Abstract
Background:
While home-based hospice care seeks to reduce suffering at the end of life (EoL), patients continue to experience a high symptom burden. High symptom burden contributes to adverse outcomes, including patient suffering, burdensome care transitions, and caregiver burden. Yet, most caregivers lack formal education in patient symptom management despite providing up to 65 h of care per week. Interventions that provide symptom support and education to caregivers could improve EoL outcomes for patients and caregivers.
Methods:
We conducted a pilot randomized controlled trial (N = 80) in a hospice organization to assess the feasibility, acceptability, and preliminary efficacy of the Improving Home Hospice Management of End-of-life Symptoms (I-HoME) intervention. This caregiver-focused intervention aims to reduce patient symptom burden through weekly tele-visits with a nurse practitioner and caregiver educational videos to provide symptom support and education.
Results:
The mean age of caregivers (N = 80) was 60.3 (standard deviation ± 12.1); with a majority being women (79%) and children of the patient (67%). In the I-HoME group (n = 40), a total of 121 of a possible 145 tele-visits (83%) were completed. Over 96% of caregivers were either satisfied or very satisfied with the tele-visits. Eighty-three percent agreed or strongly agreed that it prepared them to manage symptoms better, while 88% agreed or strongly agreed that the intervention increased their confidence in managing symptoms. The average reduction in patient symptom burden, as measured by the Edmonton Symptom Assessment Scale, for the intervention group who received all six visits was 6.6 points compared to 2.9 for the control group.
Conclusions:
The I-HoME intervention was feasible to implement in the home hospice setting and acceptable to caregivers and hospice staff. Future efficacy trials are needed to determine whether this caregiver-focused intervention, which provides symptom support and education, can measurably improve patient and caregiver outcomes in the home hospice setting.
Keywords: caregiving, hospice, symptom management
1 |. Introduction
In 2022, over 1.7 million Medicare beneficiaries received hospice services in the last year of life, with the vast majority receiving hospice care at home [1]. While home-based hospice care seeks to reduce suffering and maximize quality of life (QoL), patient symptom burden, which includes, but is not limited to, pain, shortness of breath, anxiety, and depression, remains highly prevalent [2–8] and distressing to both patients and their family caregivers. High symptom burden leads to poor QoL for patients, contributes to burdensome care transitions (e.g., hospitalization) [9, 10], and negatively impacts caregiver outcomes (e.g., burden, depression, anxiety) [11, 12].
Caregivers (often referred to as family, informal, or unpaid caregivers) provide, on average, 40–65 h of care per week for patients during the last year of life [13]. Caregiving demands during this period are significant and stressful. Among the many tasks caregivers are routinely responsible for are attending to changes in patients’ symptom status, administering medications, assisting with activities of daily living, and communicating with members of the health care team. As a result, caregivers often become the patient’s “eyes” and “ears” in the home hospice setting.
Studies have shown that high patient symptom burden negatively impacts caregiver well-being. For instance, fatigue experienced by cancer patients receiving hospice care is associated with greater levels of caregiver depression [14]. Our team found that higher symptom burden scores were significantly associated with greater caregiver burden and lower satisfaction [11, 12]. Thus, optimizing symptom management in hospice patients can not only improve patient outcomes but also have a positive impact on caregiver outcomes.
Despite our reliance on caregivers to keep home hospice patients safe and comfortable, the vast majority lack formal education and guidance in EoL symptom management. Pain management, for example, is particularly challenging for caregivers. Previous studies have identified barriers to effective pain management, which include misconceptions about using opioids (e.g., concerns about addiction or hastening death) for EoL symptoms [15–17]. Past intervention studies have tested caregiver-directed educational initiatives (e.g., brochures, videos), apps to track symptoms, cognitive-behavioral therapy, and additional supports (e.g., dedicated telephone hotline, additional visits) [18]. While some of these interventions have shown promise, many studies in the field lack rigor. In our scoping review of caregiver interventions in home palliative care, we found that 68% of the randomized controlled trials (RCTs) displayed concerns regarding bias. Further, 54 out of the 57 non-randomized studies (95%) were found to have a serious or critical risk of bias. This indicates a significant need for more rigorous and robust research, ranging from pilot studies to pragmatic trials, to identify evidence-based solutions to better support caregivers in the home hospice environment.
In this study, we describe an intervention designed by our research team to support home hospice caregivers in EoL symptom management. The intervention, Improving Home Hospice Management of End-of-life Symptoms (I-HoME), aims to reduce symptom burden through (1) weekly tele-visits conducted by a nurse practitioner (NP) interventionist and (2) caregiver educational videos. We sought to (1) establish the feasibility and acceptability of I-HoME and (2) examine the preliminary efficacy of I-HoME regarding its effects on patient symptom burden and caregiver outcomes.
2 |. Methods
2.1 |. Overview
We conducted a Stage I randomized controlled trial based on NIH’s Stage Model for Behavioral Intervention Development [19] focusing on creating and pilot testing an intervention. This study complied with the ethical rules for human experimentation stated in the Declaration of Helsinki [20], including approval of the Institutional Review Board (IRB) at Weill Cornell Medicine and VNS Health and informed consent. This study was registered with clinicaltrials.gov (NCT04243538).
2.2 |. Study Setting and Participant Eligibility Criteria
The trial was conducted in New York City, with recruitment at VNS Health Hospice, a large non-profit organization. VNS Health has 15 hospice teams, each consisting of a physician, nurse manager, nurses, social worker, and chaplain. Recruitment initially focused on one team but was later expanded to five teams to meet recruitment goals.
