Palliative Care Use in Advanced Cancer in the Garden State

Abstract

Background:

Cancer is the second leading cause of death in the United States. Patients with metastatic cancer have a high symptom burden. Major global and domestic cancer care recommendations advise integration of palliative care services for these patients. Palliative care is specialized care that can decrease cost, improve symptom burden, and improve quality of life. Patient factors driving the use of palliative care remain poorly understood but may include both physiological and psychological needs, namely, pain and depression, respectively.

Objective:

The objective of this study was to identify patient-level predictors associated with inpatient palliative care use in patients with metastatic cancer.

Methods:

This was a secondary analysis of the 2018 New Jersey State Inpatient Database. The sample was limited to hospitalized adults with metastatic cancer in New Jersey. Descriptive statistics characterized the sample. Generalized linear modeling estimated the effects of pain and depression on the use of inpatient palliative care.

Results:

The sample included 28697 hospitalizations for patients with metastatic cancer. Within the sample, 4429 (15.4%) included a palliative care consultation. There was a 9.3% documented occurrence of pain and a 10.9% rate of depression. Pain contributed to palliative care use, but depression was not predictive of an inpatient care consultation. Age, income category, and insurance status were significant factors influencing use.

Conclusion:

Understanding demographic and clinical variables relative to palliative care use may help facilitate access to palliative care for adults experiencing metastatic cancer.

Implication for Practice:

Increased screening for pain and depression may expand palliative care use for adults with metastatic cancer receiving inpatient care.

Cancer is the second leading cause of American deaths.¹ In 2021, 1.89 million Americans were diagnosed with cancer, and approximately 608000 died from the disease.² Health-related suffering in cancer patients is projected to increase significantly during the coming decades,³ and care preferences are often not achieved at the end of life.⁴ Palliative care seeks to reduce suffering and increase the achievement of patient end-of-life (EOL) goals. The World Health Organization and the American Society of Clinical Oncology advocate for palliative care access across the cancer continuum for all patients from the time of diagnosis to the EOL and into bereavement to support surviving caregivers.^5,6 From an economic standpoint, aging populations and treatment advances causing patients to live longer with advanced disease continue to escalate the overall cost of care.¹ The Centers for Disease Control and Prevention estimates that EOL care costs rose 28.3% from 2010 to 2020.¹

Inpatient palliative care consultation (IPCC) is an EOL quality indicator designated by the Institute of Medicine.⁴ An IPCC supports patients in achieving EOL care goals, decreases costs, improves symptom burden, and improves transitions to hospice.^7,8 To receive an IPCC, a patient’s provider, most often the oncologist, must place a referral for an IPCC. Oncologists, thus, are often considered the gatekeepers to palliative care for patients with cancer. Once the referral is made, the palliative care team can visit the patient for a consultation. The implementation of palliative care use is multifactorial and is dependent not only on patient factors, which this study examined, but also on provider factors, patient acceptance of care, and available system resources to provide the consult.

Unfortunately, IPCC has a low rate of use.⁹ One study found that although 19% of adult (≥18 years) inpatients met palliative care referral criteria, defined as any diagnosis of poor-prognosis cancer, New York Heart Association class IV congestive heart failure, or oxygen-dependent chronic obstructive pulmonary disease, only 39% of those eligible received IPCC.⁹ Qualitative researchers conclude that barriers to palliative services are multifactorial and often due to issues of access and logistics.¹⁰

Demographic Factors

Patients from different demographic backgrounds vary in their palliative care access and use. This study examined the potential influence of patient factors on the receipt of IPCC. Previous research has found that older age is associated with IPCC.^8,11,12 Palliative care needs are high in older adults, and older adults benefit from palliative care services.¹³ Several studies reported an association between higher income categories and increased IPCC.^11,12 One study reported that limited English proficiency resulted in unfavorable use patterns and outcomes.¹⁴ There is currently no consensus on what type of insurance is more likely to affect IPCC use.^11,15 Studies have differed in whether private insurance, Medicaid, or Medicare are associated with greater IPCC use.^11,12,15 Finally, although partnered status may impact family caregiving at the EOL,¹⁵ the relationship between partnered status and IPCC is not well understood.

