Citations

ABSTRACT Pulmonary arterial hypertension (PAH) is associated with significant symptoms and impaired health‐related quality of life (HRQOL) and survival. Specialty palliative care (SPC), an intervention delivered by a multidisciplinary team, is focused on improving HRQOL. We performed a randomized controlled trial (RCT) of SPC among adults with PAH and impaired HRQOL to determine the impact of SPC on HRQOL as assessed by PAH‐SYMPACT. PAH‐SYMPACT is a validated PAH disease‐specific patient‐reported outcome measure that assesses HRQOL across four domains: cardiopulmonary symptoms, cardiovascular symptoms, physical impact, and cognitive/emotional impact. Sixty‐two patients were included (36 usual care, 26 SPC plus usual care), but only 58% of patients randomized to SPC completed their SPC visits. In the intention‐to‐treat analysis, SPC was associated with similar changes in total, symptom, and physical impact domain scores and increased cognitive/emotional impact scores, indicative of worse HRQOL. In the per‐protocol analysis, SPC was not associated with significant changes in total or individual domain scores. According to a survey of patients who completed SPC visits, most were satisfied with their visit and were likely to recommend SPC for other patients with PAH. In conclusion, limited SPC visits in this single‐center RCT were not associated with significant improvements in HRQOL among patients with PAH and impaired HRQOL, but the majority of patients who participated in a SPC intervention were satisfied and would recommend SPC to other PAH patients. Future studies of SPC in PAH are needed to determine the appropriate patient selection and timing for SPC. www.clinicaltrials.gov (NCT03905421).

Abstract Real-time, proactive detection has become a necessity as health systems face converging pressures that legacy tools cannot solve: rising acuity, workforce strain, and persistent quality gaps that erode margins, capacity, and reputation. Clinical decision support has historically lagged because rudimentary rules-based systems drown clinicians in false alerts, eroding trust without reliably changing outcomes at the bedside. Bayesian Health’s real-time clinical intelligence platform was built to close this gap by continuously synthesizing multimodal EHR data (vitals, labs, medications, procedures, and longitudinal history) to learn patient-specific baselines and surface only the most actionable, high-precision signals to care teams. Across diverse health systems, this platform has been used to enable earlier recognition and intervention for a range of high-impact conditions, improving both bedside care and system performance. This talk will focus on concrete, real-world applications of the platform for sepsis, clinical deterioration, and palliative care needs, illustrating how earlier detection and workflow-integrated interventions drive measurable gains in clinical outcomes and margins, and why the value of early, precise detection is amplified in high-risk populations such as oncology patients, whose higher baseline acuity and treatment-related toxicities both elevate their risk of harm and make emerging deterioration harder to recognize. Case Study: Detecting Sepsis within a Oncology Patient Subpopulation. Background: Sepsis remains one of the leading causes of in-hospital mortality, and traditional rules-based alerts have struggled to deliver timely, actionable detection without overwhelming clinicians with false alarms. Oncology patients experience an elevated risk for sepsis and septic shock, and those who experience sepsis are more likely to suffer in-hospital mortality than non-cancer patients with sepsis. Early recognition of sepsis onset is challenging due to the overlap between signs of sepsis and inflammation and tumor- and treatment-related toxicities. We investigated the ability of Bayesian Health’s real-time artificial intelligence-based computational algorithm to perform timely detection of sepsis onset in hospitalized patients with cancer. This algorithm ingests data directly from the electronic health record, such as vital signs, laboratory results, medications, procedures, and patient-specific historical results, processing it in a continuous manner to flag presence of sepsis as early as possible. Methods: Analysis included 7480 adult emergency and inpatient encounters from three tertiary and quaternary academic centers (Cleveland Clinic Main Campus, Sinai Hospital of Baltimore, and Strong Memorial Hospital) and one community hospital (Cleveland Clinic Fairview Hospital) with sufficient data for inclusion (age ≥ 18, presence in an emergency or inpatient care unit, and at least one measurement of a panel of vital signs). Patients with cancer (including subgroups of solid tumor and hematologic malignancy) were identified by current or past recorded diagnoses (ICD-10 codes). All included patient records underwent a multi-tiered retrospective adjudication process. An automated screen labeled records with no evidence of infection or organ dysfunction as non-sepsis, removing them from need for manual review. Two independent physicians then adjudicated a sample of the eliminated records in addition to each remaining record for both presence of sepsis and time of onset, with a third physician reconciling disagreements. Evaluation of the algorithm included all discrete flags generated by the system during a period of time it was not in clinical use and not visible to care teams. Results: In the overall population, 3.1% of all patients and 6.0% of cancer patients were adjudicated to have had sepsis, including zero from a sample of 200 of the records marked as non-sepsis in automated screening. The detection algorithm identified sepsis with high and consistent accuracy across all patients and across cancer patients, including those with solid tumors and with hematologic malignancies (Table). A flag for sepsis occurred in 85.8% of all sepsis patients and 86.1% of cancer patients with sepsis within 4 days of sepsis onset. Sepsis was present in 24.2% of all cancer patients flagged (22.9% of solid tumor patients and 27.9% of hematologic malignancy patients). Compared to patients who were never flagged, all patients who were flagged, whether they had sepsis or not, were more likely to experience in-hospital mortality (7.9% vs 0.3%, p < 0.001) or ICU transfer (38.2% vs 7.9%, p < 0.001). This was also true within oncology patients (in-hospital mortality: 8.8% vs 0.4%, p < 0.001; ICU transfer: 34.7% vs 9.7%, p < 0.001). Implementation of an automated sepsis detection system with high accuracy in populations such as cancer patients, who have elevated risk of poor outcomes following sepsis, provides an opportunity to meaningfully improve timely delivery of care and reduce subsequent morbidity and mortality while minimizing alert fatigue for clinicians. Citation Format: Suchi Saria, Varesh Prasad, Nicasia Beebe-Wang, Ellie Zhang, Neri Cohen, David Graham, Conrad Gleber, James Morrison, Jacob Strand, Alexandra Tallman. Real-time multimodal AI enables individualized care in challenging subpopulations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr PL03-03.

