Autoimmune liver diseases (AILD), which include autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC), are rare, chronic inflammatory disorders characterized by immune-mediated injury to liver and bile ducts. These dysregulated immune responses lead to chronic hepatic inflammation that can result in multi-organ complications. Recent epidemiological studies estimate the incidence of AIH, PBC, and PSC at 1.28 [1], 1.76 [2], and 0.87 cases [3] per 100,000 person-years, respectively.
Although AIH, PBC, and PSC differ in disease characteristics, treatments, and affected populations [4,5], they share overlapping symptoms such as fatigue and are progressive diseases [6]. None of these diseases has a curative medical treatment [4], and patients often require life-long medication while experiencing uncertainty about disease progression and potential complications [7,8]. Despite advancements in treatment [4], many patients with AILD progress to cirrhosis, resulting in impaired liver function and systemic symptoms that ultimately reduce health-related quality of life (HRQOL). Examining HRQOL across all AILD subtypes may provide a more comprehensive understanding of the impact of AILD on HRQOL.
HRQOL has increasingly been recognized as a critical clinical outcome in AILD. It assesses the impact of disease on patients’ physical, psychological, and social functioning [9]. Patients with AILD have lower HRQOL compared to the general population, often reporting difficulty performing daily activities due to fatigue or pain [10,11], and psychological symptoms, including depression and anxiety [12,13]. These challenges contribute to reduced social engagement [14]. Improving HRQOL is clinically important as impaired HRQOL has been associated with disease progression [15,16] and increased mortality [17,18].
Several factors have been identified to be associated with poor HRQOL in this population. They include older age, female sex, low socioeconomic status, greater symptom burden (e.g., fatigue, pain), advanced disease severity (e.g., decompensation events), and comorbid conditions such as depression and diabetes [12,19–21]. However, most studies are limited to a cross-sectional design, and longitudinal trends of HRQOL in the AILD population remain understudied [6,20–22]. Considering the fluctuating and progressive nature of AILD, understanding longitudinal trends of HRQOL may provide essential knowledge to characterize the long-term burden of AILD. Further, identifying factors associated with poor HRQOL over time may help identify patients at higher risk for long-term HRQOL impairment.
This study aimed to assess longitudinal trends in HRQOL among patients with AILD and to identify baseline factors that are associated with HRQOL over time. Findings from this study may support early identification of patients at risk for poor HRQOL and inform the development of targeted strategies to improve long-term patient outcomes.
2Patients and methods2.1Study designThis longitudinal, single-center observational study used data from a prospective autoimmune liver registry at the Liver Research Center of Beth Israel Deaconess Medical Center. Patients who were diagnosed with AILD following the guidelines of the European Association for the Study of the Liver and the American Association for the Study of Liver Diseases were invited to participate in the registry. A total of 545 patients were enrolled between January 2018 and August 2025. At enrollment, participants completed a baseline assessment and follow-up assessments during their regular clinic visits. Among them, those aged 18 years or older who were diagnosed with AIH, PBC, or PSC, and those with overlapping conditions, were included in this study. Patients who had received a liver transplant (N = 39), did not have AIH, PBC, or PSC as primary liver disease (N = 11), did not complete the HRQOL measure (N = 38), completed only the baseline visit or did not complete follow-up HRQOL measure (N = 185), or had data inconsistencies (N = 20) were excluded from this study. This study was conducted following the guidelines of the Declaration of Helsinki and the principles of good clinical practice. All participants provided written informed consent. This study was approved by the Beth Israel Deaconess Medical Center Institutional Review Board (IRB), protocol number 2018P00019, with the most recent consent approval dated April 2, 2025, following the original approval in 2018 and annual renewals thereafter. Informed consent was obtained from all subjects involved in this study.
2.2MeasuresHealth-related quality of life. The Chronic Liver Disease Questionnaire (CLDQ) [23,24], a reliable and valid questionnaire widely used in the liver disease population, was used to assess HRQOL. Participants rated how frequently they experienced symptoms or functional limitations in six domains (Abdominal Symptoms, Activity/Energy, Emotional, Fatigue, Worry, and Systemic Symptoms) in the past 7 days from 1 (All of the time) to 7 (None of the time). Higher scores indicate better HRQOL.
