Feasibility of Implementing a Palliative Care Intervention for People with Heart Failure: Learnings from a Pilot Randomized Clinical Trial

Abstract

Background:

People with heart failure (HF) have high morbidity and mortality and may benefit from palliative care (PC).

Objective:

To pilot a randomized, clinical trial of a PC intervention for people with HF.

Design:

Participants were randomized to usual care (UC) or intervention (UC plus specialty PC) between January 2012 and December 2013. The initial PC consultation was conducted in-hospital, with six additional contacts from the PC team over six months.

Setting/Subjects:

The study was undertaken in a large, urban, academic medical center with patients (n = 30) with New York Heart Association HF Class II–IV.

Measurements:

Patients completed a survey at baseline, three and six months, assessing pain, dyspnea, depression, and quality of life (QoL).

Results:

Although there were significant improvements in mean scores from baseline to six months for pain (4.3 vs. 2.4, p = 0.05), dyspnea (3.9 vs. 2.2, p = 0.03), and QoL (59.2 vs. 42.7, p = 0.001), there were no differences between study groups over time. On average, participants in the intervention group received 5 out of 6 contacts and were satisfied with the intervention.

Conclusions:

The intervention was well accepted by patients, safe, and feasible. Our findings suggest that PC interventions for people with HF should match the PC needs of the patient. Given the trajectory of HF, studies may need to recruit outpatients and follow patients for a longer period to fully evaluate the impact of PC interventions. Clinical trials Identifier: NCT01461681.

Introduction

Randomized clinical trials have demonstrated that patients receiving palliative care (PC) in conjunction with guideline-driven heart failure (HF) therapies achieve significant improvements in quality of life (QoL) and spiritual well-being, reductions in symptom burden, and greater completion of advance care planning documentation, when compared with patients receiving usual HF treatment.^1,2 Despite these successful trials, there is a paucity of studies describing feasibility of PC interventions for patients with HF that can be invaluable for clinicians and researchers wishing to replicate studies and implement services.³ We conducted a pilot randomized clinical trial to examine the feasibility of recruitment, randomization, retention, assessment procedures, and implementation of an interdisciplinary PC intervention (Symptom Management Service–HF [SMS–HF]) alongside standard cardiology care among patients with HF.

Methods

Study design

Patients with NYHA Class II–IV HF were randomized to either the SMS–HF group or usual care (UC). Surveys were administered at baseline, and at three- and six-month follow-up (Fig. 1).

FIG. 1.

Study design.

Study setting

The study was undertaken between January 2012 and December 2013 in a large, academic medical center.

Participants

The inclusion/exclusion criteria for this study are outlined in Table 1.

Table 1.

Inclusion and Exclusion Criteria

Inclusion criteria	Exclusion criteria
HF primary diagnosis or symptomatic/active HF in current hospitalization or within prior six months	Pulmonary hypertension
NYHA Class II–IV	Right HF
≥18 Years of age	Left ventricular device
English speaking	Percutaneous coronary intervention, coronary arterial bypass graft, or valve replacement during index admission
Able to give informed consent and pass SPMSQ	Pre-/post-heart, liver, lung transplant
Medicine, Cardiology, and HF service	Homeless or live outside Bay Area
No previous PC or hospice care	Active illicit drug use

HF, heart failure; PC, palliative care; SPMSQ, short portable mental status questionnaire.

Recruitment

We recruited inpatients with actively managed HF during the past six months. Patients were screened each day by conducting a review of new inpatient admissions to the medicine and cardiology services. The research coordinator contacted the patient's attending physician before approaching the patient to obtain informed consent. Once patients passed the short portable mental status questionnaire (SPMSQ),⁴ they were randomized and completed the baseline survey. The study was approved by the UCSF Institutional Review Board (Study No. 11–06859) and registered with clinicaltrials.gov (Identifier: NCT01461681).

Intervention

Patients received intensive PC consultations provided by an interdisciplinary PC team (nurse practitioner, physician, social worker, and chaplain). The SMS–HF team provided direct care to the patient, prescribing medications for symptoms, discussing advance care planning, completing appropriate documentation, and providing psychosocial and spiritual support.⁵ The first PC consultation occurred during hospitalization. Follow-up consultations included a one-week, in-person follow-up assessment and five monthly consultations, either in person (minimum 2) or teleconference. Additional contacts with the SMS–HF team were scheduled as needed. Patients who were re-admitted to the same hospital were followed by the inpatient PC team. We used electronic health record (EHR) templates to document in-person and telephone care, communicating recommendations to the cardiology team. We developed and used standardized, evidence-based protocols for symptom management.

