**Background:** Non-alcoholic fatty liver disease (NAFLD) affects approximately 25.4% of the global population and is associated with extra-hepatic conditions including cardiovascular disease, renal disease, and malignancies. Progressive liver fibrosis is linked to increased liver-related and overall mortality, but non-invasive tools for risk stratification are needed. Vibration-controlled transient elastography (VCTE, Fibroscan®) provides liver stiffness measurement (LSM) and controlled attenuation parameter (CAP) for steatosis, but its role as a prognostic tool for mortality lacks guidelines-based recommendation.
**Methods:** This retrospective cohort linkage study included adult patients who underwent VCTE for NAFLD at four Victorian tertiary referral centres (catchment 4.26 million) between July 2008 and April 2019. Of 13,958 VCTE studies, 7079 individual records met inclusion criteria after removing duplicates and selecting the first valid study. Patients with non-NAFLD conditions (excess alcohol, viral hepatitis, autoimmune liver disease, previous transplant) were excluded. VCTE data were linked to the Victorian Death Index (VDI) and Victorian Admitted Episodes Dataset (VAED) via the Centre for Victorian Data Linkage. Cause of death was determined by combined appraisal of VDI and VAED data by a Hepatologist. The Charlson Comorbidity Index (CCI) was calculated from VAED data within 6 months of VCTE. Valid VCTE required >10 acquisitions and IQR ≤30% for LSM ≥7.1 kPa (Boursier's criteria).
**Key Results:** Of 7079 records, 6341 (89.6%) were matched via data linkage, with 5853 (92.3%) having VAED admission data. Median age was 60 years (IQR 49-69), 53.1% female, median LSM 6.1 kPa (IQR 4.7-9.4), median CAP 293 db/m (IQR 240-339). LSM ≥10 kPa (cACLD) was found in 22.9% (1624/7079). Over 22,653 person-years follow-up (median 3.6 years), there were 217 deaths. Most common causes of death: non-HCC malignancy 18.0% (n=39, median LSM 12.3 kPa), sepsis 16.1% (n=35, LSM 12.0 kPa), decompensated liver disease 15.2% (n=33, LSM 32.2 kPa), cardiac disease 15.2% (n=33, LSM 9.6 kPa), HCC 6.0% (n=13, LSM 22.3 kPa). In multivariable analysis, increased LSM (HR 1.02 per kPa, CI 1.01-1.03, p<0.001), CCI (HR 1.32 per point, CI 1.27-1.38, p<0.001), and age (HR 1.05 per year, CI 1.03-1.07, p<0.001) were independently associated with all-cause mortality. LSM ≥10 kPa was associated with mortality in multivariable analysis (HR 2.31, CI 1.73-3.09, p<0.001). CAP was not associated with mortality (HR 1.00, CI 1.0-1.0, p=0.488). Competing risk regression showed LSM remained significant for liver-related death (SHR 1.05 per kPa, CI 1.03-1.06, p<0.001).
**Clinical Implications:** This is the largest study demonstrating that VCTE LSM independently predicts all-cause mortality in NAFLD, with risk occurring along a continuum of liver stiffness. An LSM ≥10 kPa provides a useful threshold for identifying patients at heightened risk who may benefit from early intervention and concentrated health resources. The finding that CAP does not predict mortality reinforces that steatosis severity alone has limited prognostic utility as a single timepoint measure. The study's strengths include large sample size, linkage analysis reducing ascertainment bias, and adjustment for CCI. Limitations include reliance on referral-based NAFLD diagnosis (though 95.6% referred by Gastroenterologists, with 88.3% verified in validation subset), retrospective comorbidity data, and lack of serial VCTE measurements. These data support incorporating VCTE into clinical risk stratification for all-cause mortality in NAFLD patients.