Introduction
Palliative care (PC) is patient and family-centered supportive care intended to improve symptoms, reduce caregiver burden, coordinate care, and improve quality of life for patients diagnosed with serious illness. National Quality Forum (NQF, 2006) guidelines indicate the need for timely implementation of palliative services to yield the best outcome for the patient. Ideally, PC should be initiated upon initial diagnosis with a serious illness and delivered in synchrony with disease-specific treatment until symptom relief or patient death (NQF, 2006; Ferrell et al., Reference Ferrell, Temel and Temin2016; NCCN, 2020). The utilization of PC services does not impede the disease-specific medical regimen, but rather ensures that patient needs are met outside of treatment (Ferrell et al., Reference Ferrell, Temel and Temin2016). Given the benefits of PC, these services are most favorable for patients with serious illnesses, such as cancer (Taber et al., Reference Taber, Ellis and Reblin2019).
The American Society of Clinical Oncology (ASCO) states that all patients with advanced cancer should receive a referral to a multidisciplinary PC team for tailored and timely PC services (Ferrell et al., Reference Ferrell, Temel and Temin2016). Despite clear recommendations from ASCO, NQF, and others supporting the use of PC for patients with serious illness, awareness and understanding of PC remains low in the general US population, as well as among those diagnosed with serious illnesses, such as cancer (NQF, 2006; Ferrell et al., Reference Ferrell, Temel and Temin2016; Taber et al., Reference Taber, Ellis and Reblin2019; Trivedi et al., Reference Trivedi, Peterson and Ellis2019). The Health Information National Trends Survey (HINTS) data were utilized to evaluate the knowledge of PC among a nationally representative sample from the general population and found that 71% of participants had never heard of PC, 18% knew a little about PC, and only 11% could explain PC to someone else (Trivedi et al., Reference Trivedi, Peterson and Ellis2019). Males and respondents of color had a significantly lower likelihood of PC awareness (Trivedi et al., Reference Trivedi, Peterson and Ellis2019). These findings highlight the importance of PC knowledge, as a lack of awareness may act as a barrier that hinders the ability to benefit from palliative services should they be warranted in the future.
In addition, the conflation of hospice and PC is often a barrier for the provision and acceptance of PC services (Slomka et al., Reference Slomka, Prince-Paul and Webel2016). To qualify for hospice care, a patient must receive a terminal diagnosis and have a survival expectancy of less than 6 months that has been confirmed by two physicians (U.S. Department of Health and Human Services Health Care Financing Administration, 2001), whereas to qualify for PC, a patient must have a serious illness that would benefit from symptom management and quality-of-life improvement (LeMond and Allen, Reference LeMond and Allen2011). Despite this distinction, past researchers have found that 29% of physicians reported that PC and hospice were the same thing (Snyder et al., Reference Snyder, Hazelett and Allen2013). The conflation of these distinct care options may lead to physicians overlooking services or not communicating adequately with patients, leading to knowledge deficits and underutilization (Temel et al., Reference Temel, Greer and Muzikansky2010).
This study examines knowledge, understanding, and perceptions of PC among cancer survivors using survey data collected from the HINTS (HINTS 5, 2018). Given the ASCO guidelines for timely PC referral, we hypothesized that cancer survivors would have a high level of PC knowledge and accurate perceptions of the goals of PC (Ferrell et al., Reference Ferrell, Temel and Temin2016). Consistent with the general population (Trivedi et al., Reference Trivedi, Peterson and Ellis2019), women and whites (versus respondents of color) were hypothesized to have greater PC knowledge than their counterparts. It was also hypothesized that factors related to perceived health and greater health care utilization, such as those with a regular source of care (versus no regular source of care), worse self-rated health (versus better self-rated health), greater reported quality of health care (versus worse reported quality of health care), and seeking cancer information (versus not seeking cancer information) would be positively associated with PC awareness. Finally, adolescent and young adults (AYAs, diagnosed 15–39 years old; NCI, 2020) represent a unique group of cancer survivors that have previously reported inadequate health literacy and health care coverage (Parsons et al., Reference Parsons, Schmidt and Harlan2014; Richter et al., Reference Richter, Mehnert and Forstmeyer2019), as a result, we hypothesized that AYAs would have low PC awareness.
