Advertisement

Hospice Referrals and Code Status: Outcomes of Inpatient Palliative Care Consultations Among Asian Americans and Pacific Islanders With Cancer

Open AccessPublished:April 25, 2011DOI:https://doi.org/10.1016/j.jpainsymman.2011.01.010

      Abstract

      Context

      Intensive palliative care consultations for plan of care may reduce racial differences in end-of-life care.

      Objectives

      To compare cancer patients' hospice referrals and code status changes after inpatient palliative care consultations by patient ethnicity and consultation intensity.

      Methods

      This observational cohort study prospectively recorded data for all adult cancer patients receiving palliative care consultations at the largest teaching hospital in Hawaii from 2005 through 2009. Chi-squared analyses compared hospice referral and code status changes with “Do Not Attempt Resuscitation” by patient characteristics and consultation intensity (more intensive plan of care vs. pain and/or symptom management without plan of care). Multiple logistic regression models analyzed factors associated with hospice referral and code status change.

      Results

      The 1362 consultations generated 454 (33.3%) hospice referrals and 234 (17.2%) code status changes. Controlling for age, gender, Karnofsky score, and preconsultation hospital days, Asian, Pacific Islander, and “other” ethnicities demonstrated increased likelihood of hospice referral vs. whites (adjusted odds ratios [AORs] 1.46–2.34, P<0.05). Intensive plan-of-care consultations were strongly associated with hospice referral (AOR 3.08, 95% confidence interval [CI] 2.33–4.07, P<0.0001). Controlling for consultation intensity reduced the association between ethnicity and hospice referral (AORs 1.35–2.06, P=0.03, “other” ethnicity; P=nonsignificant, Asian and Pacific Islander). Intensive consultations were strongly associated with code status change (AOR 2.96; 95% CI 2.08–4.22, P<0.0001). Ethnicity was not significantly associated with code status change.

      Conclusion

      Consultation intensity was the strongest predictor of hospice referrals and code status changes and reduced the ethnic variations associated with hospice referral.

