<scp>MASLD</scp> as a Disease of Context, Beyond Metabolism

An implicit assumption has long shaped our understanding of metabolic dysfunction–associated steatotic liver disease (MASLD): that risk resides primarily within the individual and can be adequately inferred from metabolic traits such as adiposity, glycemic status and lipid abnormalities. Yet this assumption is becoming increasingly difficult to sustain. Patients with apparently similar metabolic profiles often follow strikingly different clinical trajectories, underscoring not only the biological heterogeneity of MASLD but also the limited explanatory power of risk models built exclusively around metabolic variables [1, 2]. This divergence is unlikely to be a statistical curiosity. Rather, it points to the contribution of determinants that remain invisible within conventional clinical frameworks and supports a broader reappraisal of MASLD beyond the traditional biomedical lens [3]. Against this background, the study by Danpanichkul and colleagues brings a crucial dimension into focus: social vulnerability [4]. Using a large US population-based retrospective cohort and electronic health records, the authors show that, among patients with MASLD, documented social vulnerability was associated with a significantly higher risk of major adverse liver outcomes and a broad range of cardiovascular events over 5 years, including major adverse cardiovascular events, arrhythmia and heart failure, whereas hepatocellular carcinoma (HCC) incidence did not significantly differ between groups. In the propensity score-matched analysis, hazard ratios were consistently in the range of approximately 1.5 to 1.8 across most hepatic and cardiovascular endpoints. Beyond the magnitude of these associations, the true value of the study lies in demonstrating that social vulnerability is not merely associated with disease presence, but exerts a sustained influence on longitudinal disease trajectories and the emergence of adverse clinical outcomes across multiple domains, even after adjustment for conventional clinical and metabolic factors. The consistency of these associations across both hepatic and cardiovascular outcomes suggests that social vulnerability does not act on isolated organs, but within an interconnected biological system. This is fully aligned with the contemporary view of MASLD as a multisystemic disorder embedded within a broader cardiovascular-renal-metabolic continuum, rather than a purely liver-limited condition [5]. In this sense, the study by Danpanichkul et al. provides empirical support for what has largely remained a conceptual proposition: that social context is not external to MASLD, but constitutive of its clinical expression and prognosis [3]. Within this framework, social vulnerability should not be regarded as a peripheral modifier of risk, but as a structural determinant that intersects with metabolic, behavioural and environmental pathways to amplify disease burden, in line with evidence demonstrating that social determinants directly shape risk trajectories and clinical outcomes in complex chronic diseases. Likewise, socioeconomic position has been shown to modulate systemic inflammation, neuroendocrine activation and cardiometabolic susceptibility, ultimately driving divergent trajectories even among individuals with comparable clinical profiles [6]. At the same time, the study also highlights the limitations of current data frameworks [4]. Social vulnerability was identified through administrative coding for ICD-10-CM markers and only 1.21% of the overall MASLD cohort was classified as socially vulnerable. This remarkably low prevalence almost certainly reflects the limited capacity of electronic health records to capture social adversity, rather than its true distribution in clinical populations. As the authors themselves note, individuals with more healthcare encounters are more likely to receive a social determinants of health (SDoH)-related code, whereas those with the greatest barriers to access may remain uncoded. This visibility bias, well documented in electronic health record systems, leads to a systematic underestimation of the impact of SDoH [7]. A further important limitation concerns the assessment of alcohol exposure. Excluding individuals with coded alcohol use disorder is unlikely to capture the full complexity of the MASLD-MetALD-ALD spectrum, particularly in real-world datasets where alcohol intake is often incompletely recorded [8]. In the study by Danpanichkul and colleagues, exclusion was based on diagnostic coding for alcohol use disorder and alcohol-related liver disease, an approach that improves specificity but leaves substantial room for misclassification of patients whose alcohol exposure does not reach diagnostic thresholds or is simply undocumented. This issue is especially relevant because alcohol use is unlikely to be randomly distributed across the population captured as socially vulnerable. In this context, alcohol may operate simultaneously as an unmeasured confounder and as a biological amplifier of risk, further intensifying the interaction between adverse social conditions and metabolic liver disease. A similar argument