older doubled (Korc-Grodzicki et al., 2014). It is projected that by 2044 more than half of the U.S. population will be non-Caucasian and that by 2060 almost one in four Americans will be aged 65 years or older (Colby & Ortman, 2015; Hurria et al., 2014; Korc-Grodzicki et al., 2014). With advancing age, older adults often experience multiple comorbidities, including cancer, which have the potential to cause distress. This study explores the needs of ethnically diverse older adults who receive care in an economically challenged community.

Cancer is the second leading cause of death in people aged 65 years or older (Chock et al., 2013). Historically, minorities (including African American, Asian, Hispanic, American Indian or Alaskan Native, and Asian American or Pacific Islander populations) have had higher rates of advanced disease at presentation than non-Hispanic Caucasians; they also have poorer rates of survival and experience worse physical symptoms, psychological outcomes, and social well-being related to living with cancer (Alananzeh, Levesque, Kwok, & Everett, 2016; Singh, Williams, Siahpush, & Mulhollen, 2011; Yeager et al., 2016). Nedjat-Haiem, Carrion, Ell, and Palinkas (2012) noted that economically disadvantaged Latinos with advanced cancer in underserved communities also experience poorer survival outcomes.

With developments in medical treatment, patients are living longer with advanced cancer (Brooks et al., 2013). However, as many as 78% of older adults living with cancer have at least one other condition, which adds to the burden of physical, emotional, and financial challenges faced by these patients (Cavers, Cunningham-Burley, Watson, Banks, & Campbell, 2014; Giuliani et al., 2016).

Distress, which is defined as a complex unpleasant emotional experience that interferes with patients’ ability to cope effectively with cancer, including its symptoms and treatment (Holland et al., 2013), is experienced by 20%–66% of patients with cancer (Acquati & Kayser, 2017; Admiraal, Reyners, & Hoekstra-Weebers, 2013; Ma et al., 2014). Investigation of the effects of distress on patients with cancer suggests that those who are economically disadvantaged and who belong to minority populations are more likely to report distress (Acquati & Kayser, 2017; Simonelli & Pasipanodya, 2014). Predictors of significant distress include younger age, physical symptoms, and problems with family relationships (Graves et al., 2007; Kim et al., 2013; Nelson, Balk, & Roth, 2010). Multiple researchers have reported that older adults with advanced cancer have lower levels of distress and fewer supportive care needs (SCNs) than younger adults with cancer (Fiszer, Dolbeault, Sultan, & Brédart, 2014; Kornblith et al., 2006; Sanson-Fisher et al., 2000). The reasons for these differences between older and younger patients are multifaceted; for instance, they may stem from underreporting of symptoms by healthcare professionals who are focused on assessing physical side effects or from coping strategies adopted by older patients as part of the normal aging process (Kornblith et al., 2006; Lo et al., 2010). In contrast, Akechi et al. (2012) reported that psychological distress was similar in older adults and younger adults with cancer.

SCNs are defined as requirements for patient care that are pertinent to the management of symptoms and side effects, enablement of adaption and coping, optimization of understanding and informed decision making, and minimization of functional deficits (Maguire et al., 2013). SCNs encompass the care that is provided to patients with cancer and their families throughout the cancer trajectory, beginning at prediagnosis and moving through treatment and cure to recurrence, palliation, death of the patient, and family bereavement (Harrison, Young, Price, Butow, & Solomon, 2009).

Li et al. (2013) studied the SCNs of patients with various cancer diagnoses and found that patients reported unmet needs surrounding healthcare systems, information giving, and daily physical and psychological support. Although research has focused on comparing the SCNs of patients with specific diseases, fewer studies have considered the effect of race or ethnicity on SCNs (Li et al., 2013). As reported in a study by Ashing-Giwa et al. (2004), African American, Asian, Caucasian, and Latina women shared similar concerns about physical and psychological health, possibility of cancer recurrence, body image and intimacy, and the effects of their disease on relationships with their children and others. However, important differences related to various concerns existed among patients based on their ethnicity (Im et al., 2008). For example, as noted by Ashing-Giwa et al. (2004), Caucasian women reported that support groups provided encouragement and information regarding treatment and recovery. In addition, African American breast cancer survivors noted the strength of the African American community, which led them to feelings of resilience. Asian American breast cancer survivors reported benefiting from a positive attitude and educating themselves on treatment side effects and recovery. Hispanic, African American, and Asian American survivors all reported the significance of spirituality and prayer. Giuliani et al. (2016) studied SCNs among patients with lung cancer, concluding that minority and older adult patients were more likely to report more unmet SCNs. However, other studies have reported that older patients with cancer are less likely to report more SCNs and care coordination needs (Søndergaard, Grone, Wulff, Larsen, & Søndergaard, 2013).