Most palliative care (PC) physicians will change jobs over the course of their career, yet few supports exist to guide them through this often unfamiliar process. Job transitions frequently elicit stress and uncertainty, as individuals balance personal and professional priorities, navigate complex negotiations, and assume the inherent risks of pursuing new opportunities. Yet, when approached intentionally, a job transition can also be a transformative process that allows for self-discovery, cultivation of new skills and connections, and reshaping of one’s professional identity. This article, written by a team of PC physicians who have made job transitions themselves, offers 10 tips for navigating this professional step successfully. With emphases on authentic communication, skilled mentorship, relationship building, and exploration of local culture—the article offers practical advice to help PC physicians navigate a job transition confidently, while staying grounded in their personal values and the work they find most meaningful.

1649 Background: Findings from our prior large-scale comparative effectiveness trial showed the equivalent effect of delivering early palliative care via video versus in-person on quality of life among patients with advanced non-small cell lung cancer (NSCLC). We now report on whether the two care delivery modalities were equivalent with respect to patient-reported communication with clinicians about their end-of-life (EOL) care preferences and hospice utilization. Methods: Between 6/14/2018 and 5/4/2023, we enrolled 1250 patients with newly diagnosed advanced NSCLC in a randomized trial of early palliative care across 22 US cancer centers. Patients were randomly assigned to meet with a palliative care clinician every 4 weeks from enrollment through the course of the disease, either via video or in the outpatient clinic. Participants completed self-report surveys at baseline and weeks 12, 24, 36, and 48, including an item asking if they had discussed with their clinicians the care they would want to receive if dying (yes/no); patients’ final assessments prior to death or last follow up were analyzed. We reviewed patients’ health records to collect data on hospice referral and length of stay. To test the equivalence in these outcomes, we used a binomial generalized linear model with the identity link function (pre-specified equivalence margin of ±8% for patient-reported communication about EOL care) and linear regression (pre-specified equivalence margin of ±6 days for mean length of stay in hospice). P-values were adjusted for multiplicity using a Bonferroni correction. Results: Of the 1250 enrolled participants, 888 (71.0%) completed at least one survey post baseline regarding whether they communicated with clinicians about EOL care preferences. Among those, 29.1% of the video group and 26.0% of the in-person group reported “yes,” indicating that they recalled such EOL care discussions (difference = 3.1%, 95% CI: -1.8%, 8.1%; p = 0.26 for equivalence). During the course of the trial, 733/1250 (58.6%) patients died, of whom 537/733 (73.3%) were referred to hospice. Mean lengths of hospice stay were 25.3 days for the video group versus 25.1 days for the in-person group (difference = 0.2, 95% CI: -7.0, 7.4; p = 0.46 for equivalence). When excluding outlying patients receiving hospice services > 180 days (n = 13), the mean lengths of hospice stay were 19.1 (video group) versus 19.7 (in-person group) days (difference = -0.6, 95% CI: -4.6, 3.3; p = 0.06 for equivalence). Conclusions: Although thresholds were not met to confirm equivalence statistically, the two modalities for delivering early palliative care demonstrate very similar outcomes with respect to patient-clinician communication about EOL care and hospice utilization. These findings provide further evidence of the utility of video visits for providing high quality palliative and EOL care. Clinical trial information: ClinicalTrials.gov Identifier ( NCT03375489 ) .

