Default Palliative Care Consultation for Seriously Ill Hospitalized Patients: A Pragmatic Cluster Randomized Trial.

Importance: Increasing inpatient palliative care delivery is prioritized, but large-scale, experimental evidence of its effectiveness is lacking. Objective: To determine whether ordering palliative care consultation by default for seriously ill hospitalized patients without requiring greater palliative care staffing increased consultations and improved outcomes. Design, Setting, and Participants: A pragmatic, stepped-wedge, cluster randomized trial was conducted among patients 65 years or older with advanced chronic obstructive pulmonary disease, dementia, or kidney failure admitted from March 21, 2016, through November 14, 2018, to 11 US hospitals. Outcome data collection ended on January 31, 2019. Intervention: Ordering palliative care consultation by default for eligible patients, while allowing clinicians to opt-out, was compared with usual care, in which clinicians could choose to order palliative care. Main Outcomes and Measures: The primary outcome was hospital length of stay, with deaths coded as the longest length of stay, and secondary end points included palliative care consult rate, discharge to hospice, do-not-resuscitate orders, and in-hospital mortality. Results: Of 34 239 patients enrolled, 24 065 had lengths of stay of at least 72 hours and were included in the primary analytic sample (10 313 in the default order group and 13 752 in the usual care group; 13 338 [55.4%] women; mean age, 77.9 years). A higher percentage of patients in the default order group received palliative care consultation than in the standard care group (43.9% vs 16.6%; adjusted odds ratio [aOR], 5.17 [95% CI, 4.59-5.81]) and received consultation earlier (mean [SD] of 3.4 [2.6] days after admission vs 4.6 [4.8] days; P < .001). Length of stay did not differ between the default order and usual care groups (percent difference in median length of stay, -0.53% [95% CI, -3.51% to 2.53%]). Patients in the default order group had higher rates of do-not-resuscitate orders at discharge (aOR, 1.40 [95% CI, 1.21-1.63]) and discharge to hospice (aOR, 1.30 [95% CI, 1.07-1.57]) than the usual care group, and similar in-hospital mortality (4.7% vs 4.2%; aOR, 0.86 [95% CI, 0.68-1.08]). Conclusions and Relevance: Default palliative care consult orders did not reduce length of stay for older, hospitalized patients with advanced chronic illnesses, but did improve the rate and timing of consultation and some end-of-life care processes. Trial Registration: ClinicalTrials.gov Identifier: NCT02505035.

Evaluating equity in performance of an electronic health record-based 6-month mortality risk model to trigger palliative care consultation: a retrospective model validation analysis.

OBJECTIVE: Evaluate predictive performance of an electronic health record (EHR)-based, inpatient 6-month mortality risk model developed to trigger palliative care consultation among patient groups stratified by age, race, ethnicity, insurance and socioeconomic status (SES), which may vary due to social forces (eg, racism) that shape health, healthcare and health data. DESIGN: Retrospective evaluation of prediction model. SETTING: Three urban hospitals within a single health system. PARTICIPANTS: All patients ≥18 years admitted between 1 January and 31 December 2017, excluding observation, obstetric, rehabilitation and hospice (n=58 464 encounters, 41 327 patients). MAIN OUTCOME MEASURES: General performance metrics (c-statistic, integrated calibration index (ICI), Brier Score) and additional measures relevant to health equity (accuracy, false positive rate (FPR), false negative rate (FNR)). RESULTS: For black versus non-Hispanic white patients, the model's accuracy was higher (0.051, 95% CI 0.044 to 0.059), FPR lower (-0.060, 95% CI -0.067 to -0.052) and FNR higher (0.049, 95% CI 0.023 to 0.078). A similar pattern was observed among patients who were Hispanic, younger, with Medicaid/missing insurance, or living in low SES zip codes. No consistent differences emerged in c-statistic, ICI or Brier Score. Younger age had the second-largest effect size in the mortality prediction model, and there were large standardised group differences in age (eg, 0.32 for non-Hispanic white versus black patients), suggesting age may contribute to systematic differences in the predicted probabilities between groups. CONCLUSIONS: An EHR-based mortality risk model was less likely to identify some marginalised patients as potentially benefiting from palliative care, with younger age pinpointed as a possible mechanism. Evaluating predictive performance is a critical preliminary step in addressing algorithmic inequities in healthcare, which must also include evaluating clinical impact, and governance and regulatory structures for oversight, monitoring and accountability.