Methods
Data were obtained from HINTS 5, Cycle 2, a nationally representative survey of US adult health-related information administered by the National Cancer Institute's Division of Cancer Control and Population Sciences (HINTS 5, 2018). Data were collected using a single-mode mail survey from January through May 2018 (HINTS 5, 2018). Further information can be obtained at hints.cancer.gov. This research is exempt from IRB review.
HINTS respondents who indicated that they had ever been diagnosed with cancer were retained in the sample and are referred to as “cancer survivors” in the present study. HINTS respondents were asked to describe their level of knowledge about PC in a single item with response options, including: (1) I've never heard of it; (2) I know a little bit about palliative care; and (3) I know what palliative care is and could explain it to someone else. Those who indicated PC knowledge (response options 2 or 3) were asked to respond to the following items with a Likert-type scale that included five categories: (1) strongly agree, (2) somewhat agree, (3) somewhat disagree, (4) strongly disagree, and (5) don't know. Respondents’ beliefs about the goals of PC were assessed with four statements: (1) help friends and family to cope with a patient's illness; (2) offer social and emotional support; (3) manage pain and other physical symptoms; and (4) give patients more time at the end of life. Respondents’ attitudes toward PC were assessed with five statements: (1) accepting palliative care means giving up; (2) it is a doctor's obligation to inform all patients with cancer about the option of palliative care; (3) if you accept palliative care, you must stop other treatments; (4) palliative care is the same as hospice care; and (5) when I think of “palliative care,” I automatically think of death.
Demographic characteristics collected in the HINTS include age, sex, race/ethnicity, education level, health insurance coverage, and annual household income. Respondents were asked to report their self-rated health in a single item, “In general, would you say your health is…” drawn from the Short Form 12 (SF-12) Quality of Life assessment, with five response options: (1) excellent, (2) very good, (3) good, (4) fair, and (5) poor (Ware et al., Reference Ware, Kosinski and Keller1996). Respondents were also asked to indicate if they sought cancer information and if they had a regular source of medical care, both with binary yes/no responses.
Statistical analysis
Frequencies were conducted to describe the sample and calculate the response percentages for the knowledge and attitude items. Bivariate analyses were then conducted with variables selected for their hypothesized significance to the outcome. Variables significant at p < 0.10 were included in the multivariable logistic regression model. Consistent with prior research among the overall HINTS respondents, a multivariable logistic regression model was used to examine the knowledge of PC (Trivedi et al., Reference Trivedi, Peterson and Ellis2019). This dichotomized outcome indicated that those who reported (1) “I know what palliative care is and could explain it to someone else” or “I know a little bit about palliative care” versus (0) “I've never heard of it” (Trivedi et al., Reference Trivedi, Peterson and Ellis2019). Race/ethnicity was dichotomized as white (non-Hispanic/Latino) versus respondents of color (Hispanic/Latino, Black, Asian, American Indian, and Pacific Islander) to represent ethnic and racially minoritized individuals. All tests were two-tailed, with an alpha criterion of p < 0.05. Descriptive statistics using jackknife replicate, weighted percentages were utilized in the analyses in order to generate population estimates (HINTS 5, 2018). Hot deck imputation was utilized to account for missing data (HINTS 5, 2018). All analyses were conducted using SAS statistical software (Version 9.4; SAS Institute Inc).
Results
A total of 593 individuals reported a personal history of cancer, among the overall respondents (N = 3,434). Table 1 provides unweighted frequencies and weighted percentages to characterize the sample of cancer survivors. Figure 1 provides a visual depiction of PC knowledge among cancer survivors. The majority of the sample (N = 378; 65.6%) reported that they had never heard of PC, some (N = 112; 17.8%) indicated knowing a little bit about PC, while fewer (N = 95; 16.6%) reported knowing what PC is and could explain it to someone else.
Unweighted frequencies and weighted percentages of palliative care knowledge among cancer survivors (N = 593).
Weighted characteristics of sample (N = 593)
a Non-Hispanic.