In the Northeastern United States, including New Jersey, IPCC use is less prevalent relative to other geographic regions in the United States.^8,12,16 In 2020 alone, an estimated 53340 patients were diagnosed with cancer in New Jersey, and 15710 New Jersey residents died from their cancer diagnosis.² New Jersey ranks high in EOL spending relative to other states and first in the United States for the rate of hospitalizations during the last 6 months of life.¹⁷ New Jersey has a low rate of hospice use, despite high access to palliative care, which facilitates hospice use.^18,19 Because of incongruence between New Jersey’s high EOL healthcare use during the last 6 months of life and reported access to palliative care, it is crucial to understand predictors of IPCC referral among patients with metastatic cancers.

Furthermore, New Jersey has diverse racial demographics compared with the total United States population.²⁰ Evidence of an association between race and IPCC use has been mixed, with several studies reporting both increased and decreased use by African American patients in comparison with white patients.^{8,11,12,15,16,21} Worster et al²² showed no significant association between race and IPCC use in a single urban hospital. Studies examining race as a factor in IPCC used large, diverse samples and potentially had the power to detect these differences yet reported conflicting information.^{8,11,12,15,16,21} In addition, multiple systemic reviews have provided conflicting evidence.^23-25 Ultimately, the literature is inconclusive regarding which racial groups are using IPCC. Although race and demographic background remain contributing factors in the receipt of palliative care, it is important to examine demographic factors in combination with clinical need factors.

Clinical Factors

In addition to demographic factors, clinical factors also impact IPCC. Patients with metastatic disease are known to experience a multitude of symptoms that can impact their quality of life and healthcare use.^26-30 Two commonly occurring symptoms in metastatic cancer requiring additional healthcare services are pain and depression.^31-33 Pain has been found to increase the use of IPCC.³² Although the impact of depression is known to affect overall healthcare use,³⁴ little is known about the relationship between depression and IPCC use.²²

Demographic and clinical factors may combine to impact IPCC use. Although there are mixed data in IPCC use related to race, minority patients with a cancer diagnosis have been shown to receive less symptomatic relief than their nonminority counterparts. There are documented disparities in symptomatic relief in minority patients with cancer, for both pain and psychological symptoms.^25,35,36 A study of patients with brain metastasis in the Surveillance, Epidemiology, and End Results database found that all ethnic minorities were less likely to be prescribed antidepressants and anxiolytics compared with non-Hispanic White patients.³⁶ Another study found that Black patients with advanced breast cancer were half as likely to be prescribed psychotropics (32% White, 16% African American, P < .001).³⁵ Disparities in symptom management in racial groups highlight the importance of including racial groups when examining clinical and demographic factors in IPCC.

The examination of symptomatic need factors is an important component of understanding the use of IPCC to impact identification of needs and ultimately resource distribution. The aim of this study was to examine both demographic characteristics and clinical symptoms as factors predicting IPCC. We hypothesized that the presence of pain and depression would increase referral to IPCC in patients with advanced cancer. In addition, we hypothesized that patients who were older, female, higher income, White, partnered, English speaking, and enrolled on Medicare would have increased use of palliative care.

Methods

Conceptual Framework

A synthesis of 2 theories, the theory of healthcare utilization and the theory of unpleasant symptoms, were used to describe and explain factors hypothesized to influence healthcare use, as operationalized in this study as IPCC use. According to the theory of healthcare utilization, need factors predict the use of healthcare services.³⁷ The theory purports that a combination of predisposing factors (eg, age, partnered status, race, gender, and primary language), enabling factors (eg, payer source), and need factors (eg, pain and depression) will predict IPCC use. According to the theory of unpleasant symptoms, the co-occurrence of multiple symptoms such as pain and depression is synergistic.³⁸ The combination of theories described the multifactorial association between demographic and clinical factors and IPCC.

