Abstract
Objective: To examine the relationship between admission Karnofsky Performance Status (KPS) and discharge disposition.
Background: Little is known about the relationship between functional status before hospitalization and discharge disposition.
Methods: In a retrospective cohort study of patients seen by Mount Sinai Hospital Medicine Primary Palliative Care Program (HPPC), we used demographic and clinical data to compare discharge disposition by patients' functional status before admission into the hospital.
Results: Overall, 596 patients received HPPC consults (286 [48%] female, mean age 68.4 years, median admission KPS 40% [requires hospital level care]). Of the 33 patients with a KPS ≥60% (unable to work) 30 (91%) were discharged home, whereas those 262 patients with KPS ≤30% (severely disabled) 52 (20%) were discharged home, 40 (15%) enrolled in hospice, 130 (49.5%) discharged to a facility, and 32 (12%) died in hospital.
Conclusions: Worse functional status was associated with a hospice or facility discharge and better functional status was associated with discharge home.
Key Message: This retrospective cohort study examined the relationship between KPS before hospital admission and discharge disposition in hospitalized seriously ill patients admitted to the hospital medicine service who received a HPPC consultation. The results suggest that those with a higher admission KPS (more functional) are more likely to be discharged home, whereas those with a lower KPS (less functional) are more likely to be discharged to a facility or hospice. KPS before hospital admission could guide palliative care resource allocation and discharge needs.
Keywords: admission KPS, discharge disposition, palliative care
Introduction
Palliative care is centered around improving quality of life of patients and their families facing potentially life-limiting illness.1 Typically a specialty palliative care team is an interdisciplinary team (IDT) consisting of a physician, advance practice nurse (APN), social worker (SW), chaplain, and other members, who provide a collaborative approach to patient's needs. Access to palliative care has shown to lead to improved patient-centered outcomes, including symptom management, patient and family satisfaction, and care provided in the preferred care setting.2–5 Therefore, to provide quality care, we must develop ways to provide access to the necessary palliative care services.
Barriers to delivery of palliative care include lack of specialty-trained palliative care workforce of all disciplines on the IDT. Currently there is an estimated 18,000 palliative care physician shortage just to meet hospital needs.3 This shortage is going to worsen with physicians leaving the field and not enough trainees to fill the gap.6,7 Increased funding is required to attract new physicians to the field; however, it is hard to capture productivity and need when much of a clinician's time is spent communicating while addressing symptom management and assessing goals of care. In addition to increasing number of trainee positions, developing, testing, and examining innovative models of palliative care delivery to meet patient needs in the setting of limited workforce is imperative.
To increase patient access to palliative care, we developed the Hospital Medicine Primary Palliative Care Program (HPPC), a primary palliative care model embedded into the Hospital Medicine service at Mount Sinai Hospital.8 Initially, HPPC was led by a SW who consulted on patients earlier in their disease trajectory (e.g., at time of new cancer diagnosis or first heart failure exacerbation) for an assessment of patient's values, goals of treatment, understanding of illness, and completion of appropriate advance directives. These elicited goals and values are useful in future discussions during which the clinical team may help determine patients' goals for their medical care when the patient is more ill. After its first year, the program expanded to a team of a SW and an APN to allow for symptom management. Together, with the Hospital Medicine team, the SW and APN facilitated goals of care discussions around patient's prognosis, goals, values, and treatment options. For those patients whose palliative care needs extended beyond the scope of practice of the HPPC team (e.g., refractory symptoms, conflict between clinicians about treatment options, or conflict within the family regarding treatment plan), inpatient specialty palliative care consultation service assumed the care.
Given the dynamically changing needs of patients with serious illness, it is difficult to tailor resources for these patients. Functional status may heavily impact what needs and resources are appropriate for these patients. Data show that patients with functional impairment were likely to have increased symptom burden.9 The Karnofsky Performance Status (KPS), a validated scale to assess functional status in patients with serious illness across care settings, is often used to help guide clinical decision making, as seen in patients with cirrhosis.10 Studies show that hospitalized patients with serious illnesses who have a higher KPS were more likely to go home.11,12 Yet, little is known about the relationship between admission KPS and discharge disposition. Further exploration is needed to tailor resources and deploy innovative models of palliative care delivery to increase patient access to palliative care. In this article, we will examine association between KPS before hospital admission and discharge disposition.