Despite the increasing number of ethnically diverse older adults living with advanced cancer in urban settings (McLafferty & Wang, 2009), there are less data describing the effects of distress on SCNs in this patient population. Previous research addressing the impact of distress on people diagnosed with cancer has predominantly involved Caucasian patients with breast cancer (Yoo, Levine, & Pasick, 2014). The relationships between distress and SCNs remain understudied among ethnically diverse older adults with advanced cancer. The primary aims of this cross-sectional survey were to estimate the prevalence of SCNs and distress, as well as to describe the connections among patient characteristics, SCNs, and distress in ethnically diverse older adults with advanced or recurrent cancer who receive their care in an economically challenged community.

Theoretical Framework

This study is guided by the supportive care framework for cancer care (SCFCC), which was designed to help managers and healthcare professionals assess patient needs and distribute resources appropriately (Fitch, 2008). The SCFCC is appropriate for use at the program and individual patient levels. It is based on the construct of human need, and it aids in conceptualizing how humans cope with and experience all aspects of cancer, including diagnosis and treatment (Fitch, 2008; MacIsaac, Harrison, Buchanan, & Hopman, 2011). In addition, the SCFCC is comprehensive because it addresses the trajectory of patient illness, from prediagnosis to palliation and family bereavement (Fitch, 2005, 2008). This framework separates the SCNs of patients with cancer into the following domains: emotional, informational, physical, practical, psychological, social, and spiritual. The SCFCC was employed in the current study because of the similarities between the framework’s domains and those included in the SCN Survey Short Form 34 (SCNS-SF34) (see Table 1). The SCNS-SF34 was selected because it is a validated tool commonly used to measure the SCNs of patients with cancer.

Methods

Sample and Setting

Participants were recruited from a parent study (Weiss, Kolidas, Stein, & Moadel, 2010) assessing the distress of patients and their caregivers and evaluating the feasibility of distress screening at Montiefiore Medical Center, which is located in Bronx, New York, an economically challenged urban community. In 2016 in New York’s Bronx County, where this study took place, of the more than 1.4 million residents, 16% were aged older than 60 years, 31% lived below the poverty level, and 55% were Hispanic, 29% African American, and 10% Caucasian (United States Census Bureau, 2016). Inclusion criteria for the current study were adults aged 60 years or older who were enrolled in the parent study, able to read and understand English, and diagnosed with advanced (stage III or IV) or recurrent cancer and who were not enrolled in a hospice program. Data were reported from the first 100 participants who consented to participate in the parent study after the institutional review board of Montefiore Medical Center approved an amendment allowing access to participants’ medical records and the inclusion of the SCNS-SF34 in the parent study.

A sample size of 100 patients was determined using an estimation of the prevalence of distress in patients with cancer, with a conservative assumption of 35% prevalence. Using the Clopper–Pearson exact confidence interval (CI) formula, with an expected precision of 10.5%, a sample size of 87 would produce a two-sided 95% CI with an actual width equal to 0.209 for the lower CI and an upper CI of 25.3%–45.7%. Adjusting for a 10% dropout and attrition rate because of participant withdrawal, refusal, or inability to complete survey study instruments, the minimum required sample size was 97; rounding up, 100 patients were included.

The current study reports results of the SCNS-SF34. Data from one instrument used in the parent study, the Distress Thermometer (DT), were also analyzed (National Comprehensive Cancer Network, 2003). Chart reviews (completed using paper charts and the electronic health record) were used to verify patient diagnosis, disease stage, treatment history, neighborhood of residence, and number of comorbid conditions. Participants were recruited from February 2015 to December 2016 in medical, radiation, and surgical oncology outpatient clinics. After participant completion of the informed consent process, a member of the research team sat with the participant while he or she completed the packet of study forms (a demographic form, the SCNS-SF34, and the DT).

