Abstract
Background:
Developing better acuity models to predict hospital re-admissions is vital to providing high qualitive outpatient palliative care services.
Objective:
Determine if an acuity tool could accurately predict those patients with the highest risk for hospital admission within 30 days following a palliative medicine home visit.
Design:
A retrospective observational study evaluating an outpatient practitioner-administered acuity model to determine how different acuity scores corresponded with specific follow-up intervals.
Setting/Subjects:
The study was conducted in patients’ residences, analyzing home-based outpatient palliative care visits provided by Wellstar Health System clinicians across 4 Georgia counties.
Results:
Two hundred and ninety-one patients were included in our study. 78% of patients did not experience a hospital admission within 30 days following their most recent palliative medicine home visit, while 22% (n = 64) did. The average overall acuity score was 10.93. The average acuity score for the individuals not admitted was 9.66, while the score for those admitted was 15.44.
Conclusion:
The acuity score was a significant predictor of the probability of hospitalization (P < .0001), and the odds of a hospitalization increased by 7.0% for each additional unit in the acuity score.
Keywords
Introduction
Palliative medicine (PM) focuses on symptom management and guidance with complex medical decision-making that can improve metastatic cancer survival rates, reduce cost of care, and improve patients’ quality of life perception.1-4 Previously, it was only in the hospital setting, but the emergence of home-based palliative programs has shown upstream benefits, such as reduction in hospital admissions by managing complex symptoms at home before they worsen. 5 However, there is limited data on the exact cost savings of home PM programs. If appropriately utilized, home palliative programs could decrease the cost of hospital utilization and improve patient and family experiences.6,7
The effective management of a home-based PM program with limited resources poses a significant challenge. 7 We developed our predictive acuity tool to meet the needs of medically fragile patients at the right time and place before a crisis. We conducted a thorough investigation into available tools that could assist us in determining the proper cadence for visits, while objectively considering the complexity of our patients. Due to the limited availability of tools specifically designed for the home-based palliative care population, we adopted 2 existing frameworks to inform our approach: ProHealth’s Disenrollment Protocol and the stepwise acuity tool developed by Sean Reed.8-10 These resources provided a foundational structure for the development of a scoring grid that integrates both symptom burden and patient-specific clinical characteristics to identify individuals with the highest levels of need. The scoring grid was carefully aligned with the standards set forth by the National Consensus Project (NCP), ensuring a strong emphasis on patient-centered care, comprehensive symptom management, and holistic support for patients and their families. We then assigned a numerical value to each criterion, prioritizing symptoms that we determined were higher predictors of hospital admissions.
Methods
Study Design
The outpatient acuity score emphasized the importance of evaluating a patient’s tolerance of symptoms in relation to their severity. We developed our predictive acuity tool to meet the needs of medically fragile patients at the right time and place before a crisis. We conducted a thorough investigation into available tools that could assist us in determining the proper cadence for visits, while objectively considering the complexity of our patients. The questionnaire incorporates assessments of daily living activities and goals of care discussions (Figure 1). We then assigned a numerical value to each criterion, prioritizing symptoms that we determined were higher predictors of hospital admissions.
Wellstar home palliative medicine acuity tool.
Ethical Approval
This study was reviewed by the Wellstar Health System Institutional Review Board and determined to be exempt from full review under applicable federal regulations.
Study Setting
The study was conducted within the Wellstar Health System, the largest and most integrated healthcare system in Georgia. Wellstar is a not-for-profit organization headquartered in Marietta, GA. Our study focused on outpatient palliative medical visits delivered by Wellstar clinicians in patients’ places of residence, rather than in clinic or hospital settings. These home-based visits occurred across 4 counties in Georgia: Cobb, Douglas, Cherokee, and Paulding. These counties are part of Wellstar’s core service area and represent diverse suburban and semi-rural communities within the Atlanta metropolitan region.
