Current Status of Integrated Palliative Care Among Parkinson Foundation Centers of Excellence in the United States

Abstract

Background and Objectives

To assess the current structures, knowledge, and readiness to integrate palliative care (PC) into Parkinson disease (PD) care at Parkinson's Foundation Centers of Excellence (COE) in the United States.

Methods

Three unique surveys were administered to health care professionals/staff at COEs to assess PC (1) resources, (2) knowledge and comfort, (3) clinical experience and processes, (4) barriers, and (5) readiness for implementation.

Results

Response rates for the 3 surveys were 97%, 98%, and 56%. In total, 41% of COEs have access to outpatient PC specialists, 71% have mental health counseling, 82% have support groups, and 9% had very limited PC resources. Overall, 74% of providers believed persons with advanced PD should receive PC, and knowledge of PC fundamentals was good across providers. For high-needs persons with PD (PWP), only 16% of physicians and 24% of advanced practice providers made referrals to PC specialists ≥75% of the time, while 9% and 16% never made such referrals. Limited time, space, financing, and staffing were seen as major barriers to PC implementation. In total, 37% of providers were satisfied with their COE's ability to provide PC services. Most COEs report a culture open to change and appear well-positioned to implement PC in a more comprehensive fashion.

Discussion

These results demonstrate the emergence of structures and processes to provide PC to persons with PD at COEs. They also identify concrete opportunities to strengthen integration of PC through educational, quality improvement, and advocacy efforts.

Introduction

Palliative care (PC) is an approach to improve quality of life for people living with serious illness and their families by addressing physical, emotional, social, logistical, and spiritual issues.¹ A growing literature supports the use of PC approaches to improve many outcomes important to persons with PD (PWP) and care partners. In addition, several disease societies (e.g. American Academy of Neurology, European Parkinson Disease Association) now recommend PC as part of clinical guidelines.^2-4 Notably, PC is an approach to care that can be delivered using different models depending on the targeted needs of the patient population, including primary PC (PC delivered by neurologists or primary care providers), consultative specialist PC, and integrated models.^5,6

To date, a comprehensive examination of the status of PC as administered to PWP outside of clinical trials has not been conducted. The Parkinson's Foundation Centers of Excellence (COE) network represents leading centers of care for PWP in the United States. In this study, we sought to understand the current status of PC provision at Parkinson's Foundation COE in the United States before the initiation of a large PC implementation. Through surveys, we specifically sought to understand (1) what PC resources were available at COE, (2) health care provider PC knowledge and comfort, (3) clinical experience and processes, (4) perceived barriers, and (5) readiness for implementing PC and areas to target.

Methods

The study's aims were addressed through 3 cross-sectional surveys.

Surveys/Outcome Measures

Three unique surveys were distributed by the Parkinson's Foundation to faculty and staff at the 34 COE.

Survey 1 was distributed to the medical directors and coordinators at 34 COE and explored available PC services and resources. At the time of the survey, there were 35 official COE and one declined to participate. However, one COE had a satellite clinic in another state that was included as an additional site and is now an independent COE. The term COE refers to the Parkinsons Center of Excellence itself, which may provide some or all of the PC services within the center. The institution is inclusive of the services at the COEs and also those available elsewhere in the hospital system but not available at the COE.

Survey 2 was administered to physicians, nurses/RNs, social workers, and advanced practice providers (APPs) before an online educational course on team-based palliative care offered through the Parkinson's Foundation education platform and consisted of 2 parts: (1) a 20-question knowledge assessment administered to physicians and APPs and (2) a survey of attitudes, comfort, and experience with PD.

Survey 3 was distributed to a wide range of patient-facing professionals at COE, including physicians, nurses/RNs, social workers, and APPs, and elicited perceptions of (1) PC and current PC practices at the COE, (2) the COE climate for implementation of the project, and (3) the COE practice culture. Questions pertaining to perceptions of PC and current PC practices were developed by the study team and completed by all respondents.

