Secondary Logo

Factors Influencing Clinical and Setting Pathways After Discharge From an Acute Palliative/Supportive Care Unit

Mercadante, Sebastiano, MD*; Adile, Claudio, MD*; Ferrera, Patrizia, MD*; Casuccio, Alessandra, BS

American Journal of Clinical Oncology: March 2019 - Volume 42 - Issue 3 - p 265–269
doi: 10.1097/COC.0000000000000510
Original Articles: Supportive Care

Aim: The aim of this study was to assess the factors which influence the care pathway after discharge from an acute palliative supportive care unit (APSCU).

Methods: Patients’ demographics, indications for admission, kind of admission, the presence of a caregiver, awareness of prognosis, data on anticancer treatments in the last 30 days, ongoing treatment (on/off or uncertain), the previous care setting, analgesic consumption, and duration of admission were recorded. The Edmonton Symptom Assessment Scale (ESAS) at admission and at time of discharge (or the day before death), CAGE (cut down, annoy, guilt, eye-opener), and the Memorial Delirium Assessment Scale (MDAS), were used. At time of discharge, the subsequent referral to other care settings (death, home, home care, hospice, oncology), and the pathway of oncologic treatment were reconsidered (on/off, uncertain).

Results: A total of 314 consecutive cancer patients admitted to the APSCU were surveyed. Factors independently associated with on-therapy were the lack of a caregiver, home discharge, and short hospital admission, in comparison with off-treatment, and less admission for other symptoms, shorter hospital admission, discharge at home, and better well-being, when compared with “uncertain.” Similarly, many factors were associated with discharge setting, but the only factor independently associated with discharge home was being “on-therapy.”

Conclusions: The finding of this study is consistent with an appropriate selection of patients after being discharged by an APSCU, that works as a bridge between active treatments and supportive/palliative care, according the concept of early and simultaneous care.

*Main regional center for pain relief and supportive/palliative care, La Maddalena Cancer center

Department of Sciences for Health Promotion and Mother Child Care, University of Palermo, Palermo, Italy

The authors declare no conflicts of interest.

Reprints: Sebastiano Mercadante, MD, La Maddalena Cancer Center, Anesthesia and Intensive Care Unit & Pain Relief and Palliative Care Unit, Via san Lorenzo 312, Palermo 90145, Italy. E-mails:;

Advanced cancer patients develop a relevant symptom burden during the course of disease, requiring experience in palliative care. In most countries palliative care is commonly provided at home or in hospice.1 Regrettably, palliative care is started late, as most of these patients are followed by palliative care teams in the last weeks of life. Moreover, they are often hospitalized and are receiving chemotherapy in the last month of life.2 It has been reported that the interval between palliative care referral and death was <2 months.3

In the last years it has been suggested that palliative care should be started early to support the patient during their therapeutic pathway, providing not only an adequate symptom control but also elements in the decision-making process, including discontinuation of active treatment and the appropriate choice of setting for a more specialized treatment, independently of the expected survival. Acute palliative/supportive care units (APSCUs) are inpatient units in oncological departments specialized in the treatment of complex physical and psychological symptoms.1 Several studies have reported the positive outcomes of ASPCUs, including not only the capacity to support the crises during the active treatment, in a concept of simultaneous care, or to manage pain and symptoms, but also to redirect the therapeutic trajectory, facilitating the transition to palliative care.4–8 When a patient is discharged from an ASPCU, he can continue some oncologic treatment or not, or can be referred to specialistic palliative care, particularly when he is considered to be unfit for a further oncologic treatment. However, factors influencing dispositions at discharge from a APSCU are lacking. The aim of this study was to assess the factors which influence the care pathway, including treatment and setting, after discharge from an APSCU with peculiar characteristics established about 20 years ago.

