Associations of 24-Hour Accelerometer-Measured Movement Behaviors with Chronic Liver Disease and the Mediating Role of Proteomics and Metabolomics

Background: This study aimed to investigate the associations of 24-hour movement behaviours with chronic liver disease (CLD) incidence and liver-related mortality. Methods: Data were derived from the UK Biobank cohort, comprising 86,746 participants with accelerometer-measured 24h movement behaviours. We estimated the associations of behaviours with CLD incidence and liver-related mortality using Cox proportional hazard models. We also identified behaviours-associated proteomic and metabolic signatures and their roles in mediating the associations. Results: Moderate-to-vigorous physical activity (MVPA) was inversely associated with liver-related risks, with the hazard ratios (HRs) per standard deviation (SD) increase of 0.74 (0.67-0.80) for CLD, 0.68 (0.61-0.75) for metabolic-associated fatty liver disease (MASLD), 0.81 (0.68-0.95) for cirrhosis, and 0.77 (0.64-0.94) for liver-related mortality, respectively. Sedentary behaviour (SB) was associated with higher risks of incident CLD (HR per SD: 1.12, 95% CI: 1.05-1.19), MASLD (HR per SD: 1.14, 95% CI: 1.06-1.22), cirrhosis (HR per SD: 1.15, 95% CI: 1.01-1.31), and liver-related mortality (HR per SD: 1.27, 95% CI: 1.10-1.47). Reallocating time from other behaviours to MVPA was associated with lower risks of CLD incidence and liver-related mortality. Multi-omics analyses identified distinct proteomic and metabolic signatures for each behaviour, with MVPA-related proteomic signature potentially accounting for 18.2% and metabolic signature potentially accounting for 3.0% of the association between MVPA and CLD risk. Conclusions: A higher proportion of MVPA and a lower proportion of SB over a 24-hour period was associated with lower risks of CLD incidence and liver-related mortality. Proteomic and metabolic signatures of behaviours may partly account for these associations.