Data and Population

This study was a secondary analysis of the Healthcare Cost and Utilization Project (HCUP) New Jersey 2018, a publicly available database of all hospitalizations that took place in the state in the year 2018.³⁹ In 2018, New Jersey had 70 inpatient facilities that contributed to the database.⁴⁰ Each file or case is representative of 1 hospitalization, rather than 1 patient. Therefore, this study examined the patient-level factors present during a single hospitalization that contributed to IPCC engagement. Each case has up to 30 discharge diagnoses for the International Classification of Diseases, Tenth Revision, Clinical Modification format (ICD-10-CM), which describes conditions treated in the hospitalization.⁴¹ The original sample in New Jersey contained 917250 cases. The 2018 sample was limited to patients aged 18 years and older with metastatic cancer as indicated by ICD-10-CM codes C77.x, C78.x, and C79.x. Metastatic or advanced cancer is defined as cancer that has spread beyond its original site of disease.⁴² This limitation yielded a sample of 28697 cases.

For this study, IPCC was the outcome variable. Record of IPCC was operationalized by ICD-10 code Z51.5, which indicates an encounter for palliative care. Researchers have validated that the ICD code for IPCC does accurately represent a consult performed by a palliative care provider.⁴³ Demographic factors examined in this study were age, gender, partnered status, socioeconomic status, primary language, and race. Age, gender, income quartile, and race were available in the database. Language category was included and further dichotomized for this study to those with a primary language of English and those with a primary language other than English. Partnered status, a multicategory variable in the database, was also dichotomized to partnered or not partnered. Pain and depression were operationalized as the presence of an ICD-10-CM code for each condition in any of the discharge diagnoses and were recoded to serve as indicators of the presence of the symptom based on the presence of ICD-10-CM. The ICD-10-CM codes for pain included G89.3 and R52.x. Depression conditions included major, situation, endogenous, and adjustment disorders (ICD-10-CM codes F06.3, F32.x, F33.x, F34.x, and F43.x).

Statistical Analyses

Data were inspected and cleaned. A rate of less than 5% missing data was considered acceptable, and all target variables met this threshold.⁴⁴ Descriptive statistical analyses of mean, standard deviation, and frequencies were conducted. An independent sample t test was conducted to examine the age difference between cases that did and did not receive IPCC. We performed χ² tests to investigate associations between predisposing, enabling, and need factors and IPCC. Following these analyses, variables determined to be significant at P < .05 through either χ² testing or binary logistic regression were included in the final multivariable model. Because of previous literature, race was included despite nonsignificance in χ² and individual models.

Generalized linear models or binary logistic regression models were used to analyze the relationship between the predisposing, enabling, and need factors and IPCC. Significance was set at P < .05. Statistical analyses were performed in SPSS 27.

Results

Description of the Sample

The 2018 New Jersey State Inpatient Database sample contained 28697 cases, each representing a hospitalization of an adult with documented evidence of metastatic cancer. A full sample description can be found in Table 1. Of these hospitalizations, 4429 (15.4%) involved an IPCC. The data indicated that slightly more hospitalizations of females (n = 15093, 52.6%) took place compared with males. The average age was 67.16 years (SD = 13.8). The sample was predominantly white (n = 18661, 65.6%), followed by black (n = 4395, 15.3%), Hispanic (n = 3272, 11.5%), Asian or Pacific Island (n = 1058, 3.7%), American Indian (n = 27, 0.1%) and others (n = 1042, 3.5%). Most of the sample were residents of zip code areas within the highest median income bracket (>$79000/year; n = 14931, 52%), followed by the $59000 to $78999 category (n = 6445, 22.5%), $46000 to $58999 category (n = 3661, 12.8%), and $1 to $45999 category (n = 3495, 12.2%). The primary payer source was predominantly Medicare (n = 16236, 56.6%). The metastatic cancer sample had a 9.3% (n = 2672) incidence of pain, a 10.9% (n = 3127) incidence of depression, and a 1.5% (n = 433) incidence of both pain and depression.