Telehealth Palliative Care in Nursing Homes: A Scoping Review.

OBJECTIVES: Many adults older than 65 spend time in a nursing home (NH) at the end of life where specialist palliative care is limited. However, telehealth may improve access to palliative care services. A review of the literature was conducted to synthesize the evidence for telehealth palliative care in NHs to provide recommendations for practice, research, and policy. DESIGN: Joanna Briggs Institute guidance for scoping reviews, and Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews frameworks were used to guide this literature review. SETTINGS AND PARTICIPANTS: Reviewed articles focused on residents in NHs with telehealth palliative care interventionists operating remotely. Participants included NH residents, care partner(s), and NH staff/clinicians. METHODS: We searched Medline (Ovid), Embase (Elsevier), Cochrane Library (WileyOnline), Scopus (Elsevier), CINHAL (EBSCOhost), Trip PRO, and Dissertations & Theses Global (ProQuest) in June 2021, with an update in January 2022. We included observational and qualitative studies, clinical trials, quality improvement projects, and case and clinical reports that self-identified as telehealth palliative care for NH residents. RESULTS: The review yielded 11 eligible articles published in the United States and internationally from 2008 to 2020. Articles described live video as the preferred telehealth delivery modality with goals of care and physical aspects of care being most commonly addressed. Findings in the articles focused on 5 patient and family-centered outcomes: symptom management, quality of life, advance care planning, health care use, and evaluation of care. Consistent benefits of telehealth palliative care included increased documentation of goals of care and decrease in acute care use. Disadvantages included technological difficulties and increased NH financial burden. CONCLUSIONS AND IMPLICATIONS: Although limited in scope and quality, the current evidence for telehealth palliative care interventions shows promise for improving quality and outcomes of serious illness care in NHs. Future empirical studies should focus on intervention effectiveness, implementation outcomes (eg, managing technology), stakeholders' experience, and costs.

Performance drift in a mortality prediction algorithm among patients with cancer during the SARS-CoV-2 pandemic.

Sudden changes in health care utilization during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic may have impacted the performance of clinical predictive models that were trained prior to the pandemic. In this study, we evaluated the performance over time of a machine learning, electronic health record-based mortality prediction algorithm currently used in clinical practice to identify patients with cancer who may benefit from early advance care planning conversations. We show that during the pandemic period, algorithm identification of high-risk patients had a substantial and sustained decline. Decreases in laboratory utilization during the peak of the pandemic may have contributed to drift. Calibration and overall discrimination did not markedly decline during the pandemic. This argues for careful attention to the performance and retraining of predictive algorithms that use inputs from the pandemic period.

Palliative Care Early in the Care Continuum among Patients with Serious Respiratory Illness: An Official ATS/AAHPM/HPNA/SWHPN Policy Statement.

Background: Patients with serious respiratory illness and their caregivers suffer considerable burdens, and palliative care is a fundamental right for anyone who needs it. However, the overwhelming majority of patients do not receive timely palliative care before the end of life, despite robust evidence for improved outcomes. Goals: This policy statement by the American Thoracic Society (ATS) and partnering societies advocates for improved integration of high-quality palliative care early in the care continuum for patients with serious respiratory illness and their caregivers and provides clinicians and policymakers with a framework to accomplish this. Methods: An international and interprofessional expert committee, including patients and caregivers, achieved consensus across a diverse working group representing pulmonary-critical care, palliative care, bioethics, health law and policy, geriatrics, nursing, physiotherapy, social work, pharmacy, patient advocacy, psychology, and sociology. Results: The committee developed fundamental values, principles, and policy recommendations for integrating palliative care in serious respiratory illness care across seven domains: 1) delivery models, 2) comprehensive symptom assessment and management, 3) advance care planning and goals of care discussions, 4) caregiver support, 5) health disparities, 6) mass casualty events and emergency preparedness, and 7) research priorities. The recommendations encourage timely integration of palliative care, promote innovative primary and secondary or specialist palliative care delivery models, and advocate for research and policy initiatives to improve the availability and quality of palliative care for patients and their caregivers. Conclusions: This multisociety policy statement establishes a framework for early palliative care in serious respiratory illness and provides guidance for pulmonary-critical care clinicians and policymakers for its proactive integration.