Caregiver inclusion criteria included individuals aged 18 or older, English-speaking, and providing weekly care to a home hospice patient aged 65 or older. Informed consent was obtained from caregiver participants. Due to concerns about the ability of patients with dementia [21] to consent and to avoid added burden, formal consent was not obtained from patients. We obtained a HIPAA waiver to collect patient data from VNS Health’s electronic health record (EHR).
2.3 |. Randomization and Recruitment
We randomized participants at the nurse level—nurses (not the NP interventionists) were randomly assigned (using a random number generator) to the I-HoME group or the control group. In home hospice care, nurses manage a group of patients whom they follow throughout the patient’s hospice stay, while physicians play more of an advisory/administrative role. Caregiver participants were assigned to either the intervention or the control group based on the nurse’s randomization. Our reasoning behind randomizing at the nurse level rather than the individual caregiver level was to prevent bias that might occur if nurses altered their care approach in a scenario where a nurse has two study caregivers in their caseload who are assigned to different study arms.
We identified potential caregiver participants through the electronic health record (EHR). Patient (i.e., name, age, admission date) and caregiver (i.e., name, contact number) data were obtained from the EHR. Before contacting potential participants, we reached out to the patient’s hospice nurse to screen for the appropriateness of enrolling a given caregiver. Nurses were informed that caregivers should not be enrolled if: (1) their care recipient is imminently dying, as there may not be enough time for consent and participation; or (2) the caregiver is not actively providing care. After the hospice nurse’s screening, the study team called eligible caregivers to explain the study and obtain consent from interested participants. Participants in the intervention group received a 15-min phone training session on using the Doxy.me tele-visit platform [22] before the start of the intervention. Doxy.me is a HIPAA-compliant platform that can be accessed via phone, tablet, or computer.
2.4 |. I-HoME Intervention
I-HoME is a dual-component intervention designed to reduce patient symptom burden and improve caregiver outcomes by enhancing caregiver self-efficacy and knowledge through symptom support and education. The intervention was developed using a user-centered design approach, incorporating feedback from hospice staff and caregivers to refine the intervention components before pilot testing. I-HoME consists of (1) weekly tele-visits for up to 6 weeks with a hospice NP interventionist to support caregivers in managing patient symptoms and (2) a library of eight caregiver educational videos. The NP interventionist underwent a 1-day training session before implementing I-HoME and met weekly with the study team during the study. Of note, study interventionists were NPs already embedded within the VNS Health ecosystem.
The definition of a tele-visit is that it can take place through either a video or a phone call; however, in this study, the NP interventionists were trained to prioritize video. If there were video connection issues, the interventionists could switch to a phone call at their discretion. Each tele-visit followed the same structure, with the interventionist guiding caregivers in symptom management by helping them to (1) identify EoL signs/symptoms, (2) evaluate signs/symptoms, (3) learn about and then treat the signs/symptoms, and (4) summarize the management plan. In addition, the NP interventionist also entered a template note into the patient’s chart summarizing the tele-visit, which was accessible to the patient’s hospice team.
The second component of I-HoME included eight educational videos, each lasting 2–5 min, focusing on symptom management. Developed through caregiver and clinician input, as well as a literature review [23], these videos aimed to reinforce recommendations provided by the NP interventionist during tele-visits and/or offer supplementary education on EoL symptom management. Four videos are didactic-based, while the remaining four are story-based [24]. The purpose of using both didactic and story-based videos was to provide caregivers with different content styles that would help reinforce knowledge on topics of end-of-life symptom management. The videos covered (1) how to assess pain, (2) how to assess shortness of breath, (3) pain misconceptions, (4) end-of-life signs and symptoms, (5) the nature of hospice care, (6) how to identify/manage a pain crisis—looking for signs and symptoms, (7) appropriate use of morphine at the EoL, and (8) general education about the last stages of life. After each tele-visit, the interventionist would recommend specific videos if deemed appropriate. The study team then sent a weblink to the participant’s phone via text message. Participants viewed these videos independently and at their own convenience.
Participants in the I-HoME group received the intervention along with usual hospice care.
2.5 |. Control Group
Caregiver participants randomized to the control group received usual hospice care. This included visits from a hospice nurse, social worker, and chaplain, which varied based on a patient’s needs and conditions. Figure 1 shows the overall study design.
FIGURE 1 |.
Study design. ESAS, Edmonton Symptom Assessment Scale; ZBI-12, 12-item Zarit Burden Interview; GAD-7, General Anxiety Disorder-7; PHQ-9, Patient Health Questionnaire-9; I-HoME, Improving Home hospice Management of End-of-life Symptoms; NP, nurse practitioner.
2.6 |. Measures
Study measures were collected at seven time points (i.e., baseline, and weekly from week 1 through week 6) for both the intervention and control participants. We did not blind the research assistants to treatment assignment because part of their assessment was to collect I-HoME feasibility/acceptability data. Interventionists were not blinded to the treatment assignment. An exit interview was also conducted with the intervention group at the completion of the study. Measures were collected via phone by a trained research assistant, while patient demographic data were obtained from the EHR. After each tele-visit, the NP interventionists completed a survey noting whether the visit was by video or phone, visit duration (in minutes), and the symptoms discussed.