Back to Top | Article Outline


This is a secondary analysis of a study assessing the influence of an APSCU on the therapeutic trajectory of oncological patients in a comprehensive cancer.4 Patients’ demographics, including age, sex, primary diagnosis, marital status, and educational level were collected. Karnofsky status and principal reason for admission, including pain or opioid-related problems, toxicity from chemotherapy, symptom control, reevaluation, or end of life issues, were recorded. The kind of admission, unplanned or not, or readmission was also recorded. Data regarding the caregiver, considered as a person who spent at least 4 hours per day with the patient, were recorded, as well as awareness of prognosis by semistructured interviews (complete, partial, absent).

Data on chemotherapy in the last 30 days, or other anticancer treatments (radiotherapy, surgery, target therapy, hormonal therapy, and so on), whether patients were on/off treatment or uncertain (that is the gray area of patients in which a definitive decision was not taken and postponed), the previous care setting, including home, hospital unit, day-hospital, or other hospitals, and who referred the patient to the unit, including home palliative care physicians, oncologists, other units, other hospitals, or general practitioner were also recorded. Analgesic drugs and their doses at admission and discharge were recorded. Hospital staying was also recorded. The following tools have been used: (a) The Edmonton Symptom Assessment Scale (ESAS) at admission and at time of discharge (or the day before death) (TD). ESAS is a self-reported tool assessing the intensity of most common psychological and physical symptoms on a 0 to 10 numerical scale.9 It is a valid and reliable tool for assessing the overall symptom burden, and has been found to be sensible to changes produced by a treatment10; (b) a screening test for history of alcohol dependence (CAGE: cut down, annoy, guilt, eye-opener). CAGE score has a sensitivity >90% and specificity >95% to detect alcoholism.11 Patients with alcohol problems express more symptom distress and are more likely to have a history of smoking or illegal use of drugs12; (c) the Memorial delirium Assessment Scale (MDAS), to assess patients’ cognitive status. MDAS is a validated tool to quantify the intensity of delirium,13 and has a relevant influence on symptom expression.14

At time of discharge, the subsequent referral to other care settings (death, home, home care, hospice, oncology), and the pathway of proposed oncologic treatment were collected (on/off, uncertain).

Back to Top | Article Outline

Statistical Analysis

Statistical analysis of quantitative and qualitative data, including descriptive statistics, was performed for all items. Continuous data are expressed as mean±SD, unless otherwise specified. Differences between groups were assessed by χ2 test or Fisher exact test, as needed for categorical variables, and by the independent Student t test for continuous parameters. The univariate analysis of variance was performed for parametric variables, and post hoc analysis with the Tukey test was used to determine whether there were pairwise intragroup differences. The variables significantly related to patient groups were analyzed in multinomial logistic regression model to examine the correlation between patient characteristics (independent variables), and patient groups (dependent variable). Any missing data regarding clinical variables (<2%) were excluded from statistical analysis. Data were analyzed by IBM SPSS Software 22 version (IBM Corp., Armonk, NY). All P-values were 2-sided and P<0.05 was considered statistically significant.

Back to Top | Article Outline


Three-hundred-fourteen consecutive cancer patients admitted to the APSCU were surveyed. For the different parameters taken into consideration, data were missed for <0% of patients (range, 0 to 20 patients). The characteristics of patients, stage of disease, kind of admission (urgent, readmission, etc.), information regarding people living with the patients, the presence of a caregiver, educational level, indications for admission, active treatments in the last 30 days, and patients’ and caregivers’ awareness of disease, are presented in Table 1. Most admissions were planned, although a consistent number of patients were admitted on emergency (generally from the oncological day-hospital) or transferred from other hospital units. The mean hospital stay was 6.9 days (SD, 6.3). Twelve patients (3.8%) died in the unit. Data regarding the clinical pathways suggested at discharge from the unit, including how the patient was considered for treatment purposes (on/off, or uncertain), and the referral setting, are presented in Table 2.





