Abstract
Background:
Palliative care trial recruitment of African Americans (AAs) is a formidable research challenge.
Objectives:
Examine AA clinical trial recruitment and enrollment in a palliative care randomized controlled trial (RCT) for heart failure (HF) patients and compare patient baseline characteristics to other HF palliative care RCTs.
Methods:
This is a descriptive analysis the ENABLE CHF-PC (Educate, Nurture, Advise, Before Life Ends: Comprehensive Heartcare for Patients and Caregivers) RCT using bivariate statistics to compare racial and patient characteristics and differences through recruitment stages. We then compared the baseline sample characteristics among three palliative HF trials.
Results:
Of 785 patients screened, 566 eligible patients with NYHA classification III-IV were approached; 461 were enrolled and 415 randomized (AA = 226). African Americans were more likely to consent than Caucasians (55%; PFDR = .001), were younger (62.7 + 8; PFDR = .03), had a lower ejection fraction (39.1 + 15.4; PFDR= .03), were more likely to be single (PFDR= .001), and lack an advanced directive (16.4%; PFDR < .001). AAs reported higher goal setting (3.3 + 1.3; PFDR = .007), care coordination (2.8 + 1.3; PFDR = .001) and used more “denial” coping strategies (0.8 + 1; PFDR = .001). Compared to two recent HF RCTs, the ENABLE CHF-PC sample had a higher proportion of AAs and higher baseline KCCQ clinical summary scores.
Conclusion:
ENABLE CHF-PC has the highest reported recruitment rate and proportion of AAs in a palliative clinical trial to date. Community-based recruitment partnerships, recruiter training, ongoing communication with recruiters and clinician co-investigators, and recruiter racial concordance likely contributed to successful recruitment of AAs. These important insights provide guidance for design of future HF palliative RCTs.
Trial Registration:
ClinicalTrials.gov Identifier: NCT02505425
Keywords: heart failure, RCT, recruitment strategies, KCCQ, palliative care
Introduction
Historically, clinical trials have struggled to identify the appropriate samples of seriously ill patients that palliative care can best serve1 and to enroll and retain diverse patient populations.2,3 Recruitment difficulties are especially prevalent among minority populations, including African Americans (AAs). Multiple factors contribute to the underrepresentation of AAs in palliative care research, especially for those living in rural areas or treated in non-academic health systems. Documented healthcare disparities including a lack of access to palliative care is one reason for underrepresentation of AAs in recruitment,4 particularly in the South where the proportion of AAs is highest.5 Second, cultural factors inclusive of a misunderstanding of palliative care,6 conflicting spiritual and cultural beliefs,7,8 and a mistrust of the medical system based on a history of unethical experimentation including incidents like the Tuskegee study9 have reduced research participation. Third, randomized controlled trials (RCTs) are often conducted in regions where AAs have limited health care access and are not approached as often as Caucasians for trial enrollment.10 Hence, there is a pressing need to develop strategies to recruit and retain AAs in palliative care clinical trials especially in heart failure (HF) populations where AAs are diagnosed earlier and often in a more severe disease stage when compared to their Caucasian counterparts.11 Less is known about recruitment of AA participants in palliative care trials tailored for HF since there are few palliative care clinical trials that have examined the impact of early palliative care integration.12–14
To address this need for more inclusive palliative care trials accessible to a rural and minority southern population, we adapted a successful, nurse-led telehealth early palliative care intervention, called ENABLE (Educate, Nurture, Advise, Before Life Ends) from cancer to HF,15–17 followed by a two-site geographically and racially diverse pilot trial to further adapt the intervention.18,19 In the latter pilot study, Southern patients demonstrated lower levels of “activation” as measured by the Patient Assessment of Chronic Illness Care (PACIC)20 compared with their Northern, predominantly Caucasian counterparts.18 Southern family caregivers also reported feeling less burdened than Northern participants.18 Based on results and patient, family caregiver, and clinician feedback we adapted trial recruitment and intervention materials for literacy and racial diversity18,19 in the current ENABLE CHF-PC (Educate, Nurture, Advise, Before Life Ends Comprehensive Heartcare for Patients and Caregivers; R01NR013665) RCT.21 The purpose of this analysis is to: 1) examine racial differences in RCT participants’ baseline sociodemographic and clinical characteristics, and patient reported outcomes (PROs) and 2)compare ENABLE CHF-PC participant characteristics to other recent HF palliative care trials.
Methods and Materials
Design and Setting
This a descriptive analysis of baseline data from the ENABLE CHF-PC trial (August 2016-October 2018). ENABLE CHF-PC has been described in detail elsewhere.21 Briefly, ENABLE CHF-PC comprises an in-person, outpatient palliative care consultation and nurse coach, telehealth sessions (patient = 6 sessions; caregiver = 4 sessions) followed by monthly check-in phone calls.21 The ENABLE CHF-PC sessions covered relevant palliative care topics including problem solving, self-care, symptom management, communication/decision making, life-story review, and building a legacy.21 The University of Alabama at Birmingham (UAB) and the Birmingham Veterans Affairs Medical Center (BVAMC) institutional review boards approved the study.
Participants
Trained research assistants screened clinician schedules for patients meeting these inclusion criteria: 1) ≥ 50 years old; 2) New York Heart Association Classification (NYHA) III/IV or ACC/AHA Stage C/D HF; 3) English speaking; 4) reliable telephone access; and 5) ability to complete baseline assessment by phone. Patients with dementia or significant confusion (Callahan score ≥3),22 non-correctable hearing loss, or a Diagnostic and Statistical Manual of Mental Disorders (fourth edition) Axis I diagnosis (e.g. schizophrenia, bipolar disorder, or active substance use disorder)23 were excluded since this intervention was not robust enough to meet their complex needs. Family caregivers identified by patients as “a person who knows you well and is involved in your care” and who were an adult ≥ 18 years of age were also recruited; Their data is reported elsewhere.24
Recruitment Procedures
Community-based recruitment.
