Disparities in palliative care utilization among patients with acute-on-chronic liver failure in the United States: a retrospective analysis

Introduction

Patients with decompensated liver cirrhosis or end-stage liver disease (ESLD) exhibit heterogeneous clinical trajectories. While ESLD broadly encompasses a spectrum of advanced liver disease, acute-on-chronic liver failure (ACLF) represents a distinct, rapidly progressive syndrome within this spectrum, characterized by acute organ failure and markedly higher short-term mortality (1,2). Since the early 2000s, all-cause mortality in this population has declined by an estimated 15%, attributed in part to improved understanding of ACLF pathophysiology and the broader implementation of emergency liver transplantation (LT) (3). ACLF is a distinct clinical syndrome characterized by acute hepatic decompensation in individuals with underlying chronic liver disease, accompanied by systemic inflammation and extrahepatic organ failures. Multiple professional societies have widely studied this entity; however, no uniform diagnostic definition exists. Currently, 13 separate definitions for ACLF have been proposed; however, all the definitions suggest acute, severe clinical decompensation of patients with chronic liver disease, associated with organ failures and a higher risk of mortality (4,5).

Epidemiologic estimates from cohort studies conducted between 2013 and 2020 place the prevalence of ACLF between 20% and 35% among hospitalized patients with cirrhosis (6). A recent systematic review and meta-analysis synthesizing data from 43,206 patients with ACLF and 140,835 without ACLF reported a global ACLF prevalence of 35% (95% CI: 33–38%). Alcohol-associated liver disease (ALD) was the predominant underlying etiology (45%; 95% CI: 41–50%), and bacterial infections were the most common precipitating trigger (35%) (6). Management of ACLF currently centers on supportive interventions addressing individual organ failures, predominantly within intensive care settings. While LT remains the only definitive treatment associated with favorable long-term survival, its application is limited by resource constraints, eligibility criteria, and post-transplant physiologic, psychosocial, and quality-of-life considerations (7,8). For patients deemed ineligible or unlikely to benefit from LT, palliative care (PC) services represent a critical adjunct, offering structured approaches to symptom management, psychosocial support, and care goal alignment. Despite guideline recommendations endorsing early PC integration in advanced liver disease, real-world utilization remains limited (3,9,10).

Prior investigations have identified multiple barriers contributing to PC underuse, including limited provider awareness, misperceptions equating PC with hospice, inadequate access to trained PC teams, and systemic inequities embedded within healthcare delivery structures. Furthermore, disparities in PC utilization have been documented across demographic and institutional strata, with variables such as race, ethnicity, insurance type, income quartile, and hospital teaching status independently associated with differential access (11,12). Given the high short-term mortality burden associated with ACLF and the underutilization of PC within this population, the present study focuses specifically on terminal ACLF hospitalizations to characterize demographic and clinical characteristics and predictors of PC consultation among hospitalized U.S. ACLF patients and to evaluate the impact of PC utilization on hospitalization resource metrics, including length of stay and total hospitalization charges. We present this article in accordance with the STROBE reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-2026-0029/rc).

Methods

Data source

The National Inpatient Sample (NIS) database was used to conduct this retrospective analysis. The NIS database is administered by the Healthcare Cost and Utilization Project (HCUP) and is the largest all-payer inpatient database in the United States (12,13). It is derived from a 20% stratified sample of inpatient discharges from community hospitals across participating U.S. states. After applying appropriate discharge-level weights provided by HCUP, the NIS provides nationally representative estimates of disease burden and inpatient outcomes. Each hospitalization is recorded as a unique, de-identified entry within the database. As the NIS contains de-identified patient data, institutional review board approval was not required. This study was conducted in accordance with the ethical standards of the responsible institution regarding human subjects and the principles outlined in the Declaration of Helsinki and its subsequent amendments.