Our home-based palliative medicine program included a multidisciplinary team consisting of a hospice and palliative medicine board-certified physician serving as medical director, advanced practice providers (APPs), registered nurses functioning as nurse navigators, and licensed social workers. Visits were conducted in patients private home, including assisted and independent living facilities. Medical follow-up visits included a combination of virtual and in-person encounters. The modality of each visit was determined by symptom severity, patient preference, and scheduling availability. Examples of services provided by the palliative care team included medication adjustments for pain and symptom control, advance care planning discussions to ensure treatment aligned with patient goals, and emotional support for both patients and caregivers. Social workers addressed emotional, practical, and psychosocial needs identified through the scoring tool. Their visits were conducted either telephonically or in person and could be independent or in collaboration with an APP. Nurse navigators provided follow-up phone calls to monitor symptom management, particularly in relation to medication adjustments, and offered education to patients and families regarding disease progression and care expectations. Although some patients did not receive a palliative team intervention within the 30-day window, all continued to receive care and support from their primary care providers and relevant specialists, ensuring consistent clinical oversight and access to necessary resources.
Eligibility Criteria
We conducted a retrospective observational study of patients who had home PM visits between June and December 2023. A total of 291 patients were included. Chart reviews were done via EPIC, and we determined which patients had hospital visits within 30 days of a home palliative visit, and if their code status was honored from previous visits.
Statistical Analysis
Descriptive statistics were developed for all data collected. Chi square was used to determine the significance of our acuity score, and an odds ratio was used to determine hospitalization odds. Statistical analyses were conducted through SAS version 9.4. 11
Results
Based on our analysis, 78% (n = 227) of patients did not experience a hospital admission within 30 days following their most recent palliative medicine home visit, while 22% (n = 64) did. Among females, 17.6% (n = 31.0) had a hospital admission within 30 days of the visit, compared to 28.5% (n = 33.0) among males (P = .037; Table 1).
Study Sample Characteristics for Patient’s Presenting for Outpatient Palliative Visit From June 1, 2023 to December 31, 2023.
The range of acuity scores from our questionnaire was from 1 to 45. In our patient population, the average overall acuity score was 10.93 (95% CI 9.91-11.95). The average acuity score for individuals not admitted was 9.66 (95% CI 8.58-1074), while the average acuity score for those admitted was 15.44 (95% CI 13.07-17.80). We found that the acuity score was a significant predictor of the probability of hospitalization (P < .0001) and the odds of a hospitalization increased by 7.0% (1.070, 95% CI 1.038-1.103) for each additional unit in the acuity score.
The acuity score was then broken down into 4 quartiles with scores 1 to 3 in the first quartile (or lowest 25%), 4 to 6 in the second quartile, 7 to 15 in the third quartile, and 16 to 45 in the fourth quartile. We found that the odds of admission increase as the score increases. The odds of admission for the highest quartile were 8.089 times the odds for the lowest quartile. The odds of the third quartile (those with a score between the 50% and 75%) were 4.875 times those in the lowest quartile, and the odds of the second quartile was 1.932 times the lowest quartile (Table 2).
Odds of Hospitalization by Acuity Score Quartile.
Acuity score is a composite measure used to triage outpatient palliative follow up visits based on individual patient factors. The score emphasizes tolerance of symptoms relative to severity and incorporates assessments of activities of daily living and goals-of-care discussions. Higher scores indicate greater complexity and resource needs. Quartiles were defined as follows: Quartile 1 (1-3): lowest acuity; minimal symptoms or complexity. Quartile 2 (4-6): mild acuity; some symptom burden or comorbidities. Quartile 3 (7-15): moderate acuity; multiple issues requiring management. Quartile 4 (16-45): highest acuity; severe symptoms and high complexity. Odds ratios are derived from logistic regression with Quartile 1 as the reference category.
Discussion
The focus of this study was to determine whether an outpatient palliative care acuity score could predict the likelihood of hospital admission within 30 days following a palliative medicine home visit in a home-based palliative care program. Our results show that our acuity score was statistically and clinically significant, and higher scores had higher odds of hospital admission within 30 days following a palliative medicine home visit. Multiple studies have shown that outpatient palliative care can decrease symptom burden, psychiatric issues (such as anxiety and depression), and hospitalization readmission rates.12,13 Previous studies have used questionnaires to address physical and functional distress in their patients, but a dearth of research on specific interventions for reducing or determining the risk for rehospitalization. 14 Our study was one of the first to find a predictive model between an acuity score that incorporates a distress criterion with hospital readmission risk. This is especially important to ensure that delivery of a limited resource, in this case home-based PM, is dictated by those with the highest needs and risk of readmission. This score is crucial in stratifying our patients by need and scheduling follow-up visits accordingly.
