{
  "abstract": "Background Noninvasive tests (NITs) for detecting liver fibrosis in the community remain limited by suboptimal accuracy, poor generalizability across global regions and scarce evidence supporting longitudinal risk monitoring. This study aimed to develop and validate a novel score for predicting liver fibrosis, liver-related events (LREs), and longitudinal changes in risk in community-dwelling adults.Methods In this prospective, multinational study, adults (≥18 years) were recruited from community-based populations in China, Denmark, the United States, and the United Kingdom. The LiverHealth score (free access: https://liverhealthscore.shinyapps.io/chess/) was developed in a community-based derivation cohort (n=10,790) and externally validated in three independent cohorts from China (n=9,154), Denmark (n=3,004), and the United States (n=4,653), all with liver stiffness measurement (LSM) by transient elastography. Clinically significant fibrosis was defined as LSM ≥8 kPa or histological stage ≥F2. Prognostic performance was evaluated in a UK Biobank cohort (n=454,415) with long-term follow-up, and longitudinal risk changes were assessed in a dynamic cohort (n=17,379).Results The LiverHealth score showed superior performance in predicting LSM ≥8 kPa in the derivation cohort (AUROC=0.85 [0.83-0.86]), validation cohort 1 (AUROC=0.81 [0.78-0.83]), validation cohort 2 (AUROC=0.85 [0.82-0.87]), and validation cohort 3 (AUROC=0.80 [0.78-0.82]), significantly outperforming LiverRisk, LiverPRO, CORE, FIB-4, APRI, and NFS ( IDDF2026-ABS-0348 Table 1). In the biopsy-assessment subgroup (n=2929), the AUROC was 0.91 (0.88-0.93) for histology stage ≥F2 (IDDF2026-ABS-0348 Table 2). Using rule-out and rule-in thresholds (2.535 and 2.789), participants were stratified into low-, medium-, and high-risk groups, achieving the highest sensitivities of 91.5%, 81.3%, 97.3%, and 88.7% across derivation and validation cohorts 1-3. In the prognostic cohort, compared with the low-risk group, the medium- and high-risk groups had significantly higher incidence of LREs (IDDF2026-ABS-0348 Table 3). In the dynamic cohort, risk progression was associated with a higher incidence of LREs, whereas risk regression was associated with a lower incidence (IDDF2026-ABS-0348 Figure 1). Participants classified as high-risk at either baseline or follow-up accounted for 78.4% of liver cirrhosis cases, 62.5% of liver cancer events, and 57.1% of liver-related deaths.Conclusions The LiverHealth score accurately predicted liver fibrosis and LREs in community-dwelling adults. These findings support using the LiverHealth score for early detection of liver fibrosis in community-based screening and prevention programs across diverse populations.Abstract IDDF2026-ABS-0348 Table 1Non-invasive testsAUROC (95% CI)Rule-out zoneRule-in zoneRule-out proportion, No. (%)Rule-out Sensitivity (95% CI)Rule-out Specificity (95% CI)Rule-out negative predictive value (95% CI)Rule-out false-negative rate (95% CI)Rule-in proportion, No.