Study population

We used International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes to identify adult patients hospitalized with ACLF between 2016 and 2022. ACLF was defined as the presence of two organ failures (kidney failure, cardiac failure, hepatic encephalopathy, and respiratory failure) among patients with cirrhosis. Organ failures were identified using ICD-10 diagnosis and procedure codes consistent with prior NIS-based studies of ACLF and cirrhosis-related hospitalizations (14,15). Cardiovascular failure was captured using codes for shock and invasive hemodynamic monitoring, including central venous pressure and arterial line procedures, consistent with prior claims-based ACLF algorithms. The ICD-10 codes for organ failures are provided in Appendix 1. The definition replicates the definition of ACLF grade 1 by the CANONIC study group and connotes the higher mortality risk associated with the two-organ failure threshold, as shown by the North American Consortium for the study of liver disease (16). Since the study relies on an administrative database, granular data were not available, and organ failure proxies were used to define ACLF. Only patients who suffered in-hospital mortality were included in the study.

Therefore, the analytic cohort represents terminal ACLF hospitalizations rather than all ACLF admissions. Patients with missing demographic information or mortality data were excluded from the analysis. A total of 163,060 patients met the inclusion criteria. The inclusion flow process is presented in Figure 1. Patients were stratified into two groups according to whether they received PC or not. The inclusion flow process is presented in Figure 1. ICD-10 code Z51.5 was used to identify patients who received PC. This code is used when PC services, comfort care, end-of-life care, and hospice care were utilized by the patient, regardless of whether the specialist or primary team provided the care.

Figure 1 Inclusion flow diagram. ACLF, acute-on-chronic liver failure.

Study variables

We collected data on demographic variables, including age, sex, race, median household income, insurance status, and hospital characteristics pre-specified by HCUP (region, bed size, and teaching status). The comorbidity burden was assessed using the modified Charlson Comorbidity Index (mCCI). The Charlson Comorbidity Index is a well-validated index based on ICD-10-CM codes used in large administrative data to predict mortality and hospital resource use (17-19). The modified Charlson Comorbidity Burden was created by excluding liver comorbidities from the index. Information regarding comorbidities, including diabetes, hypertension, hyperlipidemia, coronary artery disease, chronic obstructive pulmonary disease, alcohol use, obesity, smoking, and renal failure, was collected. We further collected information on underlying liver disease, including ALD, hepatitis B, hepatitis C, metabolic dysfunction-associated steatohepatitis (MASH), and hepatocellular carcinoma (HCC), as well as common decompensations of liver disease like ascites, variceal bleeding, and hepatorenal syndrome (HRS). Information regarding interventions such as blood transfusion, LT, transjugular intrahepatic portosystemic shunt (TIPS), and esophagogastroduodenoscopy (EGD) was included in the analysis. To assess resource utilization, we collected data on length of stay and total hospitalization charges. We also collected data on ACLF grade at hospitalization and categorized patients into Grade 1, Grade 2, and Grade 3.

Study outcomes

In this study, we aimed to determine the proportion of patients with ACLF-related mortality who received PC between 2016 and 2022, identify factors associated with PC utilization, and assess the impact on resource utilization. Furthermore, we conducted a sensitivity analysis excluding patients with length of stay ≤2 days to evaluate whether early in-hospital mortality influenced the observed trends in PC utilization.

Statistical analysis

Hospital-level discharge weights were used to generate national estimates. Continuous variables were compared with an independent sample t-test, whereas categorical variables were compared using the Chi-squared test. To assess trends, we calculated the proportions of patients in the study population who received PC. We evaluated the trend using the Cochrane-Armitage test. To identify factors associated with PC utilization, a multivariate logistic regression model was developed, adjusting for patient demographics, hospital characteristics, comorbidities, etiology, and decompensations of liver disease, ACLF grade, and interventions noted to have P<0.05 on multivariate analysis.

Results

Patient demographics and hospital characteristics

A total of 163,060 terminal ACLF hospitalizations ending in in-hospital mortality were included in the final analysis. Of these 75,365 patients (46.2%) received PC. Patients who received PC were predominantly aged 45–64 years (53.0%), males (60.9%), and White (67.3%). Most were insured by Medicare (42.7%). A complete list of patient demographics and hospital characteristics is presented in Table 1.

Comorbidities

Renal failure was the most prevalent comorbidity overall and was more common among patients who received PC (93.4% vs. 92.6%). Patients in the PC group had a higher prevalence of smoking (35.8% vs. 32.3%) and alcohol use (59.0% vs. 54.3%) compared with those who did not receive PC. In contrast, diabetes (28.6% vs. 32.8%), hyperlipidemia (18.0% vs. 19.2%), and coronary artery disease (13.0% vs. 15.6%) were less prevalent among patients who received PC. The majority of patients who received PC had a mCCI score of ≥3 (76.0%). A complete list of comorbidities is provided in Table 2.