Given initial difficulties recruiting for our pilot study using a study-based coordinator,18 we entrained the assistance of the UAB Recruitment and Retention Shared Facility (RRSF). The RRSF service was established in 1997 and specializes in minority recruitment to clinical trials using various strategies including racially-diverse recruitment navigators.10 We employed RRSF to conduct clinic-based recruitment and phone-based data collection.25 ENABLE CHF-PC palliative care and HF clinicians, investigators, study staff, and recruitment core staff worked together to develop standardized medical record screening, recruitment, data collection, and training procedures. Training included orientation to the electronic medical record (EMR) systems, the REDCap26 study database, a brief clinical review of advanced HF, and screening and recruitment procedures to identify eligible participants in the EMR. The ENABLE CHF-PC study staff developed a layperson-friendly algorithm of clinical indicators of NYHA Class III/IV HF to aid with screening.
Study staff also developed an UAB/ENABLE CHF-PC branded recruitment folder that included the recruitment script, consent forms, a patient-friendly schema flyer detailing study activities, a study brochure with photo inserts of the study investigators, nurse coaches, and recruiters (Figure 1), survey response sheets for phone-based data collection, and the recruiter’s business card. ENABLE CHF-PC participants received $10 incentives for completing each study data point. After recruitment began, weekly meetings between ENABLE CHF-PC study and recruitment staff were held to monitor recruitment and data collection challenges and successes.
Figure 1.
Patient Friendly Schema.
HF clinician engagement.
Heart failure clinician engagement was a priority from the study conception and pilot phase. Prior to the start of recruitment, co-investigator physician champions built critical clinician relationships across the palliative care and cardiology departments. These relationships were bolstered throughout the study by frequent communication about study progress. At study start-up, the principal investigator and co-investigators spoke at cardiology grand rounds and clinic meetings to briefly describe the trial and eligibility criteria. In addition to the study introduction, study staff provided an in-depth intervention orientation and review of the protocol-driven comprehensive palliative care consultation and telehealth session content during HF and palliative care clinic meetings. These clinician engagement sessions continued throughout the study. Clinician investigators frequently addressed issues related to recruitment and referral with insight into daily clinic operation and scheduling. The study staff developed and posted an ENABLE CHF-PC branded clinician flyer in clinics detailing a brief study overview, eligibility criteria, and ENABLE CHF-PC and recruitment study staff contact information.
Measures
Sociodemographic and clinical characteristics.
Demographic information including age, race, gender, and zip code were collected for all screened patients. Zip codes were used to determine participants’ rural vs. urban/suburban residential status by converting county-based zip codes to Rural-Urban Commuting Area (RUCA) code classifications. Additional demographics and clinical characteristics (e.g. health literacy [Rapid Estimate of Adult Literacy in Medicine/REALM])27 were collected on enrolled patients by self-report and chart review using the HF clinic note closest to date of enrollment.
Patient reported outcomes.
Trained data collectors, who were blinded to study group, collected PROs by phone. PROs included (primary outcomes) quality of life (QOL) (Kansas City Cardiomyopathy Questionnaire [KCCQ]28 and Functional Assessment of Chronic Illness Therapy-Palliative [FACIT-PAL 14]),29 mood (Hospital Anxiety and Depression Scale [HADS]),30 and (secondary outcomes) PROMIS pain intensity and pain interference scales,31 self-reported health (PROMIS SF Global Health 10),32 activation scores (PACIC),20 and reciprocal relationships (Dyadic Adjustment Scale-SF [DAS-7]).33 Two potential mediators and moderators were collected only at baseline: coping style (Brief Cope)34 and perceived social support (Multidimensional Scale of Perceived Social Support).35 The Kansas City Cardiomyopathy Questionnaire28 score ranges from 0–100 with higher scores indicating better perceived health status; clinical summary scores ≥50 are considered “fairly good” QOL36,37 and a change of 5 points is considered a clinically important difference.38,39 The FACIT-PAL 1429 is scored from 0 to 56 with higher scores indicating higher QOL. The HADS30 is scored from 0 to 21 and higher scores indicate more severe anxiety/depressive symptoms. The PROMIS pain intensity and pain interference scales32 range from 0 to 100 with higher scores indicating higher pain intensity and pain interference. Patient Assessment of Chronic Illness Care20 scores range from 0 to 5 with higher scores indicate better self-management and support of chronic condition. The DAS-7 Item Short Form33 is scored from 0 to 36 with higher scores indicating higher perceived degree of agreement in relationship adjustment.
Statistical Analysis
Of 785 patients screened, 566 met eligibility criteria. We compared information (age, sex, race, residence [rural vs urban zip code]) on all patients who passed initial chart screening to assess for indicators of systematic biases of participants vs non-participants across the stages of the recruitment. These stages included: 1) Eligible and Approached (n = 566)—those who were chart screened as potentially eligible and approached by recruitment staff, 2) Consented (n = 461)—those who were approached and signed informed consent, and 3) Declined Consent (n = 105)—those who were eligible and approached but who declined consent. Among those who were consented, we then compared those who were randomized (n = 415) vs those who “screen failed” and did not proceed to randomization (n = 46). Participants were classified as a “screen fail” if they met eligibility criteria and signed consent, but then did not complete baseline measures or were later found not to be eligible.