Questions on COE practice culture were adapted from the Practice Culture Assessment Scale,⁷ which includes 22 Likert scale items (1 = strongly disagree to 5 = strongly agree) that were used to compute scores for the subscales of change culture (10 items), work culture (8 items), and chaos (4 items). Subscale scores were created as the sum of scores (reverse scored for inversely worded items) scaled on a 0 to 100 range with higher scores representing more of the concept. Change culture items focus on how the practice approaches collaborative quality improvement, problem resolution, and change management. Work culture items assess how practice members work together for a pleasant and productive practice environment and high-quality care. Chaos items assess instability, disruption, and disorganization in the practice. Practice culture is often found to be related to the ability of the practice to implement new initiatives.⁸

Questions pertaining to the implementation climate were adapted from an Implementation Climate Scale for interventions,⁹ which included 11 questions with Likert scales scored from 0 = not at all to 4 = to a great extent covering 3 constructs (what care is expected-7 items (0–28 range); what care is supported-2 items (0–8 range), and what practices are rewarded-2 items (0–8 range). The questions pertaining to the practice culture and implementation climate were completed by nurse/RNs, social workers, chaplains, APPs, and physicians.

Analysis

The results are presented as mean with standard deviation or median with interquartile Range for continuous measures and percents for categorical measures, separately for each of the health care provider roles (e.g., physician, nurse or RN, social worker, advanced practice provider). Baseline measures were compared among the roles using analysis of variance for continuous measures and chi-square tests for categorical measures. Pairwise group comparisons were performed using the Tukey-Kramer test for analysis of variance and Bonferroni adjustment for chi-square tests to control the type I error probability. Analyses were performed using SAS v9.4. p values <0.05 were considered statistically significant. Missing data were minimal; therefore, the results are based on actual responses with no imputation of missing values.

Standard Protocol Approvals, Registrations, and Patient Consents

The Advarra Institutional Review Board provided the Parkinson's Foundation approval for the distribution of the surveys that was regulated under an existing protocol. As the surveys posed minimal risk to participants, waiver of written informed consent was approved.

Data Availability

Anonymized data not published within this article can be made available to qualified investigators on a case-by-case basis.

Results

The response rates for the 3 surveys were survey 1: 97% (34/35), survey 2: 98% (667/681), and survey 3: 56% (371/661).

Structure/Resources at Centers of Excellence

Of the 34 centers responding, the most common services that were available within the COE were support groups (82.4%), mental health counseling (70.6%), and outpatient palliative care services (41.2%; Table 1). The least common services were chaplain/spiritual counselor (17.6%) and peer navigator (17.6%). Three (8.8%) centers provided none of these services.

Table 1.

Palliative Care Resources Available to Centers of Excellence

	N	%
What is the role of the person completing this questionnaire?
Medical director	16	47.1
Coordinator	15	44.1
Medical director and coordinator	2	5.9
Which services are a part of your COE team?
Outpatient palliative care services	14	41.2
Mental health counseling	24	70.6
Chaplain or SPIRITUAL COUNSELOR	6	17.6
Support groups	28	82.4
Peer navigator	6	17.6
None of the above	3	8.8
What services are available in your institution (but outside of your COE) that you regularly use?
Outpatient palliative care services	20	58.8
Mental health counseling	21	61.8
Chaplain or spiritual counselor	17	50.0
Support groups	12	35.3
None of the above	2	5.9
What services are available in your community that you regularly use?
Outpatient palliative care services	20	58.8
Home palliative care	18	52.9
Mental health counseling	23	67.6
Chaplain or spiritual counselor	8	23.5
Support groups	29	85.3
Hospice	27	79.4
None of the above	2	5.9
Impact of potential barrier implementing outpatient palliative care at your centera
Staffing	21	61.8
Time	17	50.0
Space	15	44.1
Finances	15	44.1
Scheduling	12	35.3
Billing	9	26.5
Telehealth access	5	14.7
Telehealth support	4	11.8

^aResponses are based on the N and % that report 7 to 10 (stronger barrier) on a scale of 0–10.

The most common services available at the respondent's institution were mental health counseling (61.8%), outpatient PC services (58.8%), and chaplain/spiritual counselor (50%). The least common service available at the institution was a support group (35%). Two (5.9%) centers provided none of these services at the institution level.

The most common services regularly used by respondents through community connections were support groups (85.3%), hospice (79.4%), and mental health counseling (67.6%), while the least commonly available service in the community was a chaplain/spiritual counselor (23.5%). Two (5.9%) centers reported none of these services available in the community.