Table 1 •.

Descriptive Statistics

Variable	Total, n (%)	IPCC Absent, n (%)	IPCC Present, n (%)	Sig
Age, y	Mean, 67.1			.00
<50	2547 (8.9)
50–64	8807 (30.7)	7572 (86.0)	1235 (14.0)
65–74	8505 (29.6)	7240 (85.1)	1265 (14.9)
75–84	6159 (21.5)	5138 (83.5)	1021 (16.6)
>85	2679 (9.3)	2105 (78.6)	574 (21.4)
Race				.81
White	18661 (65.6)	15795 (84.6)	2866 (15.4)
Black	4395 (15.5)	3722 (84.7)	673 (15.3)
Hispanic	3272 (11.5)	2745 (83.9)	527 (16.1)
Asian	1058 (3.7)	900 (85.1)	158 (19.9)
Other	1042 (3.7)	885 (84.6)	157 (15.1)
Gender				.26
Male	13604 (47.4)	11539 (84.8)	2065 (15.2)
Female	15093 (52.6)	12729 (84.3)	2364 (15.7)
Payer				.00
Medicare	16236 (56.6)	13639 (84.0)	2597 (16.0)
Medicaid	2732 (9.5)	2365 (86.6)	367 (13.4)
Private	8303 (28.9)	7066 (85.1)	1237 (14.9)
Other	1426 (5.0)	1198 (84.0)	228 (16.0)
Annual median income				.00
1–45999	3494 (12.2)	2911 (83.3)	548 (16.7)
46000–58999	3661 (12.8)	3156 (86.2)	505 (11.5)
59000–78999	6445 (22.6)	5564 (86.3)	881 (13.7)
>79000	14931 (52.3)	12499 (83.7)	2432 (16.3)
Partnered				.06
Not partnered	13864 (49.2)	11670 (84.2)	2194 (15.8)	.06
Partnered	14314 (50.8)	12164 (85.0)	2150 (15.0)
Language				.65
Primary English	25779 (89.8)	21792 (84.5)	3987 (15.5)
Primary other language	2918 (10.2)	2776 (84.9)	442 (15.1)
Pain				.00
No pain	26025 (90.7)	22316 (85.7)	3709 (14.3)
Pain	2672 (9.3)	1952 (73.1)	720 (26.9)
Depression				.10
No depression	25570 (89.1)	21655 (84.7)	3915 (15.3)
Depression	3127 (10.9)	2613 (83.6)	514 (16.4)
Pain + depression				.00
No pain + depression	28264 (98.5)	23950 (98.7)	318 (1.3)
Pain + depression	433 (1.5)	4314 (73.4)	115 (26.6)

Abbreviations: IPCC, inpatient palliative care consultation; Sig, significance.

Bivariate Analysis

There were significant differences between the age for those hospitalizations who received IPCC (M = 69.06, SD = 13.6) and the age for those who did not (M = 66.81, SD = 13.86, t (28 695) = −9.96, P < .001). The results of bivariate testing between predictor variables and IPCC as determined by χ² testing are shown in Table 1. When age was analyzed as a categorical variable, increased likelihood of IPCC was associated with increasing age (χ² [4, n = 28697] = 105.85, P < .001). There were significant differences between categories of median household income in IPCC (χ² [3, n = 28532] = 68.91, P = <.001). There were significant differences between insurance category groups and IPCC use (χ² [5, n = 28697] = 34.14, P < .001). Pain was positively significantly associated with IPCC use (χ² [1, n = 28 697] = 299.19, P < .001). In hospitalizations in which there was evidence of an IPCC, 26.9% also had documentation of pain. This is in comparison with 14.3% of the hospitalizations with IPCC that did not have documentation of pain (P < .001).