To make comparisons across the above-named groups we used bivariate tests (t-tests and chi-squared tests) and measures of association, Cohen’s d for continuous variables and d-equivalent40 for categorical variables. P-values were adjusted with a False Discovery Rate (FDR) approach due to the exploratory nature of the tests. The 415 randomized participants who completed baseline assessment (sociodemographic/clinical characteristics and PROs) were compared using bivariate tests based on race (AA or Caucasian) for differences in reported characteristics.
We compared frequencies of reported characteristics of patients from two other recent studies, CASA (Collaborative Care to Alleviate Symptoms and Adjust to Illness)12 and PAL HF (Palliative Care in Heart Failure trial),14 that tested similar outpatient palliative care interventions in HF and utilized similar quality of life outcome measures to compare sample characteristics. The CASA trial was a multisite RCT providing an intervention with a nurse addressing patient symptoms and a social worker addressing psychosocial care in collaboration with the patient’s care team.12 CASA recruited patients with HF with a KCCQ score less than 70 or who reported study relevant symptoms (i.e. shortness of breath, fatigue, pain, depression).12 PAL HF was a randomized controlled trial with an interdisciplinary, evidence-based palliative care intervention for advanced HF patients and recruited participants during their hospitalization or within two weeks of discharge.14
Results
Table 1 displays characteristics of the 785 screened clinic patients. Most screened patients were male (n = 447, 57%) with a mean age of 64.8 years. Of 785 patients screened, 566 (AA n = 278; Caucasian n = 281; Other or not reported n = 7) were found to be eligible and were approached. Of the 566 eligible/approached participants, 105 (19%) declined to provide consent. Reasons for declining included “not interested” (n = 79, 75%), “passive” refusals (e.g. verbally agreeable, but behaviorally unable to commit to study activities) (n = 3, 3%), too ill (n = 2, 2%), phone issues (n = 1, 1%), too busy (n = 1, 1%), other (n = 6, 6%), or reasons not provided (n = 13, 12%). There were no statistical differences in characteristics of those who were consented/randomized (n = 415) and those who consented/not randomized (n = 46) (Table 2B). The 10% (n = 46) of “screen failures” were due to incomplete baseline data collection.
Table 1.
Screened Clinic Patients*.
| All (N = 785) | |
|---|---|
| University of Alabama at Birmingham | 617 |
| VA | 168 |
| Age, mean (SD) | 64.8 + 9.1 |
| Male, n (%) | 447 (56.9) |
| African American, n (%) | 346 (46.5) |
| Rural, n (%) | 180 (26.4) |
Table 2.
Comparisons of Patients at Different Stages of Recruitment.
| A | |||||
|---|---|---|---|---|---|
|
| |||||
| Total (N = 566) Eligible & Approached (N = 566) | Comparison of consented or declined consent |
||||
| Consented (n = 461) | Declined Consent (n = 105) | d or deq | PFDR | ||
| Age, mean (SD) | 64.3 + 8.8 | 63.8 + 8.7 | 66.6 + 8.9 | 0.312 | 0.03 |
| Male, n (%) | 315 (55.7) | 249 (54) | 66 (62.9) | 0.18 | 0.23 |
| African American, n (%) | 278 (49.1) | 252 (54.7) | 26 (24.8) | 0.57 | 0.001 |
| Rural Residence, n (%) | 150 (26.5) | 121 (26.3) | 29 (27.6) | 0.012 | 0.96 |
| B | |||||
|---|---|---|---|---|---|
|
| |||||
| Total (N = 461) Consented (N = 461) | Comparisons of those who were consented and randomized or not randomized |
||||
| Rand omized (n = 415) | Not Randomized (n = 46) | d or deq | PFDR | ||
| Age, mean (SD) | 63.8 + 8.7 | 63.8 + 8.6 | 63.1 + 9.8 | 0.08 | 0.80 |
| Male, n (%) | 249 (54) | 221 (53.3) | 28 (60.9) | 0.15 | 0.55 |
| African American, n (%) | 252 (54.7) | 226 (54.5) | 26 (56.5) | −0.04 | 0.98 |
| Rural Residence, n (%) | 121 (26.3) | 107 (25.9) | 14 (30.4) | 0.1 | 0.72 |
Of 415 randomized participants, 410 identified as Caucasian or AA and were included in this analysis comparing racial differences (Table 3). Most participants were AA (n = 226, 54%), male (n = 220, 53.7%), with a mean age of 63.8 + 8.6, and mean NYHA Class III (n = 396, 97%). While differences were found in patient baseline sociodemographic and clinical characteristics, there were no racial differences observed in primary or secondary patient reported outcomes except for activation scores (goal setting, d = 0.35; PFDR = .007; care coordination, d = 0.42; PFDR = .001) and the mediator/moderator outcome Brief Cope (denial, d = 0.44; PFDR = .001) (Table 4).
Table 3.
Randomized Patient Participant Sociodemographic Factors and Clinical Characteristics.