Demographics and clinical characteristics. Demographic variables, such as age, sex, race, ethnicity, education, marital status, and health insurance, and clinical characteristics, including AILD type, overlap syndromes, coexisting liver diseases (e.g., Metabolic-Associated Steatotic Liver Disease [MASLD]), other autoimmune diseases, laboratory values, Model for End-Stage Liver Disease (MELD) 3.0 score, liver stiffness, presence of cirrhosis, history of decompensation episodes, and comorbidities (e.g., hypertension and diabetes), were collected through medical record review.
2.3Data analysisDescriptive statistics were used to summarize baseline demographic and clinical characteristics. Means and standard deviations or medians and interquartile ranges (IQR) were reported for continuous variables, and frequencies and percentages were reported for categorical variables. To examine longitudinal trends in CLDQ total and domain scores over the follow-up period, linear mixed-effects models were used using all available CLDQ scores. Random intercepts were included for each participant to account for within-subject correlation and accommodate unequal numbers of follow-up visits per patient. To identify baseline factors associated with longitudinal CLDQ total scores, univariable and multivariable linear mixed-effects models were performed. Variables with p < 0.05 in univariable analyses and those identified in the literature as relevant factors of HRQOL (e.g., age, sex, disease severity) that had <15% missing data were included in the multivariable model. Months since diagnosis and months since baseline were controlled in all models. Model assumptions were evaluated and found to be adequately met. Normality was supported by Q-Q plots. Variance inflation factors were <5, indicating no substantial collinearity. Analyses were performed using SAS version 9.4.
3Results3.1Patient characteristicsA total of 252 patients with AILD (997 total observations) were included in the study. The mean age was 50.7 years (SD = 14.8), and the majority were female (77.2%) and White (78.5%). AIH was the most common diagnosis (53.6%), followed by PBC (25.0%) and PSC (21.4%). Approximately 20% of patients had autoimmune overlap, most commonly PBC–AIH overlap (87.5% of overlap cases). Coexisting liver disease was present in 21.6% of patients, most frequently MASLD (16.0%). Fewer than 20% of participants had cirrhosis at baseline Table 1.
Patient baseline characteristics (N = 252).
AIH: Autoimmune Hepatitis; AILD: Autoimmune Liver Disease; ALP: Alkaline Phosphatase; ALT: Alanine Transaminase; AST: Aspartate Aminotransferase; IQR: Interquartile Range; kPa: Kilopascals; MASLD: Metabolic-Associated Steatotic Liver Disease; MELD: Model for End-Stage Liver Disease; PBC: Primary Biliary Cholangitis; PSC: Primary Sclerosing Cholangitis; SD: Standard Deviation.
Mean CLDQ scores at baseline and across all follow-up visits are summarized in Table 2. At baseline, the total score and most domain scores were above 5, indicating that symptoms and functional limitations occurred less than “a little of the time.” Only the fatigue domain had a mean score below 5, indicating that fatigue-related symptoms and functional limitations occurred more than “a little bit of the time.”
Summary of CLDQ scores.
CLDQ: Chronic Liver Disease Questionnaire; SD: Standard Deviation.
The mean follow-up duration was 35.3 months (SD = 23.5), with a mean of 4.0 ± 2.2 visits, yielding a total of 745 follow-up observations. Participants with longer (>3 years) follow up duration had similar baseline CLDQ total scores compared to those with shorter (≤ 3 years) follow up (5.53 versus 5.42, p = 0.42). During follow-up, CLDQ total scores showed a small but statistically significant increase over time (β = 0.0018 per month, SE = 0.0008, p = 0.03), corresponding to an estimated improvement of approximately 0.02 points per year (Table 3). Among individual CLDQ domains, emotional and worry showed the largest improvements over time (β = 0.0024 per month, SE = 0.0010, p = 0.01; and β = 0.0057 per month, SE = 0.0013, p < 0.001, respectively), while other domains remained largely stable (Table 3). The longitudinal trajectories of the total, emotional, and worry CLDQ scores are presented in Fig. 1.
Mixed-effects model estimates of CLDQ total and domain scores over time.
CLDQ: Chronic Liver Disease Questionnaire; SE: Standard Error.
