Palliative Care in a Community-Based Serious-Illness Care Program

Abstract

Background:

Although community-based serious-illness care (CBSC) is an innovative care model, it is unclear to what extent CBSC addresses palliative care needs, particularly for those patients near death.

Objectives:

To evaluate palliative care services of a CBSC program.

Design:

Retrospective chart reviews.

Setting/Subjects:

Patients enrolled in a CBSC program in central North Carolina.

Measurement:

Descriptive statistics of palliative care needs and services, such as symptom management, psychosocial support, and advance care planning (ACP), for survivors and decedents.

Results:

Patients were seen in an 18-month time frame (n = 159). Mean enrollment in the program was 261.1 days (standard deviation 180.6). Patients' average age was 70 years, 56% were female, and 33% were black. Patients' most frequent comorbidities were dementia (32%), heart failure (32%), and diabetes (28%). Fifty patients (31%) died during the study period. Clinicians most frequently screened for pain (70%), constipation (57%), and dyspnea (57%). Of those screened positive, clinicians most frequently treated pain (92%), anxiety (84%), and constipation (83%). Among the 54% who screened positive for psychosocial distress, 82% received support. Clinicians screened 22% for spiritual needs; 4% received spiritual care. Among decedents, 66% were enrolled in hospice; 14% died in in-hospital. Decedents were more likely than survivors to have ACP (34% vs. 18%, p = 0.03) and a primary goal of comfort (40% vs. 12%, p < 0.01).

Conclusions:

A CBSC program provided palliative care services comparable with other home-based palliative care programs. Although the CBSC program does not address all domains of palliative care, it provided most with symptom management, psychosocial support, and ACP.

Introduction

Many chronically, seriously-ill patients—such as persons with multiple complex comorbidities or advanced cancer—have difficulty accessing office-based care during their illness trajectory.^1,2 Seriously-ill patients not only have medical needs but also have palliative needs—symptom management, psychosocial and spiritual support, and advance care planning (ACP).³

As this population expands, new models of community-based serious-illness care (CBSC) are offering a range of services, with notable overlap of primary care and palliative care.⁴ Although home-based primary care is often conceptualized as a longitudinal program focused on secondary prevention, home-based palliative care is often shorter term and focused on symptom distress.⁵ Little is known about how the emerging model of CBSC addresses patient needs, specifically in terms of palliative care.^6–9 Although CBSC programs may have the capacity to provide palliative care in conjunction with medical management, they lack requirements for provider training, service delivery, or quality metrics.^10–15

This study examines the extent of palliative care services provided in the REACH (Reaching out to Enhance the care of Adults in their Communities and their Homes) program, one CBSC program in central North Carolina. Understanding current delivery practices will provide a picture of current practice, and potential intervention points for future education and guideline development. As part of program evaluation, we examined the frequency of identified palliative care needs and the treatments provided for symptom management, psychosocial and spiritual support, and ACP.

Materials and Methods

Study design

We conducted retrospective chart reviews of patients enrolled in REACH between August 2014 and March 2016. The REACH program is described in detail in Daaleman et al.¹⁶ In brief, the REACH program is a referral-based service that provides serious-illness care and coordinates health care services (such as home health, durable medical equipment, social services, or hospice) for medically complex, at-risk adults in their homes or assisted-living communities. The interdisciplinary REACH team includes (1) physicians board-certified in geriatrics and/or palliative care, (2) nurse practitioners, (3) a nurse care coordinator trained in guided care,¹⁷ (3) a telephone line for urgent needs, (4) a pharmacist (PharmD), and (5) a social worker (MSW).

Two research assistants independently completed a subset of chart reviews of REACH notes and patient data in the electronic health record, reconciled differences until reviews were consistent, and conferred regularly to prevent drift and maintain data quality. This project was approved by the [redacted for blinding] Institutional Review Board.

Participants

Participants were eligible for this analysis if they received REACH services during an 18-month timeframe between August 2014 and March 2016. To access REACH services, patients had to be (1) adults referred by their primary care clinician, (2) unable to access office-based care or had a serious medical illness requiring complex coordinated medical care, and (3) live within 30 miles of [medical center redacted for blinding]. Patients enrolled in REACH tended to have high disease-management or serious-illness care needs (beyond the scope of home health or telehealth, for example).

Measures

Manual chart review at admission visit captured: patient demographics, cognitive impairment at admission, a named surrogate or advanced directives predating admission, and the setting of care (e.g., home, assisted living) at admission. We assessed comorbidities preexisting admission using the Charlson comorbidity index (CCI; range 0–37; higher scores indicate higher disease burden).¹⁸ Decedent status during the study period was confirmed by clinician report or obituary review.