      Key Words

      Introduction

      Racial differences in aggressive interventions at the end of life and hospice use have been demonstrated in previous research.
      • Johnson K.S.
      • Kuchibhatla M.
      • Tulsky J.A.
      What explains racial differences in the use of advance directives and attitudes toward hospice care?.
      • Hanchate A.
      • Kronman A.C.
      • Young-Xu Y.
      • Ash A.S.
      • Emanuel E.
      Racial and ethnic differences in end-of-life costs: why do minorities cost more than whites?.
      • Cohen L.L.
      Racial/ethnic disparities in hospice care: a systematic review.
      • Connor S.
      • Elwert F.
      • Spence C.
      • Christakis N.
      Racial disparity in hospice use in the United States in 2002.
      • Connor S.R.
      • Elwert F.
      • Spence C.
      • Christakis N.A.
      Geographic variation in hospice use in the United States in 2002.
      • Barnato A.E.
      • Anthony D.L.
      • Skinner J.
      • Gallagher P.M.
      • Fisher E.S.
      Racial and ethnic differences in preferences for end-of-life treatment.
      • Volandes A.E.
      • Paasche-Orlow M.
      • Gillick M.R.
      • et al.
      Health literacy not race predicts end-of-life care preferences.
      • Ngo-Metzger Q.
      • Phillips R.S.
      • McCarthy E.P.
      Ethnic disparities in hospice use among Asian-American and Pacific Islander patients dying with cancer.
      Inpatient pain and palliative care consultations enhance end-of-life care for diverse populations.
      • Manfredi P.L.
      • Morrison R.S.
      • Morris J.
      • et al.
      Palliative care consultations: how do they impact the care of hospitalized patients?.
      • Jenkins C.A.
      • Schulz M.
      • Hanson J.
      • Bruera E.
      Demographic, symptom, and medication profiles of cancer patients seen by a palliative care consult team in a tertiary referral hospital.
      • O'Mahony S.
      • Blank A.E.
      • Zallman L.
      • Selwyn P.A.
      The benefits of a hospital-based inpatient palliative care consultation service: preliminary outcome data.
      • Dhillon N.
      • Kopetz S.
      • Pei B.L.
      • et al.
      Clinical findings of a palliative care consultation team at a comprehensive cancer center.
      • Fernandes R.
      • Braun K.L.
      • Ozawa J.
      • et al.
      Home-based palliative care services for underserved populations.
      However, data comparing palliative care outcomes with ethnicity are limited,
      • Weng L.C.
      • Huang H.L.
      • Wilkie D.J.
      • et al.
      Predicting survival with the Palliative Performance Scale in a minority-serving hospice and palliative care program.
      particularly for inpatient settings
      • Hui D.
      • Elsayem A.
      • Palla S.
      • et al.
      Discharge outcomes and survival of patients with advanced cancer admitted to an acute palliative care unit at a comprehensive cancer center.
      and Asian American and Pacific Islander populations.
      • Fernandes R.
      • Braun K.L.
      • Ozawa J.
      • et al.
      Home-based palliative care services for underserved populations.
      • Brody A.A.
      • Ciemins E.
      • Newman J.
      • Harrington C.
      The effects of an inpatient palliative care team on discharge disposition.
      • Edrington J.
      • Sun A.
      • Wong C.
      • et al.
      Barriers to pain management in a community sample of Chinese American patients with cancer.
      • Im E.O.
      • Chee W.
      • Guevara E.
      • et al.
      Gender and ethnic differences in cancer pain experience: a multiethnic survey in the United States.
      Research on inpatient palliative care outcomes by ethnicity may clarify the role of palliative care consultations in reducing disparities in end-of-life care.
      It has been recommended that studies of inpatient palliative care teams compare different models of a hospital-based team, including models distinguished by the intensity of intervention provided.
      • Higginson I.J.
      • Finlay I.
      • Goodwin D.M.
      • et al.
      Do hospital-based palliative teams improve care for patients or families at the end of life?.
      One comparison examined patients by the indication for consultation, determined by the palliative care team as the key elements of the consultation.
      • Fromme E.K.
      • Bascom P.B.
      • Smith M.D.
      • et al.
      Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
      Most cancer patients are referred for consultations for pain or symptom management alone
      • Braiteh F.
      • El Osta B.
      • Palmer J.L.
      • Reddy S.K.
      • Bruera E.
      Characteristics, findings, and outcomes of palliative care inpatient consultations at a comprehensive cancer center.
      • Johnson C.E.
      • Girgis A.
      • Paul C.L.
      • Currow D.C.
      Cancer specialists' palliative care referral practices and perceptions: results of a national survey.
      whereas other patients receive consultations that require more time, coordination, and communication to address the plan of care. The latter consultations provide a more intense intervention
      • Higginson I.J.
      • Finlay I.
      • Goodwin D.M.
      • et al.
      Do hospital-based palliative teams improve care for patients or families at the end of life?.
      than the more frequently requested consultations for pain or symptom management.
      • Johnson C.E.
      • Girgis A.
      • Paul C.L.
      • Currow D.C.
      Cancer specialists' palliative care referral practices and perceptions: results of a national survey.
      Data are limited on the relationships between the indication for palliative care consultation, patient ethnicity, and outcomes of consultation.
      • Fromme E.K.
      • Bascom P.B.
      • Smith M.D.
      • et al.
      Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
      The more intense types of consultations for plan of care may be hypothesized to demonstrate enhanced outcomes related to plan of care, such as hospice referrals or changes in resuscitation status. Hospice care after discharge and changes in resuscitation status orders are important outcomes of inpatient palliative care consultations.
      • Hui D.
      • Elsayem A.
      • Palla S.
      • et al.
      Discharge outcomes and survival of patients with advanced cancer admitted to an acute palliative care unit at a comprehensive cancer center.
      • Fromme E.K.
      • Bascom P.B.
      • Smith M.D.
      • et al.
      Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
      • Delgado-Guay M.O.
      • Parsons H.A.
      • Li Z.
      • Palmer L.J.
      • Bruera E.
      Symptom distress, interventions, and outcomes of intensive care unit cancer patients referred to a palliative care consult team.
      • Gade G.
      • Venohr I.
      • Conner D.
      • et al.
      Impact of an inpatient palliative care team: a randomized control trial.
      • Weissman D.E.
      • Meier D.E.
      Center to Advance Palliative Care inpatient unit operational metrics: consensus recommendations.
      It may require considerable time, coordination, and communication expertise to overcome barriers to these outcomes. It is important to identify factors that may affect these outcomes for cancer patients who receive palliative care consultations, as the results will inform future studies to improve access and delivery of palliative care in diverse populations.
      The purpose of this study was to examine variation in hospice referral and change in resuscitation status after palliative care consultations by cancer patient race and ethnicity. A secondary purpose of the study was to examine whether intensity of consultation changed the relationship between these palliative care outcomes and these patient characteristics. This study analyzed data collected prospectively for all palliative care consultations for cancer patients at a major referral hospital serving an ethnically diverse patient population.

      Methods

      Design, Participants, and Setting

      This observational cohort study examined data collected prospectively on all adult cancer patients who received palliative care consultations from January 1, 2005 through December 31, 2009. Patients with a missing or noncancer admitting diagnosis were excluded (n=3758). The Pain and Palliative Care Team members collected all data. All consultations were performed at The Queen's Medical Center, a 542-bed teaching hospital, which serves as the major referral center for the Pacific basin. The study was approved by the University of Hawaii and The Queen's Medical Center Institutional Review Boards.

      Procedures

      At the time of the consultation, the consultation team recorded patient demographic characteristics, Karnofsky score,
      • Yates J.W.
      • Chalmer B.
      • McKegney F.P.
      Evaluation of patients with advanced cancer using the Karnofsky performance status.
      and preconsultation hospital length of stay (LOS) in days. At the initial consultation visit, the Pain & Palliative Care Department physician or advanced practice nurse recorded up to three indications for consultation, including pain, nonpain symptom management, and plan of care. These indications were not exclusive of one another, and patients could have one, two, or all three indications for consultation. Plan-of-care consultations required more prolonged discussions with the patient, family, and medical team to clarify goals of care, disposition, or appropriate use of medical interventions. At the close of the consultation, the consult team recorded disposition after hospital discharge, referral to hospice, and change in resuscitation code status from “full code” to “Do Not Attempt Resuscitation” (DNAR).