In univariable analyses adjusted for months since diagnosis and months since baseline (Table A.1), sex (β = 0.35, Standard error [SE] = 0.14, p = 0.01), presence of other autoimmune disease (β = −0.36, SE = 0.12, p = 0.03), coexisting liver disease (β = −0.40, SE = 0.16, p = 0.01), and ascites (β = −1.12, SE = 0.38, p = 0.003) at baseline were significantly associated with CLDQ total scores (Table 4). These variables, in addition to the variables identified from the literature as being associated with HRQOL, such as demographic and disease severity factors, were included in the multivariable model (193 participants, 753 observations). Male patients had, on average, CLDQ total scores 0.33 points higher than female patients across follow-up (β = 0.33, SE = 0.16, p = 0.04). Patients with other autoimmune disease (β = −0.34, SE = 0.13, p = 0.01), MASLD (β = −0.39, SE = 0.18, p = 0.04), and ascites (β = −1.24, SE = 0.40, p = 0.002) at baseline had, on average, 0.34, 0.39, and 1.24 points lower CLDQ total scores across follow-up, respectively (Table 5).
Univariable mixed-effects model estimates of factors associated with CLDQ total score over time.
*Reference groups: Female (Gender), White (Race), No (Other autoimmune disease), None (Other liver disease), No (Cirrhosis), No (Ascites), No (Esophageal varices), None (Number of comorbidities).
CLDQ: Chronic Liver Disease Questionnaire; MASLD: Metabolic-Associated Steatotic Liver Disease; SE: Standard Error.
Multivariable mixed-effects model estimates of factors associated with CLDQ total score over time.
*Reference groups: Female (Sex), White (Race), No (Other autoimmune disease), None (Other liver disease), No (Cirrhosis), No (Ascites), No (Esophageal varices), None (Number of comorbidities).
CLDQ: Chronic Liver Disease Questionnaire; MASLD: Metabolic-Associated Steatotic Liver Disease; SE: Standard Error.
This study examined longitudinal trends in HRQOL among patients with AILD and identified baseline factors associated with HRQOL over time. HRQOL remained stable over the follow-up period, with small improvements in total scores. The emotional and worry domains showed modest improvement, while other domains remained unchanged. Having other autoimmune disease, MASLD, and ascites at baseline was associated with poorer HRQOL, while male sex was associated with better HRQOL. These findings suggest that while the psychological burden of AILD may improve modestly over time, autoimmune and liver diseases may continue to affect patients’ long-term HRQOL.
This study contributed to the current literature on HRQOL in AILD by describing longitudinal HRQOL trends over an average follow-up of 35.3 ± 23.5 months. HRQOL remained relatively stable during follow-up, with modest improvements in psychological domains. These findings suggest that patients may gradually adapt to living with AILD. Patients with AILD may experience uncertainty about what to expect from their disease, emotional distress, and difficulty managing symptoms that can fluctuate over time, particularly earlier in the disease course [8,25,26]. Although these challenges may persist, patients may gradually learn more about their disease and develop their own ways of coping with it. Over time, they may better understand what worsens their symptoms, what helps them feel better, and how to adjust their routines and priorities in ways that are more manageable, often seeking information and connecting with others with similar experiences [7,25]. This process may help reduce emotional distress over time, contributing to their psychological HRQOL. Indeed, similar patterns of adaptation have been reported in other autoimmune diseases, including systemic lupus erythematosus [27]. However, although the observed changes were statistically significant, their magnitude was small, and their clinical significance should be interpreted with caution.
This study identified several baseline factors associated with longitudinal HRQOL in patients with AILD. Consistent with prior research, female patients were more likely to experience poorer HRQOL than male patients [6,20,22,28]. This finding may be partly explained by higher symptom burden, such as fatigue which is more frequently reported by female patients, even at similar disease severity, as well as sex-related differences in immune responses and in how symptoms are perceived and managed [5,29].
The presence of other autoimmune diseases, MASLD, and advanced liver disease indicated by ascites was associated with poorer HRQOL, whereas other comorbidities such as diabetes and coronary artery disease were not. These findings suggest that conditions related to liver function and immune dysregulation may contribute to greater symptoms and functional limitations. In particular, MASLD has been associated with worse clinical outcomes in AILD populations. For example, patients with both AIH and MASLD had higher one-year mortality and a greater 10-year risk of cirrhosis [30]. Findings of this study highlight the importance of comprehensive management of coexisting autoimmune and metabolic or advanced liver conditions to optimize HRQOL in AILD.