Performance Drift in a Mortality Prediction Algorithm during the SARS-CoV-2 Pandemic.

Research Objective: Health systems use clinical predictive algorithms to allocate resources to high-risk patients. Such algorithms are trained using historical data and are later implemented in clinical settings. During this implementation period, predictive algorithms are prone to performance changes ("drift") due to exogenous shocks in utilization or shifts in patient characteristics. Our objective was to examine the impact of sudden utilization shifts during the SARS-CoV-2 pandemic on the performance of an electronic health record (EHR)-based prognostic algorithm. Study Design: We studied changes in the performance of Conversation Connect, a validated machine learning algorithm that predicts 180-day mortality among outpatients with cancer receiving care at medical oncology practices within a large academic cancer center. Conversation Connect generates mortality risk predictions before each encounter using data from 159 EHR variables collected in the six months before the encounter. Since January 2019, Conversation Connect has been used as part of a behavioral intervention to prompt clinicians to consider early advance care planning conversations among patients with ≥10% mortality risk. First, we descriptively compared encounter-level characteristics in the following periods: January 2019-February 2020 ("pre-pandemic"), March-May 2020 ("early-pandemic"), and June-December 2020 ("later-pandemic"). Second, we quantified changes in high-risk patient encounters using interrupted time series analyses that controlled for pre-pandemic trends and demographic, clinical, and practice covariates. Our primary metric of performance drift was false negative rate (FNR). Third, we assessed contributors to performance drift by comparing distributions of key EHR inputs across periods and predicting later pandemic utilization using pre-pandemic inputs. Population Studied: 237,336 in-person and telemedicine medical oncology encounters. Principal Findings: Age, race, average patient encounters per month, insurance type, comorbidity counts, laboratory values, and overall mortality were similar among encounters in the pre-, early-, and later-pandemic periods. Relative to the pre-pandemic period, the later-pandemic period was characterized by a 6.5-percentage-point decrease (28.2% vs. 34.7%) in high-risk encounters (p<0.001). FNR increased from 41.0% (95% CI 38.0-44.1%) in the pre-pandemic period to 57.5% (95% CI 51.9-63.0%) in the later pandemic period. Compared to the pre-pandemic period, the early and later pandemic periods had higher proportions of telemedicine encounters (0.01% pre-pandemic vs. 20.0% early-pandemic vs. 26.4% later-pandemic) and encounters with no preceding laboratory draws (17.7% pre-pandemic vs. 19.8% early-pandemic vs. 24.1% later-pandemic). In the later pandemic period, observed laboratory utilization was lower than predicted (76.0% vs 81.2%, p<0.001). In the later-pandemic period, mean 180-day mortality risk scores were lower for telemedicine encounters vs. in-person encounters (10.3% vs 11.2%, p<0.001) and encounters with no vs. any preceding laboratory draws (1.5% vs. 14.0%, p<0.001). Conclusions: During the SARS-CoV-2 pandemic period, the performance of a machine learning prognostic algorithm used to prompt advance care planning declined substantially. Increases in telemedicine and declines in laboratory utilization contributed to lower performance. Implications for Policy or Practice: This is the first study to show algorithm performance drift due to SARS-CoV-2 pandemic-related shifts in telemedicine and laboratory utilization. These mechanisms of performance drift could apply to other EHR clinical predictive algorithms. Pandemic-related decreases in care utilization may negatively impact the performance of clinical predictive algorithms and warrant assessment and possible retraining of such algorithms.

Opportunities to Improve Palliative Care Delivery in Trauma Critical Illness.

BACKGROUND: Despite recommendations to integrate palliative care (PC) into care for critically ill trauma patients, little is known about current PC practices in trauma care to inform opportunities for improvement. OBJECTIVE: Describe patterns of PC delivery among a large, critically ill trauma cohort. SETTING/SUBJECTS: Retrospective cohort study of adult (≥18 years) trauma patients admitted to an intensive care unit (ICU) at an urban, level one trauma center in the United States from March 1, 2017 to March 1, 2019. METHODS: We linked the electronic medical record with the institutional trauma registry. PC process measures included a PC consult order, advance care planning (ACP) note, and hospice use. Unadjusted results are reported for the total population, decedents, and subgroups at risk for poor outcomes (age ≥55 years, Black race ≥1 pre-existing comorbidity, and severe injury) after trauma. RESULTS: Among 1309 eligible admissions, 902 (68.9%) were male, 640 (48.9%) were Black, and 654 (50.0%) were ≥55 years old. Eighty-one (6.2%) patients received a PC consult order, 66 (5.0%) had an ACP note, and 13 (1.1%) were discharged to hospice. Among decedents (N = 91; 7%), 28 (30.8%) received a PC consult order and 36 (39.6%) had an ACP note. For high-risk subgroups, PC consult orders and ACP note rates ranged from 4.5-12.8% and 4.5-11.8%, respectively. CONCLUSION: PC delivery was rare among this cohort, including those at high risk for poor outcomes. Urgent efforts are needed to identify barriers to and develop targeted interventions for high quality PC delivery in trauma ICU care.