      Definitions

      At the time of hospital admission, patients reported their race and ethnicity. These reflected Hawaii's diverse population, including Caucasian, Japanese, Okinawan, Hawaiian, part-Hawaiian, Korean, Chinese, Micronesian, Pacific Islander, Filipino, Hispanic, African American, Native American, Middle Eastern, East Indian, and mixed ethnicities. For analyses, the ethnicities were clustered into the four largest groups: Caucasian, Asian (including Japanese, Okinawan, Korean, Chinese, Filipino, East Indian, and “other Asian”), Hawaiian/part-Hawaiian/Pacific Islander/Micronesian, and other races or ethnicities (including African American, Hispanic, Native American, cosmopolitan, Middle Eastern, unknown, and declined). Plan of care as a measure of intensity of consultation was analyzed as a dichotomous “yes/no” variable.

      Outcomes

      The main outcomes of interest were hospice referrals and changes in code status to DNAR. Hospice referral included patients referred to hospice care provided at all locations, including home, nursing home, inpatient hospice facilities, and hospital hospice units and was analyzed as a dichotomous “yes/no” variable. Patients who died before completion of hospice referral or who did not accept the hospice referral were coded as “no” for hospice referral. Change in code status to DNAR as a result of the consultation was analyzed as a dichotomous “yes/no” variable and was used to approximate other goals-of-care decisions not reflected in the hospice referral numbers.

      Analyses

      Descriptive statistics of the patient population were generated, and characteristics of patients by intensity of consultation (plan-of-care indication for consultation yes or no) were compared using Chi-squared analyses for categorical variables and t-test analyses for continuous variables. Factors associated with hospice referral and change in code status were each analyzed using multiple logistic regression models, controlling for age, gender, and ethnicity. A second model for each outcome also adjusted for plan of care as a measure of the intensity of consultation, as it was hypothesized that the level of intensity of consultation would have a profound impact on outcomes. The separate models were needed to determine the associations of the other variables and the outcome, with and without the consultation intensity factor.