The presence of a depression diagnosis did not reach significance in IPCC use (χ² [1, n = 28697] = 2.70, P = .10). Hospitalizations with both diagnoses of pain and depression had a significant relationship with IPCC use (χ² [1, n = 28697] = 41.69, P < .001). The use of IPCC was 26.9% in hospitalizations with a diagnosis of pain, 16.4% in hospitalizations with a diagnosis of depression, and 26.6% in hospitalizations in which both pain and depression were documented.

Multivariate Analysis

A multivariable generalized linear model was used to estimate the effects of the factors of age, income, race, and pain on the odds of IPCC use. The omnibus test showed model significance (χ² [15, n = 28267] = 434.86, P < .001); thus, we rejected the null hypothesis. The results of the final model are in Table 2. Younger age was associated with decreased odds of IPCC occurring compared with those hospitalizations older than 85 years old (P < .001). The $46000 to $58999 income category (adjusted odds ratio [AOR] = 0.85; 95% confidence interval [CI], 0.77–0.95; P = .003) and the $59000 to $78999 income category (AOR = 0.83; 95% CI, 0.76–0.90; P = .00) were associated with decreased odds of IPCC compared with the median income category of more than $79000. The absence of documented pain resulted in decreased likelihood of IPCC (AOR = 0.41, 95% CI, 0.38–0.46; P < .001).

Table 2 •.

Final IPCC Model

IPCC	β	SE	AOR	Sig	AOR LLCI	AOR ULCI
Omnibus test	434.86 (df 15)			.00
Age, y (ref, 85+)
<50	−.79	0.09	0.45	.00	0.38	0.53
50–64	−.64	0.07	0.53	.00	0.46	0.60
65–79	−.50	0.06	0.61	.00	0.54	0.68
75–84	−.34	0.06	0.73	.00	0.63	0.80
Income (ref, >79000)
1–45999	−.07	0.06	1.07	.23	0.96	1.19
46000–58999	−.16	0.05	0.85	.00	0.77	0.95
59000–78999	−.19	0.04	0.83	.00	0.76	0.90
Race (ref, white)
Other	.03	0.09	1.03	.76	0.86	1.23
Asian	.03	0.09	1.03	.73	0.87	1.23
Hispanic	.09	0.05	1.09	.09	0.98	1.22
Black	.03	0.05	1.04	.51	0.94	1.14
Payer
Other	.17	0.09	1.18	.05	1.0	1.40
Private insurance	.09	0.05	1.09	.06	0.99	1.21
Medicaid	−.03	0.07	0.97	.67	0.84	1.12
Pain (ref, no pain)	−.87	0.05	0.42	.00	0.38	0.46

Abbreviations: AOR, adjusted odds ratio; AOR LLCI, adjusted odds ratio lower-level confidence interval; AOR ULCI, adjusted odds ratio upper-level confidence interval; df, degrees of freedom; ref, reference; Sig, significance.

Discussion

This study revealed multiple patient-level factors impacting IPCC in hospitalized patients with metastatic cancer in New Jersey, a densely populated state with high demographic variability. Overall, there was low IPCC use across patients with metastatic cancer. Results highlight that IPCC use is impacted by both clinical and demographic factors. The final generalized linear model found that patients’ age and median income status, the presence of pain, and the presence of both pain and depression were contributors in predicting IPCC use.

This model reinforces prior findings that patients with older age had an increased likelihood of IPCC use.^11,12 This may indicate that palliative care clinicians view age as an indicator of IPCC need; however, patients of all ages with life-limiting diagnoses, such as metastatic cancer, should have access to IPCC to improve symptoms and subsequently quality of life. Because palliative care clinicians are experts at tailoring care to individuals, these findings highlight the importance of efforts to increase IPCC across age groups for patients with metastatic cancer.