2.7 |. Feasibility and Acceptability Outcomes
We assessed the intervention’s feasibility and acceptability using multiple measures. From a recruitment standpoint, we analyzed recruitment and attrition rates, and the proportion of caregivers who missed tele-visit sessions. Caregiver feedback was collected weekly after each tele-visit, focusing on connection issues, satisfaction, and whether the visit addressed their symptom management concerns. We also asked caregivers about the number of educational videos that were watched and gathered insights on their perceived usefulness. Further, we contacted caregivers at the end of the study who completed at least one tele-visit to gather their reflections on the intervention’s benefits and the overall feasibility of its format and frequency. Lastly, we evaluated NP interventionists’ comfort with delivering the intervention and had them identify barriers to implementation.
2.8 |. Patient and Caregiver Outcome Measures
Patient symptom burden was the primary efficacy outcome, measured using the Edmonton Symptom Assessment Scale (ESAS), reported by the caregiver [25]. While patient-reported symptoms are ideal, approximately 50% of hospice patients have a terminal or secondary diagnosis of dementia, which raises issues about their ability to accurately report their symptom burden [21]. In addition, hospice patients who do not have dementia are often too frail/debilitated to consent/participate. Studies indicate that caregiver proxy reports can adequately reflect patient symptoms [26, 27]. A change of 3 points on the ESAS indicates a clinically significant change [25]. In addition to the ESAS, the 12-item Zarit Burden Interview (ZBI-12) [28], Patient Health Questionnaire-9 (PHQ-9) [29], and General Anxiety Disorder-7 (GAD-7) [30] scales were used to assess caregiver burden, depression, and anxiety.
2.9 |. Data Analysis
Descriptive statistics were used to analyze feasibility and acceptability outcomes. Differences between groups in baseline characteristics were assessed using one-way analysis of variance for continuous variables and chi-square tests for categorical variables. To assess the intervention’s potential efficacy, we used mixed models (MMs) to analyze specific outcomes (i.e., symptoms, caregiver burden, depression, anxiety) [31]. MMs are considered an appropriate method for intention-to-treat analysis [32]. We treated the participants’ initial status (intercept) and the change in their status (slope) as random effects. To better handle missingness in our data and detect patterns of change over time, we used all available time points to calculate the slope. SPSS 29 statistical software was used to conduct the data analyses.
3 |. Results
Figure 2 displays the study’s Consolidated Standards of Reporting Trials (CONSORT) diagram. Of the 212 eligible caregivers, 80 (38%) enrolled. We selected 80 as our sample size based on findings that 40 per group is suitable for establishing feasibility [33, 34]. Reasons for non-participation included refusal (n = 104, 79%), unreachable caregivers (n = 26, 19%), and other (n = 2, 2%). Five participants dropped out (attrition rate = 6%), with 3 withdrawing after consent while the remaining 2 could not be reached after they provided consent. Of the 80 enrolled participants, 51 (64%) completed the six-week study. The primary reason caregivers did not finish the 6-week study was patient death in hospice (n = 21), an expected outcome in this setting.
FIGURE 2 |.
Consolidated Standards of Reporting Trials (CONSORT) diagram.
3.1 |. Demographic Data
3.1.1 |. Caregivers
Table 1 shows that caregivers (N = 80) had a mean age of 60.3 (SD ± 12.1). A majority were women (n = 63, 79%), children of the patient (n = 54, 67%), and resided with the patient (n = 42, 52%). Of the 80 caregivers, 65% were non-Hispanic and racially diverse (29% Black, 42% White, 8% Asian, 21% other/mixed). The mean number of hours of care provided per day, as reported by participants, was 12.5 (SD ± 9.3) in the control and 12.7 (SD ± 8.9) in the intervention group. We asked respondents about their caregiving responsibilities: 81% provided personal care, 90% managed symptoms, and 86% administered medications. There were no significant between-group differences in baseline caregiver variables.
TABLE 1 |.
Caregiver demographic data.
| Total (N = 80) | Control arm (n = 40) | I-HoME arm (n = 40) | p | |
|---|---|---|---|---|
|
| ||||
| Age, mean (SD) | 60.3 (12.1) | 59.9 (12.9) | 60.6 (11.4) | |
| Age, median (range) | 59.5 (29–85) | 58.5 (35–83) | 60 (29–85) | |
| Gender, female, N (%) | 63 (79) | 31 (78) | 32 (80) | 0.79 |
| Ethnicity, N (%) | ||||
| Hispanic | 28 (35) | 18 (45) | 10 (25) | 0.06 |
| Non-Hispanic | 52 (65) | 22 (55) | 30 (75) | |
| Race, N (%) | ||||
| Asian | 6 (8) | 0 (0) | 6 (15) | 0.19 |
| Black | 23 (29) | 12 (30) | 11 (28) | |
| White | 34 (42) | 17 (42) | 17 (42) | |
| Other | 17 (21) | 11 (28) | 6 (15) | |
| Relationship to care recipient, N (%) | ||||
