Metabolic dysfunction-associated fatty liver disease (MAFLD), previously known as non-alcoholic fatty liver disease, is a significant global health concern affecting over one-third of the population. MAFLD is a hepatic manifestation of a multisystem disorder, leading to substantial morbidity and mortality due to its progression to end-stage liver disease and extra-hepatic complications. Currently, there are no approved drugs for MAFLD, and lifestyle modifications remain the cornerstone of management. Healthy lifestyles (HLS), encompassing healthy diet, physical activity, alcohol restriction, and smoking cessation, have been linked to reduced MAFLD risk. Sleep disturbances, increasingly prevalent in modern society, are also emerging as a contributing factor to various metabolic disorders, including MAFLD. While individual sleep behaviors (sleep duration, snoring) have been studied, comprehensive assessments of sleep quality and its interaction with HLS in relation to MAFLD are lacking. This study aims to address these gaps by investigating the complex relationships between sleep quality, traditional HLS, and MAFLD outcomes using data from two independent population-based studies: the Imaging Subcohort of South China Cohort (ISSCC) and the US National Health and Nutrition Examination Survey (US NHANES). The study will assess the mediatory role of sleep quality on the association between HLS and MAFLD, evaluate the synergistic effects of sleep and HLS, and determine if an extended lifestyle metric incorporating sleep quality improves MAFLD risk stratification.

Numerous studies have individually linked elements of a healthy lifestyle (diet, physical activity, alcohol abstinence, and no smoking) with a reduced risk of MAFLD. Similarly, research indicates a correlation between poor sleep quality and an increased risk of metabolic disorders including MAFLD. However, most studies have analyzed these factors in isolation or considered them as simple covariates. A crucial gap remains in understanding how sleep behaviors mediate the effect of traditional HLS on MAFLD risk, and the synergistic interplay between these factors. Previous research often simplifies sleep quality by focusing solely on sleep duration, neglecting the multi-faceted nature of sleep. Therefore, the use of a more comprehensive measure of sleep quality which accounts for sleep duration, insomnia, snoring, daytime sleepiness, and napping is necessary.

This study utilized data from two large, independent, population-based cohorts: the Imaging Subcohort of South China Cohort (ISSCC) and the US National Health and Nutrition Examination Survey (US NHANES). ISSCC included 5011 participants with abdominal ultrasound, while US NHANES included 3672 participants who underwent vibration-controlled transient elastography (VCTE). Sleep quality was assessed using structured questionnaires capturing various sleep behaviors (bedtime, sleep duration, insomnia, snoring, daytime sleepiness, daytime napping). A healthy sleep score was derived from these components, categorized as good, intermediate, or poor. Traditional HLS was assessed based on factors such as smoking status, alcohol consumption, physical activity levels, and diet quality, aligned with World Health Organization recommendations and clinical practice guidelines. Participants were categorized into unfavorable, average, and favorable HLS groups. A new metric, "Liver Essential 5," was created by integrating sleep quality and four traditional HLS factors, generating a score from 0 to 5, indicating a healthier lifestyle. MAFLD was defined using the new international expert consensus definition, incorporating hepatic steatosis (assessed by ultrasound in ISSCC and CAP values by VCTE in US NHANES) and at least one of the following: overweight/obesity, type 2 diabetes, or metabolic dysregulation. At-risk metabolic dysfunction-associated steatohepatitis (MASH) and significant fibrosis were also assessed in US NHANES using FibroScan-AST (FAST) score and liver stiffness. Logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for MAFLD and related outcomes, adjusting for potential confounders. Mediation analysis was conducted to quantify the mediating effect of sleep quality on the relationship between traditional HLS and MAFLD. Stratified analyses and interaction analyses were performed to further explore the relationships.

The study found a significant inverse association between traditional HLS and MAFLD in both cohorts. Sleep quality mediated 4.66–17.72% of this association, with individuals exhibiting poor sleep and unfavorable HLS having significantly higher odds of MAFLD (OR 1.72–2.25). The prevalence of MAFLD decreased with improved sleep quality in both cohorts. Liver Essential 5, a composite score incorporating sleep quality and traditional lifestyle factors, significantly improved risk stratification for MAFLD and related outcomes. Around half of participants initially classified as having favorable traditional HLS were reclassified into higher-risk categories by Liver Essential 5, showing substantially higher MAFLD prevalence. Each one-point increase in Liver Essential 5 was associated with an 18–21% reduction in MAFLD risk (OR 0.82–0.79). Similar trends were observed for at-risk MASH and significant fibrosis. Subgroup analyses revealed stronger associations between Liver Essential 5 and MAFLD in younger participants and those with metabolic syndrome in ISSCC. The associations were consistent across other subgroups, suggesting that promoting Liver Essential 5 might benefit diverse populations.

This study confirms the inverse association between traditional HLS and MAFLD, highlighting the importance of lifestyle interventions. The substantial mediation effect of sleep quality underscores the need to incorporate sleep into comprehensive lifestyle assessments and interventions for MAFLD prevention and management. The development and validation of Liver Essential 5 demonstrates a significant improvement in risk stratification compared to traditional HLS alone. The reclassification of individuals initially considered low-risk based on traditional HLS further emphasizes the added value of integrating sleep quality. The consistency of the findings across two diverse cohorts strengthens the generalizability of the results. Future interventions should consider the synergistic effects of sleep quality and traditional HLS in managing MAFLD risk.

Sleep quality offers a valuable addition to traditional HLS for risk stratification in MAFLD. The Liver Essential 5 metric provides a more comprehensive approach to evaluating lifestyle's impact on liver health. Future research should focus on evaluating the efficacy of interventions that target both sleep quality and traditional lifestyle factors in preventing and treating MAFLD.

The study's cross-sectional design limits causal inference. Self-reported data on lifestyle and sleep quality are susceptible to measurement error, although validated questionnaires were used. Differences in MAFLD definitions and sleep score calculations between cohorts might affect direct comparison. The assumption of equal effect magnitudes of lifestyle factors within Liver Essential 5 might not be entirely accurate. Missing data on covariates were addressed through multiple imputation, but residual confounding remains a possibility despite using directed acyclic graphs. Future prospective studies and randomized controlled trials are warranted to confirm these findings and assess the effectiveness of comprehensive lifestyle interventions.