To assess palliative care services, we created a structured chart abstraction tool to capture data on symptom management, psychosocial distress and spiritual needs, ACP, and end-of-life care. Symptom management was based on the McCorkle symptom distress scale assessed at index visit and reported as frequencies: (1) whether each symptom was screened for each patient and (2) if the symptom was identified with any severity, whether the patient received medication/nonmedication treatment.¹⁹ Likewise, research staff recorded whether psychosocial distress and spiritual care were (1) screened and (2) treated/addressed if screened positive.

ACP was measured at the index visit and subsequent REACH visits through code status; presence of a discussion of the primary goal of care; if present, what the goal was; changes in the primary goal of care; presence of a formal ACP note or medical orders for scope of treatment (MOST form); hospice enrollment; and in-hospital death among decedents.

Analysis

We used descriptive statistics to report baseline characteristics and elements of palliative care. We examined differences by decedent versus survivor status using bivariate with t-tests for continuous variables and chi-square for categorical variables. Significance was considered p < 0.05 (STATA 14; College Station, TX).

Results

Participant characteristics

We reviewed records from all 159 adults enrolled in REACH during the 18-month study period. Median length of program engagement was 237 days (interquartile range [IQR]: 117–379). Fifty patients (31%) died during follow-up care; median length of program engagement for survivors was 284 (IQR: 186–407) and for decedents was 75.5 (IQR: 27–237). Patients averaged 70 years of age, 56% were female, and 33% were black, with a mean CCI of 3.2 (Table 1).

Table 1.

Descriptive Characteristics of REACH Patients Reported as n (%) Unless Otherwise Noted

	Total (n = 159)	Surviving enrollee (n = 109)	Decedents (n = 50)	p
Age, mean (SD)	70.4 (17.0)	68.2 (17.7)	75.3 (14.3)	0.01^*
Female	89 (56.0)	63 (57.8)	26 (52.0)	0.49
Race
White	100 (62.9)	62 (56.9)	38 (76.0)	0.03^*
Black	52 (32.7)	43 (39.5)	9 (18.0)
Other	7 (4.4)	4 (3.7)	3 (6.0)
Hispanic/Latinx ethnicity	1 (0.6)	1 (0.9)	0 (0.0)	0.50
Marital status
Single	10 (6.3)	10 (9.2)	0 (0.0)	0.01^*
Married	67 (42.1)	44 (39.5)	24 (48.0)
Divorced	24 (15.1)	21 (19.3)	3 (6.0)
Widowed	38 (23.9)	20 (18.4)	18 (36.0)
Other	20 (12.6)	15 (13.8)	5 (10.0)
CCI score
0	18 (11)	12 (11.0)	6 (12.0)	0.08
1–2	57 (36)	47 (43.1)	10 (20.0)
3–4	38 (24)	26 (23.9)	12 (24.0)
≥5	46 (29)	24 (22.0)	22 (44.0)
CCI, individual comorbidities
Dementia	51 (32.1)	32 (29.4)	19 (38.0)	0.28
Congestive heart failure	50 (31.5)	34 (31.2)	16 (32.0)	0.92
Diabetes	45 (28.3)	36 (33.0)	9 (18.0)	0.05
Moderate–severe renal disease	43 (27.0)	27 (24.8)	16 (32.0)	0.34
Chronic pulmonary disease	40 (25.2)	28 (25.7)	12 (24.0)	0.82
Cerebrovascular disease	28 (23.9)	25 (22.9)	13 (26.0)	0.67
Cancer	23 (14.5)	7 (6.4)	16 (32.0)	<0.01
Peripheral disease	13 (8.2)	8 (7.3)	5 (10.0)	0.57
Liver disease	12 (7.6)	7 (6.4)	5 (10.0)	0.43
Hemiplegia	8 (5.0)	7 (6.4)	1 (2.0)	0.24
Connective tissue disease	4 (2.5)	2 (1.8)	2 (4.0)	0.42
Peptic ulcer disease	4 (2.5)	2 (1.8)	2 (4.0)	0.42
Cognitive impairment	105 (66.0)	64 (58.7)	41 (82.0)	<0.01^*
Named surrogate decision maker before admission	116 (73.0)	73 (67.0)	43 (86.0)	0.01^*
Advance directive before admission	47 (29.6)	27 (24.8)	20 (40.0)	0.05
Living will	22 (13.8)	13 (11.9)	9 (18.0)	0.30
HCPOA	36 (22.6)	21 (19.3)	15 (30.0)	0.13
MOST	7 (4.4)	2 (1.8)	5 (10.0)	0.02^*
Five wishes	1 (0.6)	1 (0.9)	0 (0.0)	0.50
Long-term care	16 (10.1)	10 (9.2)	6 (12.0)	0.58
Before admission visit	6 (3.8)	4 (3.7)	2 (4.0)
During REACH enrollment	10 (3.8)	6 (5.5)	4 (8.0)

Values given in table represented as n (%).