      Results

      Demographic and clinical characteristics by patient race and ethnicity are displayed in Table 1. A total of 1362 patients with cancer received palliative care consultations during the five calendar years from 2005 through 2009. Fifty-two percent of the patients were male. Forty percent of the patients had the more intensive consultations for plan of care. In 69% of the intensive plan-of-care consultations, the palliative care team also provided pain and/or symptom management services (data not shown).
      Table 1Characteristics of Hospitalized Adult Cancer Patients Receiving Palliative Consultations by Ethnicity
      Patient CharacteristicTotal Sample (n=1362)White (n=396) (29.1%)Asian (n=611) (44.9%)Pacific Islander
      Pacific Islanders: Hawaiian n=197 (14.6%), Pacific Islanders n=46 (3.4%), and Micronesians n=50 (3.7%).
      (n=293) (21.5%)
      Other Ethnicity
      Other ethnicity includes the following: African American n=17 (1.3%), Hispanic n=25 (1.9%), and “other” n=11 (0.8%) (Native American, “cosmopolitan,” unknown, Middle Eastern, and declined). Ethnicity was missing for nine patients (0.7%).
      (n=62) (4.6%)
      P-value
      Less than 65 years831 (61.2%)231 (58.3%)320 (52.6%)238 (81.2%)42 (67.7%)<0.0001
      65–74 years290 (21.3%)102 (25.8%)131 (21.5%)43 (14.7%)14 (22.6%)
      Older than 74 years240 (17.6%)63 (15.9%)159 (26.1%)12 (4.1%)6 (9.7%)
      Male702 (51.5%)208 (52.5%)319 (52.2%)140 (47.8%)35 (56.5%)0.5
      Female660 (48.5%)188 (47.5%)292 (47.8%)153 (52.2%)27 (43.6%)
      Plan-of-care consult
      Plan-of-care consultations provided more intensive communication and care coordination interventions; managed plan of care alone (n=166 [12.2%]), plan of care+pain (n=201 [14.8%]), plan of care+symptoms (n=55 [4.0%]), and plan of care, pain, and symptoms (n=110 [8.1%]). Patients without plan-of-care consultations had less intensive consultations for pain (n=651 [47.8%]), symptoms (n=22 [1.6%]), or pain+symptoms (n=135 [9.9%]).
      532 (39.7%)120 (31.0%)274 (45.3%)112 (39.0%)26 (42.6%)0.0001
      Not plan of care808 (60.3%)267 (69.0%)331 (54.7%)175 (61.0%)35 (57.4%)
      Mean Karnofsky score42.0±21.044.9±21.140.8±21.340.5±20.043.2±20.80.01
      Preconsult days LOS7.2±16.36.7±13.67.4±18.17.8±16.85.2±7.40.7
      a Pacific Islanders: Hawaiian n=197 (14.6%), Pacific Islanders n=46 (3.4%), and Micronesians n=50 (3.7%).
      b Other ethnicity includes the following: African American n=17 (1.3%), Hispanic n=25 (1.9%), and “other” n=11 (0.8%) (Native American, “cosmopolitan,” unknown, Middle Eastern, and declined). Ethnicity was missing for nine patients (0.7%).
      c Plan-of-care consultations provided more intensive communication and care coordination interventions; managed plan of care alone (n=166 [12.2%]), plan of care+pain (n=201 [14.8%]), plan of care+symptoms (n=55 [4.0%]), and plan of care, pain, and symptoms (n=110 [8.1%]). Patients without plan-of-care consultations had less intensive consultations for pain (n=651 [47.8%]), symptoms (n=22 [1.6%]), or pain+symptoms (n=135 [9.9%]).
      The Asian and Pacific Islander ethnicity groups had lower mean Karnofsky scores than the white and “other” races or ethnicities. Significantly more of the nonwhite patients received plan-of-care consultations compared with the white patients (39.0%–45.3% vs. 31.0%). More patients in the Pacific Islander group were less than 65 years old whereas more Asian patients were older than 74 years. There were no significant differences in gender or preconsult hospital LOS by ethnicity.
      Outcomes of palliative care consultation are shown in Table 2. Palliative care consultations resulted in 454 patients (33.3%) referred to hospice, and code status of 234 (17.2%) patients changed to DNAR. More of the nonwhite cancer patients had hospice referral (34.1%–45.2% vs. 27.0%, P=0.005). More of the older patients were referred to hospice than the younger patients (40.0% of those older than 74 years compared with 29.1% of those under 65, P=0.0002, data not shown). More patients with plan-of-care consultations had hospice referral (51.3% vs. 21.5% without plan-of-care consultation, P<0.0001, data not shown). The mean Karnofsky score of patients referred to hospice was significantly lower than those not referred to hospice (35.2 vs. 45.5, P<0.0001, data not shown). However, the preconsult LOS was not significantly different by hospice referral status (7.3 days vs. 7.1 days for patients not referred to hospice, P=0.9, data not shown).
      Table 2Outcomes of Palliative Care Consultations Among Adult Hospitalized Cancer Patients by Ethnicity
      CharacteristicTotal Sample (n=1362)White (n=396) (29.1%)Asian (n=611) (44.9%)Pacific Islander (n=293) (21.5%)Other Ethnicity (n=62) (4.6%)P-value
      Hospice referral
      Referred to hospice group did not include patients who died in the hospital before start of hospice care.
      454 (33.3%)107 (27.0%)219 (35.8%)100 (34.1%)28 (45.2%)0.005