Back to Top | Article Outline

Discharge Care Pathway (On-Off-Uncertain)

In the univariate analysis low Karnofsky level (P<0.0005), primary tumor (breast, P=0.028), the presence of the caregiver (P=0.004), less awareness of disease (P=0.001), indications for admission (toxicity due to anticancer therapies, P=0.019, other symptoms, P=0.006, and end of life care P=0.002), anticancer treatment in the last 30 days (0.001), referral setting (home care, P=0.004), MDAS at T0 and MDAS at TD (0.001 and <0.0005, respectively), weakness at TD (P<0.0005), depression at TD (P<0.0005), drowsiness at TD (P=0.004), appetite at T0 and TD (P=0.030 and 0.009, respectively), poor well-being at TD (P=0.003), global ESAS at TD (P=0.016), hospital staying (P=0.012), and discharge settings (home care, P<0.0005), were associated with the decision to definitely stop oncological treatments at discharge (off therapy). Age, sex, education, stage of disease, CAGE-positivity, smoking habits, the kind of admission (planned or not, readmission), amount of opioid rugs (OME), did not influence the discharge therapeutic pathway.

The multivariate regression analysis regarding the treatment pathway (on-off-uncertain) is showed in Table 3. Being “on-therapy” was independently associated with a lack of caregiver, home discharge, and shorter hospital admission, in comparison with off-treatment. In comparison with “uncertain” group, “on-treatment” was independently associated less admission for other symptoms, shorter hospital admission, discharge at home, and better well-being.



Back to Top | Article Outline

Discharge Setting

In the univariate analysis younger age (P=0.009), higher Karnofsky level (P<0.0005), stage of disease (no metastatic disease, P=0.038), patient awareness (P<0.0005), admission for uncontrolled pain (P=0.011), referral setting (home, P<0.0005), lower MDAS at T0 and TD (P=0.007 and 0.001, respectively), on-treatment (P<0.0005), no transfer from other units, P=0.012), higher pain intensity at T0 (P<0.0005), lower values of weakness (P<0.0005), nausea (P=0.004), depression (P<0.0005), anxiety (P=0.002), drowsiness (P<0.0005), dyspnea (P=0.030), insomnia (P=0.004), appetite (P=0.001), poor well-being (P<0.0005), and global ESAS at TD (P<0.0005) were associated with home discharge. The multivariate regression analysis is showed in Table 4 (discharge at home vs. home care, hospice or other). The only parameter independently associated with discharge home was “on-therapy.”



Back to Top | Article Outline


APSCU serves a very heterogenous population with different characteristics, in terms of stage of disease ongoing treatments, and reasons for admission, covering a broad range of cancer patients, just from time of diagnosis until the end of life. Thus, many patients are still receiving and possibly will continue anticancer therapies, have a short hospital staying, and are unlikely to die in hospital. Differently from what it was reported in hospitals in United States where half of them provide palliative care services,15 in most European countries, particularly Italy, most hospitals do not offer such opportunities, as palliative care is available as home care and hospice care services only, mainly not linked with the oncological setting, and accepting patients with a short survival in the last weeks of life.16 Instead, an early palliative care intervention may have more beneficial effects on quality of life and symptom intensity among patients with advanced cancer than among those given usual/standard cancer care alone, and may increase use of approach-oriented coping, which was associated with higher quality of life and reduced depressive symptoms.17

This study provided a series of data influencing decision-making process at discharge from an APSCU, regarding either the oncological treatment and the care setting. In comparison to patients considered off-therapy, patients considered to benefit from continuing anticancer treatment (on) were more likely to do not have a caregiver, according to the definition adopted in this study, to be discharged home without specialized home care, and to have a shorter hospitalization. When compared with the gray area of patients defined “uncertain,” in which a definitive decision was not taken and postponed, patients “on-treatment” were more likely to have a shorter admission, to be discharged home, to have less admission for other symptoms, and a better poor well-being at discharge. Of interest, many other parameters taken into consideration in this study, including age, sex, education, stage of disease, CAGE-positivity, smoking habits, the kind of admission (planned or not, readmission), amount of opioid rugs (OME), did not influence the discharge disposition. The various variables examined did non influence the setting discharge, as the only variable associated with home discharge (without specialized home care service), was “on-therapy.” Conversely, off-therapy patients were more likely to be discharged to palliative home care or hospice care services. This finding is consistent with an appropriate selection of patients after being discharged by an APSCU, that works as a bridge between active treatment and supportive/palliative care, according the concept of early and simultaneous care. The therapeutic and settings pathways support the need of specialized beds in an oncological department, where it is possible reevaluate patients and provide the right direction, other than controlling symptoms.