| All* (N = 410) | Caucasian (n = 184) | African American (n = 226) | Effect Size d or deq | PFDR** | |
|---|---|---|---|---|---|
| Age, mean (SD) | 63.8 (8.6) | 65.2 (9.1) | 62.7 (8) | 0.29 | 0.03 |
| Male, n (%) | 220 (53.7) | 101 (54.9) | 119 (52.7) | 0.03 | 0.81 |
| Site, n (%) | 0.21 | 0.14 | |||
| UAB Cardiology/HF | 292 (71.2) | 130 (70.7) | 162 (71.7) | ||
| Birmingham VA | 85 (20.7) | 45 (24.5) | 40 (17.7) | ||
| Religion, n (%) | 0.025 | 0.17 | |||
| Protestant | 377 (92) | 165 (89.7) | 212 (93.8) | ||
| None | 13 (3.2) | 8 (4.4) | 5 (2.2) | ||
| Attend Religious Services, n (%) | 0.22 | 0.09 | |||
| Occasionally | 169 (41.2) | 70 (38) | 99 (43.8) | ||
| Regularly | 197 (48) | 85 (46.2) | 112 (49.6) | ||
| Ever prayed for your own health, n (%) | 370 (90.2) | 162 (88) | 208 (92) | 0.13 | 0.33 |
| Marital Status, n (%) | 0.39 | 0.001 | |||
| Never married | 51 (12.4) | 10 (5.4) | 41 (18.1) | ||
| Married or living with partner | 197 (48) | 106 (57.6) | 91 (40.3) | ||
| Divorced or separated | 107 (26.1) | 43 (23.4) | 64 (28.3) | ||
| Widowed | 54 (13.2) | 25 (13.6) | 29 (12.8) | ||
| Lives alone, n (%) | 89 (21.7) | 42 (22.8) | 47 (20.8) | 0.04 | 0.80 |
| Work status, n (%) | 0.18 | 0.22 | |||
| Employed | 30 (7.3) | 15 (8.2) | 15 (6.6) | ||
| Retired/Homemaker | 173 (42.2) | 89 (48.4) | 84 (37.2) | ||
| Not employed/Disability | 206 (50.2) | 80 (43.5) | 126 (55.7) | ||
| Education, n (%) | 0.09 | 0.63 | |||
| ≤High school, GED | 190 (46.3) | 78 (42.4) | 112 (49.6) | ||
| ≥Some college/technical school | 219 (53.4) | 106 (57.6) | 113 (50) | ||
| Rural Residence, n (%) | 107 (26.1) | 63 (34.3) | 44 (19.5) | 0.33 | 0.009 |
| Literacy in medicine (REALM-SF), n (%) | 0.01 | 0.53 | |||
| ≤Eighth grade | 37 (9) | 13 (7.1) | 24 (10.6) | ||
| ≥High school | 366 (89.3) | 170 (92.4) | 196 (86.7) | ||
| Medical Insurance, n (%) | 0.39 | 0.001 | |||
| Private/commercial | 74 (18) | 35 (19) | 39 (17.3) | ||
| Medicare/Medicaid | 119 (29) | 35 (19) | 84 (37.2) | ||
| Medicare + Private | 135 (32.9) | 78 (42.4) | 57 (25.2) | ||
| Uninsured | 35 (8.5) | 11 (6) | 24 (10.6) | ||
| Smoking Habits, n (%) | 0.04 | 0.81 | |||
| Never smoked | 175 (42.7) | 75 (40.8) | 100 (44.3) | ||
| Used to smoke | 35 (8.5) | 91 (49.5) | 108 (47.7) | ||
| Ever used other tobacco products | 199 (48.5) | 17 (9.2) | 18 (8) | 0.25 | 0.05 |
| Alcoholic drinks per week, n (%) | 0.17 | 0.23 | |||
| None | 359 (87.6) | 168 (91.3) | 191 (84.5) | ||
| 1–4 | 42 (10.2) | 13 (7.1) | 29 (12.8) | ||
| Years since advanced HF diagnosis, mean (SD) | 5.1 (5.2) | 5.5 (5.7) | 4.7 (4.8) | 0.15 | 0.23 |
| Last recorded Ejection Fraction*, mean (SD) | 41.2 (15.8) | 43.7 (15.9) | 39.1 (15.4) | 0.3 | 0.03 |
| NYHA Class III, n (%) | 396 (96.6) | 182 (98.9) | 214 (94.7) | 0.18 | 0.14 |
| Charlson Comorbidity Index, n (%) | 0.15 | 0.30 | |||
| 1–2 | 170 (41.5) | 83 (45.1) | 87 (38.5) | ||
| 3–4 | 154 (37.6) | 70 (38) | 84 (37.2) | ||
| ≥5 | 86 (21) | 31 (16.9) | 55 (24.3) | ||
| Seattle HF Model, mean (SD) | |||||
| 1-year survival % | 86.5 (12.7) | 86.4 (12.4) | 86.6 (13) | 0.02 | 0.97 |
| 2-year survival % | 75.9 (18.8) | 75.7 (18) | 75.9 (19.5) | 0.01 | 0.97 |
| 5-year survival % | 52.6 (24.2) | 51.7 (23.6) | 53.3 (24.7) | 0.06 | 0.70 |
| Hospital Days, last 2 months, mean (SD) | 2.7 (6) | 2.5 (5.1) | 2.8 (6.6) | 0.05 | 0.75 |
| ICU Days, last 2 months, mean (SD) | 0.6 (2.6) | 0.7 (2.8) | 0.6 (2.4) | 0.04 | 0.81 |
| ED visits, last 2 months, mean (SD) | 0.4 (0.8) | 0.4 (0.7) | 0.5 (0.9) | 0.13 | 0.34 |
| Hospice program, last 2 months, n (%) | 24 (5.9) | 11 (6) | 13 (5.8) | 0.02 | 0.999 |
| Completed an AD, n (%) | 103 (25.1) | 66 (35.9) | 37 (16.4) | 0.37 | <0.001 |
| DNR order, n (%) | 65 (15.9) | 42 (22.8) | 23 (10.2) | 0.33 | 0.003 |
UAB = University of Alabama at Birmingham; HF = heart failure; VA = Veteran’s Affairs; GED = General Educational Development; REALM-SF = The Rapid Estimate of Adult; Literacy in Medicine—Short Form; NYHA = New York Heart Association; ICU = intensive care unit; ED = Emergency Department; AD = advanced directive; DNR = do not resuscitate.