Statistically significant p<0.05.

CCI, Charlson comorbidity index; HCPOA, health care power of attorney; MOST, medical orders for scope of treatment; REACH, Reaching out to Enhance the care of Adults in their Communities and their Homes; SD, standard deviation.

Symptom management

Documentation indicated that REACH clinicians most frequently screened at the index visit for pain (n = 112, 70%), constipation (n = 91, 57%), and dyspnea (n = 91, 57%). Clinicians more often screened decedents than survivors for constipation (n = 41, 82% vs. n = 50, 46%, p < 0.01), appetite (n = 29, 58% vs. n = 32, 29%, p < 0.01), nausea (n = 21, 42% vs. n = 21, 19%, p < 0.01), decreased level of consciousness (n = 23, 46% vs. n = 9, 8%, p < 0.01), and agitated delirium (n = 23, 46% vs. n = 8, 7%, p < 0.01).

Among those who screened positive for symptoms, clinicians most commonly treated for pain (n = 22, 92% of the 24 who screened positive), anxiety (n = 27, 84% of the 32 who screened positive), constipation (n = 30, 83% of the 36 who screened positive), and agitated delirium (n = 10, 83% of the 12 who screened positive). Among those who screened positive, clinicians more often treated decedents than survivors for decreased level of consciousness (55% [6/11] vs. 50% [2/4], p < 0.01) and agitated delirium (90% [8/9] vs. 67% [2/3], p < 0.01; Fig. 1).

FIG. 1.

Symptom management among REACH patients. (a) REACH screening and treatment for palliative care needs among the total population (n = 159). (b) REACH screening and treatment for symptoms among survivors (n = 109). (c) REACH screening and treatment for palliative care needs among decedents (n = 50). REACH, Reaching out to Enhance the care of Adults in their Communities and their Homes.

Psychosocial support

Clinicians screened 54% of patients for psychosocial distress. Among the 76 who screened positive, 82% (n = 62) received treatment or support. Clinicians more often screened decedents than survivors for spiritual needs (35% vs. 15%, p < 0.01). Clinicians addressed spiritual care in two (1% of total) patients, both of whom were decedents (p = 0.04) (Table 2).

Table 2.

Psychosocial Distress, Spiritual Care, and Advance Care Planning among REACH Patients

	Total (n = 159)	Survivors (n = 109)	Decedents (n = 50)	p
Psychosocial distress screened and treated
Screened	85 (53.5)	62 (56.9)	23 (46.0)	0.20
Treated	62 (81.6)	49 (83.1)	13 (76.5)	0.02^*
Spiritual care needs screened and treated
Screened	35 (22.0)	16 (15)	19 (38.0)	<0.01^*
Treated	2 (1)	0	2 (4)	0.04^*
Code status at admission
Full code	100 (62.9)	81 (74.3)	19 (38.0)	<0.01^*
Limited code^*	7 (4.4)	5 (4.6)	2 (4.0)
DNR/DNI	52 (32.7)	23 (21.1)	29 (58.0)
Code status change from full code to either limited code or DNR/DNI at initial visit or within three months	26 (16.5)	13 (11.9)	13 (26.0)	0.03^*
Discussion of primary goal of care (cumulative)	100 (62.9)	62 (56.9)	38 (76.0)	<0.01^*
Prolonging life	12 (7.6)	9 (8.3)	3 (6.0)
Supporting function	45 (28.3)	35 (32.1)	10 (20.0)
Improving comfort	33 (20.8)	13 (11.9)	20 (40.0)
Multiple	10 (6.3)	5 (4.6)	5 (10.0)
Not discussed	59 (37.1)	47 (43.1)	12 (24.0)
Goals of care change from prolonging life to limited code or DNR/DNI or from limited code to DNR/DNI at or within three months of admission	20 (12.6)	5 (4.6)	15 (30.0)	<0.01^*
Advance care planning note (cumulative)	37 (23.3)	20 (18.4)	17 (34.0)	0.03^*
At admission visit	25 (15.9)	14 (13.1)	11 (22.0)
At subsequent visit	15 (9.5)	6 (5.5)	9 (18.0)
MOST form (cumulative)	15 (9.4)	6 (5.5)	9 (18.0)	0.01^*
Before admission visit	5 (3.1)	2 (1.8)	3 (6.0)
At admission visit	6 (3.8)	2 (1.8)	4 (8.0)
At subsequent visit	9 (5.7)	4 (3.7)	5 (10.0)
Hospice enrollment (cumulative)	41 (25.8)	8 (7.3)	33 (66.0)	<0.01^*
Already enrolled in hospice	13 (8.2)	3 (2.8)	10 (20.0)
REACH-facilitated enrollment during admission visit	5 (3.1)	2 (1.8)	3 (6.0)
REACH-facilitated enrollment at subsequent visit	11 (6.9)	1 (0.9)	10 (20.0)
Non-REACH facilitated enrollment	12 (7.5)	2 (1.8)	10 (20.0)
Death at hospital			7 (14.0)
Death upon discontinuation of life-sustaining treatment			2 (4.0)