(%)Rule-in Sensitivity (95% CI)Rule-in Specificity (95% CI)Rule-in positive predictive value (95% CI)Rule-in false-positive rate (95%CI)Derivation cohort (n=10,790)LiverHealth0.85 (0.83-0.86)45.0%91.5% (89.5-93.4)48.1% (45.9-50.4)98.5% (98.2-98.8)8.5% (6.6-10.5)13.9%59.0% (55.6-62.4)89.9% (88.8-91.0)33.0% (30.4-35.6)10.1% (9.0-11.2)LiverRisk0.73 (0.71-0.75)95.1%19.1%(16.3-21.8)96.3% (95.9-96.7)93.4% (93.2-93.6)81.0% (78.2-83.7)0.7%5.6% (4.0-7.2)99.7% (99.6-99.8)59.8% (48.5-71.2)0.3% (0.2-0.4)CORE0.70 (0.68-0.72)84.9%37.6% (34.3-41.0)86.8% (86.1-87.5)94.3% (94.0-94.6)62.4% (59.0-65.7)0.8%5.5% (3.9-7.1)99.6% (99.4-99.7)51.0% (40.7-61.3)0.4% (0.3-0.6)FIB-40.60 (0.58-0.62)70.7%41.6% (38.2-45.0)71.8% (70.8-72.7)93.6% (93.2-93.9)58.4% (55.0-61.8)4.1%11.4% (9.2-13.5)96.5% (96.1-96.9)21.5% (17.9-25.2)3.5% (3.1-3.9)APRI0.61 (0.59-0.64)93.0%21.1% (18.3-23.9)94.2% (93.7-94.6)93.4% (93.2-93.6)78.9% (76.1-81.7)0.4%2.8% (1.7-3.9)99.8% (99.7-99.9)51.3% (36.7-65.9)0.2% (0.1-0.3)NFS0.72 (0.70-0.74)59.7%69.1% (65.9-72.3)62.2% (61.2-63.2)96.0% (95.5-96.4)30.9% (27.7-34.1)5.9%22.3% (19.2-25.4)95.5% (95.-96.0)29.6% (26.1-33.1)4.5% (4.0-4.9)Validation cohort 1 (n=9,154)LiverHealth0.81 (0.78-0.83)61.0%81.3% (77.1-85.4)62.6% (61.6-63.6)98.9% (98.6-99.1)18.7% (14.6-22.9)4.5%31.2% (26.3-36.2)96.6% (96.2-97.0)26.1% (22.4-29.9)3.4% (3.0-3.8)LiverRisk0.75 (0.73-0.78)97.1%16.9% (12.9-20.8)97.7% (97.4-98.0)96.8% (96.7-96.9)83.1% (79.2-87.1)0.3%3.2% (1.3-5.1)99.8% (99.8-99.8)44.2% (24.3-64.1)0.2% (0.2-0.2)CORE0.74 (0.71-0.76)83.3%48.4% (43.0-53.7)84.5% (83.8-85.2)97.7% (97.5-97.951.6% (46.3-57.0)0.6%7.0% (4.3-9.7)99.6% (99.5-99.7)41.4% (28.9-54.0)0.4% (0.3-0.5)FIB-40.62 (0.59-0.66)69.3%49.1% (43.9-54.3)70.0% (69.0-71.0)97.3% (97.0-97.6)50.9% (45.7-56.1)2.6%10.1% (6.9-13.2)97.7% (97.4-98.0)14.6% (10.3-18.9)2.3% (2.0-2.6)APRI0.69 (0.66-0.72)95.6%19.2% (15.0-23.4)96.2% (95.8-96.6)96.8% (96.7-97.0)80.8% (76.6-85.0)0.2%2.9% (1.1-4.7)99.9% (99.8-100)47.6% (25.9-69.3)0.1% (0.0-0.2)NFS0.68 (0.64-0.71)77.6%48.7% (43.3-54.0)78.6% (77.8-79.5)97.5% (97.3-97.8)51.3% (46.0-56.7)1.3%5.4% (3.0-7.9)98.8% (98.6-99.0)15.1% (8.8-21.4)1.2% (1.0-1.4)Validation cohort 2 (n=3,004)LiverHealth0.85 (0.82-0.87)20.7%97.3% (95.4-99.2)22.6% (21.1-24.2)98.7% (97.9-99.6)2.7% (0.8-4.6)29.0%77.3% (72.5-82.1)76.3% (74.7-77.9)26.1% (24.3-27.9)23.7% (22.1-25.3)LiverRisk0.81 (0.78-0.84)90.9%41.3% (35.8-46.8)94.5% (93.6-95.3)93.7% (93.1-94.3)58.7% (53.2-64.2)1.7%12.6% (8.8-16.3)99.5% (99.3-99.7)73.5% (61.5-85.5)0.5% (0.3-0.7)LiverPROa0.78 (0.76-0.81)52.1%81.6% (77.2-86.0)55.8% 54.0-57.6)96.6% (95.8-97.4)18.4% (14.0-22.8)9.0%36.6% (31.1-42.1)94.0% (93.1-94.9)39.8% (34.7-44.9)6.0% (5.1-6.9)CORE0.79 (0.76-0.82)65.4%74.5% (69.5-79.5)69.7% (67.9-71.5)96.2% (95.5-96.6%)25.5% (20.5-30.5)3.9%25.6% (20.6-30.6)98.5% (98.0-98.9)64.4% (56.0-72.7)1.5% (1.1-2.0)FIB-40.66 (0.62-0.70)66.9%56.7% (51.0-62.4)69.4% (67.7-71.2)93.6% (92.8-94.5)43.3% (37.6-49.0)2.7%13.3% (9.4-17.3)98.4% (97.9-98.9)47.5% (36.6-58.5)1.6% (1.1-2.1)APRI0.71 (0.68-0.75)91.7%32.1% (26.8-37.5)94.2% (93.3-95.1)92.8% (92.2-93.3)67.9% (62.5-73.2)0.5%4.3% (2.0-6.6)99.9% (99.8-100)79.8% (58.6-100)0.1% (-0.0-0.2)NFS0.73 (0.69-0.7657.9%33.1% (27.7-38.6)92.4% (91.4-93.4)92.7% (92.2-93.3)66.9% (61.4-72.3)3.4%13.4% (9.5-17.3)97.7% (97.1-98.3)38.8% (29.6-48.0)2.3% (1.7-2.9)Validation cohort 3 (n=4,653)LiverHealth0.80 (0.78-0.82)43.2%88.7% (85.9-91.4)47.2% (45.6-48.8)97.1% (96.4-97.8)11.3% (8.6-14.1)16.4%53.5% (49.1-58.0)88.2% (87-89.2)35.9% (33.2-38.6)11.8% (10.8-12.8)LiverRisk0.70 (0.67-0.72)95.3%16.8% (13.4-20.1)96.8% (96.2-97.4)90.4% (90.0-90.7)83.2% (79.9-86.6)0.8%4.1% (2.3-5.9)99.6% (994-99.8)57.3% (39.4-75.1)0.4% (0.2-0.6)LiverPROa0.69 (0.67-0.72)70.5%52.0% (47.6-56.3)73.3% (71.9-74.7)92.5% (91.8-93.1)48.0% (43.7-52.4)4.9%16.9% (13.5-20.2)96.3% (95.7-97.0)36.2% (30.3-42.2)3.7% (3.0-4.3)CORE0.67 (0.65-0.70)87.8%28.2% (24.0-32.3)89.8% (88.8-90.7)91.0% (90.5-91.4)71.8% (67.7-76.0)0.9%5.6% (3.5-7.6)99.7% (99.5-99.9)67.0% (52.1-81.8)0.3% (0.1-0.5)FIB-40.59 (0.57-0.62)73.5%37.2% (32.9-41.6)74.8% (73.5-76.2)90.6% (90.0-91.2)62.8% (58.4-67.1)2.4%6.0% (3.9-8.1)98.0% (97.6-98.5)28.0% (19.6-36.3)1.9% (1.5-2.4)APRI0.61 (0.58-0.64)96.3%12.7% (9.7-15.7)97.5% (96.9-98.0)90.0% (89.7-90.3)87.3% (84.3-90.3)0.4%2.0% (0.8-3.2)99.8% (99.6-100)51.6% (28.5-74.7)0.2% (0.0-0.4)NFS0.70 (0.68-0.73)57.9%68.0% (63.8-72.3)61.2% (59.6-62.7)93.9% (93.2-94.7)32.0% (27.7-36.2)7.0%20.6% (16.7-24.5)94.6% (93.9-95.4)32.2% (27.2-37.2)5.4% (4.6-6.1)Cutoff values: LiverHealth, <2.535 (rule-out) and ≥2.789 (rule-in); LiverRisk, <8 (rule-out) and >12 (rule-in); CORE, <0.004 (rule-out) and >0.05 (rule-in); FIB-4, <1.3 (rule-out) and >2.67 (rule-in); APRI, <0.5 (rule-out) and >1.5 (rule-in); NFS, <-1.455 (rule-out) and >0.676 (rule-in). Data are presented as n (%) or % (95% CI). Results were pooled across 10 imputed datasets using Rubin’s rules. AUROC, area under the receiver operating characteristic curve; NPV, negative predictive value; PPV, positive predictive value; CI, confidence interval; LSM, liver stiffness measurement; FIB-4, Fibrosis-4 index; APRI, aspartate aminotransferase-to-platelet ratio index; NFS, NAFLD fibrosis score.Abstract IDDF2026-ABS-0348 Table 2Summary of the model’s performance in predicting clinically significant fibrosis (histological stage ≥F2) in the biopsy-assessment subgroup (n=2929)Area under the receiver operating characteristic curve (95% CI)0.91 (0.88-0.93)Brier score (95%CI)0.0328 (0.0276-0.0380)Rule-out proportion, No (%)21.2%Rule-out sensitivity (95% CI)99.2% (98.1-100.0)Rule-out specificity (95% CI)22.1% (20.6-23.7)Rule-out negative predictive value (95% CI)99.8% (99.5-100.0)Rule-out false negative rate (95% CI)0.8% (0-1.9)Rule-in proportion, No. (%)27.8%Rule-in sensitivity (95% CI)89.1% (84.0-94.2)Rule-in specificity (95% CI)75.0% (73.4-76.6)Rule-in positive predictive value (95% CI)14.1% (13.0-15.2)Rule-in false