Available literature regarding discharge disposition after admission to an APSCU is poor. In Italy almost all inpatients palliative care units are represented by hospices, where survival is short (about 18 d) and almost all patients die. From these data, it appears that hospice admission is only one way for end of life treatments. Thus, patients admitted to hospice receive specialized palliative care only for 2 to 3 weeks before death, implying an inacceptable timing for patients with several problems presumed to be present early during the course of disease.2

In a pioneer study, 59 and 41 patients discharged from an acute palliative care were transferred home and to hospice, respectively. Younger patients, a better functional status, being married, and with a better cognitive status, evaluated with Mini-Mental State Examination, were more likely to be transferred home. However, data regarding the trajectory of oncological treatments was unavailable.18 In a more recent study, patients admitted to an inpatient palliative care unit, 45% of them were discharged, with a mean hospital staying of 15 days, and 55% died. Karnofsky status and palliative performance scale were higher for discharged patients.19 No correlation was found between the pattern of sociodemographic variables. Moreover, the inpatient mortality was quite different from that reported in an unit like that presented in this study and previous surveys (3% to 7% of patients).4,5 In a study performed with a large sample of advanced cancer patients admitted to an acute palliative care unit, 958 (33%) patients died during admission, while 592 (20%), 855 (29%), 404 (14%) 56 (2%), and 33 (1%) patients were sent home, home hospice, inpatient hospice, long-term care facilities, and other institutions, respectively. Home discharge was associated with a longer survival in comparison with hospice transfers. Younger age and direct admission to the acute palliative care unit were independently associated with home discharge.15 Data regarding the care pathway were not available, and other possible determinants of discharge, including symptom severity and Karnofsky status, were unavailable. Considering the different inpatient death rate (about 1/3), it is likely that many of these patients were severely ill in comparison with patients reported in this study. In a study including inpatients and patients seen for palliative care consultation, almost all patients were discharged to palliative care services. Stable care trajectories were found for patients with a short survival, a median of 24 days.20 In patients with unplanned admission to an oncological unit, discharge location was influenced by a substantial physical and psychological symptom burden. However, data refers to a general oncologic ward.21

Globally, from data of literature it seems that most palliative care units admit very advanced cancer patients with short survival and a relatively high mortality rate during admission. Thus, existing data are hard to compare with data of the present study. Despite calls for integrated palliative care, this frequently occurs late. The application of targeted cancer-specific transition points, for example an APSCU, to trigger integration of palliative care as a standard part of quality oncological care and to provide appropriate discharge disposition, are of paramount importance.18

For these reasons, patients admitted to our APSCU may not representative of patients with cancer, and results cannot be generalizable. However, this model, that seems to deliver the earliest palliative care, should be widespread as it allows a precious weapon in the context of the concept simultaneous care. APSCU admission is not only useful for symptom control , toxicities, end of life care, but also to allow an appropriate patients’ assessment according to more strict palliative care criteria and a multidisciplinary evaluation of the individual therapeutic options. In other words, APSCU may enhance coordination of care. providing a junction to choose the best pathway in terms of oncologic treatment and care setting. Another limitation is represented by the lack of information about survival. However, many patients lived at long distance from this regional center, and there was no way to gather this information.

It would be relevant to compare different models, for example external consultation or mobile teams of palliative care in oncological wards. For instance, it has been shown that APSCU allows a better planning admission to hospital and a better care trajectory in comparison to a busy oncological ward in a similar population of advanced cancer patients.22,23 Thus, specialized palliative care beds may facilitate the management of advanced cancer patients in the context of a busy oncological ward.