      Code status change234 (17.2%)55 (13.9%)114 (18.7%)57 (19.5%)8 (12.9%)0.1
      Discharge disposition
      Home569 (43.8%)193 (50.3%)220 (37.7%)131 (47.3%)25 (44.6%)0.02
      Institution
      Nursing home, rehabilitation hospital, other hospital, or care home.
      65 (5.00%)20 (5.2%)32 (5.5%)10 (3.6%)3 (5.4%)
      Hospice
      Hospice group includes patients discharged to home hospice, hospice at nursing homes, free-standing inpatient hospice facilities, and hospice care in the hospice.
      309 (23.8%)76 (19.8%)155 (26.6%)61 (22.0%)17 (30.4%)
      Died357 (27.5%)95 (24.7%)176 (30.2%)75 (27.1%)11 (19.6%)
      a Referred to hospice group did not include patients who died in the hospital before start of hospice care.
      b Nursing home, rehabilitation hospital, other hospital, or care home.
      c Hospice group includes patients discharged to home hospice, hospice at nursing homes, free-standing inpatient hospice facilities, and hospice care in the hospice.
      There were no significant differences in code status changes by ethnicity. More patients with plan-of-care consultation had code status changed to DNAR (29.1% vs. 9.4% with no plan-of-care consultation, P<0.0001, data not shown). Of the patients with code status changed to DNAR, 51.9% died in the hospital, 36.1% were discharged to hospice care in any setting, 2.6% were discharged to a nursing home facility without hospice, and 9.4% were discharged home without hospice (P<0.0001, data not shown). The mean Karnofsky score of patients with code status changed to DNAR was significantly lower than those without code status change (31.6 vs. 44.2, P<0.0001, data not shown). The mean preconsult LOS for patients with code status changed to DNAR was significantly longer than those without code status change (10.3 days vs. 6.5 days, P=0.001, data not shown).
      Table 3 shows the multiple logistic regression analyses of factors associated with the outcome of hospice referral. Controlling for age, gender, Karnofsky score, and preconsult hospital LOS, Asian, Pacific Islander, and “other” ethnicities remained significantly associated with an increased likelihood of hospice referral compared with white ethnicity. After controlling for plan-of-care consultations, the likelihood of hospice referral was significant only for “other ethnicities,” with P=0.03 and was nonsignificant for Asian and Pacific Islander ethnicities.
      Table 3Factors Associated With Outcome of Referral to Hospice: Multiple Logistic Regression Models
      FactorModel 1: Plan-of-Care Consult Type Not IncludedModel 2: Plan-of-Care Consult Type Included
      OR (95% CI)P-valueOR (95% CI)P-value
      Oldest age group1.23 (0.89–1.69)0.21.16 (0.83–1.61)0.4
      Male gender1.05 (0.82–1.35)0.71.06 (0.82–1.36)0.7
      White ethnicity1.00 (reference)1.00 (reference)
      Asian1.46 (1.08–1.98)0.021.35 (0.98–1.85)0.06
      Islander1.50 (1.05–2.16)0.031.46 (1.01–2.11)0.05
      Other ethnicity2.34 (1.26–4.35)0.0072.06 (1.08–3.90)0.03
      Karnofsky score
      Karnofsky score and LOS are continuous variables, that is, for every 10-point increase in Karnofsky, there was a 3% decrease in likelihood of hospice referral for Model 1.
      0.97 (0.96–0.98)<0.00010.98 (0.97–0.99)<0.0001
      Preconsult LOS (days)0.99 (0.99–1.00)0.30.99 (0.98–1.00)0.2
      Plan-of-care indication3.08 (2.33–4.07)<0.0001
      a Karnofsky score and LOS are continuous variables, that is, for every 10-point increase in Karnofsky, there was a 3% decrease in likelihood of hospice referral for Model 1.
      Using multiple logistic regression models controlling for the above covariates, ethnicity was not associated with the likelihood of a change in code status to DNAR (Table 4). Plan-of-care consultations were strongly associated with increased likelihood of code status change.
      Table 4Factors Associated With Outcome of Change in Code Status (to DNAR): Multiple Logistic Regression Models
      FactorModel 1: Plan-of-Care Consult Type Not IncludedModel 2: Plan-of-Care Consult Type Included
      OR (95% CI)P-valueOR (95% CI)P-value
      Oldest age group0.69 (0.45–1.05)0.080.64 (0.42–0.99)0.04
      Male gender1.21 (0.89–1.65)0.21.23 (0.90–1.68)0.2
      White ethnicity1.00 (reference)1.00 (reference)
      Asian1.39 (0.95–2.05)0.091.27 (0.85–1.88)0.2
      Islander1.35 (0.87–2.11)0.21.29 (0.82–2.04)0.3
      Other ethnicity1.11 (0.48–2.55)0.80.93 (0.40–2.17)0.9
      Karnofsky score
      Karnofsky score and LOS are continuous variables, that is, for every 10-point increase in Karnofsky, there was a 4% decrease in likelihood of change in code status for Model 1.
      0.96 (0.95–0.97)<0.00010.97 (0.96–0.98)<0.0001
      Preconsult LOS (days)1.01 (0.99–1.02)0.21.01 (0.99–1.01)0.3
      Plan-of-care indication2.96 (2.08–4.22)<0.0001
      a Karnofsky score and LOS are continuous variables, that is, for every 10-point increase in Karnofsky, there was a 4% decrease in likelihood of change in code status for Model 1.