Instruments

The SCNS-SF34 is a 34-item self-administered, cancer-specific questionnaire designed to assess patients’ perceived level of unmet supportive care across five domains: health system and information (11 items), patient care and support (5 items), physical and daily living (5 items), psychological (10 items), and sexuality (3 items) (Bonevski et al., 2000). In addition, the SCNS-SF34 measures the magnitude of perceived needs using a five-point Likert-type scale ranging from 1 (no need or not applicable) to 5 (high need) and has two broad categories of “no need” and “need” (Boyes, Girgis, & Lecathelinais, 2009). Internal validity has been demonstrated, with a Cronbach alpha of 0.87–0.96 (McElduff, Boyes, Zucca, & Girgis, 2004). Reliability has also been demonstrated, with a Cronbach alpha of 0.86–0.95. The SCNS-SF34 was used to determine patients’ levels of need and how patients prioritize their needs. This instrument was chosen because it aligns with the construct of human needs, which is the foundation of the SCFCC. One domain of the SCFCC, the spiritual domain, is not included in the SCNS-SF34. The SCNS-SF34 score was standardized using a summated score; standardized domain scores ranged from 0–100, with larger values indicating greater perceived need.

The DT is a standardized measure consisting of a single question that asks the respondent to report his or her distress level on a scale of 0 (no distress) to 10 (extreme distress). A score of 4 or greater on the DT was considered to represent clinically significant distress (Holland et al., 2013). The DT has been validated for detection of distress in patients with cancer. Ma et al. (2014) performed a meta-analysis that reported a pooled sensitivity of 0.81 (95% CI [0.79, 0.82]) and a pooled specificity of 0.72 (95% CI [0.71, 0.72]) at the cutoff score of 4.

Analysis

Participant characteristics were summarized numerically using descriptive statistics. To assess the correlation between SCNs and distress, the Spearman rank correlation coefficient was used. The distress score, which ranges from 0–10, was dichotomized into less than 4 (low distress) and 4 or greater (high distress); a binary logistic regression model was then fitted to assess risk for high distress. The Kruskal–Wallis test was used to determine if the distress and SCNs differed by self-reported race or ethnicity. After observation of statistical significance with the Kruskal–Wallis test, post hoc pairwise comparisons were made among all possible groups using the Mann–Whitney U test. Type I error was adjusted using a Bonferroni correction. SCN scores did not meet a normality assumption; as a result, the authors performed different power transformations (e.g., logarithm, square root). Even after transformation, scores did not meet the normality assumption, so the authors dichotomized each SCN domain based on its median score. In addition, each dichotomized variable was assessed for its association with the patient characteristics of gender, race or ethnicity, children in the home, number of comorbidities, and neighborhood location (measured by a surrogate indicator, zip code) using logistic regression models. Statistical analyses were completed using IBM SPSS Statistics, version 23.0.

Results

Participant Characteristics

The mean age of participants was 70.7 years (SD = 7.39, range = 60–91 years). Sixty-nine participants were men, and most (n = 56) were African American. Participants who preferred not to identify with any race or ethnicity and those who self-identified with more than one race or ethnicity were classified as mixed/other. Participants predominantly resided in Bronx County (n = 80); however, all participants received their care in this underserved community. The mean number of comorbid conditions was 3.9 (range = 0–9). The five most common comorbid conditions were hypertension (experienced by 74 participants), hypercholesterolemia (50 participants), non-insulin–dependent diabetes mellitus (34 participants), coronary artery disease (23 participants), and chronic kidney disease (23 participants). See Table 2 for additional sample characteristics.

Distress

Sixty-seven participants did not have clinically significant distress, and 33 participants reported clinically significant distress, as indicated by a distress score of 4 or greater. The five most prevalent unmet SCNs were in the physical and daily living domain (i.e., concerns about not being able to do things, reported by 51 participants; pain, reported by 46 participants; and lack of energy or tiredness, reported by 40 participants) and in the psychological domain (i.e., fear of cancer spreading, reported by 52 participants, and concerns about the worries of those close to me, reported by 44 participants).

Relationships Between Supportive Care Needs and Distress

Relationships between each of the five domains of the SCNS-SF34 and distress were assessed. Spearman rank correlations showed a weak positive relationship of distress to the health system and information domain (r_s = 0.36) and a moderately positive relationship to the patient care and support domain (r_s = 0.41). Results indicate that participants who reported distress were more likely to have needs in the health system and information domain and in the patient care and support domain. A strongly positive relationship was noted between distress and the physical and daily living domain (r_s = 0.63), as well as a moderately positive relationship between distress and the psychological domain (r_s = 0.59), as seen in Table 3. Overall, findings suggest that patients with distress are very likely to have SCNs.

As hypothesized, patients in the high-distress group tended to have significantly higher scores in the following domains: health system and information (U = 576.5, p = 0.000); physical and daily living (U = 549.5, p = 0.000); and psychological (U = 332.5, p = 0.000) (see Table 4).