| Spouse/partner | 14 (17) | 6 (15) | 8 (20) | 0.69 |
| Child | 54 (68) | 28 (70) | 26 (65) | |
| Relative/friend | 12 (15) | 6 (15) | 6 (15) | |
| Living with care recipient, N (%) | ||||
| Yes | 42 (53) | 19 (48) | 23 (58) | 0.37 |
| No | 38 (47) | 21 (52) | 17 (42) | |
| Employment, N (%) | ||||
| Full-time | 26 (32) | 14 (35) | 12 (30) | 0.36 |
| Part-time | 19 (24) | 6 (15) | 13 (32) | |
| Not employed/retired | 35 (44) | 20 (50) | 15 (38) | |
| Income, N (%) | ||||
| $25,000 or less | 11 (14) | 8 (20) | 3 (8) | 0.45 |
| $25,001–$50,000 | 22 (28) | 11 (28) | 11 (28) | |
| $50,001–$75,000 | 18 (23) | 8 (20) | 10 (25) | |
| $75,001 or greater | 28 (35) | 13 (32) | 15 (39) | |
| Marital status, N (%) | ||||
| Single | 23 (29) | 13 (32) | 10 (25) | 0.78 |
| Married | 39 (49) | 19 (48) | 20 (50) | |
| Divorced | 12 (15) | 6 (15) | 6 (15) | |
| Widowed | 6 (7) | 2 (5) | 4 (10) | |
| Education, N (%) | ||||
| High school or lower | 10 (13) | 6 (15) | 4 (10) | 0.98 |
| Some college | 21 (26) | 10 (25) | 11 (28) | |
| College degree | 19 (24) | 10 (25) | 9 (22) | |
| Graduate school | 30 (37) | 14 (35) | 16 (40) | |
| Number of hours of caregiving provided by caregiver, mean (SD) | 12.6 (9.0) | 12.5 (9.3) | 12.7 (8.9) | |
| Caregiving duties, N (%) | ||||
| Personal care | 65 (81) | 34 (85) | 31 (78) | 0.39 |
| Manage symptoms | 72 (90) | 37 (92) | 35 (88) | 0.45 |
| Administering medications | 69 (86) | 36 (90) | 33 (82) | 0.33 |
| Communicate with providers | 74 (92) | 37 (92) | 37 (92) | 1.00 |
| Number of caregivers caring for patient, mean (SD) | 3.1 (1.7) | 2.9 (1.6) | 3.3 (1.7) | |
3.1.2 |. Patients
Patient demographic data (Table 2) showed that the average patient age was 86.9 (SD ± 8.1). A majority were female (n = 59, 74%), and 35% (n = 28) had a terminal diagnosis of cancer. There were no significant differences in patient sex, ethnicity, race, terminal diagnosis, or length of stay between the intervention and control groups. However, the control group had a higher proportion of dementia diagnoses (70% vs. 43%, p = 0.013) compared to the I-HoME group.
TABLE 2 |.
Patient demographic data.
| Total (N = 80) | Control arm (n = 40) | I-HoME arm (n = 40) | p | |
|---|---|---|---|---|
|
| ||||
| Age, mean (SD) | 86.9 (8.1) | 86.5 (8.0) | 87.3 (8.2) | |
| Gender, female, N (%) | 59 (74) | 32 (80) | 27 (68) | 0.20 |
| Ethnicity, N (%) | ||||
| Hispanic | 27 (34) | 18 (45) | 9 (23) | 0.73 |
| Non-Hispanic | 52 (65) | 22 (55) | 30 (75) | |
| Not reported | 1 (1) | 0 (0) | 1 (2) | |
| Race, N (%) | ||||
| Asian | 4 (5) | 0 (0) | 4 (10) | 0.11 |
| Black | 21 (26) | 10 (25) | 11 (28) | |
| White | 27 (34) | 12 (30) | 15 (37) | |
| Other/not reported | 28 (35) | 18 (45) | 10 (25) | |
| Hospice diagnosis, N (%) | ||||
| Cancer | 28 (35) | 13 (33) | 15 (37) | 0.56 |
| Dementia | 22 (28) | 11 (28) | 11 (28) | |
| Cardiac disease | 7 (9) | 5 (12) | 2 (5) | |
| Stroke/neurologic | 17 (21) | 8 (20) | 9 (23) | |
| Lung disease | 3 (4) | 2 (5) | 1 (2) | |
| Renal disease | 1 (1) | 1 (2) | 0 (0) | |
| Other | 2 (2) | 0 (0) | 2 (5) | |
| Length of stay in hospice (days), mean (SD) | 198 (177) | 233 (201) | 163 (144) | |
3.2 |. Feasibility Outcomes
In the I-HoME group, 121 out of 145 potential tele-visits (83%) were completed (accounting for patient attrition), with 83 via video and 38 via phone. The number of potential tele-visits was calculated from caregiver dropout and patient attrition (e.g., death). As would be expected in hospice care, some patients died before caregivers could finish the 6-week study. In such cases, we did not continue with the intervention after a patient’s death, as it would not be appropriate in the clinical context. Thirteen reported issues with accessing, connecting, or maintaining the video visit at some point during the intervention, citing connection problems (n = 8), platform (Doxy.me) issues (n = 5), hardware issues (n = 2), and/or a preference for phone visits (n = 2).
The average duration of the tele-visits was 23 min. The most common symptoms addressed were delirium/altered mental status (79%), pain (79%), and shortness of breath (69%). In the 121 tele-visits conducted, the hospice NP interventionists reported that they reviewed medications (86%), provided education on symptom management (91%), and recommended/adjusted medications (13%).
A total of 60 videos were recommended by the interventionist, with the most common being How to Assess Pain (n = 15), End of Life Signs and Symptoms (n = 13), and How to Assess Shortness of Breath (n = 12). Caregivers reported watching 23 (38%) of the videos. Reasons for not watching included lack of time (n = 18) and forgetting (n = 1).
All NP interventionists (n = 3) reported feeling comfortable delivering the I-HoME intervention by the third tele-visit session. All NPs (n = 3) reported that the primary barrier to implementing I-HoME was coordinating their clinical schedules with the caregivers’ availability.