Usual care

The patients randomized to UC received guideline-driven HF treatment.

Procedure

Study data were collected by using patient surveys and by review of the EHR.

Patient survey

The baseline survey was completed during hospital admission and took 30–35 minutes to complete. The follow-up surveys were mailed to all participants. Patients who did not return the survey after two weeks were contacted by telephone and declared nonresponsive after 5 attempts.

We collected demographic information (age, gender, ethnicity, marital status, income, educational level) and HF severity by using the NYHA Classification.⁶ QoL was assessed by using the Minnesota Living with HF Questionnaire (MLHFQ),⁷ and the Functional Assessment of Chronic Illness Therapy was assessed with the PC (FACIT-PAL) subscale.⁸ Symptom-specific instruments included: the Brief Pain Inventory (BPI),⁹ the Hospital Anxiety and Depression Scale (HADS),¹⁰ the Edmonton Symptom Assessment Scale (ESAS),¹¹ The Brief Fatigue Inventory (BFI),¹² and the BORG scale for dyspnea.¹³ Patients also rated their satisfaction with care.

Electronic health records

Assessment of HF etiology, ejection fraction, medications, documentation of advance care planning discussions, and number of readmissions during the six-month assessment period was obtained from the EHR. Advance care planning documentation included: identification of a surrogate decision maker, preferences for resuscitation, and advance care planning documents such as an advance directive or Physician Order for Life Sustaining Treatments (POLST) form.

Randomization

Patients were randomized within blocks of six to minimize potential bias and confounding.¹⁴ A member of the research team with no contact with the study patients conducted a random assignment procedure to prevent any bias in the allocation to groups.¹⁵

Statistical analysis

Descriptive statistics were used to examine the distribution of measures. Chi-square (χ²) analysis and analysis of variance examined bivariate associations. Change in outcome variables (QoL, depression, anxiety, pain severity, and dyspnea) repeated at baseline, three- and six-months, and differences between study groups were assessed by using linear mixed-effects models. This approach adjusts for sources of variation and correlation among repeated measures within a subject. Further, the analysis of repeated measurements ensures subjects with missing data can still be included. Confidence intervals and p-values were calculated for the overall average effects. An alpha of <0.05 determined statistical significance. McNemar tests were undertaken to compare responses to questions regarding patient satisfaction with care at baseline and at the six-month follow-up. The Statistical Package for the Social Sciences (SPSS) for Mac version 23¹⁶ was used for all analysis. The statistician was blinded to group coding.

Results

Patient characteristics

Overall, 1757 patients were screened for eligibility between March 2012 and March 2013, of whom 117 were eligible and 39 were enrolled (33% participation rate) (Fig. 2). Patients were randomly assigned to UC (n = 17) or intervention (n = 22). Three patients in UC and six patients in the intervention group dropped out. A total of 30 patients (14 patients UC and 16 intervention) completed follow-up surveys.

FIG. 2.

Consort flow diagram.

Patients were 65 years of age, 16 were men, 13 were white, with self-reported NYHA Class I (n = 7), II (n = 9), III (n = 11), or IV (n = 3). Patients had been living with HF for nine years, and 10 had a mean ejection fraction >50%. The intervention group had significantly more women than the UC group (69%, n = 11 vs. 28%, n = 4; p = 0.03) (Table 2).

Table 2.

Participant Characteristics

	Usual care	Intervention
Characteristics	N = 14	N = 16	p
Age, mean (SD)	59 (19)	71 (18)	0.08
Sex, % (n)			0.03
Female	28 (4)	69 (11)
Race, % (n)			0.9
White	43 (6)	44 (7)
No. of people at home with the patient, mean (SD)	2.2 (1.6)	1.4 (1.4)	0.1
Education, % (n)			0.3
Some college	36 (5)	19 (3)
4+ Years	64 (9)	81 (13)
Years with HF, mean (SD)	8 (5)	10 (13)	0.5
Ejection fraction, % (n)	N = 8	N = 15	0.7
>50%	38 (3)	47 (7)

SD, standard deviation.

Prescribed medications and guideline-driven heart-related therapies

There were no differences between the two study groups in terms of guideline-driven HF therapies received or medications for other conditions (Table 3).