Underlying liver disease and liver-related decompensations

The most common underlying liver disease among patients who received PC was ALD (56.0%) followed by MASH (29.3%) and hepatitis C (14.6%). Patients who received PC had a higher prevalence of liver-related decompensations, including ascites (59.2% vs. 52.7%), HRS (22.0% vs. 18.6%) and variceal bleeding (11.3% vs. 10.4%), as compared to patients who did not receive PC. A complete list of underlying liver disease and liver-related decompensations is presented in Table 3.

In-hospital interventions and ACLF grades

The burden of in-hospital interventions, including TIPS (1.3% vs. 0.6%) and EGD (16.5% vs. 15.5%), was higher among patients with PC utilization. The highest utilization of PC was seen in patients with Grade 1 ACLF (65.8%) followed by Grade 2 (32.9%) and Grade 3 (1.3%). A complete list of interventions and ACLF grades is presented in Table 4.

Resource utilization

The mean length of stay was comparable between the two groups (9.04±0.09 vs. 8.82±0.09 days) (P=0.08). However, mean total hospitalization charges were lower among patients receiving PC ($190,179.5±2,516.2 vs. $197,410.4±2,813.2) (P=0.03), as compared to those who did not receive PC. In the adjusted analysis, no significant difference in length of stay was observed between the two groups (adjusted coefficient, 0.17; 95% CI: −0.07 to 0.41; P=0.17). However, PC utilization was independently associated with lower total hospitalization charges (adjusted coefficient, −$7,942.93; 95% CI: −$14,065.26 to −$1,820.59; P=0.01).

Additionally, we conducted a sensitivity analysis excluding patients with length of stay ≤2 days to evaluate whether early in-hospital mortality influenced the observed trends in PC utilization. After excluding these patients, the difference in the length of stay was significant, with patients receiving PC demonstrating a shorter length of stay (adjusted coefficient, −0.55; 95% CI: −0.85 to −0.25; P<0.001). Similarly, total hospitalization charges were lower among patients receiving PC (adjusted coefficient, −$22,605.75; 95% CI: −$30,811.82 to −$14,399.67; P<0.001).

Yearly trends of PC utilization

A significant upward temporal trend in PC utilization was observed from 2016 through 2022, increasing from 40.7% to 48.7% (P<0.001) (Figure 2). Even after excluding patients with a length of stay ≤2 days, the increasing temporal trend remained statistically significant.

Figure 2 Yearly trends in palliative care utilization between 2016 and 2022.

Multivariate analysis of factors associated with PC utilization

In multivariate logistic regression, several demographic and clinical factors were independently associated with PC utilization. Female sex was associated with higher odds of PC utilization compared with males [adjusted odds ratio (aOR) 1.06, 95% confidence interval (CI): 1.01–1.11; P=0.01]. Compared with White patients, the adjusted odds of receiving PC were significantly lower among Blacks (aOR 0.71, 95% CI: 0.66–0.76; P<0.001), Hispanics (aOR 0.67, 95% CI: 0.63–0.72; P<0.001) and Asian/Pacific Islanders (aOR 0.85, 95% CI: 0.73–0.99; P=0.04). Age category and insurance status were not significantly associated with PC utilization (P>0.05). A graded association was observed across income quartiles, with progressively higher odds of PC utilization in the second (aOR 1.18), third (aOR 1.26), and highest quartiles (aOR 1.32), each compared with the lowest quartile (P<0.001). Compared with the Northeast, the Midwest and West had 33% and 11% higher odds of PC utilization, respectively. Admission to a teaching hospital (aOR 1.46, 95% CI: 1.37–1.56; P<0.001) were associated with increased PC utilization.