Palliative Care Knowledge and Comfort

A 20-question assessment of knowledge about PC for PD was administered in survey 2 to physicians (n = 243) and APPs (n = 46) before an online educational course. The mean number of correct responses was 14.1 (SD 2.6) for physicians and 13.7 (SD 2.2) for APPs. The most frequent incorrectly answered questions were regarding (1) common barriers to pain management, (2) patients' qualification for hospice, and (3) strategies for discussing difficult emotions. The most correctly answered questions regarded the complete definition of PC, strategies to improve serious illness conversations, and the timing appropriateness of a proxy in decision-making.

The COE team members' comfort level related to PC was assessed, and detailed results by role are presented in Table 2.

Table 2.

Proportion of Parkinson Foundation Center of Excellence Team Members Comfort Level Related to Palliative Care Rated Somewhat or Very Comfortable

	Physician (n = 239)	Nurse/RN (n = 45)	Social worker (n = 39)	Advanced practice provider (n = 46)
Discussing palliative care, %	79.5	84.5	94.9	63.0
Discussing hospice, %	70.7	73.3	92.3	60.9
Referring to hospice, %	74.1	71.1	87.2	65.2
Assessing caregivers for burnout, %	66.5	82.2	89.7	63.1
Providing support for caregivers, %	70.2	88.9	97.4	73.9
Assessing Parkinson patients with pain, %	64.0	75.6	43.6	45.7
Completing advanced directives, %	46.9	48.9	76.9	34.8
Discussing prognosis, %	84.1	53.3	53.8	69.6
Talking about difficult emotions, %	72.8	73.3	97.4	71.7
Assessing spiritual wellbeing, %	29.7	48.9	79.5	41.3
Providing spiritual support, %	23.4	53.3	66.7	30.4
Giving bad news, %	75.7	35.6	79.5	58.7
Managing behavioral issues in persons with dementia, %	78.3	80.0	79.5	65.2

Most providers (79%–93%) felt comfortable (defined as “somewhat comfortable” or “very comfortable”) discussing PC with patients/family. Similarly, most providers reported feeling comfortable discussing hospice, but 30% of physicians and 50% APPs felt uncomfortable/neutral (defined as “very uncomfortable,” “somewhat uncomfortable,” or “not uncomfortable/comfortable”). Comfort with “assessing caregiver burnout” and “providing support for caregivers” was reported more often by nurses (RNs) and social workers (SWs) than by physicians and APPs. Least comfort reported by physicians was for assessing spiritual well-being (27%) and providing spiritual support (23%).

Clinical Experience and Processes to Deliver Palliative Care

There was general agreement among HCPs (range 70%–82%) that patients with advanced stage PD should receive PC at the COE, but only 28.3% of physicians and 37% of APPs thought it appropriate for all patients with PD. Approximately one-third of respondents felt that “PC is provided frequently at this center” (range 31%–46%), and approximately one-third (range 25%–42%) felt “very satisfied with center's ability to assist with PC issues.”

Assessment of Palliative Care Pillars

Respondents were asked about the 5 PC Pillars: “approximately how many PD patients that you see personally receive the following from you:” (1) systematic management of nonmotor symptoms, (2) management of patient grief and psychosocial issues, (3) caregiver support, (4) annual advance care planning, and (5) timely patient and family referrals to PC specialist (including hospice) when needed.

Systematic Management of Nonmotor Symptoms

Most physicians (58%) and APPs (72%) routinely used (defined as providing to >75% of their PD patients) a screening tool to assess nonmotor symptoms. While most physicians (69%) routinely managed nonmotor symptoms, only 6% of physicians and 4% of APPs managed pain routinely (Table 3).

Table 3.