      Discussion

      This study examined hospice referral and change in code status outcomes after inpatient palliative care consultations in a cancer patient population that had more than 70% nonwhite race or ethnicity. Nonwhite cancer patients receiving palliative care consultations were more likely to be referred to hospice than white patients, in contrast to previous studies describing reduced hospice use among nonwhites.
      • Johnson K.S.
      • Kuchibhatla M.
      • Tulsky J.A.
      What explains racial differences in the use of advance directives and attitudes toward hospice care?.
      • Cohen L.L.
      Racial/ethnic disparities in hospice care: a systematic review.
      • Ngo-Metzger Q.
      • Phillips R.S.
      • McCarthy E.P.
      Ethnic disparities in hospice use among Asian-American and Pacific Islander patients dying with cancer.
      However, this relationship between race or ethnicity and hospice referral was markedly reduced after consultation intensity (plan of care vs. without plan of care) was included in the multiple logistic regression model. Thus, consultation intensity was the strongest predictor of hospice referral, not ethnicity.
      Previous work has demonstrated that inpatient palliative care consultations result in increased likelihood of discharge to hospice,
      • Brody A.A.
      • Ciemins E.
      • Newman J.
      • Harrington C.
      The effects of an inpatient palliative care team on discharge disposition.
      • Reville B.
      • Miller M.N.
      • Toner R.W.
      • Reifsnyder J.
      End-of-life care for hospitalized patients with lung cancer: utilization of a palliative care service.
      earlier hospice referrals,
      • Brody A.A.
      • Ciemins E.
      • Newman J.
      • Harrington C.
      The effects of an inpatient palliative care team on discharge disposition.
      and longer time spent receiving hospice care.
      • Brody A.A.
      • Ciemins E.
      • Newman J.
      • Harrington C.
      The effects of an inpatient palliative care team on discharge disposition.
      • Gade G.
      • Venohr I.
      • Conner D.
      • et al.
      Impact of an inpatient palliative care team: a randomized control trial.
      Ethnicity was not significantly associated with hospital death or discharge home among patients in an acute palliative care unit.
      • Hui D.
      • Elsayem A.
      • Palla S.
      • et al.
      Discharge outcomes and survival of patients with advanced cancer admitted to an acute palliative care unit at a comprehensive cancer center.
      There have not, to the authors' knowledge, been previous studies examining the relationship between patient ethnicity and hospice referrals among patients receiving inpatient palliative care consultations.
      The races and ethnicities in this study correspond to Hawaii's general population and include groups underrepresented in research. More Asian patients were older whereas more Hawaiian, Pacific Islander, and Micronesian patients were younger. Hawaiian, Micronesian, and Pacific Islander patients have more cancer at younger ages than white patients and may present at more advanced stages of disease, with higher mortality rates.
      • Braun K.L.
      • Fong M.
      • Gotay C.
      • Pagano I.S.
      • Chong C.
      Ethnicity and breast cancer in Hawaii: increased survival but continued disparity.
      • Braun K.L.
      • Fong M.
      • Gotay C.C.
      • Chong C.D.
      Ethnic differences in breast cancer in Hawai'i: age, stage, hormone receptor status, and survival.
      • Clegg L.X.
      • Li F.P.
      • Hankey B.F.
      • Chu K.
      • Edwards B.K.
      Cancer survival among US whites and minorities: a SEER (Surveillance, Epidemiology, and End Results) Program population-based study.
      • Hashibe M.
      • Gao T.
      • Li G.
      • Dalbagni G.
      • Zhang Z.F.
      Comparison of bladder cancer survival among Japanese, Chinese, Filipino, Hawaiian and Caucasian populations in the United States.
      • Miller B.A.
      • Chu K.C.
      • Hankey B.F.
      • Ries L.A.
      Cancer incidence and mortality patterns among specific Asian and Pacific Islander populations in the U.S.
      • Palafox N.A.
      • Tsark J.
      Cancer in the US Associated Pacific Islands (UASPI): history and participatory development.
      The advanced stages of disease may partly explain why more Hawaiian, Micronesian, and Pacific Islander patients had hospice referrals than white patients, although it is unclear why the hospice referrals also were increased for the Asian and “other” ethnicity patients compared with the white patients.
      Race and ethnicity were not significantly associated with code status change whereas consultation intensity was strongly associated with increased likelihood of code status change. A recent study of intensive care unit (ICU) patients with cancer found lower rates of DNAR code status among nonwhite patients in the medical ICU than white patients,
      • Cardenas-Turanzas M.
      • Gaeta S.
      • Ashoori A.
      • Price K.J.
      • Nates J.
      Demographic and clinical determinants of having do not resuscitate orders in the intensive care unit of a comprehensive cancer center.
      but palliative care consultations were not examined. Another study found greater use of invasive interventions at the end of life among blacks and Hispanics
      • Hanchate A.
      • Kronman A.C.
      • Young-Xu Y.
      • Ash A.S.
      • Emanuel E.
      Racial and ethnic differences in end-of-life costs: why do minorities cost more than whites?.
      but also did not examine palliative care consultations.
      There have been studies examining the focus of palliative care consultations in terms of symptom management, goals of care, family support, discharge planning, and other areas.
      • Manfredi P.L.
      • Morrison R.S.
      • Morris J.
      • et al.
      Palliative care consultations: how do they impact the care of hospitalized patients?.
      • Fromme E.K.
      • Bascom P.B.
      • Smith M.D.
      • et al.
      Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
      In one study, pain or symptom management consultations had lower in-hospital mortality than consultations for clarifying prognosis, goals of care, family support, or terminal care.
      • Fromme E.K.
      • Bascom P.B.
      • Smith M.D.
      • et al.
      Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
      Thus, plan-of-care communication and coordination has been used previously to classify and compare palliative care consultations and may indicate intensity of palliative care consultations. In this study, plan-of-care consultations resulted in more hospice referrals and code status changes than consultations without plan of care, which has not been previously demonstrated to the authors' knowledge.
      There are several limitations in this study. As an observational study, patients were not randomized to plan-of-care consultations vs. consultations without plan of care, focused on pain and symptom management. It is possible that other factors not included in these models could explain the relationships between intensity of consultation and outcomes of consultation. However, the plan-of-care indication provides a natural separation between more intense consultations and the more straightforward consultations for pain and/or symptom management alone and provides two models of palliative care for this study.
      The palliative care service in this study is truly a dual pain management service and palliative care service. The hospital nurse-run pain management service was modified into a combined pain and palliative care service after a palliative care-trained physician, a social worker, and an additional nurse were added. There is not a separate pain management service or provider for acute or chronic pain issues. Thus, this dual service may more easily distinguish between consultations for pain or symptom management alone vs. those for plan of care as one of the indications, even among cancer patients.
      Stage or severity of cancer was not collected or able to be examined, although data on Karnofsky score and preconsult LOS in days were included to provide insight into the patient's location in the cancer disease trajectory. Data on insurance status were not available in this de-identified clinical database, and because race or ethnicity is often a surrogate for socioeconomic class and other factors, this is an important area of future research. However, Hawaii has a low rate of uninsured relative to the national average,

      The Henry Kaiser Family Foundation. State health facts. Available from http://www.statehealthfacts.org. Accessed January 4, 2011.