Higher comorbidity burden (mCCI) was associated with greater odds of PC utilization (aOR 1.25, 95% CI: 1.10–1.42; P<0.001). Among comorbidities, diabetes (aOR 0.89; P<0.001), coronary artery disease (aOR 0.84; P<0.001), and chronic obstructive pulmonary disease (aOR 0.94; P=0.04) were associated with lower odds of PC utilization, whereas hypertension (aOR 1.07; P=0.01) and smoking (aOR 1.12; P<0.001) were associated with higher odds. Liver-related decompensations such as ascites (aOR 1.17; P<0.001) and HRS (aOR 1.11; P<0.001) were associated with higher odds of PC utilization. Among underlying liver diseases, ALD (aOR 1.22, P=0.001) and HCC (aOR 1.37; P<0.001) were significantly associated with higher PC utilization, whereas there was no significant association between PC utilization and MASH and viral hepatitis. Regarding in-hospital interventions, patients undergoing LT (aOR 0.51; P=0.001) and blood transfusion (aOR 0.84; P<0.001) were associated with lower odds of PC utilization, while TIPS and EGD showed higher odds, but showed no statistical significance (P>0.05). There was no significant association seen between the grades of ACLF and PC utilization in our analysis. Adjusted associations between covariates and PC utilization are summarized in Table 5 and displayed in Figure 3.

Figure 3 Adjusted odds of palliative care utilization among terminal ACLF hospitalizations: multivariable logistic regression (forest plot). Forest plot displaying aORs and 95% CIs from the multivariable logistic regression model evaluating factors associated with palliative care utilization. Points represent aORs and horizontal bars represent 95% CIs plotted on a logarithmic scale. The vertical reference line indicates no association (aOR =1.0). Reference categories are labeled as “(Ref)” and are plotted at aOR =1.0. Estimates to the right of 1.0 indicate higher odds of palliative care utilization, whereas estimates to the left of 1.0 indicate lower odds. ACLF, acute-on-chronic liver failure; ALD, alcohol-associated liver disease; aOR, adjusted odds ratio; CI, confidence interval; COPD, chronic obstructive pulmonary disease; EGD, esophagogastroduodenoscopy; HCC, hepatocellular carcinoma; HRS, hepatorenal syndrome; MASH, metabolic dysfunction-associated steatohepatitis; mCCI, modified Charlson Comorbidity Index; TIPS, transjugular intrahepatic portosystemic shunt.

Discussion

In this national cohort of 163,060 adult hospitalizations with ACLF ending in in-hospital mortality from 2016 to 2022, PC was documented in fewer than half of patients. Therefore, the observed 46.2% utilization rate reflects underuse of PC among patients who ultimately died during the index hospitalization (P<0.001). This finding is important because the cohort represents terminal ACLF hospitalizations, in which symptom control, prognostic communication, goals-of-care discussions, and family support are especially relevant. Although PC utilization increased over time from 40.7% in 2016 to 48.7% in 2022, the fact that more than half of terminal ACLF hospitalizations had no documented PC encounter underscores a persistent gap in end-of-life care delivery. Prior studies in ESLD and ACLF have similarly shown that PC remains inconsistently used despite high symptom burden, intensive care unit (ICU) utilization, transplant ineligibility, and objective PC needs; however, our study extends these observations by focusing specifically on a national terminal ACLF cohort (20-22).

Demographic and socioeconomic disparities in PC utilization were evident. Black and Hispanic patients had significantly lower adjusted odds of receiving PC than White patients (both P<0.001), consistent with previously reported racial disparities in end-of-life care (20,23). In contrast, insurance status was not independently associated with PC use after adjustment, suggesting that inequities may reflect broader structural, institutional, and socioeconomic barriers rather than payer type alone. These findings align with Mafi and Soldera’s systematic review, which highlighted persistent racial, socioeconomic, provider-level, and transplant-related barriers to timely PC integration in ESLD and ACLF. Together, our results suggest that inequitable PC delivery in terminal ACLF is part of a broader national pattern in advanced liver disease rather than an isolated finding (24).

Beyond patient-level disparities, hospital-level factors further suggest unequal access to PC. Patients treated at teaching hospitals had higher odds of receiving PC (aOR 1.46; P<0.001), likely reflecting greater specialist availability, clinician training, institutional resources, and established referral practices. Because specialty PC availability differs substantially across care settings, standardized referral pathways may help reduce institutional variability (25). In ACLF, practical triggers for PC involvement may include ICU admission, multi-organ failure, mechanical ventilation, transplant ineligibility, or removal from transplant consideration, consistent with guideline recommendations supporting early PC integration in critically ill patients with cirrhosis and ACLF (1,25).