positive rate (95% CI)25.0% (23.4-26.6)Abstract IDDF2026-ABS-0348 Table 3Liver-related events (HR [95% CI])Liver-related hospitalization (HR [95% CI])Liver Cirrhosis (HR [95% CI])Liver Cancer (HR [95% CI])Liver-related Death (HR [95% CI])All-cause Death (HR [95% CI]Prognostic cohort Low-risk1 (reference)1 (reference)1 (reference)1 (reference)1 (reference)1 (reference) Medium-risk1.66 (1.58-1.74)1.65 (1.57-1.73)4.08 (3.36-4.95)2.49 (1.71-3.61)4.45 (3.47-5.72)1.89 (1.84-1.93) High-risk5.25 (4.99-5.52)5.10 (4.85-5.38)49.15 (41.16-58.69)34.72 (25.09-48.0452.5 (41.64-66.2)3.83 (3.72-3.95)Dynamic cohortLow to Low1 (reference)1 (reference)1 (reference)1 (reference)1 (reference)1 (reference)Low to Medium1.77 (1.20-2.60)1.73 (1.17-2.57)3.23 (0.20-51.65)Not estimable3.22 (0.45-22.89)2.22 (1.77-2.78)Medium to Low1.05 (0.54-2.02)1.07 (0.55-2.08)Not estimableNot estimableNot estimable1.69 (1.21-2.37)Medium to Medium1.39 (1.00-1.94)1.41 (1.01-1.97)9.12 (1.10-75.76)Not estimable1.52 (0.21-10.77)2.47 (2.05-2.96)Medium to High4.23 (2.75-6.51)4.36 (2.83-6.71)96.18 (12.18-759.16)Not estimable5.29 (0.48-58.33)4.42 (3.39-5.77)High to Medium2.95 (1.52-5.71)3.03 (1.56-5.88)21.40 (1.34-342.12)Not estimableNot estimable2.31 (1.46-3.67)High to High5.87 (4.03-8.56)6.04 (4.13-8.82)189.21 (25.33-1413.43)Not estimable34.41 (7.15-165.64)6.00 (4.75-7.59)Abstract IDDF2026-ABS-0348 Figure 1",
  "authors": [
    {
      "affiliations": [
        "Zhongda Hospital, Medical School, Southeast University, China"
      ],
      "name": "Shanghao Liu"
    },
    {
      "affiliations": [
        "Odense University Hospital, Denmark"
      ],
      "name": "Katrine Lindvig"
    },
    {
      "affiliations": [
        "Affiliated Yancheng Hospital, School of Medicine, Southeast University, China"
      ],
      "name": "Zhenyu Dai"
    },
    {
      "affiliations": [
        "Lianyungang Clinical College of Nanjing Medical University, China"
      ],
      "name": "Hui Shi"
    },
    {
      "affiliations": [
        "Zhongda Hospital, Medical School, Southeast University, China"
      ],
      "name": "Bicheng Ye"
    },
    {
      "affiliations": [
        "Zhongda Hospital, Medical School, Southeast University, China"
      ],
      "name": "Yilin Zhang"
    },
    {
      "affiliations": [
        "The Chinese University of Hong Kong, China"
      ],
      "name": "Terry Cheuk-Fung Yip"
    },
    {
      "affiliations": [
        "Qingdao Public Health Clinical Center, China"
      ],
      "name": "Yuxia Qi"
    },
    {
      "affiliations": [
        "People’s Hospital of Huanghua, China"
      ],
      "name": "Xibin Liu"
    },
    {
      "affiliations": [
        "Taihe Hospital, China"
      ],
      "name": "Yijun Tang"
    },
    {
      "affiliations": [
        "Lishui Central Hospital, China"
      ],
      "name": "JianSong Ji"
    },
    {
      "affiliations": [
        "Guangxi Hospital Division of The First Affiliated Hospital, Sun Yat-sen University, China"