Proportion of Parkinson Foundation Center of Excellence Team Members Providing Palliative Care Pillars to at Least 75% of Their Parkinson Patients

	Physician (n = 142)	Nurse/RN (n = 31)	Social worker (n = 26)	Advanced practice provider (n = 25)
Systematic management of nonmotor symptoms, %
Use of screening tool to assess nonmotor symptoms	58.5	41.9	23.1	72.0
Management of select nonmotor symptoms	73.2	29.0	19.2	68.0
Management of nearly all nonmotor symptoms	69.0	48.4	19.2	72.0
Pain management	6.3	16.1	3.9	4.0
Coordination of nonmotor symptom management	28.9	29.0	11.5	24.0
Patient grief and psychosocial issues, %
Management of depression or anxiety	44.4	32.3	34.6	40.0
Screening for grief, guilt, demoralization, or spiritual concerns	14.8	16.1	34.6	16.0
Direct support for difficult but normal emotions	15.5	9.7	42.3	8.0
Referrals to chaplain or counselor	4.2	6.5	15.4	8.0
Caregiver support, %
Use of scale or checklist to screen care partners for burnout or other issues	2.1	6.5	15.4	0.0
Direct support for family care partners	12.7	22.6	53.9	4.0
Referrals of care partners to outside services	14.8	16.1	46.2	0.0
Annual advance care planning, %
Anticipatory guidance	21.1	12.9	26.9	16.0
A review and discussion of advance care planning	4.2	9.7	30.8	8.0
Completion of advance care planning documents	2.1	9.7	3.9	8.0
Timely patient and family referrals to PC specialist (including hospice) when needed, %
% Patients you see demonstrate advanced or end-stage PD	2.1	3.2	3.9	4.0
Among end-stage patients, how many are referred by you for hospice services	8.5	16.1	19.2	4.0
% Patients seen by you demonstrate moderate to high palliative care needs	6.3	3.2	3.9	8.0
Among high-need patients, how many do you refer to a specialist	16.2	9.7	15.4	24.0

Patient Grief and Psychosocial Issues

Mood symptoms were managed routinely by 44% of physicians and 40% of APPs, but screening for grief, guilt, demoralization, and spiritual concerns occurred less commonly (physicians 14%, APPs 16%). SWs were most likely (42%) to routinely direct support for difficult but normal emotions, whereas less than 15% of physicians, APPS, and RNs were. SWs (15%) were also more likely to refer to chaplains/counselors routinely, whereas approximately a third of physicians, APPs, and RNs reported doing so none of the time.

Caregiver Support

Approximately half of the respondents reported never using a scale/checklist to screen care partners for burnout/other issues. However, when burnout was suspected, caregivers were reported to be referred routinely for outside services by physicians 15%, APPs 0%, RNs 16%, and SWs 46% (Table 3).

Annual Advance Care Planning

While anticipatory guidance is routinely provided by 21% of physicians and 16% of APPs, a review/discussion of advanced care planning is offered much less often (4% and 8%, respectively). SWs (30%) standout as routinely discussing advanced care planning.

Timely Patient and Family Referrals to PC Specialist (Including Hospice) When Needed

While 78% of physicians reported that <50% of their patients had moderate to high PC needs, 55% of physicians reported referring <50% of these patients to a specialist, and 9% referred none of their high needs PC patients to a specialist (data not shown). Of the end-stage patients, most physicians (70%) reported referring <50% of patients to hospice, and 7% of physicians referred none of these patients.

Perceived Barriers

In Survey 1, when asked to rate (on scale of 0–10) barriers to implementing PC at their center, respondents reported highest concerns (≥7, indicating stronger barrier) for staffing (62%), time (50%), space (44%), finances (44%), scheduling (35%), billing (26%), telehealth access (15%), and telehealth support (12%) (Table 1).

Other practitioners and staff (Survey 3) reported “time in schedule,” “lack of workflows to accommodate PC,” and “lack of key services/specialists” as the most common top 3 concerns, with no differences from physicians, RNs, SWs, and APPs (Table 4).

Table 4.