Total of 410/415 participants included in analysis. Five participants identified as a race other than Caucasian or African American.
PFDR: P-values adjusted with a False Discovery Rate approach, 10% false discover rate.
Table 4.
Randomized Patient Participants’ Baseline Patient Reported Outcomes by Race.
| All (N = 410) Mean (SD) | Caucasian (n = 184) Mean (SD) | African American (n = 226) Mean (SD) | Effect Size d or deq | PFDR | ||
|---|---|---|---|---|---|---|
| KCCQ Clinical Summary | 52.5 (21) | 50.5 (21) | 54.2 (20.9) | 0.17 | 0.221 | |
| FACIT-Pal 14 | 36.4 (9.5) | 35.5 (9.5) | 37.1 (9.5) | 0.16 | 0.227 | |
| HADS Anxiety | 6.7 (3.6) | 6.7 (3.4) | 6.7 (3.8) | 0.00 | 0.974 | |
| HADS Depression | 5.7 (4.3) | 5.9 (4.1) | 5.6 (4.4) | 0.05 | 0.725 | |
| PROMIS Pain Intensity | 46 (10.6) | 45.9 (10.6) | 46 (10.6) | 0.01 | 0.974 | |
| PROMIS Pain Interference | 55.1 (10.8) | 55.3 (10.8) | 54.8 (10.8) | 0.05 | 0.835 | |
| PROMIS SF Global | ||||||
| Physical Health T Score | 38.3 (8) | 37.7 (8) | 38.8 (8) | 0.14 | 0.300 | |
| Mental Health T Score | 45.3 (8.7) | 44.6 (8.6) | 45.8 (8.7) | 0.15 | 0.287 | |
| PACIC Summary Score | 3.2 (1) | 3 (1) | 3.3 (1.1) | 0.28 | 0.035 | |
| Goal Setting | 3.1 (1.3) | 2.9 (1.2) | 3.3 (1.3) | 0.35 | 0.007 | |
| Care Coordination | 2.6 (1.2) | 2.3 (1.1) | 2.8 (1.3) | 0.42 | 0.001 | |
| Brief Cope-Denial | 0.7 (0.9) | 0.4 (0.7) | 0.8 (1) | 0.44 | 0.001 | |
| MSPSS | 69.2 (10.8) | 70.1 (10.8) | 68.4 (10.8) | 0.16 | 0.287 | |
| DAS7-SF | 27.2 (5.1) | 28.2 (5) | 26.3 (5) | 0.38 | 0.055 |
Table 5 details ENABLE CHF-PC sample baseline characteristics compared to samples from two other recent palliative care HF trials—CASA12 and PAL-HF.14 Main differences included a higher proportion of AAs, more NYHA class III participants, and higher KCCQ scores indicating higher perceived baseline QOL in the ENABLE CHF-PC sample.
Table 5.
Comparisons among Palliative/HF Clinical Trials: ENABLE CHF-PC, CASA, and PAL-HF.
| ENABLE CHF-PC (N = 415) | CASA (n = 314) | PAL-HF (n = 150) | |
|---|---|---|---|
| Age, mean (SD) | 63.8 (8.6) | 65.5 (11.4) | 70.85 (12.9)* |
| African American, n (%) | 226 (54) | 64 (20) | 62 (41) |
| Male n (%) | 221 (53) | 247 (79) | 79 (53) |
| NYHA Class III, n (%) | 401 (97) | 150 (47) | 112 (75) |
| NYHA Class IV, n (%) | 11 (3) | 44 (14) | 20 (13) |
| Baseline KCCQ Clinical Summary Score, mean (SD) | 52.5 (21) | 46.9 (19.2)* | 33.7 (18.2)* |
| 12–16 Weeks | 58.2 (22.7) | 51.75 | 56.15 (23.8) |
| 24 Weeks | 58.6 (24.4) | 52.45 | 57.6 (22.8) |
| FACIT-PAL 14, mean (SD) | 36.4 (9.5) | NA | NA |
| FACIT-PAL, mean (SD) | NA | NA | 119.3 (26.1)* |
| HADS**, mean (SD) | |||
| Anxiety | 6.7 (3.6) | NA | 6.45 (4.7) |
| Depression | 5.7 (4.3) | NA | 6.65 (4.1) |
| PHQ-9, median (IQR) | NA | 9 (5–14) | NA |
ENABLE CHF-PC = Educate, Nurture, Advise, Before Life Ends Comprehensive Heartcarefor Patients and Caregivers; CASA = Collaborative Care to Alleviate Symptoms and Adjust to Illness; PAL-HF = Palliative Care in Heart Failure; NYHA = New York Heart Association; KCCQ = Kansas City Cardiomyopathy Questionnaire; FACIT Pal = Functional Assessment of Chronic Illness Therapy-Palliative; HADS = Hospital Anxiety and Depression Scale; PHQ-9 = Patient Health Questionnaire 9.
pooled SDs from intervention and usual care groups
PAL-HF HADS baseline measurement at 2 weeks.
Discussion
To our knowledge, the ENABLE CHF-PC trial has the largest recruitment of AA participants in a HF palliative care trial reported to date. Over half of the ENABLE CHF-PC sample identified as AA and when approached, Caucasian patients were more likely to decline consent. Identified racial differences in sociodemographic characteristics included the number of AA participants who had completed an advanced directive before the study (16.4%; PFDR < .001). As AA participants are also less likely to have end of life care congruent with their preferences,41 our study highlights the remaining need for targeted, appropriate interventions for AA patients to promote the advance care planning component of palliative care.42 Additionally, while AA’s baseline scores did not significantly differ in study primary or secondary outcomes, we observed higher PACIC goal setting (3.3 + 1.3; d = 0.35; PFDR = .007), care coordination scores (2.8 + 1.3; d = 0.42; PFDR = .001), and denial coping styles (0.8 + 1; d = 0.44; PFDR = .001). African American participants were more activated in their health care, but less likely to complete an advanced directive.