Using electronic health records and claims data to identify high-risk patients likely to benefit from palliative care.

OBJECTIVES: Palliative care has been demonstrated to have positive effects for patients, families, health care providers, and health systems. Early identification of patients who are likely to benefit from palliative care would increase opportunities to provide these services to those most in need. This study predicted all-cause mortality of patients as a surrogate for patients who could benefit from palliative care. STUDY DESIGN: Claims and electronic health record (EHR) data for 59,639 patients from a large integrated health care system were utilized. METHODS: A deep learning algorithm-a long short-term memory (LSTM) model-was compared with other machine learning models: deep neural networks, random forest, and logistic regression. We conducted prediction analyses using combined claims data and EHR data, only claims data, and only EHR data, respectively. In each case, the data were randomly split into training (80%), validation (10%), and testing (10%) data sets. The models with different hyperparameters were trained using the training data, and the model with the best performance on the validation data was selected as the final model. The testing data were used to provide an unbiased performance evaluation of the final model. RESULTS: In all modeling scenarios, LSTM models outperformed the other 3 models, and using combined claims and EHR data yielded the best performance. CONCLUSIONS: LSTM models can effectively predict mortality by using a combination of EHR data and administrative claims data. The model could be used as a promising clinical tool to aid clinicians in early identification of appropriate patients for palliative care consultations.

Palliative Care Team Perceptions of Standardized Palliative Care Referral Criteria Implementation in Hospital Settings.

Background: Standardized referral criteria can aid in identifying patients who would benefit from palliative care consultation. Little is known, however, on palliative care team members' perceptions of these criteria. Objective: Describe palliative care programs' reasons for referral criteria implementation and their perception of the benefits or disadvantages of its use. Design: Online survey of National Palliative Care Registry™ participants who use standardized referral criteria. Results: Fifty-three programs participated. Late referrals (64.2%) were the most commonly cited reason for referral criteria implementation. The majority (77.4%) felt that referral criteria lead to positive outcomes, including earlier referrals for palliative care-appropriate patients (71.7%). Increases in staff workload and inappropriate referrals were identified as disadvantages of referral criteria use.* Conclusion: Palliative care program members identified both benefits and disadvantages of referral criteria use, but felt they had mostly productive results. *Correction added on March 18, 2021 after first online publication of December 18, 2020: In the Results section of the abstract, the third sentence was changed from "Increases in clinical volume and inappropriate referrals were identified as disadvantages of referral criteria use." to "Increases in staff workload and inappropriate referrals were identified as disadvantages of referral criteria use."

Prognosticating Outcomes and Nudging Decisions with Electronic Records in the Intensive Care Unit Trial Protocol.

Expert recommendations to discuss prognosis and offer palliative options for critically ill patients at high risk of death are variably heeded by intensive care unit (ICU) clinicians. How to best promote such communication to avoid potentially unwanted aggressive care is unknown. The PONDER-ICU (Prognosticating Outcomes and Nudging Decisions with Electronic Records in the ICU) study is a 33-month pragmatic, stepped-wedge cluster randomized trial testing the effectiveness of two electronic health record (EHR) interventions designed to increase ICU clinicians' engagement of critically ill patients at high risk of death and their caregivers in discussions about all treatment options, including care focused on comfort. We hypothesize that the quality of care and patient-centered outcomes can be improved by requiring ICU clinicians to document a functional prognostic estimate (intervention A) and/or to provide justification if they have not offered patients the option of comfort-focused care (intervention B). The trial enrolls all adult patients admitted to 17 ICUs in 10 hospitals in North Carolina with a preexisting life-limiting illness and acute respiratory failure requiring continuous mechanical ventilation for at least 48 hours. Eligibility is determined using a validated algorithm in the EHR. The sequence in which hospitals transition from usual care (control), to intervention A or B and then to combined interventions A + B, is randomly assigned. The primary outcome is hospital length of stay. Secondary outcomes include other clinical outcomes, palliative care process measures, and nurse-assessed quality of dying and death.Clinical trial registered with clinicaltrials.gov (NCT03139838).