Results of the analysis showed a significant difference in distress and the psychological domain by self-reported race or ethnicity. Because distress and the psychological domain differed by race, the authors conducted six pairwise comparisons among the groups of African American, Caucasian, Hispanic, and mixed/other using the Mann–Whitney U test. A Bonferroni correction was used to control for type I error, and the adjusted alpha level was revised to p = 0.008. After correction, results were no longer significant.

The estimated odds of high needs in the health system and information domain are 1.41 times greater in patients for every additional comorbidity after adjusting for gender, race, child-raising, and neighborhood (see Table 5). An association was observed between needs in the physical and daily living domain and comorbid disease. The estimated odds of high needs in the physical and daily living domain are 43% greater for every additional comorbidity after adjusting for gender and race (see Table 6). In addition, gender and race are associated with needs in the sexuality domain; the odds of high needs in this domain are 3.5 times greater for men than for women (see Table 7). No associations were observed between the psychological domain and patient characteristics or between the patient care and support domain and patient characteristics.

Discussion

The purpose of this study was to estimate the prevalence of distress and SCNs, as well as to describe relationships among patient characteristics, SCNs, and distress, in ethnically diverse older adults with advanced or recurrent cancer at a large urban academic medical center. One of the major strengths of this research is that more than half of the participants reported being of an ethnic minority, which better captures the perspectives of an understudied population. The current findings support prior research on patients with advanced cancer, namely that those who are economically disadvantaged and who self-report minority status are more likely to report distress (Acquati & Kayser, 2017), as well as that more psychological needs are associated with higher distress (Uchida et al., 2011). Such findings are in contrast to earlier studies reporting that older adults experience lower levels of distress than younger adults (Graves et al., 2007; Kim et al., 2013; Lambert et al., 2013; Tuinman, Van Nuenen, Hagedoorn, & Hoekstra-Weebers, 2015).

The seminal research of Zabora, BrintzenhofeSzoc, Curbow, Hooker, and Piantadosi (2001) showed that African Americans reported more distress than Caucasians, but the difference was small. In addition, in a matched sample of African American and Caucasian men with prostate cancer, Nelson et al. (2010) determined that African American men reported the same levels of distress as Caucasian men but exhibited more resilience than their Caucasian counterparts. In the current sample, distress was not affected by self-reported race; these data suggest that age and advanced disease are more strongly associated with distress than with race or ethnicity.

As noted, many previous studies of distress and SCNs have focused on Caucasian women with breast cancer. Even when multiple cancer sites are examined, findings continue to represent primarily Caucasian patients, or race is not included among the patient characteristics at all (Admiraal et al., 2013; Baken & Woolley, 2011; Smith et al., 2007; Snyder et al., 2008). The current sample also reported higher median scores in all five domains of the SCNS-SF34 compared to the validated distributions of summated scales for the SCNS-SF34 (McElduff et al., 2004).

Previous research is inconclusive about the most prevalent SCNs experienced by patients. Similar to findings in the current study, McDowell, Occhipinti, Ferguson, Dunn, and Chambers (2010), in an investigation involving 439 adults aged older than 20 years with mixed cancers in Australia, reported that the five most prevalent unmet needs were in the physical and daily living domain and in the psychological domain; lack of energy was the most commonly experienced need, followed by the patient’s inability to do things he or she could once do, fears about the cancer spreading, concerns about the worries of those close to the patient, and uncertainty about the future. In contrast, Park and Hwang (2012) reported that among 52 women with recurrent breast cancer in Korea, the most common unmet SCNs were in the health system and information domain (being informed about cancer, which is under control or diminishing; being informed about things the patient can do to help him- or herself to get well) and in the patient care and support domain (waiting a long time for clinic appointments). Results from the current study demonstrated a weak relationship between needs in the physical and daily living domain and distress, controlling for patient characteristics (race, gender, treatment status), which is consistent with previous findings.

In the current study, the odds of high needs in the sexuality domain were greater for men than for women. This finding should be interpreted with caution, given the high prevalence of prostate cancer diagnoses in this sample (n = 38). A systematic review of the literature regarding unmet SCNs of men with prostate cancer showed that men with prostate cancer report high psychological and emotional needs and high interpersonal and intimacy needs (53% and 65%, respectively) (Paterson, Robertson, Smith, & Nabi, 2015). Men with prostate cancer, from middle age onward, experience significant psychosexual needs throughout the cancer trajectory (O’Brien et al., 2011). The current authors’ findings are consistent with prior research involving patients with prostate cancer.