The ENABLE CHF-PC trial evolved from a cancer study and was developed based on a series of pilot studies which examined the need for tailoring or adaptation to accommodate both HF patients and a racially diverse culture. This descriptive analysis provides insights into how to enhance AA recruitment and ultimately reduce disparities in serious illness. Strategies that may have contributed to ENABLE CHF-PC’s recruitment and enrollment success include: a partnership with UAB’s community-based RRSF, use of recruitment scripts and in-person trainings consisting of multiple meetings with clinic physicians and recruiters, involving clinicians as co-investigators, in-person meetings to discuss outcomes and challenges, participation incentives, and possible racial concordance between patients and the recruitment team. Previous studies have shown that racial concordance is a potential facilitator to recruiting AA populations.43,44 Hence, our partnership with UAB’s RRSF and their racially diverse team of recruiters may have contributed to ENABLE CHF-PC’s successful recruitment of a high proportion of AAs.
Recruitment for ENABLE CHF-PC might have also been successful because the content resonated with AA patients in the Deep South. Based on the pilot studies, we learned that the materials needed to be simpler, less textually dense, with more pictures and branded to UAB. Other factors such as financial incentives could have contributed to improving participant recruitment and retention45,46; however, recommended participant incentive amount and frequency in RCTs remains unclear,45 and it remains difficult to determine how much incentives contributed.
It is notable that the ENABLE CHF-PC population had higher QOL scores at baseline compared to CASA and PAL-HF studies with similar interventions and QOL measures. Specifically, PAL HF predominately recruited participants directly from the hospital or shortly after discharge,14 and CASA screened patients prior to enrollment to confirm that their KCCQ score was below a threshold or who were reporting significant symptoms. The ENABLE CHF-PC participants were recruited solely from ambulatory advanced HF clinics in an academic medical center and affiliated BVAMC outpatient center. In palliative care trials, careful thought is needed to further identify HF patients with high symptom burden. Future studies comparing patients receiving palliative care services and their documented KCCQ and other HF QOL instrument scores are needed to further understand which patients may most benefit from palliative care services.
One limitation of the trial was that while ENABLE CHF-PC eligibility criteria aimed for a diverse and symptomatic population, our final sample comprised mostly NYHA class III patients with relatively good QOL. It is possible that as a tertiary referral center, patients who were more ill were less likely to travel to the academic clinic. Palliative care integration into HF is still novel with few palliative care trials providing evidence to support best practices, and future work is needed to further identify the most appropriate subpopulations of HF patients and the best timing for palliative care integration. Establishing thresholds for study entry related to physical and psychological symptoms (e.g. anxiety and depression) may be an important consideration in future palliative care trials.
Conclusions
In summary, this trial has the highest reported recruitment rate and proportion of AAs in a palliative clinical trial to date. Both our earlier pilot work and this trial provided important insights about recruitment and intervention tailoring to a diverse southern population. Our success in recruiting a high proportion of AA participants was likely due to collaborating with experienced recruiters, ongoing communication with community-engaged recruiters and clinician co-investigators, and recruiter racial concordance. Given the high percentage of persons with early stage HF and better-perceived QOL, it is important to consider minimum thresholds of distress to target for the most effective deployment of palliative care resources. These considerations provide important insights for future HF palliative RCTs.
Acknowledgments
We would like to thank the UAB Department of Cardiology clinicians and staff (especially Renzo Loyaga-Rendon MD and Deepak Acharya MD) and the UAB Division of Geriatrics, Gerontology and Palliative Care (especially Richard Taylor DNP and Cathy Casey NP) for their support of the ENABLE CHF-PC study. In addition, we thank Oladele Osisami, Kiana Minor, Julie Schach, James Mapson, Diane Williams, Jacques DeBrow, Cynthia Y Johnson, and Tawny Martin for their recruitment and data collection efforts.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institutes of Health (NIH)/NINR [R01NR013665]. JNDO is supported by the NIH/NINR (R00NR015903) and the National Palliative Care Research Center. Rachel Wells is a postdoctoral trainee supported by a AHRQ funded Health Systems, Outcomes, and Effectiveness Training Program (T32 HS013852). Deborah Ejem is supported by the NIH/NINR (3R01NR013665). Macy Stockdill is supported by the NIH/NINR (1F31NR018782). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
- 1.McMillan SC, Weitzner MA. Methodologic issues in collecting data from debilitated patients with cancer near the end of life. Oncol Nurs Forum. 2003;30(1):123–129. doi: 10.1188/03.onf.123-129 [DOI] [PubMed] [Google Scholar]
- 2.Kirchhoff KT, Kehl KA. Recruiting participants in end-of-life research. Am J Hospice Palliat Care. 2007;24(6):515–521. doi: 10.1177/1049909107300551 [DOI] [PubMed] [Google Scholar]