PC utilization was independently associated with lower total hospitalization charges and did not prolong length of stay in the primary analysis. After excluding patients with a length of stay ≤2 days, PC was associated with both lower charges and shorter hospitalization (P<0.001), suggesting that the observed cost difference was not explained solely by early deaths before PC could be initiated. These findings align with prior studies in ESLD, decompensated cirrhosis, and ALD, showing that PC is associated with lower hospitalization costs, fewer resource-intensive interventions, and reduced procedure burden (1,12,20,26-28). However, because PC timing, ICU use, procedures after PC consultation, and goals-of-care decisions are not captured in the NIS, these associations should be interpreted as resource-use signals rather than causal effects.

A notable finding was the lower odds of PC utilization among patients undergoing LT (aOR 0.51; P=0.001). This may reflect the persistent perception that PC is incompatible with transplant-directed or aggressive care, despite its role in symptom management, prognostic communication, and complex decision-making. Prior studies have shown that patients declined or delisted for transplantation rarely receive adequate PC, and that patient-, caregiver-, and provider-level barriers continue to delay referral in ESLD and transplant settings (29-31). In terminal ACLF hospitalizations, this finding may represent missed opportunities for concurrent supportive care, particularly when the prognosis is uncertain.

Together, these findings support the need for earlier and more standardized PC integration in ACLF. Persistent cultural, structural, and clinician-level barriers may delay referral, particularly when transplant eligibility remains uncertain. Standardized referral criteria, such as MELD ≥20, ICU admission, inoperable HCC, multi-organ failure, or transplant ineligibility, may improve consistency in identifying patients who could benefit from PC and reduce provider-driven variability (32). Integrating such referral triggers into routine ACLF care may promote more timely, equitable, supportive care while reducing unnecessary resource-intensive interventions.

The lack of association between ACLF grade and PC utilization was unexpected, as higher ACLF grades generally indicate greater illness severity and supportive care needs. This may be because all patients in this cohort died during hospitalization, making the entire population clinically high-risk and limiting the ability of ACLF grade to further distinguish PC need. It may also reflect that PC referral is influenced more by institutional practices, transplant status, clinician judgment, and service availability than by ACLF grade alone.

This study has limitations inherent to analyses using the NIS. The NIS does not contain laboratory values or key severity markers, preventing calculation of MELD and limiting risk adjustment in ACLF/ESLD; therefore, diagnosis-based proxies may not fully capture physiologic severity or clinical trajectory. Although our ICD-10-CM-based definition was designed to identify ACLF hospitalizations by requiring cirrhosis with at least two coded organ failures, misclassification remains possible. Specifically, the NIS does not provide sufficient clinical detail or temporality to distinguish acute renal dysfunction from chronic kidney disease, or acute hepatic encephalopathy from chronic or recurrent encephalopathy in patients with advanced cirrhosis. As a result, some patients with advanced decompensated cirrhosis and chronic organ dysfunction may have been classified as ACLF, and our findings should be interpreted as applying to an administratively defined ACLF cohort rather than clinically adjudicated ACLF. Administrative coding may also introduce misclassification and temporal variation due to evolving coding practices, variable code specificity for liver disease etiologies, and changes in comorbidity capture across NIS years. In addition, the NIS does not provide hospital-level information on specialty PC service availability; therefore, we could not determine how many patients were treated at hospitals without formal PC programs. Absence of ICD-10-CM code Z51.5 should also not be interpreted as absence of goals-of-care discussions, comfort-focused care, or primary palliative communication by ICU, hepatology, or primary teams, as these may occur without formal PC consultation and may not be consistently coded. Accordingly, documented PC utilization may be underestimated, particularly in hospitals without specialty PC services or with variable documentation practices. Finally, important clinical and social determinants, including social support, substance use history, transplant candidacy, outpatient PC access, and authorization processes, are not captured, and deidentification precludes tracking patient trajectories across care settings. Restricting the cohort to in-hospital mortalities enhances the clinical relevance of the study by focusing on terminal ACLF hospitalizations, where PC is highly pertinent for symptom management and goals-of-care alignment. However, this design limits generalizability to the broader ACLF population and precludes assessment of PC needs, timing, and outcomes across the full ACLF trajectory. Therefore, findings should be interpreted as associations within terminal inpatient ACLF hospitalizations rather than causal effects across the full care continuum.