Methods
Study design and data collection
We conducted a retrospective cohort study of all patients who received a HPPC consultation from January 18, 2018 to May 30, 2019. We collected data about the patients in the cohort using the palliative care clinical database and hospital administrative data. These data included patient demographics, primary serious illness, primary insurance, hospital length of stay, and discharge disposition. We also collected the KPS of the patient before hospital admission, which was determined by the history obtained by the HPPC clinician at time of their initial consultation. The HPPC clinician specifically asked the patients to reflect on their functional status before the acute change that resulted in hospital admission. KPS was selected instead of the Eastern Cooperative Oncology Group (ECOG) performance status and the Palliative Performance Scale (PPS) because ECOG is less sensitive in identifying functional status and is validated only specifically for the cancer population and PPS is a modified version of the KPS.6 The KPS scale ranges from 0 to 100, with 100% indicating full normal activity and 0% marking death. The relationship between KPS before admission and discharge disposition was examined.
Statistical analysis
We described patient characteristics, KPS before admission, and discharge disposition using means and percentages. We conducted bivariate analyses to compare the relationship of KPS before admission and discharge disposition.
The Icahn School of Medicine at Mount Sinai Institutional Review Board approved the study.
Results
Over 16 months, a total of 596 patients were seen by the HPPC team. The patient population had a mean age of 68.4 years, median KPS before admission was 40%, 286 (48%) of patients were female, 194 (33%) had been hospitalized in the past six months, and 261 (36%) had a diagnosis of advanced cancer and the remaining two-thirds had a serious illness caused by organ failure (e.g., heart failure, liver failure, or dementia). The median time from hospital admission to HPPC consultation was three days, the median number of days followed by HPPC was 2.6 days and the median hospital length of stay of 11.2 days. Overall, 349 (59%) had Medicare and 127 (21%) had Medicaid (Table 1).
Table 1.
Patient Characteristics of Population Seen by Hospital Medicine Primary Palliative Care Program (N = 596)
| N (%) | |
|---|---|
| Female | 286 (48%) |
| Mean age, years (median) | 68.4 (70) |
| Mean admission KPS (median) | 37% (40%) |
| Solid cancer | 216 (36%) |
| Mean hospital LOS, days (median) | 17.1 (11.2) |
| Mean time to palliative care involvement, days (median) | 5.9 (3) |
| Mean palliative care LOS, days (median) | 3.5 (2.6) |
| Hospital admission in prior six months | 194 (33%) |
| Insurance | |
| Medicare or Medicare HMO | 349 (59%) |
| Medicaid or Medicaid HMO | 127 (21%) |
| HMO contract | 28 (5%) |
| PPO contract | 24 (4%) |
| Worker's comp | 4 (0.7%) |
Table reports the demographics of sample of patients seen by Hospital Medicine Primary Palliative Care Program.
HMO, Health care maintenance organization; KPS, Karnofsky Performance Status; LOS, length of stay; PPO, preferred provider organization.
We compared the discharge disposition of patients with a KPS ≥60% (able to care for themselves completely or required occasional assistance), those with a KPS 40% to 50%, (required considerable assistance or were disabled), and those with KPS ≤30%, (severely disabled to moribund) (p < 0.001%). Overall, all 33 patients with a KPS ≥60% were discharged alive, of which 30 (91%) were discharged home without services, 1 (3%) home with hospice, and 2 (6%) to a skilled nursing facility (Fig. 1). Of the 300 patients with a KPS of 40% to 50%, 176 (59%) patients were discharged home without services, 23 (7%) home with hospice, 45 (15%) to a long-term care or skilled nursing facility, and 32 (11%) to a hospice or palliative care facility. A total of 9 (3%) patients died before discharge. Finally, 16 patients (5%) were transferred from hospital medicine (e.g., intensive care unit).
FIG. 1.
KPS comparison. Color image is available online.
Of 263 patients with KPS ≤30%, 52 (20%) patients were discharged home without services, 40 (15%) to home with hospice, 41 (15.5%) to a facility without services and 89 (34%) to a facility with hospice or palliative care. Thirty-two patients (12%) died before discharge and 9 (3.5%) patients were transferred to a different hospital clinical service (Table 2).
Table 2.
Admission Karnofsky Performance Status and Discharge Disposition
| |
Karnofsky performance status |
p |
||
|---|---|---|---|---|
| Discharge disposition | 60+ (n = 33) | 40–50 (n = 300) | <30 (n = 263) | <0.001 |
| Death | 0 | 9 (3%) | 32 (12%) | |
| Facility-with hospice or palliative care | 1 (3%) | 32 (11%) | 89 (34%) | |
| Facility without services | 1 (3%) | 44 (15%) | 41 (15.5%) | |
| Home-with hospice | 1 (3%) | 23 (7%) | 40 (15%) | |
| Home-with or without palliative care | 30 (91%) | 176 (59%) | 52 (20%) | |
| Transfer to another service | 0 | 16 (5%) | 9 (3.5%) | |
Table reports discharge disposition based on patient's KPS at the time of admission.