The palliative care needs of lung transplant candidates.

BACKGROUND: Little is known about the palliative care needs of patients awaiting lung transplantation. The aim of this study was to describe these needs in patients undergoing evaluation for or awaiting lung transplantation. METHODS: Cross-sectional survey using an adapted version of the Needs at the End-of-life Screening Tool (NEST-13) at a US-based transplant program. RESULTS: Among the 111 participants, 83.5% were White, 60.0% were female, and almost three-quarters had either restrictive or obstructive lung disease. The greatest palliative care needs included difficulty being physically active (mean: 7.9/10; SD: 2.6; median: 9.0), physical symptoms (mean: 7.4/10; SD: 2.6; median: 8.0), missing work due to illness (mean: 6.2/10; SD: 4.0; median: 8.0), and concerns that life might end (mean: 5.1/10; SD: 3.6; median: 5.0). Participants reported that religious/spiritual beliefs contribute to their sense of purpose (mean: 4.1/10; SD: 3.9) but had few unmet needs in this area (mean: 0.9/10; median: 0.0). Only 6.4% reported seeing a palliative care specialist, and 48.2% were unsure what a palliative care specialist is. CONCLUSION: There are substantial palliative care needs among lung transplant candidates, particularly physical symptoms and end-of-life concerns. These findings support integrating palliative care and end-of-life discussions in the management of lung transplant candidates.

"I Don't Have Time to Sit and Talk with Them": Hospitalists' Perspectives on Palliative Care Consultation for Patients with Dementia.

BACKGROUND/OBJECTIVES: Specialty palliative care for hospitalized patients with dementia is widely recommended and may improve outcomes, yet rates of consultation remain low. We sought to describe hospitalists' decision-making regarding palliative care consultation for patients with dementia. DESIGN: Descriptive qualitative study. SETTING: Seven hospitals within a national nonprofit health system. PARTICIPANTS: Hospitalist physicians. MEASUREMENTS: Individual semistructured interviews. We used thematic analysis to explore factors that influence hospitalists' decision to consult palliative care for patients with dementia. RESULTS: A total of 171 hospitalists were eligible to participate, and 28 (16%) were interviewed; 17 (61%) were male, 16 (57%) were white, and 18 (64%) were in practice less than 10 years. Overall, hospitalists' decisions to consult palliative care for patients with dementia were influenced by multiple factors across four themes: patient, family caregiver, hospitalist, and organization. Consultation was typically only considered for patients with advanced disease, particularly those receiving aggressive care or with family communication needs (navigating conflicts around goals of care and improving disease and prognostic understanding). Hospitalists' limited time and, for some, a lack of confidence in palliative care skills were strong drivers of consultation. Palliative care needs notwithstanding, most hospitalists would not request consultation if they perceived families would be resistant to it or had limited availability or involvement in caregiving. Additional barriers to referral at the organization level included a hospital culture that conflated palliative and end-of-life care and busy palliative care teams at some hospitals. CONCLUSION: Hospitalists described a complex consultation decision process for involving palliative care specialists in the care of patients with dementia. Systematic identification of hospitalized patients with dementia most likely to benefit from palliative care consultation and strategies to overcome modifiable family and organization barriers are needed. J Am Geriatr Soc 68:2365-2372, 2020.

Risk Factors for Long-term Mortality and Patterns of End-of-Life Care Among Medicare Sepsis Survivors Discharged to Home Health Care.