Back to Top | Article Outline


1. Bruera E, Hui D. Integrating supportive and palliative care in the trajectory of cancer: establishing goals and models of care. J Clin Oncol. 2010;28:4013–4017.
2. Mercadante S, Vitrano V. Palliative care in Italy: problems areas emerging from the literature. Minerva Anestesiol. 2010;76:1060–1071.
3. Cheng WW, Willey J, Palmer JL, et al. Interval between palliative care referral and death among patients treated at a comprehensive cancer center. J Palliat Med. 2005;8:1025–1032.
4. Mercadante S, Adile C, Caruselli A, et al. The palliative-supportive care unit in a comprehensive cancer center as crossroad for patients’ oncological pathway. PLoS One. 2016;11:e0157300.
5. Mercadante S, Villari P, Ferrera P. A model of acute symptom control unit: Pain Relief and Palliative Care Unit of La Maddalena Cancer Center. Support Care Cancer. 2003;11:114–119.
6. Rigby A, Krzyanowska M, Le LW, et al. Impact of opening an acute palliative care unit on administrative outcomes for a general oncology ward. J Clin Oncol. 2008;113:3267–3274.
7. Smith T, Coyne P, Cassel B, et al. A high volume specialist palliative care unit and team may reduce in-hospital end-of life care costs. J Palliat Med. 2003;6:699–705.
8. Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet. 2014;383:1721–1730.
9. Bruera E, Kuehn N, Miller MJ, et al. The Edmonton Symptom Assessment Scale (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care. 1991;7:6–7.
10. Chang VT, Hwang SS, Feurman M. Validation of the Edmonton Symptom Assessment Scale. Cancer. 2000;88:2164–2171.
11. Dhalla S, Kopec JA. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clin Invest Med. 2007;1:33–41.
12. Parsons HA, Delgado-Guay MO, El Osta B, et al. Alcoholism screening in patients with advanced cancer: impact on symptom burden and opioid use. J Palliat Med. 2008;11:964–968.
13. Lawlor PG, Nekolaichuk C, Gagnon B, et al. Clinical utility, factor analysis, and further validation of the memorial delirium assessment scale in patients with advanced cancer: assessing delirium in advanced cancer. Cancer. 2000;88:2859–2867.
14. de la Cruz M, Ransing V, Yennu S, et al. The frequency, characteristics, and outcomes among cancer patients with delirium admitted to an acute palliative care unit. Oncologist. 2015;20:1425–1431.
15. 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;12:49–57.
16. Mercadante S, Valle A, Sabba S, et al. Pattern and characteristics of advanced cancer patients admitted to hospices in Italy. Support Care Cancer. 2013;21:935–939.
17. Haun MW, Estel S, Rücker G, et al. Early palliative care for adults with advanced cancer. Cochrane Database Syst Rev. 2017;6:CD011129. Doi:10.1002/14651858.CD011129.pub2.
18. Fainsinger RL, Demoissac D, Cole J, et al. Home versus hospice inpatient care: discharge characteristics of palliative care patients in an acute care hospital. J Palliat Care. 2000;16:29–34.
19. Masel EK, Huber P, Schur S, et al. Coming and going: predicting the discharge of cancer patients admitted to a palliative care unit: easier than thought? Support Care Cancer. 2015;23:2335–2339.
20. Kotzsch F, Stiel S, Heckel M, et al. Care trajectories and survival after discharge from specialized inpatients palliative care; results from an observational fllow-up study. Support Care Cancer. 2015;23:627–634.
21. Lage DE, Nipp RD, D’ Arpino SM, et al. Predictors of posthospital transitions of care in patients with advanced cancer. J Clin Oncol. 2018;36:76–82.
22. Mercadante S, Marchetti P, Adile C, et al. Characteristics and care pathways of advanced cancer patients in a palliative-supportive care unit and an oncological ward. Support Care Cancer. 2018;26:1961–1966.
23. Collins A, Sundararajan V, Burchell J, et al. Transition points for the routine integration of palliative care in patients with advanced cancer. J Pain Symptom Manage. 2018;56:185–194.

advanced cancer; palliative care; supportive care; discharge disposition

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.