Table 3.

Medications (Prescription and Over the Counter) Included in the Medication List of Patients

		Usual care	Intervention
Total	% (n)	% (n)	% (n)	χ ²	p
Guideline-driven HF therapies
CRT device	27.7 (8)	35.7 (5)	20.0 (3)	0.9	0.3
ACE-I/ARB	48.3 (14)	35.7 (5)	60.0 (9)	1.7	0.2
Diuretics	75.9 (22)	64.3 (9)	86.7 (13)	2.0	0.2
Spironolactone/eplerenone	27.6 (8)	28.6 (4)	26.7 (4)	0.01	0.9
Beta-blockers	58.6 (17)	50.0 (7)	66.7 (10)	0.8	0.4
Medications for other conditions
Cholesterol-lowering medication	62.1 (18)	50.0 (7)	73.3 (11)	1.7	0.2
Anti-anginal	17.2 (5)	14.3 (2)	20.0 (3)	0.2	0.7
Diabetes medication	13.8 (4)	14.3 (2)	13.3 (2)	0.006	0.9
Antidepressants	24.1 (7)	28.6 (4)	20.0 (3)	0.3	0.6
Pain medication (NSAIDs and opioids)	37.9 (11)	21.4 (3)	53.3 (8)	3.1	0.08
Anxiety medication	3.4 (1)	7.1 (1)	0 (0)	1.1	0.3
Constipation	27.6 (8)	28.6 (4)	26.7 (4)	0.01	0.9

CRT, cardiac resynchronization therapy devices; NSAIDs, non-steroidal anti-inflammatory drugs.

Clinical outcomes

At baseline, there were no differences between the UC or intervention groups for heart-related QoL (p = 0.9), QoL (p = 0.2), depression (p = 0.4), anxiety (p = 0.4), pain severity (p = 0.7), and dyspnea (p = 0.4).

Patients in both groups reported improvements in all outcomes except depression across the three assessment periods: heart-related QoL (p = 0.001), QoL (p = 0.03), anxiety (p = 0.003), severe pain (p = 0.05), dyspnea (p = 0.03), and depression (p = 0.15) (Table 4). There were no differences between the two groups over the course of the study.

Table 4.

Change in Symptom Scores from Baseline, Three-Month, and Six-Month Assessment between Usual Care and Intervention Groups

	Usual care	Intervention		Month ^a		Group × month ^b
	N = 14	N = 16	Month
Symptoms	Mean (95% CI)	Mean (95% CI)	Mean	F	p	F	p
MLHFQ (0–105)^c
Baseline	60.0 (46.8–73.2)	58.4 (46.1–70.8)	59.2	9.5	0.001	0.2	0.9
Three months	41.8 (28.0–55.6)	41.1 (30.1–52.0)	41.5
Six months	45.4 (31.3–59.4)	40.0 (27.1–52.8)	42.7
FACIT-PAL (0–184)
Baseline	115 (102–128)	122 (110–134)	118.5	3.9	0.03	0.6	0.6
Three months	128 (112–144)	130 (117–144)	129.0
Six months	126 (113–143)	128 (112–140)	127.0
Depression (8–21)
Baseline	6.5 (4.4–8.6)	5.4 (3.4–7.4)	5.9	1.9	0.15	1.2	0.3
Three months	4.3 (2.1–4.5)	4.9 (3.1–6.7)	4.6
Six months	5.9 (3.6–8.1)	4.7 (2.6–6.8)	5.3
Anxiety (8–21)
Baseline	7.4 (4.9–9.9)	5.9 (3.5–8.2)	6.6	7.6	0.003	2.1	0.1
Three months	4.2 (2.2–6.2)	4.8 (3.1–6.5)	9.0
Six months	4.9 (2.2–7.5)	5.4 (3.0–7.9)	5.2
Pain severity (0–10)
Baseline	4.1 (2.2–5.9)	4.6 (2.5–6.6)	4.3	3.8	0.05	0.05	0.95
Three months	2.5 (0.2–4.7)	2.5 (0.6–4.2)	2.5
Six months	2.2 (0.2–4.1)	2.5 (0.4–4.5)	2.4
Dyspnea (0–10)
Baseline	4.5 (2.7–6.2)	3.4 (1.8–5.0)	3.9	4.0	0.03	2.1	0.1
Three months	3.3 (1.7–4.8)	2.4 (1.3–3.5)	2.9
Six months	2.4 (1.1–3.8)	1.9 (0.6–3.2)	2.2

Examines the overall means between the two interventions.