      ],
      "name": "Wenjuan Wang"
    },
    {
      "affiliations": [
        "Sir Run-Run Shaw Hospital, China"
      ],
      "name": "Xiao Liang"
    },
    {
      "affiliations": [
        "Yichun People’s Hospital, China"
      ],
      "name": "Huizhen Fan"
    },
    {
      "affiliations": [
        "General Hospital of Ningxia Medical University, China"
      ],
      "name": "Yang Bo"
    },
    {
      "affiliations": [
        "Bozhou People’s Hospital, China"
      ],
      "name": "Liping Wang"
    },
    {
      "affiliations": [
        "Puxian People’s Hospital, China"
      ],
      "name": "Puqiang Liu"
    },
    {
      "affiliations": [
        "Chinese Academy of Sciences, China"
      ],
      "name": "Guoqing Zhang"
    },
    {
      "affiliations": [
        "Zhongda Hospital, China"
      ],
      "name": "Chuan Liu"
    },
    {
      "affiliations": [
        "Odense University Hospital, Denmark"
      ],
      "name": "Julie Astono"
    },
    {
      "affiliations": [
        "Xuzhou Infectious Diseases Hospital, China"
      ],
      "name": "Kai Hu"
    },
    {
      "affiliations": [
        "Baiyin Central Hospital, China"
      ],
      "name": "Yan Yang"
    },
    {
      "affiliations": [
        "Shenzhen Third People’s Hospital, China"
      ],
      "name": "Qing He"
    },
    {
      "affiliations": [
        "The Second Affiliated Hospital of Guangzhou Medical University, China"
      ],
      "name": "Hui Yang"
    },
    {
      "affiliations": [
        "Mengchao Hepatobiliary Hospital of Fujian Medical University, China"
      ],
      "name": "Yongyi Zeng"
    },
    {
      "affiliations": [
        "Huzhou Central Hospital, China"
      ],
      "name": "Qiang Yan"
    },
    {
      "affiliations": [
        "Medical Department of the People’s Hospital of Homo sapiens, China"
      ],
      "name": "Xiongwei He"
    },
    {
      "affiliations": [
        "Zhengzhou Central Hospital Affiliated to Zhengzhou University, China"
      ],
      "name": "Xingguo Xiao"
    },
    {
      "affiliations": [
        "Affiliated Yancheng Hospital, School of Medicine, Southeast University, China"
      ],
      "name": "Jianxiang Song"
    },
    {
      "affiliations": [
        "Lianyungang Clinical College of Nanjing Medical University, China"
      ],
      "name": "Rong Hu"
    },
    {
      "affiliations": [
        "Qingdao Public Health Clinical Center, China"
      ],
      "name": "Xiuli Han"
    },
    {
      "affiliations": [
        "People’s Hospital of Huanghua, China"
      ],
      "name": "Haiying Feng"
    },
    {
      "affiliations": [
        "Taihe Hospital, China"
      ],
      "name": "Zhongji Meng"
    },
    {
      "affiliations": [
        "Department of Oncology Center The Fifth Affiliated Hospital of Wenzhou Medical University, China"
      ],
      "name": "Yanru Xie"
    },
    {
      "affiliations": [