Of those who had heard of PC (N = 207; 34.4%), most strongly agreed that the goals of PC were to: manage pain and other physical symptoms (71.9%), offer social and emotional support (56.4%), help friends and family cope with a patient's illness (47.3%), and give patients more time at the end of life (27.3%; Table 2). The majority of respondents strongly disagreed with the following statements: accepting palliative care means giving up (61.6%), if you accept palliative care, you must stop other treatments (49.7%), and palliative care is the same as hospice care (29.0%). Respondents strongly agreed that it is a doctor's obligation to inform all patients with cancer about the option of palliative care (52.3%). The majority of respondents somewhat agreed with when I think of “palliative care,” I automatically think of death (39.3%).
Frequencies and weighted percentages of palliative care knowledge and attitudes (N = 207)a
a Unweighted frequencies and weighted percentages displayed.
Table 3 provides weighted bivariate and multivariable models of correlates of PC knowledge. In bivariate analyses, survivors who were female (versus male), more highly educated, those with a greater annual household income, and those who sought information about cancer (versus those who did not seek information) were significantly more likely to know about PC. Survivors less than 40 years of age (versus those age 40+) and those with greater self-rated health (versus worse self-rated health) were significantly less likely to know about PC. In multivariable analysis, males, those with a lower education level, and those with worse self-rated health were significantly less likely to report knowledge of PC. Survivors with a lower annual household income and those who did not seek cancer information were marginally less likely to know about PC (p < 0.10). No other variables retained significance in the adjusted model.
Weighted bivariate and multivariable models of correlates of knowledge of palliative care (N = 593)
a Includes Black or African-American (non-Hispanic), Hispanic, Asian (non-Hispanic), and other (non-Hispanic).
+ p < 0.10.
* p < 0.05.
** p < 0.01.
*** p < 0.001.
Discussion
The findings of the present study indicate a lack of knowledge regarding PC among cancer survivors. Over 65% of respondents indicated that they had never heard of PC (as compared with 71% previously reported in the general population; Trivedi et al., Reference Trivedi, Peterson and Ellis2019). The similarity in PC knowledge between cancer-affected individuals and the general population is concerning, as cancer survivors have experienced a serious illness and should have greater awareness of PC (Ferrell et al., Reference Ferrell, Temel and Temin2016). ASCO guidelines indicate the timely receipt of palliative services at any stage of a serious illness, thus having cancer should heightens survivor's awareness of PC in comparison with the general population (NQF, 2006; Temel et al., Reference Temel, Greer and Muzikansky2010; Ferrell et al., Reference Ferrell, Temel and Temin2016; NCCN, 2020).
Among survivors who reported PC knowledge, understanding of the goals of PC were accurate and perceptions of PC were positive with the exception of “Palliative care is the same as hospice care.” Half of the participants reported agreement with the statement or a lack of knowledge about this distinction, indicating that many survivors do not understand the differences between PC and hospice care. Therefore, even among those who do report knowledge of PC, further education is needed to ensure the accurate distinction between the two modes of supportive care. Providers have also reported the name “palliative care” to be a barrier to referral for early-stage survivors, such that physicians are more likely to direct survivors to care referred to as supportive care rather than PC (Fadul et al., Reference Fadul, Elsayem and Palmer2009). This suggests that there may be a lack of knowledge, comfort, and/or acceptance of PC within the medical community, and that physicians likely represent a key barrier in access and knowledge of PC services among survivors.
While having a regular source of care and quality of health care were not associated with knowledge of PC, those with greater self-rated health were significantly less likely to know about PC, while those who sought information about cancer were significantly more likely to know about PC. Respondents with greater self-rated health may be less likely to report knowledge of PC due to less frequent physician's appointments or a decreased need for PC services in comparison with those with worse self-rated health. On the other hand, respondents who seek cancer information may be more likely to report knowledge of PC due to greater engagement in their health and/or self-interest in available resources.
Demographic factors such as sex and race/ethnicity proved to be associated with PC knowledge. Female cancer survivors were more than twice as likely to report PC knowledge than males after controlling for current age at survey, education level, annual household income, self-rated health, and seeking cancer information, consistent with the general population (Trivedi et al., Reference Trivedi, Peterson and Ellis2019). It has been shown that females have greater health literacy than males, which may explain why females have greater PC knowledge (Zhou & Wang, Reference Zhou and Wang2019). Among cancer survivors, this discrepancy may also be explained by studies finding that females have a more accurate understanding of their illness than males (Fletcher et al., Reference Fletcher, Prigerson and Paulk2013) which may lead to the underutilization of PC by males (Kumar et al., Reference Kumar, Casarett and Corcoran2012). Future studies should aim to examine these gender differences in order to adequately address and increase PC use among male cancer patients.