3.3 |. Acceptability Outcomes
Of the 121 tele-visits conducted, we completed 113 post tele-visit surveys, with 8 surveys not completed due to non-response from the caregiver. Out of the 113 responses, 112 (99%) reported that the tele-visit session adequately or somewhat adequately addressed their symptom management concerns. Additionally, 109 (96%) reported being either satisfied or very satisfied with the tele-visit session. Of the 23 watched videos, 17 (74%) found them helpful.
3.4 |. Post-Study Feasibility and Acceptability Measures
Out of the 31 caregiver participants who received at least one tele-visit (5 caregivers dropped out, and 4 patients died before the caregiver received a tele-visit) and were contacted post-study completion, we obtained responses from 24. Among those responding, 23 (96%) found the I-HoME intervention helpful, and 21 (88%) had no scheduling issues. Additionally, 22 (92%) felt that the frequency of the tele-visits was adequate, and 23 (96%) found the delivery of the intervention via tele-visit feasible. When asked about their preference for visit format, 18 (75%) preferred video visits, 3 (12%) preferred phone, and 3 (12%) were indifferent. Regarding the possibility of conducting two tele-visits per week, 15 (63%) felt it would be manageable, although some cited time constraints.
3.5 |. Patient and Caregiver Outcomes
In the I-HoME group, 83% of caregivers agreed or strongly agreed that the intervention enhanced their knowledge and understanding of EoL symptoms. Additionally, 83% agreed or strongly agreed that it prepared them to better manage the care recipient’s symptoms. Finally, 88% of caregivers agreed or strongly agreed that I-HoME increased their confidence in managing care recipients’ symptoms.
The average reduction in total ESAS score for the intervention group, at the 6-week timepoint (n = 17), was 3.8 points compared to 2.9 points in the control group (n = 34) (Table 3). Of note, a proportion of caregivers (n = 7) who completed surveys at the 6-week mark did not receive all six tele-visits due to various reasons (e.g., caregiver availability). Among the subset of caregivers who received all six tele-visits (n = 10), the reduction of ESAS score was 6.6 points. A reduction of 3 points on the ESAS indicates a clinically meaningful difference [25]. However, when conducting mixed-model analyses looking at outcomes between the two arms, we found no statistically significant differences in changes in ESAS scores (p = 0.22), caregiver burden (p = 0.35), depression (p = 0.72), or anxiety (p = 0.28) between the two groups.
TABLE 3 |.
Symptom burden scores between control and I-HoME arm among those who completed the 6-week study.
| Total ESAS Score (range: 0–90) mean (SD) | Baseline | Week 1 | Week 2 | Week 3 | Week 4 | Week 5 | Week 6 |
|---|---|---|---|---|---|---|---|
|
| |||||||
| Control | 39.0 (13.9) | 38.0 (16.7) | 34.6 (15.1) | 34.7 (15.9) | 34.9 (16.5) | 37.0 (16.1) | 36.1 (15.6) |
| I-HoME | 43.7 (17.6) | 40.7 (19.6) | 42.5 (16.8) | 42.3 (16.1) | 38.8 (19.2) | 44.1 (17.8) | 39.9 (20.1) |
Note: Data reflect study timepoints, not the number of tele-visits conducted.
4 |. Discussion
This study highlights some of the successes as well as the challenges we encountered designing and implementing a caregiver-directed intervention in the home hospice setting. Although we faced challenges (e.g., recruitment rate, patient attrition, feasibility of delivering educational videos), overall, caregivers found the I-HoME intervention to be feasible and acceptable, and reported increased knowledge and confidence in managing symptoms Table 3.
We found moderate feasibility for the I-HoME intervention, achieving a recruitment rate of 38%, which is consistent with other studies involving caregivers in home hospice settings [35–39]. Over the course of our study, we streamlined recruitment/study processes (e.g., providing a remote consent option and conducting phone rather than in-person training), which helped increase our recruitment rate. We explored whether recruiting caregivers from the start of hospice enrollment to 4 weeks afterward would improve our recruitment rate but found no specific time point that was more effective. A common barrier to participation was the busy schedules of caregivers [40], with our sample reporting an average of 12.5 h of caregiving per day. Further, 64% of our sample reported working either full-time or part-time. Therefore, hospice organizations and researchers must consider these time constraints when designing interventions for caregivers in the home hospice setting. It is essential to design caregiver-directed interventions and study protocols that minimize participant burden.
However, once caregivers were recruited, most adhered to their weekly tele-visits and completed the weekly assessments, as reflected in our caregiver attrition rate of just 6%. Tele-visits averaged 23 min, and weekly surveys took 10–15 min to complete, which caregivers found manageable. From an organizational perspective, hospice administrators/staff reported no disruption or burden to clinical workflow. Further, the NPs interventionists were part of the hospice organization, which allowed for a smooth integration into the clinical workflow. We found no differences in the frequency of nursing visits between the two groups during the study period. From a hospice organizational perspective, if these types of interventions prove to be effective, future research will be necessary to assess the practicality of allocating resources to implement such interventions within the hospice payment system. From a caregiver resource perspective, while we provided hardware and internet connection if needed, most caregivers already had these resources. However, this may differ for rural or low-income caregivers with limited access to reliable internet or cellular data. Future studies should investigate feasibility in these settings.
