“She Would Be Flailing Around Distressed”: The Critical Role of Home-Based Palliative Care for Patients with Advanced Cancer

Introduction

With oncologic treatment options advancing, many patients with cancer are moved to pursue cancer-directed therapies late into their disease.^1,2

Identifying and attending to all of patients' and caregivers' struggles can be difficult during brief oncology clinic visits. Many of these patients may benefit from specialized palliative care (PC).^3–5 However, they typically are unable to access hospice given ongoing disease-directed therapy and getting to a PC clinic is challenging for the homebound.

Home-based palliative care (HBPC) may fill a service gap for these patients.^6,7 Community Bridges (CBs) was a pilot HBPC program for patients with metastatic cancer and limited functional status or complex care needs. This report describes CBs and future opportunities.

Methods

CBs provided specialty HBPC to functionally limited patients with cancer and complex care needs seen at a large urban academic medical center (AMC). The CB team included 0.1 MD full-time equivalents (FTE) and 0.1 NP FTE from the AMC and 0.1 RN from a partnering social services agency. CB providers communicated through weekly team meetings, phone, and secure e-mails. CB's RN had read-only access to the AMC electronic medical record (EMR).

The program was advertised through bimonthly e-mails to potential referring providers and the CBs' research coordinator announcements at Phase I trial study meetings. Oncologist and PC provider referrals were sent by e-mails to the CB research coordinator. Nonhospice patients ≥18 years old with metastatic cancer who were homebound and met ≥1 of the following criteria were eligible: (1) progression on ≥3 lines of chemotherapy, (2) poor functional status (ECOG ≥3, capable of only limited self-care; confined to bed or chair more than 50% of waking hours), (3) missed a PC clinic appointment because was “too sick” or hospitalized, or (4) were a Phase I clinical trial participant.

CB patients received 2 months of home visits (≥3, 45–90 minutes home visits) and phone support during regular business hours. Overnight and weekends, patients contacted their oncologist. Visit frequency and assignment of CB providers were determined based on individual patient need and provider availability. CB providers comanaged patients with their oncology and primary care teams and communicated with them by EMR, e-mail, and less frequently, phone.

Data were collected at baseline and two, three, and eight weeks after the initial home visit using the following: (1) the ESAS^8,9; (2) the Zarit 12-Question Caregiver Burden Survey ¹⁰ ; (3) a 12-question patient satisfaction survey; and (4) qualitative caregiver phone interviews. Patients and caregivers completed telephone exit interviews with a trained research assistant. A CB team focus group explored key features and lessons learned using the Predisposing, Enabling, and Reinforcing factors framework. ¹¹

Descriptive statistics were calculated. Analyses used SAS version 9.4 (SAS Institute, Inc., Cary, NC). The AMC's IRB approved this study.

Results

During the 6-month program, 47 patients were referred. Twenty-three patients were excluded (12 lived outside of CBs' 18 × 10 mile territory, 5 were not seen regularly at the AMC, 4 did not speak English, and 2 had ECOG <3). Of the 24 eligible patients, 17 enrolled (5 transitioned to hospice before CB enrollment and 2 refused enrollment). Of enrolled patients, 59% were referred by PC providers and 29% by oncologists; 52% were referred at hospital discharge; and 59% were receiving cancer-directed therapy at enrollment.

Table 1 shows baseline characteristics. Mean age was 66, 71% were female, and 65% were white. Seventy-one percent of patients had thoracic, gynecologic, or gastrointestinal cancers.

Baseline symptom burden was high (Table 2). At baseline, patients reported moderate to severe pain (29%), fatigue (64%), and depression (39%); 29% reported poor quality of life. Forty-three percent of patients required help with ≥1 activity of daily living (ADL). Caregiver burden was high, with an average baseline Zarit score of 20.6 ± 12 (n = 5) and final score of 16.4 ± 6.8 (n = 5).

Table 3 lists the CB team's activities. All patients underwent a comprehensive assessment using the validated measures above. Advance care planning (including conversations on advance directives, POLST forms, and/or preferred place of death) (88.2%) and hospice (76.5%) were discussed with most patients. Advance directives or POLSTs were completed with 17% of patients.

Twenty-four percent of patients died within the first month of program enrollment; 35% and 59% died within two and six months of program enrollment, respectively. Six patients (35%) were referred by CB and enrolled in hospice during the two-month data collection period. Average hospice length of stay was 9.5 days.

Patient and caregiver program satisfaction was high. Both groups felt CB was very helpful in providing care that was not otherwise available to them (caregivers 75% [n = 6] and patients 60% [n = 3]). Patients reported that CB helped them navigate the medical system and reduce stress. All caregivers (n = 8) would recommend CB to others and reported CBs helped their loved ones regain control of their care. One caregiver noted that without CB she would have been “flailing around distressed…[I] knew I needed help but didn't know how to get it.”

Three themes, identified during the CB team focus group, highlight the role of CB with this patient population.

Discussion

CBs was an HBPC program to address the needs of patients with advanced cancer and declining functional status or complex care needs. CB patients had high symptom burden, mortality, and caregiver burden.

Our population was similar to studies describing the high symptom burden of advanced cancer patients.^1,12,13 While these patients may derive benefit from HBPC outside the treatment limitations of hospice, only one-quarter of National Comprehensive Cancer Network member institutions offer any HBPC. ¹⁴

CBs filled a service gap for this critically ill patient population in need of intensive PC services at home, but not on hospice. Most patients and caregivers agreed that CBs provided care that was not otherwise available to them. Through HBPC, CB providers were able to gain an accurate understanding of their patients' home lives, quickly address crises, and serve as “credible witnesses” to oncologists.

Our team's diverse clinical and interpersonal skillsets allowed us to delegate tasks to manage time-sensitive patient situations and crises as they arose. Association with our AMC's established PC programs helped secure buy in when marketing CB to oncologists and support a comanagement model.

CB included fewer patients than planned. Twelve referred patients lived outside of the program's geographic reach. Video TeleMedicine is a promising technology to address this issue. Despite established relationships with oncologists and aggressive marketing, oncologist referrals were lower than expected. Opportunities exist for further education of referring providers on the unique services offered by PC and differences between PC and hospice. Our small patient population, alongside high mortality rate, limited our ability to accurately report symptom scores over time. Hidden costs of the program included travel time, the duration of home visits, and unanticipated administrative demands.

HBPC programs may bridge the gap for patients desiring ongoing cancer treatment, but in need end-of-life services. To understand whether a model like CBs, which uniquely targets patients late in their disease, is sustainable and advantageous, an evaluation of healthcare utilization and cost is needed within a larger population. Additional next steps include examining whether the experience of HBPC varies across patients with different types of cancer and exploring how the role of CB providers as “credible witnesses” influences medical decision making and patient outcomes.