Parkinson Foundation Center of Excellence Team Member Survey Administered by Invitation by Role

	Physician (n = 164)	Nurse/RN (n = 35)	Social worker (n = 29)	Advanced practice provider (n = 27)	Other (n = 116)	p Value
Demographics
Female, %	48.2*†‡a	97.1*	93.1†	96.3‡	83.2a	<0.001
Age, y, mean (SD)	43.6 (11.1)	44.1 (14.0)	42.1 (13.2)	42.6 (10.0)	44.0 (13.2)	0.94
Time working in health care (post-training), y, mean (SD)	11.5 (11.4)	16.2 (12.8)	13.5 (10.6)	15.6 (10.5)	14.6 (11.2)	0.07
Time working in this movement disorders center, y, mean (SD)	8.6 (9.1)	6.3 (6.9)	7.4 (9.8)	8.3 (7.3)	7.5 (6.9)	0.57
Full-time employee, %	89.0	91.4	75.9	77.8	77.7	0.05
Personal experience as a care partner, %	15.9	14.3	20.7	14.8	20.5	0.80
Perception of palliative care at center, %
Patients/care partners frequently (>50% of the time) request PC	13.9*	20.0	42.9*†	7.4	17.1†	0.003
Patients frequently referred for PC outside of this center	12.0	20.0	17.9	14.8	10.8	0.60
PC Provided frequently at this center	31.0	31.4	46.4	33.3	31.5	0.61
Very satisfied with center's ability to assist with PC issues	42.7	38.2	39.3	25.9	39.8	0.60
Totally agree that a center should provide PC to
All PD patients	28.3*	51.5	53.6	37.0	45.2*	0.01
Select PD patients	53.3	57.6	50.0	51.9	54.8	0.98
Advanced stage PD patients	72.4	72.7	82.1	70.4	72.1	0.85
Seriously thinking about ways to do this more effectively in the next year
Screen/manage nonmotor symptoms	20.4*†	53.1*	51.9†	26.9	NA	<0.001
Screen/support emotional/spiritual challenges	60.6	62.5	55.6	50.0	NA	0.73
Assess/provide support to care partners	52.1	53.1	33.3	65.4	NA	0.13
Complete/review advance care planning	64.1	59.4	59.3	69.2	NA	0.84
Refer patients to hospice or specialist PC	50.0	53.1	51.9	61.5	NA	0.76
Practice culture at center
Improvement and change culture (0–100), median (IQR)	80.0 (7.5–92.5)	78.8 (65.0–96.3)	72.5 (65.0–85.0)	81.3 (65.0–90.0)	76.3 (65.0–90.0)	0.14
Work relationships (0–100), median (IQR)	81.3(75.0–90.6)	81.3 (71.9–90.6)	75.0 (65.6–84.4)	76.6 (68.8–90.6)	75.0 (65.6–90.6)	0.12
Chaos (0–100), median (IQR)	18.8 (6.3–31.3)*	25.0 (18.8–34.4)	31.3 (18.8–43.8)*	25.0 (12.5–37.5)	25.0 (12.5–37.5)	0.01
Implementation climate at center
Expectations (0–28), mean (SD)	17.9 (4.8)*†	13.3 (6.3)*‡ab	19.1 (5.0)‡c	17.9 (4.5)ad	9.1 (6.9)†bcd	<0.001
Support (0–8), mean (SD)	4.4 (1.9)	5.1 (2.0)	5.4 (1.3)	4.7 (1.7)	4.4 (2.5)	0.08
Rewards (0–8), mean (SD)	3.4 (2.0)*	4.3 (2.6)	4.7 (1.7)*	3.4 (2.3)	3.8 (2.6)	0.04
Top three concerns about starting/providing PC at this center, %
Time in schedule	45.1*	28.6	44.8†	48.2‡	14.7*†‡	<0.001
Don't have workflows set up to accommodate PC	38.4*	37.1	44.8†	40.7	16.4*†	0.001
Lack key services/clinical specialists	30.5	28.6	20.7	25.9	17.2	0.14
Top 3 areas of interest for clinicians, %
Logistics of integrating PC into practice	54.9	34.3	44.8	59.3	NA	0.11
Discussing prognosis/end-of-life with patients/family	39.6*	48.6	58.6	74.1*	NA	0.004
Initiating advance care planning conversations	40.9	42.9	62.1	59.3	NA	0.08

Abbreviations: IQR = interquartile range; NA = not available; PC = palliative care; PD = Parkinson disease.

Other role includes Other patient care (Neuropsychologist, PharmD) (n = 26), Other allied health clinicians (n = 18), Administration/management (n = 12), Coordinators (n = 11), Research (n = 11), Administrative support or front desk/scheduling (n = 10), Medical assistant (n = 9), Psychologist (n = 5), Unknown (n = 4), LPN (n = 3), Volunteers or patient advocates (n = 3), Chaplain (n = 2), Mental health counselor (n = 2).