- 3.Hui D, Glitza I, Chisholm G, Yennu S, Bruera E. Attrition rates, reasons, and predictive factors in supportive care and palliative oncology clinical trials. Cancer. 2013;119(5):1098–1105. doi: 10.1002/cncr.27854 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Center to Advance Palliative Care and National Palliative Research Center. America’s care of serious illness: a state-by-state report care on access to palliative care in our nation’s hospitals. NY, NY: 2019. [Google Scholar]
- 5.Rastogi S, Johnson T, Hoeffel E, Drewery M. The Black population: 2010. US census bureau. In: Government Printing Office. https://www.census.gov/prod/cen2010/briefs/c2010br-06.pdf; 2011 [Google Scholar]
- 6.Johnson KS, Kuchibhatla M, Tulsky JA. Racial differences in self-reported exposure to information about hospice care. J Palliat Med. 2009;12(10):921–927. doi: 10.1089/jpm.2009.0066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Johnson KS. Racial and ethnic disparities in palliative care. J Palliat Med. 2013;16(11):1329–1334. doi: 10.1089/jpm.2009.0066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Johnson KS, Elbert-Avila KI, Tulsky JA. The influence of spiritual beliefs and practices on the treatment preferences of African Americans: a review of the literature. J Am Geriat Soc. 2005; 53(4):711–719. doi: 10.1111/j.1532-5415.2005.53224.x [DOI] [PubMed] [Google Scholar]
- 9.Hong E-M, Yi H, Johnson KJ, Adamek ME. Facilitators and barriers for advance care planning among ethnic and racial minorities in the US: a systematic review of the current literature. J Immigr Minor Health. 2018;20(5):1277–1287. doi: 10.1007/s10903-017-0670-9 [DOI] [PubMed] [Google Scholar]
- 10.Durant RW, Wenzel JA, Scarinci IC, et al. Perspectives on barriers and facilitators to minority recruitment for clinical trials among cancer center leaders, investigators, research staff, and referring clinicians: enhancing minority participation in clinical trials (EMPaCT). Cancer. 2014;120(S7):1097–1105. doi: 10.1002/cncr.28574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Carnethon MR, Pu J, Howard G, et al. Cardiovascular health in African Americans: a scientific statement from the American Heart Association. Circ. 2017;136(21):e393–e423. doi: 10.1161/cir.0000000000000534 [DOI] [PubMed] [Google Scholar]
- 12.Bekelman DB, Allen LA, McBryde CF, et al. Effect of a collaborative care intervention vs usual care on health status of patients with chronic heart failure: the CASA Randomized Clinical Trial. JAMA Intern Med. 2018;178(4):511–519. doi: 10.1001/jamainternmed.2017.8667 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sidebottom AC, Jorgenson A, Richards H, Kirven J, Sillah A. Inpatient palliative care for patients with acute heart failure: outcomes from a randomized trial. J Palliat Med. 2015;18(2): 134–142. doi: 10.1089/jpm.2014.0192 [DOI] [PubMed] [Google Scholar]
- 14.Rogers JG, Patel CB, Mentz RJ, et al. Palliative care in heart failure: The PAL-HF Randomized, Controlled Clinical Trial. J Am Coll Cardiol. 2017;70(3):331–341. doi: 10.1016/j.jacc.2017.05.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Dionne-Odom JN, Kono A, Frost J, et al. Translating and testing the ENABLE: CHF-PC concurrent palliative care model for older adults with heart failure and their family caregivers. J Palliative Med. 2014;17(9):995–1004. doi: 10.1089/jpm.2013.0680 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II randomized controlled trial. JAMA. 2009;302(7):741–749. doi: 10.1001/jama.2009.1198 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bakitas M, Tosteson T, Li Z, et al. The ENABLE III randomized controlled trial of concurrent palliative oncology care. J Clin Oncol. 2014;32(15):9512–9512. doi: 10.1200/jco.2014.32.15_suppl.9512 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bakitas M, Dionne-Odom JN, Pamboukian SV, et al. Engaging patients and families to create a feasible clinical trial integrating palliative and heart failure care: results of the ENABLE CHF-PC pilot clinical trial. BMC Palliat Care. 2017;16(1):45. doi: 10.1186/s12904-017-0226-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Akyar I, Dionne-Odom JN, Bakitas MA. Using patients and their caregivers feedback to develop ENABLE CHF-PC: an early palliative care intervention for advanced heart failure. J Palliat Care. 2018. doi: 10.1177/0825859718785231 [DOI] [PubMed] [Google Scholar]
- 20.Glasgow RE, Edward HW, Judith S, Lisa DM, Robert JR, Sarah MG. Development and validation of the Patient Assessment of Chronic Illness Care (PACIC). Medical Care. 2005;43(5): 436–444. doi: 10.2307/3768396 [DOI] [PubMed] [Google Scholar]
- 21.Wells R, Stockdill ML, Dionne-Odom JN, et al. Educate, Nurture, Advise, Before Life Ends Comprehensive Heartcare for Patients and Caregivers (ENABLE CHF-PC): study protocol for a randomized controlled trial. Trials. 2018;19(1):422. doi: 10.1186/s13063-018-2770-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Callahan CM, Unverzagt FW, Hui SL, Perkins AJ, Hendrie HC. Six-item screener to identify cognitive impairment among potential subjects for clinical research. Med Care. 2002;40(9):771–781. doi: 10.1097/01.Mlr.0000024610.33213.C8 [DOI] [PubMed] [Google Scholar]
- 23.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association; 2000. [Google Scholar]
- 24.Dionne-Odom J, Ejem D, Wells R, et al. Effects of early palliative care for family caregivers of persons with advanced heart failure: The ENABLE CHF-PC Randomized Controlled Trial. J Pain Symptom Manage. 2020;59(2):467–468. doi: 10.1001/jamainternmed.2020.2861 [DOI] [Google Scholar]