Importance: Despite a growing recognition of the increased mortality risk among sepsis survivors, little is known about the patterns of end-of-life care among this population. Objective: To describe patterns of end-of-life care among a national sample of sepsis survivors and identify factors associated with long-term mortality risk and hospice use. Design, Setting, and Participants: This cohort study assessed sepsis survivors who were Medicare fee-for-service beneficiaries discharged to home health care using national Medicare administrative, claims, and home health assessment data from 2013 to 2014. The initial and final primary analyses were conducted in July 2017 and from July to August 2019, respectively. Exposures: Sepsis hospital discharge and 1 or more home health assessments within 1 week. Main Outcomes and Measures: Outcomes were 1-year mortality among all sepsis survivors and hospitalization in the last 30 days of life, death in an acute care hospital, and hospice use among decedents. Multivariate logistic regression was used to identify factors associated with 1-year mortality and hospice use. Results: Among 87 581 sepsis survivors who were Medicare fee-for-service beneficiaries discharged to home health care, 49 323 (56.3%) were aged 75 years or older, 69 499 (79.4%) were non-Hispanic white, and 48 472 (55.3%) were female. Among the total survivors, 24 423 (27.9%) people died within 1 year of discharge, with a median (interquartile range) survival time of 119 (51-220) days. Among these decedents, 16 684 (68.2%) were hospitalized in the last 30 days of life, 6560 (26.8%) died in an acute care hospital, and 12 573 (51.4%) were enrolled in hospice, with 5729 (45.6%) using hospice for 7 or fewer days. Several factors were associated with 1-year mortality, including a cancer diagnosis (odds ratio [OR], 3.66; 95% CI, 3.50-3.83; P < .001), multiple dependencies of activities of daily living or instrumental activities of daily living (OR, 2.80; 95% CI, 2.57-3.05; P < .001), and an overall poor health status (OR, 2.21; 95% CI, 2.01-2.44; P < .001) documented on home health assessment. Among the decedents, cancer was associated with hospice use (OR, 2.25; 95% CI, 2.11-2.41; P < .001), patients aged 85 years or older (OR, 1.49; 95% CI, 1.37-1.61; P < .001), and living in an assisted living setting (OR, 1.93; 95% CI, 1.69-2.19; P < .001). Conclusions and Relevance: The findings of this study suggest that death within 1 year after sepsis discharge may be common among Medicare beneficiaries discharged to home health care. Although 1 in 2 decedents used hospice, aggressive care near the end of life and late hospice referral were common. Readily identifiable risk factors suggest opportunities to target efforts to improve palliative and end-of-life care among high-risk sepsis survivors.

Electronic Health Record Mortality Prediction Model for Targeted Palliative Care Among Hospitalized Medical Patients: a Pilot Quasi-experimental Study.

BACKGROUND: Development of electronic health record (EHR) prediction models to improve palliative care delivery is on the rise, yet the clinical impact of such models has not been evaluated. OBJECTIVE: To assess the clinical impact of triggering palliative care using an EHR prediction model. DESIGN: Pilot prospective before-after study on the general medical wards at an urban academic medical center. PARTICIPANTS: Adults with a predicted probability of 6-month mortality of ≥ 0.3. INTERVENTION: Triggered (with opt-out) palliative care consult on hospital day 2. MAIN MEASURES: Frequencies of consults, advance care planning (ACP) documentation, home palliative care and hospice referrals, code status changes, and pre-consult length of stay (LOS). KEY RESULTS: The control and intervention periods included 8 weeks each and 138 admissions and 134 admissions, respectively. Characteristics between the groups were similar, with a mean (standard deviation) risk of 6-month mortality of 0.5 (0.2). Seventy-seven (57%) triggered consults were accepted by the primary team and 8 consults were requested per usual care during the intervention period. Compared to historical controls, consultation increased by 74% (22 [16%] vs 85 [63%], P < .001), median (interquartile range) pre-consult LOS decreased by 1.4 days (2.6 [1.1, 6.2] vs 1.2 [0.8, 2.7], P = .02), ACP documentation increased by 38% (23 [17%] vs 37 [28%], P = .03), and home palliative care referrals increased by 61% (9 [7%] vs 23 [17%], P = .01). There were no differences between the control and intervention groups in hospice referrals (14 [10] vs 22 [16], P = .13), code status changes (42 [30] vs 39 [29]; P = .81), or consult requests for lower risk (< 0.3) patients (48/1004 [5] vs 33/798 [4]; P = .48). CONCLUSIONS: Targeting hospital-based palliative care using an EHR mortality prediction model is a clinically promising approach to improve the quality of care among seriously ill medical patients. More evidence is needed to determine the generalizability of this approach and its impact on patient- and caregiver-reported outcomes.