        "Guangxi Hospital Division of The First Affiliated Hospital, Sun Yat-sen University, China"
      ],
      "name": "Guo Zhang"
    },
    {
      "affiliations": [
        "Sir Run-Run Shaw Hospital, China"
      ],
      "name": "Zhengao Xu"
    },
    {
      "affiliations": [
        "Yichun People’s Hospital, China"
      ],
      "name": "Jianwen Sheng"
    },
    {
      "affiliations": [
        "General Hospital of Ningxia Medical University, China"
      ],
      "name": "Jing Wang"
    },
    {
      "affiliations": [
        "Bozhou People’s Hospital, China"
      ],
      "name": "Shuli Hao"
    },
    {
      "affiliations": [
        "Puxian People’s Hospital, China"
      ],
      "name": "Jianping Zhang"
    },
    {
      "affiliations": [
        "Xuzhou Infectious Diseases Hospital, China"
      ],
      "name": "Hui Cheng"
    },
    {
      "affiliations": [
        "Baiyin Central Hospital, China"
      ],
      "name": "Lingxiao Wang"
    },
    {
      "affiliations": [
        "Baiyin Central Hospital, China"
      ],
      "name": "Pinghua Sun"
    },
    {
      "affiliations": [
        "Baiyin Central Hospital, China"
      ],
      "name": "Jintian Shu"
    },
    {
      "affiliations": [
        "Shenzhen Third People’s Hospital, China"
      ],
      "name": "Fengjuan Chen"
    },
    {
      "affiliations": [
        "The Second Affiliated Hospital of Guangzhou Medical University, China"
      ],
      "name": "Yajuan Zhao"
    },
    {
      "affiliations": [
        "Mengchao Hepatobiliary Hospital of Fujian Medical University, China"
      ],
      "name": "Zuxiong Huang"
    },
    {
      "affiliations": [
        "Huzhou Central Hospital, China"
      ],
      "name": "Zhaowei Tong"
    },
    {
      "affiliations": [
        "Medical Department of the People’s Hospital of Homo sapiens, China"
      ],
      "name": "Jiangwen Liu"
    },
    {
      "affiliations": [
        "Zhengzhou Central Hospital Affiliated to Zhengzhou University, China"
      ],
      "name": "Yanhong Liu"
    },
    {
      "affiliations": [
        "The Chinese University of Hong Kong, China"
      ],
      "name": "Vincent Wai-Sun Wong"
    },
    {
      "affiliations": [
        "Odense University Hospital, Denmark"
      ],
      "name": "Maja Thiele"
    },
    {
      "affiliations": [
        "Zhongda Hospital, Medical School, Southeast University, China"
      ],
      "name": "Gao-Jun Teng"
    },
    {
      "affiliations": [
        "Odense University Hospital, Denmark"
      ],
      "name": "Aleksander Krag"
    },
    {
      "affiliations": [
        "Zhongda Hospital, Medical School, Southeast University, China"
      ],
      "name": "Xiaolong Qi"
    }
  ],
  "title": "IDDF2026-ABS-0348 Development and Validation of liverHealth score for prediction of liver fibrosis and liver-related events in community-dwelling adults: a prospective, multinational study",
  "uid": "4b5c70a6-a4f9-53a0-bbee-11ffd9ebdd2c"
}