- 25.Ladia MAJ, Sison OT, Anonuevo CA, Alejandria MM. Community-based recruitment for clinical trials poses the need for social and ethical considerations. J Clin Epidemiol. 2018;102: 78–86. doi: 10.1016/j.jclinepi.2018.06.002 [DOI] [PubMed] [Google Scholar]
- 26.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2): 377–381. doi: 10.1016/j.jbi.2008.08.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Arozullah AM, Yarnold PR, Bennett CL, et al. Development and validation of a short-form, rapid estimate of adult literacy in medicine. Medical Care. 2007;1026–1033. doi: 10.2307/40221577 [DOI] [PubMed] [Google Scholar]
- 28.Green CP, Porter CB, Bresnahan DR, Spertus JA. Development and evaluation of the Kansas City Cardiomyopathy Questionnaire: a new health status measure for heart failure. J Am Coll of Cardiol. 2000;35(5):1245–1255. doi: 10.1016/s0735-1097(00)00531-3 [DOI] [PubMed] [Google Scholar]
- 29.Shinall MC, Ely EW, Karlekar M, Robbins SG, Chandrasekhar R, Martin SF. Psychometric properties of the FACIT-Pal 14 administered in an outpatient palliative care clinic. Am J Hospice Palliat Care. 2018;35(10):1292–1294. doi: 10.1177/1049909118763793 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Snaith RP, Zigmond AS. The hospital anxiety and depression scale. Br Med J. 1986;292(6516):344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Amtmann D, Cook KF, Jensen MP, et al. Development of a PRO-MIS item bank to measure pain interference. PAIN. 2010;150(1): 173–182. doi: 10.1016/j.pain.2010.04.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hays RD, Bjorner JB, Revicki DA, Spritzer KL, Cella D. Development of physical and mental health summary scores from the patient-reported outcomes measurement information system (PROMIS) global items. Qual Life Res. 2009;18(7):873–880. doi: 10.1007/s11136-009-9496-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hunsley J, Best M, Lefebvre M, Vito D. The seven-item short form of the Dyadic Adjustment Scale: further evidence for construct validity. Am J Fam Ther. 2001;29(4):325–335. doi: 10.1080/01926180152588734 [DOI] [Google Scholar]
- 34.Carver CS. You want to measure coping but your protocol’ too long: consider the brief cope. Int J Behav Med. 1997;4(1):92. doi: 10.1207/s15327558ijbm0401_6 [DOI] [PubMed] [Google Scholar]
- 35.Zimet GD, Dahlem NW, Zimet SG, Farley GK. The Multidimensional Scale of Perceived Social Support. J Pers Assess. 1988; 52(1):30–41. doi: 10.1207/s15327752jpa5201_2 [DOI] [PubMed] [Google Scholar]
- 36.Mastenbroek MH, Pedersen SS, Meine M, Versteeg H. Distinct trajectories of disease-specific health status in heart failure patients undergoing cardiac resynchronization therapy. Qual Life Res. 2016;25(6):1451–1460. doi: 10.1007/s11136-015-1176-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Arnold SV, Li Z, Vemulapalli S, et al. Association of transcatheter mitral valve repair with quality of life outcomes at 30 days and 1 year: analysis of the transcatheter valve therapy registry. JAMA Cardiol. 2018;3(12):1151–1159. doi: 10.1007/s11136-015-1176-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Butler J, Khan MS, Mori C, et al. Minimal clinically important difference in quality of life scores for patients with heart failure and reduced ejection fraction. Eur J Heart Fail. 2020;22(6): 999–1005. doi: 10.1002/ejhf.1810 [DOI] [PubMed] [Google Scholar]
- 39.Joseph SM, Novak E, Arnold SV, et al. Comparable performance of the Kansas City cardiomyopathy questionnaire in patients with heart failure with preserved and reduced ejection fraction. Circ Heart Fail. 2013;6(6):1139–1146. doi: 10.1016/s0735-1097(00)00531-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Rosenthal R, Rubin DB. r equivalent: a simple effect size indicator. Psychol Methods. 2003;8(4):492–496. doi: 10.1037/1082-989X.8.4.492 [DOI] [PubMed] [Google Scholar]
- 41.Sanders JJ, Robinson MT, Block SD. Factors impacting advance care planning among African Americans: results of a systematic integrated review. J Palliat Med. 2016;19(2):202–227. doi: 10.1089/jpm.2015.0325 [DOI] [PubMed] [Google Scholar]
- 42.Ejem DB, Barrett N, Rhodes RL, et al. Reducing disparities in the quality of palliative care for older African Americans through improved advance care planning: study design and protocol. J Palliat Med. 2019;22(S1):90–100. doi: 10.1089/jpm.2019.0146 [DOI] [PubMed] [Google Scholar]
- 43.George S, Duran N, Norris K. A systematic review of barriers and facilitators to minority research participation among African Americans, Latinos, Asian Americans, and Pacific Islanders. Am J Pub Health. 2014;104(2):e16–31. doi: 10.2105/ajph.2013.301706 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Frierson GM, Pinto BM, Denman DC, Leon PA, Jaffe AD. Bridging the gap: racial concordance as a strategy to increase African American participation in breast cancer research. J Health Psychol. 2019;24(11):1548–1561. doi: 10.1177/1359105317740736 [DOI] [PubMed] [Google Scholar]
- 45.Treweek S, Pitkethly M, Cook J, et al. Strategies to improve recruitment to randomised trials. Cochrane Database Syst Rev. 2018;2:Mr000013. doi: 10.1002/14651858.MR000013.pub6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Brueton VC, Tierney JF, Stenning S, et al. Strategies to improve retention in randomised trials: a Cochrane systematic review and meta-analysis. BMJ Open. 2014;4(2):e003821. doi: 10.1136/bmjopen-2013-003821 [DOI] [PMC free article] [PubMed] [Google Scholar]