Values are mean (SD) or median (IQR) for continuous variables and percent (%) for categorical variables.

p values test for differences among the available roles. Matching *†‡abcd symbols indicate significantly different pairwise comparisons using post hoc multiple comparisons tests controlling for the type I error rate.

Readiness for Implementation of Palliative Care and Areas to Target

In general, respondents reported a positive “culture for improvement and change” (range 72–81 of 100), including topics such as considerate and responsive leadership, openness to change, and being quality-driven. “Work relationships” were reported similarly (range 75–81 of 100) and included topics such as team members feeling valued, understood, included, supported, and having a collegial environment.

Average scores for feeling “rewarded” (i.e., receiving recognition and appreciation) for providing PC were reported similarly by physicians and APPs (score 3.4 of 8).

Physicians and APPs rated “expectation” to directly or indirectly provide or facilitate PC as 17.9 out of 28, suggesting an understanding that HCPs are expected to provide PC (Table 4).

The top 3 overall areas of interest for additional education and coaching were 1) logistics of integrating PC into practice, 2) discussing prognosis/end-of-life with patients/family, and 3) initiating advance care planning conversations. Interest in discussing prognosis/end-of-life with patients/family was higher for APPs than physicians (74.1% vs 39.6%, p value = 0.004) (Table 4).

Discussion

Our results highlight several areas of strength in the provision of PC services as well as areas to target for improving the access and quality of PC. Regarding strengths, most COE have resources and faculty/staff that can help in a team-based approach to providing PC. Notably, 32% are now offering at least some integrated specialist outpatient PC clinics. In addition, general knowledge of PC is good on average, most providers believe that this approach is an important part of care, and the cultures at most COE are supportive of change and clinical improvement. Regarding areas to focus to improve PC, many aspects of the PC approach, including pain management, caregiver support, psychosocial and spiritual support, advance care planning, and referral for specialist PC and hospice, are not currently well-integrated into COE care and may benefit from educational and quality improvement initiatives. COEs also note systems barriers to PC that if addressed could promote this model of care including time for providers to spend on PC issues, staffing, and reimbursement models.

The classic Donabedian model for evaluating the quality of care delivery evaluates 3 components: (1) structure, (2) processes, and (3) outcomes.¹⁰ Regarding structure, one of the key components for PC is the presence of faculty and staff with expertise in PC or disciplines that support key components (e.g. chaplains and social workers). While having certain services available within the COE may be essential (e.g., personnel with PC expertise), other services (e.g., home PC, hospice) may be available only at the institution or community level. Of the 34 COEs, most had at least some of the key components for provision of PC; however, 3 provided none of these services at the COE, and 2 COEs reported no services at the institution or community levels. This highlights that while most COEs are well-equipped to integrate PC, some COEs are lacking key structural components.

A key aspect of PC workforce development is attitudes toward PC, knowledge of PC, and comfort with the PC approach. Performance on the knowledge assessment showed that physicians (70.5% correct) and APPs (68.5% correct) possess fair to good knowledge regarding PC. The most frequent incorrectly answered questions show that better tools are needed to address pain, to better define qualification for hospice, and understand strategies to discuss difficult emotions.

Most (79%–93%) providers reported feeling comfortable (defined as ‘somewhat comfortable’ or ‘very comfortable’) discussing PC with patients/family, with the least comfort reported by APPs. Despite this reported comfort, a prior study found that physicians rarely discussed end-of-life expectations and that even when these discussions occurred, the caregivers usually initiated the conversation.¹¹ A possible reason why providers reported feeling more comfortable than expected could be due to a cognitive bias known as “illusory superiority” or the “Dunning-Kruger effect.” This refers to the tendency for individuals to overestimate their abilities or knowledge in a given area, particularly when they have limited experience or understanding of the topic. As they gain more knowledge and experience, they become more aware of their limitations, and their confidence decreases accordingly. It remains to be seen how implementation of educational resources would affect the level of comfort reported by providers.