Conclusions

Our analysis reveals that despite high mortality and critical illness among patients with terminal ACLF hospitalizations in a population with near-certain mortality, fewer than half of patients received PC, indicating a persistent failure of system-level integration rather than simple underuse, highlighting a substantial gap in end-of-life care delivery. PC utilization was inequitably distributed across race, income, hospital region, teaching status, and selected clinical factors, and was associated with lower hospitalization charges without prolonging length of stay. These findings support the need for standardized, equity-focused PC referral pathways, improved clinician training, and broader institutional PC capacity to ensure timely and goal-concordant care for patients with terminal ACLF.

Acknowledgments

This study was previously presented in abstract form at the American College of Gastroenterology, held in Phoenix, Arizona, in October 2025.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-2026-0029/rc

Peer Review File: Available at https://apm.amegroups.com/article/view/10.21037/apm-2026-0029/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-2026-0029/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the ethical standards of the responsible institution regarding human subjects and the principles outlined in the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Karvellas CJ, Bajaj JS, Kamath PS, et al. AASLD Practice Guidance on Acute-on-chronic liver failure and the management of critically ill patients with cirrhosis. Hepatology 2024;79:1463-502. [Crossref] [PubMed]
2. Arroyo V, Moreau R, Jalan R. Acute-on-Chronic Liver Failure. N Engl J Med 2020;382:2137-45. [Crossref] [PubMed]
3. Sacleux SC, Saliba F. How to Optimize the Results of Liver Transplantation for Acute-on-Chronic Liver Failure. Life (Basel) 2023;13:1152. [Crossref] [PubMed]
4. Sarin SK, Choudhury A, Sharma MK, et al. Acute-on-chronic liver failure: consensus recommendations of the Asian Pacific association for the study of the liver (APASL): an update. Hepatol Int 2019;13:353-90. [Crossref] [PubMed]
5. Bajaj JS, O'Leary JG, Reddy KR, et al. Survival in infection-related acute-on-chronic liver failure is defined by extrahepatic organ failures. Hepatology 2014;60:250-6. [Crossref] [PubMed]
6. Mezzano G, Juanola A, Cardenas A, et al. Global burden of disease: acute-on-chronic liver failure, a systematic review and meta-analysis. Gut 2022;71:148-55. [Crossref] [PubMed]
7. Li X, Zhang L, Pu C, et al. Liver transplantation in Acute-on-Chronic liver failure: Timing of transplantation and selection of patient population. Front Med (Lausanne) 2022;9:1030336. [Crossref] [PubMed]
8. Schulz MS, Gu W, Schnitzbauer AA, et al. Liver Transplantation as a Cornerstone Treatment for Acute-On-Chronic Liver Failure. Transpl Int 2022;35:10108. [Crossref] [PubMed]
9. Philips CA, Kedarisetty CK. Palliative Care for Patients with End-Stage Liver Disease. J Clin Exp Hepatol 2023;13:319-28. [Crossref] [PubMed]
10. Patel K, Tandon P, Hernaez R. Palliative Care in the Patient With Acute-on-Chronic Liver Failure. Clin Liver Dis (Hoboken) 2022;19:198-202. [Crossref] [PubMed]
11. Hawley P. Barriers to Access to Palliative Care. Palliat Care 2017;10:1178224216688887. [Crossref] [PubMed]
12. Sohal A, Chaudhry H, Sharma R, et al. Recent Trends in Palliative Care Utilization in Patients With Decompensated Liver Disease: 2016-2020 National Analysis. J Palliat Med 2024;27:335-44. [Crossref] [PubMed]
13. HCUP-US NIS Overview [Internet]. [cited 2025 Nov 9]. Available online: https://hcup-us.ahrq.gov/nisoverview.jsp
14. Allen AM, Kim WR, Moriarty JP, et al. Time trends in the health care burden and mortality of acute on chronic liver failure in the United States. Hepatology 2016;64:2165-72. [Crossref] [PubMed]