Cirrhosis itself was not significantly associated with HRQOL in this study. Previous research in AILD, most of which were cross-sectional, has shown that decompensated cirrhosis or a history of decompensation episodes is associated with poor HRQOL [20,21], whereas compensated cirrhosis is not [19,28]. Further, studies that did not specify whether cirrhosis was compensated or decompensated have documented poorer HRQOL in patients with cirrhosis [31]. Our findings build on this evidence by showing that baseline ascites, a decompensation episode, was associated with poorer HRQOL over time, suggesting that HRQOL may be influenced more by the symptomatic burden of hepatic decompensation than by fibrosis stage. Ascites is a highly symptomatic complication that can cause abdominal discomfort, pain, dyspnea, and emotional distress, while also requiring ongoing management with diuretics and, in some cases, paracentesis [32]. Prior cross-sectional studies have also shown that ascites is independently associated with poorer HRQOL in patients with cirrhosis [33]. Some patients with compensated cirrhosis, however, may remain asymptomatic, which may explain why cirrhosis itself was not associated with HRQOL in this study [34]. Esophageal varices, although also a decompensation event, were not associated with HRQOL in this study. This may reflect their relatively lower symptomatic burden unless bleeding occurs and the availability of effective therapeutic interventions. Differences in symptomatic burden across complications, rather than decompensation alone, may explain their differential impact on HRQOL. Other decompensation events with substantial symptomatic burden, such as variceal hemorrhage and hepatic encephalopathy, may also adversely affect HRQOL, but this was not evaluated in this study due to the small number of cases.
This study’s longitudinal design with repeated HRQOL assessments over an extended follow-up period enabled us to examine changes in HRQOL. However, a few limitations should be noted. Since the study used data from a single-center registry in the United States, the findings may not be generalizable to broader populations. A substantial number of patients were excluded due to missing follow-up HRQOL data, which may have introduced selection bias. For example, individuals who remained in follow-up may have differed in disease stability or engagement in care. As a result, the findings may not fully reflect the experiences of all patients with AILD. A notable limitation of this study is that psychosocial factors, such as depression, anxiety, and social support, were not included. These factors are known to influence HRQOL [6,28,35], and their absence may have limited our ability to fully explain variability in HRQOL over time. Only variables applicable across all AILD subtypes were included in the analysis. While our analysis did not demonstrate significant differences by AILD subtypes, the influence of subtype-specific factors was not examined in this study. Lastly, since this study examined associations between baseline factors and long-term HRQOL, the findings do not imply causal relationships. Changes in those baseline factors over time and their influences on HRQOL were not assessed.
5ConclusionsIn conclusion, HRQOL in patients with AILD appeared relatively stable over time, with modest improvements observed in psychological domains such as emotion and worry. Female sex, coexisting autoimmune disease, MASLD, and ascites at baseline were associated with poorer HRQOL over time. These findings highlight the need for tailored support for patients who may experience impaired HRQOL due to autoimmune comorbidities or liver disease severity. Future research is needed to characterize HRQOL trajectories and examine how clinical and psychosocial factors influence HRQOL over time in larger, more diverse samples to advance understanding of HRQOL in this population.
Funding statementThe analysis for this study was supported by institutional funds from Northeastern University.
CRediT authorship contribution statementDami Ko: Conceptualization, Project administration, Data curation, Formal analysis, Software, Visualization, Writing – original draft, Writing – review & editing. Vilas Patwardhan: Conceptualization, Methodology, Supervision, Writing – original draft, Writing – review & editing. Natalia Rojas-Amaris: Conceptualization, Data curation. Victoria Abadi-Ron: Conceptualization, Data curation. Ana Marenco-Flores: Conceptualization, Data curation. Romelia Barba: Software, Data curation. Esli Medina-Morales: Data curation. Leandro Sierra: Data curation. Behnam Saberi: Methodology, Supervision, Writing – original draft, Writing – review & editing. Alan Bonder: Conceptualization, Methodology, Project administration, Supervision, Data curation, Writing – original draft, Writing – review & editing.
Alan Bonder consults for Intercept Pharmaceuticals, Ipsen, ChemomAb Ltd., GSK, and Guidepoint. He has received a grant from Gilead Sciences. He has also been the primary investigator for trials sponsored by Gilead Sciences, Cara Therapeutics, Mirum Pharmaceuticals Inc., CymaBay Therapeutics Inc., Genfit, ChemomAb Ltd., and Intercept Pharmaceuticals. He has received royalties or holds licenses from UpToDate and DynaMed. He serves on the editorial committee of DynaMed, and the Clinical Liver Disease Journal for AASLD, and has given expert testimony for Expert Review, Inc. He has also served as a medical reviewer for Pfizer. The other authors do not have conflicts of interest relevant to this work.
The authors thank Darleen Lessard, MS, for statistical support. Statistical analyses were supported by institutional funds from Northeastern University.