      and barriers to hospice utilization in Hawaii have been postulated to include lack of familiarity with hospice and cultural or organizational barriers.
      • Braun K.L.
      • Onaka A.T.
      • Horiuchi B.Y.
      Advance directive completion rates and end-of-life preferences in Hawaii.
      DNAR was used as a proxy outcome for goals-of-care decisions not resulting in hospice referrals and provides information only on changes in resuscitation status and not on other important outcomes, such as ventilator withdrawal, rehospitalization, tube-feeding wishes, family distress, or caregiver satisfaction.
      Racial and ethnic variations in palliative care consultation intensity and outcomes may be related to differences in patient age, diagnoses, and presentations of illness among the different ethnicities, although this study controlled for age, performance status, and preconsultation hospital LOS. Data on cancer patients not receiving palliative care consultations were not available, and comparison of outcomes by ethnicity and referral to palliative care are needed. Physician biases in referral patterns, communication issues, cultural expectations, or other factors may underlie these variations and are important to examine to improve quality of end-of-life care for all patients.

      Conclusions

      Patients who received palliative care consultations for plan of care were more likely to have subsequent hospice referrals and code status changes. These findings suggest that providing palliative care consultations for plan of care may enhance end-of-life care in diverse populations. Future work is needed to examine referral patterns in diverse populations and to understand the potential for palliative care consultations to reduce ethnic disparities in end-of-life care.

      Disclosures and Acknowledgments

      This research was supported by The John A. Hartford Foundation Center of Excellence in Geriatrics, University of Hawaii and Research Centers in Minority Institutions Award 1 R25 RR019321 , from the National Center for Research Resources, National Institutes of Health . The authors declare no conflicts of interest.
      Portions of this study were presented as a paper presentation at the American Academy of Hospice and Palliative Medicine Annual Conference in Boston in March 2010.
      The authors acknowledge James Davis, PhD, for his assistance with statistical analyses; Kamal Masaki, MD, for her assistance with study design; and Aida Wen, MD, for her feedback on the manuscript.