We found that the feasibility of delivering educational videos was limited. During our user-centered design process, we received feedback from hospice staff expressing concerns about sharing these videos without input from hospice staff, as some topics may not be relevant to caregivers and could cause distress. Consequently, we decided to share the videos at the discretion of the NP interventionist. However, this approach likely limited the videos’ usage. In a separate study we conducted concurrently, where we piloted educational videos for Black hospice caregivers, we found a trend toward enhanced caregiver preparedness and knowledge. We also found that most preferred access to the videos at the start of hospice enrollment [41]. Reflecting on the results of these two studies, along with other research in the literature [42], it seems that showing the educational videos at the start of the home hospice experience, in a more standardized way, could give caregivers a foundational understanding of hospice symptom management and might have led to better caregiver outcomes. Further, delivering this intervention element in this manner would allow the NP interventionists to focus on more detailed, patient-specific symptom concerns during the weekly tele-visits.
Patient attrition, primarily due to hospice deaths, impacted the number of tele-visits delivered. It is crucial to account for this attrition when designing future efficacy and effectiveness trials, especially in caregiver intervention studies within the home hospice setting. In our exit interviews, we asked caregivers if it would be feasible to condense the intervention to two tele-visits per week instead of one per week. We found that 63% of respondents were open to this approach, suggesting that conducting two tele-visits per week could increase the likelihood of caregivers completing all six tele-visits before experiencing attrition due to patient death. However, this change could also introduce new feasibility challenges. Future intervention efforts must balance the frequency of administering an intervention with caregivers’ ability to participate, considering their time constraints.
Overall, we found the I-HOME intervention to be highly acceptable. Anecdotally, caregivers shared that they gained personal benefits from the program and appreciated the additional support it provided. Through our exit interviews, many participants shared that the intervention enhanced their preparedness and self-efficacy for managing EoL symptoms. We believe this speaks to the fact that caregivers are looking for more guidance, education, and support during this phase of caregiving, which many are thrust into without formal training or education.
This study has several limitations. First, we conducted the study at one non-profit hospice organization in an urban setting. Different hospice organizations may have different workflows that may impact implementation. Therefore, future studies looking at piloting this in rural, small, and/or for-profit hospices should be explored. Second, while we did not formally assess fidelity, we conducted initial training sessions with the interventionists, provided a manual of operations, and had weekly meetings with the NP interventionists to address any study-related concerns. Lastly, the average length of stay of the patients in our study was longer when compared to the national average, and therefore, these results may not reflect the experiences of caregivers of patients with shorter hospice stays.
In summary, we found the I-HoME intervention feasible and acceptable for home hospice caregivers. Future steps include conducting a randomized controlled trial powered to evaluate the efficacy of the intervention. Given the sparse number of evidence-based interventions to support caregivers in this phase of caregiving, additional studies are needed to advance the field of caregiving at the EoL.
Summary.
- Key points
- Caregivers provide a significant amount of care (~12 h a day) to patients receiving home hospice care.
- We found the I-HoME intervention to be feasible and acceptable for caregivers of home hospice patients.
- This study highlights some of the successes and challenges encountered when designing and implementing a caregiver-directed intervention in the home hospice setting.
- Why does this paper matter?
- Caregivers (often referred to as family, informal, or unpaid caregivers) provide substantial care to home hospice patients; however, many caregivers undertake this role without formal training or education in symptom management.
- It is crucial to develop interventions that support caregivers, which can translate to better care for patients at the end of life and improve caregiver well-being.
Funding:
This study was supported by the National Institute on Aging (1K76AG059997–01A1, K24AG053462, and P30AG022845) and the National Center For Advancing Translational Sciences of the National Institutes of Health (UL1TR002384). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Conflicts of Interest
The authors declare no conflicts of interest.
This paper was presented at the American Geriatrics Society Annual Meeting in May 2025.
References
- 1.“Hospice Facts and Figures,” 2024. Edition. National Hospice and Palliative Care Organization, (2024), 1–21, https://www.nhpco.org/hospice-care-overview/hospice-facts-figures/. [Google Scholar]