applies to the wider environment in which MASLD develops. If documented social vulnerability is associated with worse liver and cardiovascular outcomes despite balanced baseline metabolic and clinical characteristics after matching, then part of the excess risk may plausibly reflect exposures that remain poorly captured in administrative datasets, including alcohol use, food insecurity and other structural constraints that shape daily health behaviours. From a broader perspective, these considerations point beyond individual behaviour alone. Social vulnerability does not arise in isolation; it is embedded within commercial, food and policy environments that differentially distribute harmful exposures and limit the capacity for prevention [9]. Within this perspective, contemporary models suggest that the burden of cardiometabolic disease reflects the interaction between social inequality and commercial environments that systematically promote exposure to risk, particularly through alcohol and ultra-processed foods [10, 11]. In this sense, MASLD may be understood not only as a metabolic or even syndemic condition, but also as the biological expression of a wider structural context in which social adversity and unhealthy commercial environments converge to sustain cardio-metabolic and hepatic risk (Figure 1). This does not displace the biomedical model; rather, it expands it by situating individual biology within the system that shapes exposure, vulnerability and resilience. Social vulnerability is also materially expressed through food insecurity, a key determinant of dietary quality and, therefore, of metabolic risk. Socioeconomically disadvantaged environments are consistently associated with energy-dense, nutrient-poor diets that promote insulin resistance and accelerate disease progression [12]. In this context, food should be understood as an active biological modulator, exerting effects comparable to those of pharmacological interventions [13, 14]. Beyond diet and alcohol, MASLD should increasingly be interpreted within a broader exposomic framework, shaped by environmental, behavioural and social exposures that are unequally distributed across populations [15]. This constellation of exposures acquires particular relevance in the Anthropocene, an era defined by rapid environmental transformation and a profound mismatch between human biology and the environments in which it now operates [16]. From this perspective, the liver may be understood as an integrative organ of cumulative exposures, effectively acting as a biological sensor of inequality [1]. The absence of significant differences in HCC incidence observed in the study does not contradict this model; rather, it likely reflects the relatively greater influence of biological determinants in carcinogenesis, contrasted with the stronger impact of SDoH on disease progression and access to therapeutic interventions-domains in which disparities are well established [17]. The study's methodological limitations—including visibility bias in administrative coding and the absence of objective alcohol measurement—must be acknowledged [4]. Yet, rather than undermining its conclusions, these limitations underscore the structural inadequacies of current systems in capturing critical dimensions of disease. Indeed, the study does more than identify an association; it highlights the need to systematically incorporate SDoH into risk stratification and to design targeted interventions for high-risk populations [18]. The clinical implications are therefore unavoidable. Risk stratification in MASLD can no longer rely solely on traditional metabolic variables—an approach increasingly recognized as insufficient to capture the full complexity of cardiometabolic risk [5]. Recent evidence consistently demonstrates that SDoH directly influence exposure, disease progression and clinical outcomes, necessitating their integration into both risk assessment and clinical practice [19]. Accordingly, systematic identification of social vulnerability, the development of structured tools for its measurement, and its incorporation into clinical decision-making should now be viewed as strategic priorities in MASLD care. This also requires moving beyond an exclusively individual-centred model to incorporate interventions targeting the broader environment—including food security, alcohol exposure and living conditions- as integral components of disease management. Ultimately, MASLD is not only a disease of metabolism, but a disease shaped by context. Recognizing, measuring and addressing the social and structural conditions that shape its trajectory is not a theoretical ambition; it is a clinical necessity. To ignore these determinants is to accept a model of medicine capable of describing risk with precision, yet unable to effectively address its most fundamental drivers of disease. The authors have nothing to report. The authors have nothing to report. The authors declare no conflicts of interest. This article is linked to Danpanichkul et al. papers. To view this article, visit https://doi.org/10.1111/liv.70550. This work is an editorial. It does not provide original data. This section does not apply.