15. Celdir MG, Wehby G, Prakash S, et al. July effect in hospitalized cirrhosis patients: A US nationwide study using difference-in-differences analysis. PLoS One 2025;20:e0316445. [Crossref] [PubMed]
16. Moreau R, Jalan R, Gines P, et al. Acute-on-chronic liver failure is a distinct syndrome that develops in patients with acute decompensation of cirrhosis. Gastroenterology 2013;144:1426-37, 1437.e1-9.
17. Sundararajan V, Henderson T, Perry C, et al. New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol 2004;57:1288-94. [Crossref] [PubMed]
18. Bannay A, Chaignot C, Blotière PO, et al. The Best Use of the Charlson Comorbidity Index With Electronic Health Care Database to Predict Mortality. Med Care 2016;54:188-94. [Crossref] [PubMed]
19. Charlson ME, Carrozzino D, Guidi J, et al. Charlson Comorbidity Index: A Critical Review of Clinimetric Properties. Psychother Psychosom 2022;91:8-35. [Crossref] [PubMed]
20. Patel AA, Walling AM, Ricks-Oddie J, et al. Palliative Care and Health Care Utilization for Patients With End-Stage Liver Disease at the End of Life. Clin Gastroenterol Hepatol 2017;15:1612-1619.e4. [Crossref] [PubMed]
21. Poonja Z, Brisebois A, van Zanten SV, et al. Patients with cirrhosis and denied liver transplants rarely receive adequate palliative care or appropriate management. Clin Gastroenterol Hepatol 2014;12:692-8. [Crossref] [PubMed]
22. Vieira da Silva SN, Baptista P, Fonseca Silva I, et al. Palliative Care in Advanced Liver Disease: Similar or Different Palliative Care Needs in Patients with a Prospect of Transplantation? Prospective Study from a Portuguese University Hospital and Transplantation Center. GE Port J Gastroenterol 2023;30:204-12.
23. Puentes JCP, Rocha H, Nicolau S, et al. Effectiveness of the MELD/Na Score and the Child-Pugh Score for the Identification of Palliative Care Needs in Patients with Cirrhosis of the Liver. Indian J Palliat Care 2018;24:526-8. [Crossref] [PubMed]
24. Mafi VIP, Soldera J. Palliative care for end-stage liver disease and acute on chronic liver failure: A systematic review. World J Methodol 2024;14:95904. [Crossref] [PubMed]
25. Rogal SS, Hansen L, Patel A, et al. AASLD Practice Guidance: Palliative care and symptom-based management in decompensated cirrhosis. Hepatology 2022;76:819-53. [Crossref] [PubMed]
26. Barnes A, Woodman RJ, Kleinig P, et al. Hepatobiliary and Pancreatic: Early palliative care referral in patients with end‐stage liver disease is associated with reduced resource utilization. J Gastroenterol Hepatol 2020;35:840-5.
27. Gupta K, Hans B, Khan A, et al. A retrospective study on use of palliative care for patients with alcohol related end stage liver disease in United States. World J Hepatol 2022;14:1817-29. [Crossref] [PubMed]
28. May P, Normand C, Cassel JB, et al. Economics of Palliative Care for Hospitalized Adults With Serious Illness: A Meta-analysis. JAMA Intern Med 2018;178:820-9. [Crossref] [PubMed]
29. Donlan J, Ufere NN, Indriolo T, et al. Patient and Caregiver Perspectives on Palliative Care in End-Stage Liver Disease. J Palliat Med 2021;24:719-24. [Crossref] [PubMed]
30. Beck KR, Pantilat SZ, O'Riordan DL, et al. Use of Palliative Care Consultation for Patients with End-Stage Liver Disease: Survey of Liver Transplant Service Providers. J Palliat Med 2016;19:836-41. [Crossref] [PubMed]
31. Deng LX, Bischoff KE, Kent DS, et al. Frailty is strongly associated with self-reported symptom burden among patients with cirrhosis. Eur J Gastroenterol Hepatol 2021;33:e395-400. [Crossref] [PubMed]
32. Walling AM, Schreibeis-Baum H, Pimstone N, et al. Proactive case finding to improve concurrently curative and palliative care in patients with end-stage liver disease. J Palliat Med 2015;18:378-81. [Crossref] [PubMed]