- 2.Kutner JS, Kassner CT, and Nowels DE, “Symptom Burden at the End of Life: Hospice Providers’ Perceptions,” Journal of Pain and Symptom Management 21, no. 6 (2001): 473–480, 10.1016/S0885-3924(01)00281-0. [DOI] [PubMed] [Google Scholar]
- 3.Zeppetella G, O’Doherty CA, and Collins S, “Prevalence and Characteristics of Breakthrough Pain in Cancer Patients Admitted to a Hospice,” Journal of Pain and Symptom Management 20, no. 2 (2000): 87–92, 10.1016/S0885-3924(00)00161-5. Erratum in Journal of Pain and Symptom Management 21, no. 3 (2001): 265. [DOI] [PubMed] [Google Scholar]
- 4.Brandt HE, Ooms ME, Deliens L, van der Wal G, and Ribbe MW, “The Last Two Days of Life of Nursing Home Patients—A Nationwide Study on Causes of Death and Burdensome Symptoms in The Netherlands,” Palliative Medicine 20, no. 5 (2006): 533–540, 10.1191/0269216306pm1164oa. [DOI] [PubMed] [Google Scholar]
- 5.Potter J, Hami F, Bryan T, and Quigley C, “Symptoms in 400 Patients Referred to Palliative Care Services: Prevalence and Patterns,” Palliative Medicine 17, no. 4 (2003): 310–314, 10.1191/0269216303pm760oa. [DOI] [PubMed] [Google Scholar]
- 6.Kutner JS, Bryant LL, Beaty BL, and Fairclough DL, “Time Course and Characteristics of Symptom Distress and Quality of Life at the End of Life,” Journal of Pain and Symptom Management 34, no. 3 (2007): 227–236, 10.1016/j.jpainsymman.2006.11.016. [DOI] [PubMed] [Google Scholar]
- 7.Conill C, Verger E, Henríquez I, et al. , “Symptom Prevalence in the Last Week of Life,” Journal of Pain and Symptom Management 14, no. 6 (1997): 328–331, 10.1016/S0885-3924(97)00263-7. [DOI] [PubMed] [Google Scholar]
- 8.Klinkenberg M, Willems DL, van der Wal G, and Deeg DJH, “Symptom Burden in the Last Week of Life,” Journal of Pain and Symptom Management 27, no. 1 (2004): 5–13, 10.1016/j.jpainsymman.2003.05.008. [DOI] [PubMed] [Google Scholar]
- 9.Phongtankuel V, Scherban BA, Reid MC, et al. , “Why Do Home Hospice Patients Return to the Hospital? A Study of Hospice Provider Perspectives,” Journal of Palliative Medicine 19, no. 1 (2016): 51–56, 10.1089/jpm.2015.0178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Phongtankuel V, Paustian S, Reid MC, et al. , “Events Leading to Hospital-Related Disenrollment of Home Hospice Patients: A Study of Primary Caregivers’ Perspectives,” Journal of Palliative Medicine 20, no. 3 (2017): 260–265, 10.1089/jpm.2015.0550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Phongtankuel V, Teresi JA, Eimicke JP, et al. , “Identifying the Prevalence and Correlates of Caregiver-Reported Symptoms in Home Hospice Patients at the End of Life,” Journal of Palliative Medicine 23, no. 5 (2020): 635–640, 10.1089/jpm.2019.0324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Phongtankuel V, Reid MC, Czaja SJ, et al. , “Caregiver-Reported Quality Measures and Their Correlates in Home Hospice Care,” Palliative Medicine Reports 1, no. 1 (2020): 111–118, 10.1089/pmr.2020.0055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ornstein KA, Kelley AS, Bollens-Lund E, and Wolff JL, “A National Profile of End-Of-Life Caregiving in the United States,” Health Affairs 36, no. 7 (2017): 1184–1192, 10.1377/hlthaff.2017.0134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Han CJ, Chi NC, Han S, et al. , “Communicating Caregivers’ Challenges With Cancer Pain Management: An Analysis of Home Hospice Visits,” Journal of Pain and Symptom Management 55, no. 5 (2018): 1296–1303, 10.1016/j.jpainsymman.2018.01.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Morita T, Miyashita M, Shibagaki M, et al. , “Knowledge and Beliefs About End-Of-Life Care and the Effects of Specialized Palliative Care: A Population-Based Survey in Japan,” Journal of Pain and Symptom Management 31, no. 4 (2006): 306–316, 10.1016/j.jpainsymman.2005.09.004. [DOI] [PubMed] [Google Scholar]
- 16.Gardiner C, Gott M, Ingleton C, Hughes P, Winslow M, and Bennett MI, “Attitudes of Health Care Professionals to Opioid Prescribing in End-Of-Life Care: A Qualitative Focus Group Study,” Journal of Pain and Symptom Management 44, no. 2 (2012): 206–214, 10.1016/j.jpainsymman.2011.09.008. [DOI] [PubMed] [Google Scholar]
- 17.Oliver DP, Wittenberg-Lyles E, Washington K, et al. , “Hospice Caregivers’ Experiences With Pain Management: I’m Not a Doctor, and I Don’t Know if I Helped Her Go Faster or Slower,” Journal of Pain and Symptom Management 46, no. 6 (2013): 846–858, 10.1016/j.jpainsymman.2013.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Alshakhs S, Park T, McDarby M, et al. , “Interventions for Family Caregivers of Patients Receiving Palliative/Hospice Care at Home: A Scoping Review,” Journal of Palliative Medicine 27, no. 1 (2024): 112–127, 10.1089/jpm.2023.0160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rounsaville BJ, Carroll KM, and Onken LS, “A Stage Model of Behavioral Therapies Research: Getting Started and Moving on From Stage I,” Clinical Psychology (New York) 8, no. 2 (2001): 133–142, 10.1093/clipsy.8.2.133. [DOI] [Google Scholar]
- 20.World Medical Association Declaration of Helsinki, “Recommendations Guiding Physicians in Biomedical Research Involving Human Subjects,” Journal of the American Medical Association 277, no. 11 (1997): 925, 10.1001/jama.1997.03540350075038. [DOI] [PubMed] [Google Scholar]
- 21.Harrison KL, Cenzer I, Ankuda CK, Hunt LJ, and Aldridge MD, “Hospice Improves Care Quality for Older Adults With Dementia in Their Last Month of Life,” Health Affairs 41, no. 6 (2022): 821–830, 10.1377/hlthaff.2021.01985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Doxy.me, “Telemedicine Solutions,” (2018), www.doxy.me. [Google Scholar]