Values given in table represented as n (%).

Limited code = anything neither full code without limitations nor DNR/DNI with orders against other treatments (e.g., full code but orders not to tube feed; DNR, but would want treatments until then).

Advance care planning

Clinicians completed goals of care conversations in 63% of patients. REACH clinicians more often discussed goals of care with decedents than survivors (76% vs. 57%, p < 0.01). Decedents most frequently had a primary goal of comfort (40%); survivors most frequently had a primary goal of supporting function (32%). Thirteen percent of all patients and 30% of decedents (vs. 5% of surviving patients; p < 0.01) shifted their primary goal during REACH enrollment, none of whom shifted toward more curative treatment. Sixty-six percent of those who died enrolled in hospice. Seven patients (14% of decedents) died in the hospital (Table 2).

Discussion

This study reports the delivery of select palliative care services on admission to a CBCS program. The population served by this program is seriously ill with a high burden of chronic conditions, including a group with generally worsening disease burden, nearing the end of life.^20,21 Although the CBSC program did not provide comprehensive palliative care to all patients, clinicians provided many patients—particularly those approaching death—with elements of symptom management, psychosocial support, and ACP.

Unlike home health or hospice, CBSC practices have the potential to combine palliative care with complex medical management. This combination may be one way of improving outpatient disease management without compromising the values of patients who are not yet ready to enroll in hospice or to forgoing life-prolonging treatments.²² For example, this CBSC program addressed pain and symptom management for survivors nearly as frequently as for those who died during the study period, indicating the model may effectively provide guideline-concordant symptom management across severity and trajectory of illness.¹⁵ Other elements of palliative care—such as spiritual care—may require more training or different interdisciplinary team composition.²³

Clinicians in this CBSC program enrolled two-thirds of decedents in hospice. Furthermore, the low rate of in-hospital death may be due to low symptom exacerbation and high-quality ACP, which is consistent with utilization patterns for patients enrolled in other community-based palliative care programs.^13,24 REACH provided care comparable with at least one other home-based palliative care program.²⁵ One other such program had nearly universal ACP and goals-of-care conversations, indicating that these elements of palliative care could likely be further enhanced in REACH with systematic implementation of ACP.²⁶

Limitations

Our results encompass one program in a single geographic region. Given limited sample size, we were unable to fully determine which patients benefited most from which services. The retrospective noncomparative study design limits the interpretation of results in terms of generalizability. However, we provide a picture of comprehensive care for seriously ill patients over time.

Conclusion

The REACH CBCS program provides added palliative care benefit to seriously ill patients concurrently receiving community-based medical care. Although only one program, these results are promising when considering new and hybrid models of meeting palliative care needs.

Footnotes

Acknowledgments

The authors acknowledge (redacted for blinding) for his support with data collection.

Funding Information

This study was funded by University of North Carolina.

Author Disclosure Statement

No competing financial interests exist.

References

Ornstein

, Leff

, Covinsky

, et al.: Epidemiology of the homebound population in the United States. JAMA Intern Med, 2015; 175:1180–1186.

Qiu

, Dean

, Liu

, et al.: Physical and mental health of homebound older adults: An overlooked population. J Am Geriatr Soc, 2010; 58:2423–2428.

Ventura

, Burney

, Brooker

, et al.: Home-based palliative care: A systematic literature review of the self-reported unmet needs of patients and carers. Palliat Med, 2014; 28:391–402.