Examines the interaction between the “Intervention Group” by “Month.”

Higher score reflects a poorer QoL.

95% CI, 95% confidence interval; FACIT-PAL, Functional Assessment of Chronic Illness Therapy with the PC; MLHFQ, Minnesota Living with HF Questionnaire; QoL, quality of life.

Intervention adherence

The 16 patients involved in the intervention received on average 5 of the 6 components of the intervention (range: 1–6, Table 5). Nine patients received additional consultations (range: 1–4).

Table 5.

Components of the Intervention Received by Patients on the Symptom Management Service–Heart Failure Group (0 = No, 1 = Yes)

		In-person consultations		Telephone consultations						Study completion
Patient	One-week follow-up	First	Second	First	Second	Third	Total consultations	Additional consultations	Spiritual assessment	AD documented	POLST completed
1	1	1	1	1	1	1	6	0	1	1	0
2	1	1	1	1	1	1	6	3^a	1	1	0
3	1	1	1	1	1	1	6	4^a,b	1	0	0
4	0	1	1	1	1	1	5	1	1	0	0
5	0	1	1	1	1	0	4	0	1	0	0
6	0	1	1	1	1	1	5	1	1	1	0
7	0	1	1	1	1	1	5	1	1	0	0
8	1	1	1	1	0	0	4	0	1	0	1
9	1	1	1	1	1	1	6	1	1	1	0
10	1	1	1	1	1	1	6	0	1	1	0
11	1	1	1	1	1	1	6	1	1	0	0
12	1	1	1	1	1	1	6	1	1	1	0
13	0	1	1	1	1	0	4	1	1	1	0
14	1	0	0	0	0	0	1	0	1	1	0
15	0	1	1	1	0	0	3^c	0	1	0	0
16	0	0	0	1	0	0	1^c	0	1	0	0
Total	10	14	14	15	12	14			16	8	1

In person.

Hospital consultation.

Discontinued participation.

AD, advance directives; POLST, Physician Order for Life Sustaining Treatments.

Spiritual assessment and advance care planning documentation

All patients in the intervention group received a spiritual assessment. At the completion of the study, eight (50.0%) intervention group patients had advance care planning documentation and one completed a POLST form (Table 5). In contrast, only three (21.4%) patients receiving UC had advance care planning (ACP) documentation. However, this difference between the study groups was not significant (χ² = 0.2, p = 0.15).

Satisfaction with care

Patient satisfaction with the care they received was documented at the initial assessment as well as during three- and six-month follow-up assessments. We found no difference in satisfaction with care between the two study groups at any of these three assessment periods.

Discussion

Patients hospitalized with HF had mild-to-moderate symptoms at baseline and were able to report symptoms and QoL throughout the six-month assessment period regardless of study group. Although it was time consuming for patients to complete the surveys and engage with the PC team, the intervention was acceptable to patients and feasible.

This trial provided additional information to inform future studies and service development, especially given PC workforce shortages. For example, some patients who were documented to have NYHA HF Class II at time of screening reported only NYHA HF Class I at initial assessment and many improved significantly by discharge and may not have needed an intensive, six-month PC intervention. Although patients had poorer QoL and symptom scores at baseline, most participants reported improvements. It is important to target the right patients for a specialty PC intervention that is tailored to their needs. We developed a standardized intervention that was comprehensive and intensive, which may not have been necessary for all patients. Matching PC interventions to patient need, including primary PC, might be more appropriate. Though more challenging to implement, given the trajectory of HF, a longer follow-up period may be needed to fully evaluate the impact of the PC intervention. Our six-month follow-up, which was practical for our pilot study, seemed too short to assess the full impact of advance care planning, long-term symptom management, and psychosocial and spiritual support.

Our survey relied on self-report, resulting in potential error associated with poor recall.¹⁷ We attempted to mitigate poor recall by assessing the cognitive capabilities of each patient. As a result, many of the sickest patients who might have been more likely to benefit were excluded. Future studies may consider getting consent and assessing outcomes from surrogates or using outcome measures based on observation rather than patient report. Our survey was long, which may have increased responder burden, particularly for sicker patients. Reducing the number of items or using short-form instruments may mitigate this burden. Given the small number of participants, there was an imbalance of randomization. There were more women and a greater proportion of patients were on ACE-I/ARB and pain medications in the intervention group. Future studies may need to use block randomization to ensure equal allocation across these characteristics and adjust for these between-group differences in multivariate modeling.