Our study found that patients in the highest and lowest income brackets were most likely to engage in IPCC in New Jersey hospitals. Increased rates of palliative care use were present in hospitalizations in the lowest median income category ($1–$45999, 16.3%) and the highest income category (>$79000, 16.7%) compared with hospitalizations in the $46000–$58999 (11.5%) and $59000–$78999 (13.7%) income categories. This finding differs from prior research, which found that the highest income category had the greatest use of IPCC.^11,12 Reasons for lower use of IPCC in middle-income categories may include a lack of availability of healthcare resources and patient perceptions of availability of payment for services. More information is necessary to understand this relationship fully; however, patients in the highest and lowest socioeconomic groups may have access to resources that facilitate IPCC use. Patients in the middle-income categories may have fewer resources to engage in additional care, such as IPCC, even though such care could benefit symptom reduction, goal consideration, and total cost burden.

This study found inferior odds of IPCC when either Medicaid or private insurance was the primary payer. These findings confirmed previous work that Medicare recipients were the most likely to receive IPCC and patients with private insurance or Medicaid had a lower likelihood of IPCC.^11,15 However, this finding was attenuated when the model was adjusted for age. It is likely that controlling for age removed the favorable effect of Medicare as this program is almost exclusively available to older people living in New Jersey. This study highlighted the need for education to both patients and providers regarding IPCC benefits regardless of insurance types. Patients increasingly have financial concerns regarding costs of care, and nurses’ knowledge of this barrier is increasingly important.

This study did not find significant differences in IPCC among racial groups. This supports the previous study by Worster et al,²² which found no variation in access to IPCC based on race. However, because of the rates of symptom occurrence across demographics, our finding may not indicate that patients in New Jersey are receiving equitable care across racial groups, and more research is needed to understand the relationship. Although rates of pain detection were equitable across racial groups, depression was not. Depression incidence showed variability across racial groups in this study. White patients had the highest occurrence of documented depression (12.7%), followed by black (7.3%), Hispanic (9.0%), Asian (4.0%), and other (7.3%) (P = .00) patients. Variability in rates of diagnosis of depression between racial groups is reported in the previous literature,^45,46 and previous studies report that the expression of symptoms in ethnic minorities may occur differently than in white patients,⁴⁵ thus leading to underdetection by clinicians. For example, Zhang et al⁴⁷ reported that African Americans were more likely to report insomnia, fatigue, and irritability than sadness. Racial minorities may have decreased identification of symptomatic need factors and treatment of such symptoms.^35,36,45 Underdiagnosis in racial groups may occur because of variability in the presentation of depression, combined with the coupling of oncologic symptoms of fatigue, low energy, sadness, and increased sleep time.

The lack of significance of depression in the study may be related to underdiagnosis of depression. Screening tools are available; however, concerns exist regarding the validation and consistency of use with patients at the EOL.⁴⁸ Efforts must be made to increase consistent depression screening practices to identify psychiatric need factors of depression in all patients with metastatic cancer. Screening tools should be validated in diverse demographics of palliative care populations to confirm efficacy in appropriately identifying symptoms. Following confirmation of screening tools, studies should reexamine the relationship between psychological and physiologic need factors and their relationship to IPCC. Consistency in screening practice across clinical settings may lead to increased identification of patients who could derive benefit from IPCC.

Ultimately, this study highlighted the need to educate both patients and clinicians on the availability and benefit of palliative care services to increase use. Further research is necessary to focus on the benefit of palliative care services among age categories, socioeconomic status, insurance type, and symptomatic need factors. Both researchers and clinicians should continue to strive to decrease barriers to service for all patients. Clinician education must continue to inform screening practices across patient groups and understanding and decreasing barriers to care. Healthcare policy must work to provide available palliative care resources and train adequate clinicians to respond to patient clinical need factors.