Additional demographic factors such as education level and income were associated with PC knowledge, indicating that those who are more highly educated and those with a greater annual household income are more likely to report knowledge of PC. These findings suggest a disparity by socioeconomic status, as those who are less educated and have a lower income may be more at risk for underprovision of PC. Prior research has found that the level of education has an inverse relationship with cancer mortality, indicating that those who are less educated are disproportionately affected by their cancer (Albano et al., Reference Albano, Ward and Jemal2007). Similarly, numerous researchers have linked low income to worse cancer outcomes including mortality (Gupta et al., Reference Gupta, Wilejto and Pole2014; Stanbury et al., Reference Stanbury, Baade and Yu2016; van den Berg et al., Reference van den Berg, Büttner and van den Braak2019). Attention to the socio-demographic factors that place certain groups at risk for underprovision of PC may aid in efforts to mitigate socioeconomic status-driven cancer disparities.
Bivariate analysis revealed that respondents less than 40 years of age (AYA range) versus older respondents, were significantly more likely to know about PC; however, this was not significant in the multivariable analysis. Little research currently exists on PC awareness among AYAs; however, in the general population, the average age of participants who had heard of PC is considerably older than the AYA age range (52 years; Taber et al., Reference Taber, Ellis and Reblin2019). While not significant in multivariable analysis, univariate findings reveal that further analyses are needed in larger samples to examine age-specific effects.
The present study is not without limitations. The study had a low number of non-white participants limiting power for robust subgroup analysis. Therefore, we grouped respondents of color who may be more likely to experience healthcare discrimination (Campesino et al., Reference Campesino, Saenz and Choi2012). The sample characteristics suggest that respondents primarily had access to care, with 95.26% having insurance, 84.43% having a regular source of care, and 77.18% seeking cancer information. Despite this, the overall knowledge of PC remained low. Clinical factors regarding cancer type, stage, and treatment status were not collected, and patients with advanced-stage cancer who had PC referral may not be represented in the sample (e.g., survivorship bias; Nitecki et al., Reference Nitecki, Diver and Kamdar2018). Therefore, we could not distinguish treatment severity or between respondents on active treatment versus those in remission and/or cancer-free. Furthermore, the sample represents the long-term survivor population as the mean time since diagnosis was 13.83 years, allowing us to infer that these are long-term survivors who may have had less aggressive disease at diagnosis.
In summary, the findings of this study serve to quantify gaps in PC awareness among cancer survivors and identify groups at-risk for low knowledge. While there is a concerning gap in PC knowledge among cancer patients, improving awareness of PC is amenable to intervention. For example, increased clinical counseling to cancer patients about the benefits of PC and reducing stigma surrounding seeking services can increase exposure to PC and eliminate barriers to utilization of care (Lane et al., Reference Lane, Ramadurai and Simonetti2019; Shen and Wellman, Reference Shen and Wellman2019). In addition, physicians can target groups that are most vulnerable and develop effective and culturally appropriate modes of raising awareness of PC. Overall, increasing discussions among providers and patients regarding PC and/or PC referral guidelines would likely increase acceptance and utilization of PC services.
Author contributions
JS: Conceptualization, formal analysis, writing — original draft, reviewing, and editing. MBN: Conceptualization, writing — reviewing and editing. EMM: Conceptualization, writing — reviewing and editing. CYO: Conceptualization, writing — reviewing and editing. JEM: Conceptualization, writing — reviewing and editing. DRF: Conceptualization, writing — reviewing and editing. KAM: Conceptualization, formal analysis, writing — reviewing and editing.
Funding statement
JS, MBN, EMM, CYO, JEM, DRF, and KAM do not have funding information to declare.
Disclosures
JS, MBN, EMM, CYO, JEM, DRF, and KAM do not have information to disclose.