Limitations

The current study had some limitations. The theoretical framework that was used to guide this study included a spiritual domain, which was not addressed by the domains of the SCNS-SF34. Therefore, it was not possible to determine if relationships exist among spirituality, distress, and SCNs. Spirituality has been found to be an important factor in coping with cancer for older adults and minorities (Mollica & Nemeth, 2015; Sterba et al., 2014; Wenzel et al., 2012).

Convenience sampling and self-report survey were used, which limits generalizability of the study results (Acharya, Prakash, Saxena, & Nigam, 2013). Because a cross-sectional design that included patients at multiple places on the cancer trajectory was used, the number of SCNs and levels of distress experienced may have been related to time since diagnosis or time since completing anticancer therapy. The authors did compare participants currently receiving treatment to those not undergoing treatment and found no significant differences related to number of SCNs or levels of distress.

Implications for Nursing

Oncology nurses must take a comprehensive and holistic approach to the assessment of patient needs when caring for older adults with advanced or recurrent cancer. Nurses should encourage patients to communicate and prioritize their needs. Assumptions that older adults with advanced cancer only express SCNs in the physical domain and do not report distress no longer hold. Findings from the current study show that all domains must be assessed, including sexuality, and patient distress must be addressed. A critical component of assessment is understanding how patients experience support, which is essential in helping patients adapt to the life changes associated with a cancer diagnosis and subsequent treatment (Grange, Matsuyama, Ingram, Lyckholm, & Smith, 2008). The SCNS-SF34 is available for download as a PDF (https://bit.ly/2Jlz5u9), along with instructions on scoring. It can be used for research purposes or in clinical practice and is easily administered at any point in the cancer trajectory (McElduff et al., 2004). The use of a standardized measure aids in documentation and communication of needs to other members of the healthcare team. The SCNS-SF34 can be used as a quality assurance tool to help oncology program managers allocate resources for additional programs, guided by patient-reported needs.

Because most patients experience multiple SCNs coupled with distress, complex care coordination and involvement of an interprofessional team may be required. Identifying and addressing such needs can prevent patient distress, poor quality of life, and dissatisfaction with care (Alananzeh et al., 2016), as well as resultant increases in healthcare utilization and costs (Paterson et al., 2015). These data from the current study did not show a relationship between distress and race or ethnicity; therefore, the impact of multiple comorbidities may have been greater than the impact of race or ethnicity on quality of life for this group of older adults with advanced or recurrent cancer.

Conclusion

The population of ethnically diverse older adults living with advanced or recurrent cancer is growing rapidly. The current authors found that distress is not affected by self-reported race or ethnicity among a sample of ethnically diverse older adults receiving care in an economically challenged neighborhood. Nurses and other members of the healthcare team should incorporate the assessment of SCNs and distress among older adults into routine care and strive toward further understanding of the impact of distress and SCNs on the lives of members of this population. Additional research is needed to determine if spirituality, personal relationships, or relationships with the healthcare team affect reported distress or SCNs.

About the Author(s)

Hilda Haynes-Lewis, PhD, ANP, ANP-BC, AOCNP®, is an adult nurse practitioner at Montefiore Medical Center in Bronx, NY; Margaret F. Clayton, PhD, APRN, is an associate professor in the College of Nursing at the University of Utah in Salt Lake City; Shankar Viswanathan, DrPH, MSc, is an assistant professor and Alyson Moadel-Robblee, PhD, is the director of the Psychosocial Oncology Program, both in the Department of Epidemiology and Population Health at the Albert Einstein College of Medicine in Bronx, NY; and Lauren Clark, RN, PhD, FAAN, is a professor in the College of Nursing and Michael Caserta, PhD, is a professor in the Gerontology Interdisciplinary Program, both at the University of Utah in Salt Lake City. Clayton has previously consulted for the Patient-Centered Outcomes Research Institute and is supported by funding from the National Cancer Institute. Haynes-Lewis completed the data collection. Haynes-Lewis, Viswanathan, and Moadel-Robblee provided statistical support. Haynes-Lewis, Clayton, Viswanathan, and Clark provided the analysis. Haynes-Lewis, Clayton, Viswanathan, Moadel-Robblee, and Clark contributed to the manuscript preparation. All authors contributed to the conceptualization and design. Haynes-Lewis can be reached at hayneslewis.np@gmail.com, with copy to ONFEditor@ons.org. (Submitted July 2017. Accepted February 7, 2018.)