Specialty PC has limited availability. Therefore, it is important to find the “sweet spot” where specialty PC significantly improves patient care and to know where PC needs can be met by the patient's other clinicians and system-wide interventions as for advance care planning. The importance of providing PC for patients with HF is becoming increasingly recognized as an invaluable resource and proven in clinical trials.¹⁸ Feasibility studies such as the one described in this article can inform future studies and guide clinical service implementation.

Footnotes

Acknowledgments

The authors would like to thank the National Palliative Care Research Center and the Alafi Family Foundation for funding this study. They thank the patients, primary care physicians, cardiologists, and nurses for their involvement and for allowing them to recruit patients. They would also like to thank the specialist PC team for their commitment to this project. This study was registered with (Identifier: NCT01461681). Funding was provided by the National Palliative Care Research Center.

Author Disclosure Statement

No competing financial interests exist.

References

Rogers

, Patel

, Mentz

, et al.: Palliative care in heart failure: The PAL-HF randomized, controlled clinical trial. J Am Coll Cardiol, 2017; 70:331–341.

Sidebottom

, Jorgenson

, Richards

, et al.: Inpatient palliative care for patients with acute heart failure: Outcomes from a randomized trial. J Palliat Med, 2015; 18:134–142.

Avery

, Williamson

, Gamble

, et al.: Informing efficient randomised controlled trials: Exploration of challenges in developing progression criteria for internal pilot studies. BMJ Open, 2017; 7:e013537.

Pfeiffer

: A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. J Am Geriatr Soc, 1975; 23:433–441.

Steinhauser

, Alexander

, Byock

, et al.: Seriously ill patients’ discussions of preparation and life completion: An intervention to assist with transition at the end of life. Palliat Support Care, 2009; 7:393–404.

The Criteria Committee of the New York Heart Association: Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels, 9th ed. Boston, MA: Little Brown & Co., 1994.

Rector

, Cohn

: Assessment of patient outcome with the Minnesota Living with Heart Failure questionnaire: Reliability and validity during a randomized, double-blind, placebo-controlled trial of Pimobendan. Pimobendan Multicenter Research Group. Am Heart J, 1992; 124:1017–1025.

Lyons

, Bakitas

, Hegel

, et al.: Reliability and validity of the Functional Assessment of Chronic Illness Therapy-Palliative care (FACIT-Pal) scale. J Pain Symptom Manage, 2009; 37:23–32.

Cleeland

, Ryan

: Pain assessment: Global use of the Brief Pain Inventory. Ann Acad Med Singapore, 1994; 23:129–138.

10.

Zigmond

, Snaith

: The hospital anxiety and depression scale. Acta Psychiatr Scand, 1983; 67:361–370.

11.

Nekolaichuk

, Watanabe

, Beaumont

: The Edmonton Symptom Assessment System: A 15-year retrospective review of validation studies (1991—2006). Palliat Med, 2008; 22:111–122.

12.

Mendoza

, Wang

, Cleeland

, et al.: The rapid assessment of fatigue severity in cancer patients: Use of the Brief Fatigue Inventory. Cancer, 1999; 85:1186–1196.

13.

Carvalho

, Bocchi

, Guimaraes

: The Borg scale as an important tool of self-monitoring and self-regulation of exercise prescription in heart failure patients during hydrotherapy. A randomized blinded controlled trial. Circ J, 2009; 73:1871–1876.

14.

Efird

: Blocked randomization with randomly selected block sizes. Int J Environ Res Public Health, 2011; 8:15–20.

15.

Hulley

, Cummings

, Browner

, et al.: Designing Clinical Research, 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2001.

16.

IBM SPSS Statistics for Macintosh, Version 22.0 [computer program]. Armonk, NY: IBM Corp., 2013.

17.

Bhandari

, Wagner

: Self-reported utilization of health care services: Improving measurement and accuracy. Med Care Res Rev, 2006; 63:217–235.

18.

Jaarsma

, Beattie

, Ryder

, et al.: Palliative care in heart failure: A position statement from the palliative care workshop of the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail, 2009; 11:433–443.