Discussion
Given the limited workforce and resources, KPS before admission may help guide allocation of palliative care resources and determine the appropriate model of delivery in relation with patient discharge disposition and needs. As the patients' functional status worsened (decreased KPS), there was a decrease in the percentage of patients who were discharged home without services and an increase in the percentage of patients who were discharged home with hospice or to a facility. Based on these findings, it may be possible to allocate different clinical models to patients with serious illness based on their functional status before admission. More specifically, the SW-only model may be sufficient for patients with KPS ≥60%, a joint SW and APN model for patients with KPS ≤40% (worse function) who are more likely be symptomatic, and a specialty-level palliative care team for patients with a KPS ≤30%.
After this study, the Australian-modified KPS (AKPS), which has been validated for use across care settings,13 is integrated into the admission nurse assessment at Mount Sinai Hospital. In the future, the AKPS can be used as a trigger to capture patients earlier in the admission with either unmet palliative care needs, physical and occupational therapy services to prevent functional decline during the hospitalization, or social work services for upstream discharge planning based on needs.
There are limitations to this study. The data was collected from a quaternary care center in New York City with a well-established specialty palliative care program and may not be generalizable. Also, we only included patients admitted to Hospital Medicine who may not represent all patients with serious illness. We did not collect data on other factors that may impact discharge disposition such as acuity of medical needs, bed availability at facilities, insurance coverage, patient preferences, and caregiver resources at home. Further research is needed to examine allocation of palliative care services using all patients admitted to the hospital. By optimizing and tailoring models of palliative care delivery, more patients with serious illness will receive the benefits of palliative care.
Authors' Contributions
All authors have reviewed and approved this article before submission.
Funding Information
The funding sources had no role in the study's design, conduct, analysis, or reporting. L.P.G. received additional support from the National Institutes of Health, National Institute on Aging (K23AG049930).
Author Disclosure Statement
No competing financial interests exist.
References
- 1. Kelley AS, Morrison RS: Palliative care for the seriously ill. N Engl J Med 2015;373:747–755 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. El-Jawahri A, Greer JA, Temel JS: Does palliative care improve outcomes for patients with incurable illness? A review of the evidence. J Support Oncol 2011;9:87–94 [DOI] [PubMed] [Google Scholar]
- 3. Meier DE: Increased access to palliative care and hospice services: Opportunities to improve value in health care. Milbank Q 2011;89:343–380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kaasa S, Loge JH, Aapro M, et al. : Integration of oncology and palliative care: A Lancet Oncology Commission. Lancet Oncol 2018;19:e588–e653 [DOI] [PubMed] [Google Scholar]
- 5. Higginson IJ, Finlay I, Goodwin DM, 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 [DOI] [PubMed] [Google Scholar]
- 6. Kamal AH, Maguire JM, Meier DE: Evolving the palliative care workforce to provide responsive, serious illness care. Ann Intern Med 2015;163:637–638 [DOI] [PubMed] [Google Scholar]
- 7. Kamal AH, Wolf SP, Troy J, et al. : Policy changes key to promoting sustainability and growth of the specialty palliative care workforce. Health Aff (Millwood) 2019;38:910–918 [DOI] [PubMed] [Google Scholar]
- 8. Berglund K, Chai E, Moreno J, et al. : A social worker-led primary palliative care model for hospitalized patients admitted to the hospital medicine service. Palliative Med Rep 2020;1:234–241 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. de Kock I, Mirhosseini M, Lau F, et al. : Conversion of Karnofsky Performance Status (KPS) and Eastern Cooperative Oncology Group Performance Status (ECOG) to Palliative Performance Scale (PPS), and the interchangeability of PPS and KPS in prognostic tools. J Palliat Care 2013;29:163–169 [PubMed] [Google Scholar]
- 10. Tandon P, Reddy KR, O'Leary JG, et al. : A Karnofsky performance status–based score predicts death after hospital discharge in patients with cirrhosis. Hepatology 2017;65:217–224 [DOI] [PubMed] [Google Scholar]
- 11. Serper M, Bittermann T, Rossi M, et al. : Functional status, healthcare utilization, and the costs of liver transplantation. Am J Transplant 2018;18:1187–1196 [DOI] [PubMed] [Google Scholar]
- 12. Runacres F, Gregory H, Ugalde A: Restorative care for palliative patients: A retrospective clinical audit of outcomes for patients admitted to an inpatient palliative care unit. BMJ Support Palliat Care 2016;6:97–100 [DOI] [PubMed] [Google Scholar]
- 13. Abernethy AP, Shelby-James T, Fazekas BS, et al. The Australia-modified Karnofsky Performance Status (AKPS) scale: A revised scale for contemporary palliative care clinical practice [ISRCTN81117481]. BMC Palliat Care 2005;4:7. [DOI] [PMC free article] [PubMed] [Google Scholar]