Limitations

This study had several limitations worth noting. Secondary analyses, by their nature, are limited by the data points and the rigor of the original data collection. The data analyzed in this study were collected by each inpatient facility and submitted to the Agency for Health Research and Quality. The dataset is maintained and made available for research and analytic purposes.³⁹ The ability to obtain additional data points, which may impact the dependent variable, is limited. Another limitation in this study was the accuracy of the provider billing codes. In addition, the Agency for Health Research and Quality data are based on cases, and although each case represents 1 hospitalization, an individual patient could be represented multiple times. Furthermore, the code for “encounter for palliative care” is based on the ICD-10-CM code. The use of this indicator code only signifies that an IPCC occurred; it does not quantify the amount or type of palliative care service provided to the patient. However, using this code as an indicator for palliative care consult is validated to be specific for IPCC in the inpatient setting.⁴³ The use of ICD-10-CM codes for pain and depression may have led to underdetection of both pain and depression. Patients may have subclinical depressive symptoms, which may not be coded or treated in practice. Another limitation is that the dataset is unable to link multiple hospitalizations of an individual patient, but rather provides a snapshot in time of patient acuity, and in the case of this study, current diagnoses and use of IPCC.

A further limitation is due to the compiling of all hospitals in 1 data source. This does not allow for the adjustment of provider-specific and institution-specific policies and procedures surrounding the use of IPCC. Likewise, administrative data sources, such as the one used in this study, use billing codes and thus do not include important clinical and contextual factors that potentially influence the outcomes. In the case of this study, these factors include, but are not limited to, the lack of information about the disease and response to current or past therapy, as well as each patient’s prior involvement with palliative care services before hospitalization.

Another limitation of this study was that the dataset only encompassed New Jersey cases, and therefore, findings cannot be generalized to other regions. A national database would give access to information on regional differences in providing palliative care and differences between regional populations. Future studies could use the National Inpatient Sample to report on regional differences and confirm current findings.

Barriers and facilitators to oncology palliative care include system, patient, and oncologist factors. Although this study focused on the demographic and symptom patient factors, it is unable to discern those patients who did not receive IPCC because of oncologist hesitancy to provide a referral or patient refusing the consultation. Although efforts have been made to implement triggers or checklists into oncology workflow to improve referral, the use of such tools varies.⁴⁹ Ultimately, IPCC rely on the combination of the perceived need by the patient and the evaluated need of the oncologist. It is necessary for all members of the team to participate in the screening of symptomatic need factors to identify patients who would benefit from IPCC.

Implications

This study confirmed the complexity of IPCC patterns and who is using care. First, study findings indicate that multiple demographic factors influence IPCC. Income category and payer source may be a barrier to IPCC, but understanding the causes of use or avoidance is limited. Additional survey-based and qualitative studies are necessary to examine barriers to palliative care and provider referral patterns. This study also added information on the relationship between pain and depression and IPCC, but further work is necessary to understand the implications of these relationships. It is important to understand the true incidence of pain and depression in IPCC patients; comprehensive screening could help ensure referral to services to support the level of patient need. Nurses and other team members must have valid and reliable scales to screen for clinical need factors uniformly across demographic groups. Education of nurses and providers may be necessary to ensure the identification of symptoms in patients with metastatic cancer. Increasing rigor and consistency of screening strategies to determine patient needs could result in more patients receiving IPCC use, and as a result, improve symptom burden in metastatic cancer patients.

This study highlighted treatment inconsistencies among patients receiving care in varying age, income categories, insurance categories, and symptomatic need factors. Nurses caring for patients with advanced cancers must educate all patients and families on palliative care benefits. Nursing education must provide adequate content regarding the EOL process to allow nurses to provide equitable care across populations.

Conclusion

Patients with metastatic cancer experience high levels of symptom burden. Inpatient palliative care consultation may improve these symptoms, but it is necessary to first identify patients needing such care. Using a combination of demographic and clinical factors to appropriately determine the need will assist in identifying patients. Screening for both psychological and physiologic need factors in oncology care may be necessary to change referral patterns. This study demonstrated the need for further research on predisposing, enabling, and need factors associated with IPCC so that future work can focus on eliminating barriers to inpatient palliative care use.