We found a general agreement among HCPs that the COE should provide PC to patients with advanced stage PD. Approximately one-third of respondents felt that “PC is provided frequently at this center,” and approximately one-third felt “very satisfied with center's ability to assist with PC issues.” While this is reassuring that PC efforts are already ongoing at some COEs, it also demonstrates that many HCPs recognize the unmet need of providing PC at their COE.

Some responses that varied by role of the HCP are worthy of note, as they highlight the diverse clinical experiences and processes to deliver PC. For example, SWs were 3 times more likely than physicians to report that patients/partners “frequently (>50% of the time)” request PC. These findings emphasize the importance of a team-based approach. For example, while these results may be a reflection of the providers' own PC perceptions and experiences, they may also indicate that patients/partners are more likely to discuss PC with the SW than the physician, possibly due to SW currently being more involved in nonmedical aspects of care. SWs may also be more keen to discuss PC, as they report more comfort in discussing PC than physicians (Table 2). In addition, patients/caregivers may perceive physicians/APPs to prioritize medical management over discussing PC, as was found in our recent qualitative study (data currently unpublished).

Our data also showed that among patients who reach end stage or demonstrate to have moderate to high PC needs, referrals to a specialist or hospice were made most or all (51%–100%) of the time by a minority (16%–38%) of respondents. This may be explained by qualitative research which showed that providers felt inexperienced or lacked education about PD, palliative/hospice care, or both, which led to uncertainty about timing and appropriateness of referral to PC, and lack of effective communication between physicians and patients.^12,13

Physicians and APPs were notably less comfortable with assessing caregiver burnout and assessing spiritual well-being; both domains in which SWs reported highest level of comfort. Caregiver burnout was never formally assessed by many of the respondents (42%–60%). One explanation for this finding is that caregiver burnout may be overlooked by most care providers either due to time constraints, lack of knowledge regarding the prevalence of this issue, or the unavailability of resources to address it. Another explanation may be that while most practitioners do not use a checklist/scale to quantify caregiver burnout, they do assess it in a less formal way, perhaps through a simple discussion with the caregiver. Routinely appraising caregiver burden is critical to ensuring not only the caregiver's well-being but also the patient's. Caregivers may experience sleeplessness, fatigue, depression, anxiety, and guilt and may not be forthcoming with their complaints so as to avoid making the patient feel that they are a “burden.” Therefore, the health care team should be certain to learn and use the appropriate screening tools and provide resources to address issues, when identified.

Process measures tend to be a fair proxy for quality of care in that a lack of systematic process tends to lead to high variability in the quality of care provided. As an example, despite increasing knowledge on the presence and importance of nonmotor symptoms, these symptoms still tend to be undertreated and underrecognized by clinicians.¹⁴ By using a systematic approach to this issue (use of screening checklist), integrated PC can significantly improve symptom burden in PC.¹⁵ Even with nonmotor symptoms, which tended to be the highest attended pillar of PC, care was often not comprehensive (pain was not included despite its prominence as a common and bothersome nonmotor symptom¹⁶), and checklists or templates were used routinely by only half of respondents.

Physicians and APPs reported managing depression/anxiety most or all of the time, but screening for grief, guilt, demoralization, and spiritual concerns was low across all roles. SWs seem to play the predominant role in addressing difficult but-normal emotions and referring to chaplains/counselors. Surprisingly, approximately a third of physicians, APPs, and RNs reported never referring to a chaplain/counselor. These findings may reflect not only the availability of resources or the practice culture but also personal religious beliefs and practices.

Formal annual advance care planning was reported to occur most of the time by nearly half of the physicians. Approximately a third of respondents reported referring patients to hospice or PC services most of the time, while majority refer only some of the time. A small number of physicians (6% and 9%) reported never referring end-stage patients to hospice or for PC services. Emerging evidence clearly supports the use of PC and hospice services for patients with advanced PD; therefore, understanding the reasons why physicians refer high-needs/end-stage patients to PC/hospice is a key step in improving access for these services.

The differences in the practice of these Pillars underscore the importance of a team-based approach. While physicians/APPs more frequently report managing depression/anxiety, nearly a third of them never referred the patient to a chaplain/counselor, whereas this was a common practice reported among SWs.

In addition, these differences highlight education gaps in the training/experience of practitioners who may be unaware of available resources and services which may complement the care of the patient in a team-based approach.