Background Heart failure self‐care can contribute to a high daily workload and treatment burden. The goal of this cohort study was to assess the characteristics and outcomes associated with burden of treatment (BoT). Methods Surveys comprising validated instruments to measure BoT and other constructs were mailed to patients with heart failure in Southeastern Minnesota. Participants were divided into tertiles by BoT scores. Associations of clinical variables with BoT were examined using multinomial logistic regression. Associations of BoT with mortality and hospitalizations were examined using Cox proportional hazard regression and Andersen–Gill models, respectively. Results A total of 609 participants (mean age 76.3 years, 60.3% men, 55.2% urban, 64.3% preserved ejection fraction) completed surveys. Higher BoT was associated with worse health status, more depressive symptoms, lower resilience, less social support, lower medication adherence, and worse health literacy. Mean±SD follow‐up was 14.4 (4.1) months. Estimated 1‐year mortality (8.3% [95% CI, 4.3%–12.1%], 11.0% [95% CI, 6.5%–15.2%], 16.0% [95% CI, 10.8%–21.0%]) and 1‐year mean cumulative hospitalizations (0.57 [95% CI, 0.45–0.72], 0.83 [95% CI, 0.66–1.05], 1.15 [95% CI, 0.93–1.42]) increased across patients reporting low, medium, and high BoT, respectively. Adjustment for health status eliminated any significant association of BoT with risks of death and hospitalization (adjusted hazard ratio [HR], 1.10 [95% CI, 0.58–2.07] and 1.09 [95% CI, 0.74–1.61], respectively, highest versus lowest BoT tertile). Conclusions BoT in heart failure varies by clinical and psychosocial factors. Higher BoT identifies patients at increased risk of adverse health outcomes due to their worse health status. These findings can serve as a foundation for interventions to minimize workload and improve quality of life.

Background: Benefits of hospice care, such as improvement in quality of life and reduced costs, depend on duration of enrollment in hospice services, making timely hospice referral essential. Patient-, provider-, and system-level factors associated with hospice referral timing and length of stay (LOS) are unclear. 1 – 6 Objective: To review existing hospice referral patterns to identify patient- and system-level factors associated with hospice LOS. Methods: We performed a retrospective review of all adult patients referred from our institution (located in Rochester, Minnesota, USA) to a nonprofit hospice agency between 2013 and 2017. The primary dependent variable was hospice LOS. Patient-level variables included demographic characteristics, place of residence, and hospice diagnosis. System-level variables included referral setting and provider-specific variables, such as title and gender. Statistical analyses were performed using multivariate logistic regression. Results: A total of 2072 patients were referred to hospice during the study period. Patient-level factors associated with LOS included hospice diagnosis and place of residence. Patients referred for cancer had a higher odds of a long LOS, while patients from long-term care facilities had a higher odds of a short LOS. System-level factors associated with LOS included provider gender and title. Referral by a female provider or by a physician, rather than an advanced practice provider, was associated with a lower odds of a short LOS. Conclusion: Based on a review of hospice referral patterns, the integration of hospice care into subspecialty practices, long-term care facilities, and advanced practice education could be an effective strategy to improve hospice LOS.