Although a dearth of research exists on other triage models, 1 pilot study did create a triage coding system based off the Edmonton Symptom Assessment System (ESAS). Like our acuity score, the study divided patients into 3 different groups based on symptoms and frequency of outpatient visits was determined based on severity of symptoms.15,16 The higher the priority group, the more home visits by a palliative team. They found that as the number of outpatient home visits increased, the symptoms addressed by ESAS decreased. There was also a statistically significant decrease in deaths in the hospital in the higher priority group, indicating that this scoring system, just like our acuity scale, was beneficial in decreasing rehospitalization. 17 Another scoring system called the Urgency of Need in Palliative Care (RUN-PC) has also been found to have good reliability in patients in triaging specific needs of palliative patients in the hospital.18,19 However, no other palliative scoring system was found to help triage patients and decrease rehospitalization.
It is important to determine the best interventions to reduce hospitalizations in patients seeking palliative or end-of-life care. One model, called the Cumulative Complexity Model (CuCoM), emphasizes how a patient centered model is an amalgamation of different clinical and social factors that are impacted by patient workload of demands and a patient’s ability to address these demands. 20 In other words, understanding personal limitations, such as access to care, is crucial to providing quality care and reducing early hospital readmissions. One study found that good discharge planning was effective at decreasing readmissions but also supported the model that the most effective intervention was complex and supportive patient care.21,22 The transitional Care Palliative-End Stage Heart Failure program (TCP-ESHF) found that improving continuum of care between inpatient and outpatient palliative services was done by communication between the palliative team and other specialties involved in the patient’s treatment, thus helping to reduce hospital readmissions and improve patients’ quality of life. 22
Limitations
This study was conducted within a single health system, which limits external validity and generalizability to other organizational contexts. Replication may be challenging because most health care systems do not currently offer a home-based palliative care program, making our model relatively unique within the current care delivery landscape. Structural differences—such as resource allocation, staffing models, and integration with primary and specialty care—may influence feasibility and outcomes if implemented elsewhere. Additionally, the single-site design restricts our ability to account for variability in patient demographics, geographic factors, and institutional practices that could affect acuity scoring and hospitalization risk.
Conclusion
Our acuity tool was designed to manage patients’ needs in a timely manner and accurately predict those patients with the highest risk of 30-day hospital readmissions. This tool is one of the first of its kind in the home-based palliative medicine setting. 9 It can help home-based palliative programs triage their limited resources to the highest-need patients to reduce hospitalizations and prevent “crisis care.” Moving forward, we will focus on improving the acuity tool to improve clarity, usability, and its ability to predict hospitalization risk. One proposed modification involves revising ambiguous criteria such as “pain is not at goal” to include more actionable language, for example, “pain medications require adjustments.” This change aims to reduce subjectivity and improve consistency in clinical decision-making. We also plan to evaluate the tool as an educational resource for palliative care fellows, using it to guide decisions regarding visit frequency and follow-up prioritization. Incorporating the tool into training may strengthen fellows’ ability to assess patient needs and allocate resources effectively. Finally, we are exploring the feasibility of extending tool utilization to our inpatient palliative care team as a triage mechanism for determining the urgency of post-discharge home visits. This approach could streamline transitions of care, reduce gaps in service delivery, and ensure timely support for patients at highest risk of unmet needs following hospitalization. Future studies should examine the impact of these adaptations on hospitalization rates, patient satisfaction, and overall program efficiency. This ongoing research will continue to refine and improve our acuity tool, ensuring its effectiveness in the palliative medicine field to best meet the needs of our patients.
Footnotes
Author Note
This work was presented at the Center to Advance Palliative Care (CAPC) Annual Conference, 2025 and Hospice and Palliative Nurses Association Conference, 2025.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.