Most notable barriers to care reported by COE directors and coordinators (survey 1) included inadequate staffing, time, space, and finances. While finances may seem like the common link between all these barriers, it is important to recognize that in a systems-based approach, additional resources may not be necessary, as development of processes and reorganization of existing resources may provide the necessary means to achieve an integrated PC program. Case in point, practitioners and staff (in survey 3) specified insufficient “time in schedule” and “lack of workflows to accommodate PC” as top concerns for implementing PC. However, in a team-based approach, members of the care team would divide the work based on their role and expertise and provide holistic management without overlapping with roles of other providers, thereby making the overall care more efficient and effective.

Finally, our results also showed that the practice and implementation culture at these centers is overall positive and the centers are receptive to change. In general, the COE Network appears to be well-poised to implement changes related to better integrating PC into the care currently provided. This aligns with findings from other survey studies of health care workers that the overwhelming majority (90%) of respondents believe that PD patients/caregivers have palliative needs, and most (76%) said that these needs are “never” or only “sometimes” met.¹² Encouragingly, nearly all respondents (97%) were open to further education regarding PC in PD. Our findings reveal that many providers do not feel rewarded and appreciated for providing palliative-related services, thus creating a culture where providing such services is recognized and incentivized is critical. Nonetheless, most providers feel that they are “expected” to provide PC, and their willingness to learn provides a fertile foundation for improvement and optimism for the large-scale implementation of primary PC at these centers.

Some limitations are worth noting. First, our participants included only highly specialized Centers of Excellence which are typically equipped with more resources (funding, personnel, multidisciplinary expertise). The knowledge, comfort, and perception of providers at these centers may not be representative of the general neurology practices where many PD patients receive care throughout the United States or other parts of the world. However, this is a relevant first step in understanding care as these academic centers provide training and often set standards of practice for other academic and community clinics. Second, the response rate for Survey 3 was under 56% and may introduce bias, with a particular risk of those people most interested in PC being more likely to respond. Finally, these surveys capture self-report but may not reflect actual skills, practice patterns, or outcomes. However, these perceptions still provide an important window into care at COE and opportunities for changing practice and culture.

Parkinson's Foundation COE in the United States have great potential to make integrated PC a new standard within their network and are well on their way with certain structures (e.g., presence of a social worker), pillars of PC (e.g., nonmotor symptoms management), and knowledge. Further work at COE, some already underway, will be needed to advance provider perspectives on integrated PC and to tackle some aspects of care that are currently underaddressed, including psychosocial support and referrals to specialist PC including at end of life. This study may also serve as a benchmark for the Parkinson's Foundation and other organizations looking to improve integrated PC.

Appendix. Authors

Name	Location	Contribution
Umer Akbar, MD	Department of Neurology, Brown University / Rhode Island Hospital, Providence, RI	Drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data
Sandhya Seshadri, PhD, MA, MS	Department of Neurology, University of Rochester, NY	Drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data
Megan Dini, MA	Parkinson's Foundation, New York	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data
Peggy Auinger, MS	Center for Health & Technology, University of Rochester, NY	Drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data
Sally A. Norton, PhD, RN	University of Rochester School of Nursing, NY	Drafting/revision of the manuscript for content, including medical writing for content
Jodi S. Holtrop, PhD	Dissemination and Implementation Science Program, Adult and Child Center for Outcomes Research and Delivery Science (ACCORDS), University of Colorado School of Medicine	Study concept or design
Benzi M. Kluger, MD, MS	Department of Neurology, University of Rochester, NY	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data

Study Funding

The study was funded by a PCORI grant (DI-2019C2-17499) and a NIH R01 grant (R01NR016037).

Disclosure

The authors report no relevant disclosures. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.

TAKE-HOME POINTS

• → Growing literature supports the use of palliative care in management of people with Parkinson disease, but the extent and scope of current practice is unknown.

• → Survey of PD Centers of Excellence showed that many structural elements of PC—including social workers, chaplains, and nurses—are available at most COEs or the institutions.

• → Perceived barriers to PC included staffing, time, space, finances, scheduling, and billing.

• → In general, respondents reported a positive culture for improvement and change demonstrating a fertile substrate for implementation of PC at COEs.