1559 Background: Timely access to specialty Palliative Care (SPC) services provides significant benefits to hospitalized patients with cancer, including improvements in quality of life. However, many patients with cancer who could benefit from SPC suffer from lack of timely referral or do not receive SPC services. A previously published trial of an artificial intelligence/machine learning (AI/ML) model to predict need for SPC at Mayo Clinic increased timely SPC consultation of hospitalized patients overall. We sought to review the performance of this algorithm in patients with cancer across an expanded hospitalized population. Methods: The study population consisted of all patients admitted into a Mayo Clinic hospital in Minnesota, Wisconsin, Arizona and Florida between January 2020 and September 2023. Due to the size of the cohort a case-control sample of three controls for every case was created. Patients were considered to be a patient with cancer if they had at least one billing diagnosis up to 1 year prior to their index hospitalization from any of the 5 HCC categories denoting cancer: Metastatic Cancer and Acute Leukemia (HCC 8); Lung and Other Severe Cancers (HCC 9); Lymphoma and Other Cancers (HCC 10); Colorectal, Bladder, and Other Cancers (HCC 11); and Breast, Prostate, and Other Cancers and Tumors (HCC 12). The training data set consisted of 107,076 patient encounters with a total of 8,355,090 time periods of constant risk. An AI/ML model using gradient boosting methods which contained 269 variables (both static and time-varying) of various classes with SPC consultation treated as a time-to-event outcome was trained. Results: Due to the longitudinal nature of the prediction, performance was assessed using the max score AUC (using the max score a patient received during a given encounter prior to an event or discharge) to produce the AUC. The model had an overall AUC of 0.932 (0.929, 0.935 – 95% CI). We saw a decrease in performance of the AI/ML model in the Oncology population with an overall performance of 0.818 (0.806, 0.83). Some of the most influential variables were previous SPC visit, age, metastatic disease, acute leukemia diagnosis, and pain scores. Conclusions: An AI/ML model can effectively predict the need for an inpatient PC consult in a population of hospitalized patients with cancer. [Table: see text]

11108 Background: Advances in cancer treatment have led to a growing population of patients living with metastatic cancer. Symptom burden for these patients is understudied. Our objective is to report the prevalence and severity of SPPADE symptoms (sleep disturbance, pain, physical function impairment, anxiety, depression, and low energy /fatigue) in a large cohort of patients with metastatic cancer. Methods: These data are drawn from the Enhanced, EHR-facilitated Cancer Symptom Control (E2C2) stepped wedge pragmatic trial evaluating a symptom management intervention for cancer patients (NCT03892967). This analysis was limited to those identified with metastatic disease using ICD- codes and a natural language processing algorithm prior to or within 30 days of their baseline symptom survey. Symptom severity was measured on a 0-10 scale with scores of 1-3 categorized as “mild”, 4-6 as “moderate”, and ≥7 as “severe.” A SPPADE composite score was calculated by summing six individual symptom scores and is an indicator of total symptom burden. We report descriptive statistics for each symptom by severity category. We also test for differences in SPPADE composite score by race, disability status, employment status, education and cancer type using Kruskal-Wallis tests. Results: 12,824 patients with metastatic disease completed a survey between March 2019 and January 2023. Discretized score distributions for the SPPADE symptoms are reported in the Table. The SPPADE composite mean score for the cohort was 17.6 (SD +/- 12.0). Significant differences were found in total symptom burden by sociodemographic and clinical characteristics ( p<0.01): Higher total symptom burden (Mean [SD])was experienced by individuals who were American Indian/Alaska Native (AI/AN) (23.8[13.5]) or African American (AA) (20.5[14.7]), disabled (24.0[12.8]) or unemployed (21.1[13.0]), and had a high school education (20.0[13.3]) or less than high school education (18.5[12.4]). There were also statistically significant differences ( p<0.01) in symptom burden across cancer types, with patients with metastatic lung cancer reporting the highest symptom burden (20.0[12.2]) and those with metastatic melanoma reporting the lowest (14.7[11.4]). Conclusions: SPPADE symptoms are common in patients with metastatic cancer, especially fatigue, sleep disturbance, and physical function impairment. Symptom burden is especially high in AI/AN, AA, disabled/unemployed, and less educated patients. Further research is needed to tailor symptom interventions among those living with metastatic disease. [Table: see text]