Teno

, Montgomery

, Valuck

, et al.: Accountability for community-based programs for the seriously Ill. J Palliat Med, 2018; 21:S-81.

Ritchie

, Leff

: Population health and tailored medical care in the home: The roles of home-based primary care and home-based palliative care. J Pain Sympt Manag, 2017; 55:1041–1046.

Stall

, Nowaczynski

, Sinha

: Systematic review of outcomes from home-based primary care programs for homebound older adults. J Am Geriatr Soc, 2014; 62:2243–2251.

Edes

, Kinosian

, Vuckovic

, et al.: Better access, quality, and cost for clinically complex veterans with home-based primary care. J Am Geriatr Soc, 2014; 62:1954–1961.

Brian Cassel

, Kerr

, McClish

, et al.: Effect of a home-based palliative care program on healthcare use and costs. J Am Geriatr Soc, 2016; 64:2288–2295.

Sudat

, Franco

, Pressman

, et al.: Impact of home-based, patient-centered support for people with advanced illness in an open health system: A retrospective claims analysis of health expenditures, utilization, and quality of care at end of life. Palliat Med, 2017; 32:485–492.

10.

Leff

, Carlson

, Saliba

, Ritchie

: The invisible homebound: Setting quality-of-care standards for home-based primary and palliative care. Health Aff, 2015; 34:21–29.

11.

Leff

, Weston

, Garrigues

, et al.: Home-based primary care practices in the United States: Current state and quality improvement approaches. J Am Geriatr Soc, 2015; 63:963–969.

12.

Smith

, Soriano

, Boal

: Brief communication: National quality-of-care standards in home-based primary care. Ann Intern Med, 2007; 146:188–192.

13.

Lustbader

, Mudra

, Romano

, et al.: The impact of a home-based palliative care program in an accountable care organization. J Palliat Med, 2017; 20:23–28.

14.

Malik

, Goldman

, Kevork

, et al.: Engagement of primary care physicians in home palliative care. J Palliat Care, 2017; 32:3–10.

15.

Ferrell

, Connor

, Cordes

, et al.: The national agenda for quality palliative care: The National Consensus Project and the National Quality Forum. J Pain Sympt Manag, 2007; 33:737–744.

16.

Daaleman

, Ernecoff

, Kistler

, et al.: The impact of a community-based serious illness care program on healthcare utilization and patient care experience. J Am Geriatr Soc, 2019; 67:825–830.

17.

Boult

, Reider

, Frey

, et al.: Early effects of “Guided Care” on the quality of health care for multimorbid older persons: A cluster-randomized controlled trial. J Gerontol A Biol Sci Med Sci, 2008; 63:321–327.

18.

Charlson

, Szatrowski

, Peterson

, Gold

: Validation of a combined comorbidity index. J Clin Epidemiol, 1994; 47:1245–1251.

19.

McCorkle

, Young

: Development of a symptom distress scale. Cancer Nurs, 1978; 1:373–378.

20.

Dhingra

, Barrett

, Knotkova

, et al.: Symptom distress among diverse patients referred for community-based palliative care: Sociodemographic and medical correlates. J Pain Sympt Manag, 2018; 55:290–296.

21.

Gomes

, Calanzani

, Gysels

, et al.: Heterogeneity and changes in preferences for dying at home: A systematic review. BMC Palliat Care, 2013; 12:7.

22.

Parikh

, Bowman

, Dahlin

, et al.: Scalable principles of community-based high-value care for seriously ill individuals: Diamonds in the rough. Healthc (Amst), 2017; 5:12–16.

23.

Puchalski

, Ferrell

, Virani

, et al.: Improving the quality of spiritual care as a dimension of palliative care: The report of the Consensus Conference. J Palliat Med, 2009; 12:885–904.

24.

Wright

, Youens

, Moorin

: Earlier initiation of community-based palliative care is associated with fewer unplanned hospitalizations and emergency department presentations in the final months of life: A population-based study among cancer decedents. J Pain Sympt Manag, 2018; 55:745–754.

25.

Riggs

, Breuer

, Dhingra

, et al.: Hospice enrollment after referral to community-based, specialist palliative care: Impact of telephonic outreach. J Pain Sympt Manag, 2017; 54:219–225.

26.

Chen

, Thorsteinsdottir

, Cha

, et al.: Health care outcomes and advance care planning in older adults who receive home-based palliative care: A pilot cohort study. J Palliat Med, 2015; 18:38–44.