      References

        • Johnson K.S.
        • Kuchibhatla M.
        • Tulsky J.A.
        What explains racial differences in the use of advance directives and attitudes toward hospice care?.
        J Am Geriatr Soc. 2008; 56: 1953-1958
        • Hanchate A.
        • Kronman A.C.
        • Young-Xu Y.
        • Ash A.S.
        • Emanuel E.
        Racial and ethnic differences in end-of-life costs: why do minorities cost more than whites?.
        Arch Intern Med. 2009; 169: 493-501
        • Cohen L.L.
        Racial/ethnic disparities in hospice care: a systematic review.
        J Palliat Med. 2008; 11: 763-768
        • Connor S.
        • Elwert F.
        • Spence C.
        • Christakis N.
        Racial disparity in hospice use in the United States in 2002.
        Palliat Med. 2008; 22: 205-213
        • Connor S.R.
        • Elwert F.
        • Spence C.
        • Christakis N.A.
        Geographic variation in hospice use in the United States in 2002.
        J Pain Symptom Manage. 2007; 34: 277-285
        • Barnato A.E.
        • Anthony D.L.
        • Skinner J.
        • Gallagher P.M.
        • Fisher E.S.
        Racial and ethnic differences in preferences for end-of-life treatment.
        J Gen Intern Med. 2009; 24: 695-701
        • Volandes A.E.
        • Paasche-Orlow M.
        • Gillick M.R.
        • et al.
        Health literacy not race predicts end-of-life care preferences.
        J Palliat Med. 2008; 11: 754-762
        • Ngo-Metzger Q.
        • Phillips R.S.
        • McCarthy E.P.
        Ethnic disparities in hospice use among Asian-American and Pacific Islander patients dying with cancer.
        J Am Geriatr Soc. 2008; 56: 139-144
        • Manfredi P.L.
        • Morrison R.S.
        • Morris J.
        • et al.
        Palliative care consultations: how do they impact the care of hospitalized patients?.
        J Pain Symptom Manage. 2000; 20: 166-173
        • Jenkins C.A.
        • Schulz M.
        • Hanson J.
        • Bruera E.
        Demographic, symptom, and medication profiles of cancer patients seen by a palliative care consult team in a tertiary referral hospital.
        J Pain Symptom Manage. 2000; 19: 174-184
        • O'Mahony S.
        • Blank A.E.
        • Zallman L.
        • Selwyn P.A.
        The benefits of a hospital-based inpatient palliative care consultation service: preliminary outcome data.
        J Palliat Med. 2005; 8: 1033-1039
        • Dhillon N.
        • Kopetz S.
        • Pei B.L.
        • et al.
        Clinical findings of a palliative care consultation team at a comprehensive cancer center.
        J Palliat Med. 2008; 11: 191-197
        • Fernandes R.
        • Braun K.L.
        • Ozawa J.
        • et al.
        Home-based palliative care services for underserved populations.
        J Palliat Med. 2010; 13: 413-419
        • Weng L.C.
        • Huang H.L.
        • Wilkie D.J.
        • et al.
        Predicting survival with the Palliative Performance Scale in a minority-serving hospice and palliative care program.
        J Pain Symptom Manage. 2009; 37: 642-648
        • Hui D.
        • Elsayem A.
        • Palla S.
        • et al.
        Discharge outcomes and survival of patients with advanced cancer admitted to an acute palliative care unit at a comprehensive cancer center.
        J Palliat Med. 2010; 13: 49-57
        • Brody A.A.
        • Ciemins E.
        • Newman J.
        • Harrington C.
        The effects of an inpatient palliative care team on discharge disposition.
        J Palliat Med. 2010; 13: 541-548
        • Edrington J.
        • Sun A.
        • Wong C.
        • et al.
        Barriers to pain management in a community sample of Chinese American patients with cancer.
        J Pain Symptom Manage. 2009; 37: 665-675
        • Im E.O.
        • Chee W.
        • Guevara E.
        • et al.
        Gender and ethnic differences in cancer pain experience: a multiethnic survey in the United States.
        Nurse Res. 2007; 56: 296-306
        • Higginson I.J.
        • Finlay I.
        • Goodwin D.M.
        • et al.
        Do hospital-based palliative teams improve care for patients or families at the end of life?.
        J Pain Symptom Manage. 2002; 23: 96-106
        • Fromme E.K.
        • Bascom P.B.
        • Smith M.D.
        • et al.
        Survival, mortality, and location of death for patients seen by a hospital-based palliative care team.
        J Palliat Med. 2006; 9: 903-911
        • Braiteh F.
        • El Osta B.
        • Palmer J.L.
        • Reddy S.K.
        • Bruera E.
        Characteristics, findings, and outcomes of palliative care inpatient consultations at a comprehensive cancer center.
        J Palliat Med. 2007; 10: 948-955
        • Johnson C.E.
        • Girgis A.
        • Paul C.L.
        • Currow D.C.
        Cancer specialists' palliative care referral practices and perceptions: results of a national survey.
        Palliat Med. 2008; 22: 51-57
        • Delgado-Guay M.O.
        • Parsons H.A.
        • Li Z.
        • Palmer L.J.
        • Bruera E.
        Symptom distress, interventions, and outcomes of intensive care unit cancer patients referred to a palliative care consult team.
        Cancer. 2009; 115: 437-445
        • Gade G.
        • Venohr I.
        • Conner D.
        • et al.
        Impact of an inpatient palliative care team: a randomized control trial.
        J Palliat Med. 2008; 11: 180-190
        • Weissman D.E.
        • Meier D.E.
        Center to Advance Palliative Care inpatient unit operational metrics: consensus recommendations.
        J Palliat Med. 2009; 12: 21-25
        • Yates J.W.
        • Chalmer B.
        • McKegney F.P.
        Evaluation of patients with advanced cancer using the Karnofsky performance status.
        Cancer. 1980; 45: 2220-2224
        • Reville B.
        • Miller M.N.
        • Toner R.W.
        • Reifsnyder J.
        End-of-life care for hospitalized patients with lung cancer: utilization of a palliative care service.
        J Palliat Med. 2010; 13: 1261-1266
        • Braun K.L.
        • Fong M.
        • Gotay C.
        • Pagano I.S.
        • Chong C.
        Ethnicity and breast cancer in Hawaii: increased survival but continued disparity.
        Ethn Dis. 2005; 15: 453-460
        • Braun K.L.
        • Fong M.
        • Gotay C.C.
        • Chong C.D.
        Ethnic differences in breast cancer in Hawai'i: age, stage, hormone receptor status, and survival.
        Pac Health Dialog. 2004; 11: 146-153
        • Clegg L.X.
        • Li F.P.
        • Hankey B.F.
        • Chu K.
        • Edwards B.K.
        Cancer survival among US whites and minorities: a SEER (Surveillance, Epidemiology, and End Results) Program population-based study.
        Arch Intern Med. 2002; 162: 1985-1993
        • Hashibe M.
        • Gao T.
        • Li G.
        • Dalbagni G.
        • Zhang Z.F.
        Comparison of bladder cancer survival among Japanese, Chinese, Filipino, Hawaiian and Caucasian populations in the United States.
        Asian Pac J Cancer Prev. 2003; 4: 267-273
        • Miller B.A.
        • Chu K.C.
        • Hankey B.F.
        • Ries L.A.
        Cancer incidence and mortality patterns among specific Asian and Pacific Islander populations in the U.S.
        Cancer Causes Control. 2008; 19: 227-256
        • Palafox N.A.
        • Tsark J.
        Cancer in the US Associated Pacific Islands (UASPI): history and participatory development.
        Pac Health Dialog. 2004; 11: 8-13
        • Cardenas-Turanzas M.
        • Gaeta S.
        • Ashoori A.
        • Price K.J.
        • Nates J.
        Demographic and clinical determinants of having do not resuscitate orders in the intensive care unit of a comprehensive cancer center.
        J Palliat Med. 2011; 14: 45-50
      1. The Henry Kaiser Family Foundation. State health facts. Available from http://www.statehealthfacts.org. Accessed January 4, 2011.

        • Braun K.L.
        • Onaka A.T.
        • Horiuchi B.Y.
        Advance directive completion rates and end-of-life preferences in Hawaii.
        J Am Geriatr Soc. 2001; 49: 1708-1713