- 23.Cruz-Oliver DM, Talamantes M, and Sanchez-Reilly S, “What Evidence Is Available on End-Of-Life (EOL) Care and Latino Elders? A Literature Review,” American Journal of Hospice and Palliative Care 31, no. 1 (2014): 87–97, 10.1177/1049909113480841. [DOI] [PubMed] [Google Scholar]
- 24.Cruz-Oliver DM, Abshire M, Budhathoki C, Oliver DP, Volandes A, and Smith TJ, “Reflections of Hospice Staff Members About Educating Hospice Family Caregivers Through Telenovela,” American Journal of Hospice and Palliative Care 38, no. 2 (2021): 161–168, 10.1177/1049909120936169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hui D and Bruera E, “The Edmonton Symptom Assessment System 25 Years Later: Past, Present, and Future Developments,” Journal of Pain and Symptom Management 53, no. 3 (2017): 630–643, 10.1016/j.jpainsymman.2016.10.370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Kroenke K, Stump TE, and Monahan PO, “Agreement Between Older Adult Patient and Caregiver Proxy Symptom Reports,” Journal of Patient-Reported Outcomes 6, no. 1 (2022): 50, 10.1186/s41687-022-00457-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kutner JS, Bryant LL, Beaty BL, and Fairclough DL, “Symptom Distress and Quality-Of-Life Assessment at the End of Life: The Role of Proxy Response,” Journal of Pain and Symptom Management 32, no. 4 (2006): 300–310, 10.1016/j.jpainsymman.2006.05.009. [DOI] [PubMed] [Google Scholar]
- 28.Bédard M, Molloy DW, Squire L, Dubois S, Lever JA, and O’Donnell M, “The Zarit Burden Interview: A New Short Version and Screening Version,” Gerontologist 41, no. 5 (2001): 652–657, 10.1093/geront/41.5.652. [DOI] [PubMed] [Google Scholar]
- 29.Kroenke K, Spitzer RL, and Williams JBW, “The PHQ-9: Validity of a Brief Depression Severity Measure,” Journal of General Internal Medicine 16, no. 9 (2001): 606–613, 10.1046/j.1525-1497.2001.016009606.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Spitzer RL, Kroenke K, Williams JBW, and Löwe B, “A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7,” Archives of Internal Medicine 166, no. 10 (2006): 1092–1097, 10.1001/archinte.166.10.1092. [DOI] [PubMed] [Google Scholar]
- 31.Raudenbush SW and Bryk AS, Hierarchical Linear Models: Applications and Data Analysis Methods, 2nd ed. (Sage Publications, 2002). [Google Scholar]
- 32.Chakraborty H and Gu H, A Mixed Model Approach for Intent-To-Treat Analysis in Longitudinal Clinical Trials With Missing Values (Research Triangle Park (NC): RTI Press, 2009), http://www.ncbi.nlm.nih.gov/books/NBK538904/. [PubMed] [Google Scholar]
- 33.Hertzog MA, “Considerations in Determining Sample Size for Pilot Studies,” Research in Nursing and Health 31, no. 2 (2008): 180–191, 10.1002/nur.20247. [DOI] [PubMed] [Google Scholar]
- 34.Teresi JA, Yu X, Stewart AL, and Hays RD, “Guidelines for Designing and Evaluating Feasibility Pilot Studies,” Medical Care 60, no. 1 (2022): 95–103, 10.1097/MLR.0000000000001664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Demiris G, Oliver DP, Washington K, and Pike K, “A Problem-Solving Intervention for Hospice Family Caregivers: A Randomized Clinical Trial,” Journal of the American Geriatrics Society 67, no. 7 (2019): 1345–1352, 10.1111/jgs.15894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Parker Oliver D, Demiris G, Washington K, Kruse RL, and Petroski G, “Hospice Family Caregiver Involvement in Care Plan Meetings: A Mixed-Methods Randomized Controlled Trial,” American Journal of Hospice and Palliative Care 34, no. 9 (2017): 849–859, 10.1177/1049909116661816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mayahara M, Wilbur J, Fogg L, Breitenstein SM, and Miller AM, “Feasibility of e-Pain Reporter: A Digital Pain Management Tool for Informal Caregivers in Home Hospice,” Journal of Hospice and Palliative Nursing 21, no. 3 (2019): 193–199, 10.1097/NJH.0000000000000548. [DOI] [PubMed] [Google Scholar]
- 38.Grande GE, Austin L, Ewing G, O’Leary N, and Roberts C, “Assessing the Impact of a Carer Support Needs Assessment Tool (CSNAT) Intervention in Palliative Home Care: A Stepped Wedge Cluster Trial,” BMJ Supportive and Palliative Care 7 (2015): 326–334, 10.1136/bmjspcare-2014-000829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Latter S, Hopkinson JB, Lowson E, et al. , “Supporting Carers to Manage Pain Medication in Cancer Patients at the End of Life: A Feasibility Trial,” Palliative Medicine 32, no. 1 (2018): 246–256, 10.1177/0269216317715197. [DOI] [PubMed] [Google Scholar]
- 40.Cruz-Oliver DM, Milner GE, Abshire Saylor M, et al. , “Recruitment, Retention and Intervention Delivery Experiences With Hospice Family Caregivers,” American Journal of Hospice and Palliative Care 40, no. 4 (2023): 409–415, 10.1177/10499091221110855. [DOI] [PubMed] [Google Scholar]
- 41.Park T, Sloan DH, Cruz-Oliver DM, et al. , “Black Caregivers’ Symptom Management, Cultural, and Religious Experiences With Home Hospice Care,” Journal of Pain and Symptom Management 66, no. 2 (2023): 116–122.e1, 10.1016/j.jpainsymman.2023.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Cruz-Oliver DM, Milner GE, Mensh K, et al. , “Promising Impact of Telenovela Intervention for Caregivers of Hospice Patients: A Pilot Study,” American Journal of Hospice and Palliative Care 41, no. 12 (2024): 1400–1407, 10.1177/10499091241228835. [DOI] [PubMed] [Google Scholar]