Gastroenterology · Medical Literature
[Correspondence] The spectrum of steatotic liver disease in the USA
Abstract
I commend Zobair M Younossi and colleagues1 on their excellent study providing evidence for type 2 diabetes and binge drinking being leading mortality risk factors in steatotic liver disease. When interpreting premature death rates in patients with steatotic liver disease, however, different amounts of exposure to tobacco smoke have to be taken into account. On average, current smoking is associated with dying 12 years prematurely compared with never smoking.2 Major drivers of premature death of people who smoke are cancer, chronic lung disease, and cardiovascular and cerebrovascular diseases.
Clinical relevance
Does tobacco smoking status represent a significant, underappreciated confounding variable that must be accounted for when interpreting premature death rates reported in studies of steatotic liver disease (SLD) in the United States? __SNAPSHOT__ {"studyDesign":"Published correspondence (letter to the editor); not a primary research study. Offers methodological commentary on the Younossi et al. cross-sectional and longitudinal analysis of SLD in the USA (Lancet Gastroenterol Hepatol, March 2026).","setting":"Not applicable (correspondence); the index study (Younossi et al., 2026) was conducted in the USA using national data.","sampleSize":"Not applicable (correspondence). The index study referenced by Scherübl involved a large US national cohort; specific n not extractable from the abstract provided.","population":"Adults in the USA with steatotic liver disease (SLD), including MASLD, MetALD, and ALD subtypes, as defined by contemporary consensus nomenclature.","intervention":"Not applicable. The correspondence highlights tobacco smoke exposure as a potential confounding variable in mortality analyses of SLD populations.","comparator":"Not applicable. The correspondence implicitly compares current smokers versus never-smokers with respect to life expectancy.","primaryOutcome":"Not applicable as a primary study. The correspondence raises the issue of differential tobacco exposure as an uncontrolled confounder of premature mortality estimates in SLD patients.","followUp":"Not reported (correspondence).","fundingSource":"Not reported."}
Key findings
This is a correspondence piece by Scherübl responding to the Younossi et al. (Lancet Gastroenterol Hepatol, 2026) analysis, which identified type 2 diabetes and binge drinking as leading mortality risk factors in SLD. Scherübl argues that differential tobacco smoke exposure must be accounted for when interpreting premature death rates in SLD cohorts. The author cites published evidence that current smoking is associated with dying approximately 12 years prematurely compared with never smoking—a figure consistent with pooled cohort data from NEJM Evidence (Patel et al.) showing mean life-years lost of approximately 12 years for current male smokers versus never-smokers across major disease categories including vascular disease, cancer, and respiratory disease. The CDC additionally notes life expectancy for smokers is at least 10 years shorter than for nonsmokers. The major drivers of smoking-attributable premature death cited in the correspondence are cancer, chronic lung disease, and cardiovascular and cerebrovascular disease—causes that substantially overlap with competing causes of mortality in SLD populations. No new statistics, effect sizes, p-values, or confidence intervals are introduced by the correspondence itself; all cited mortality estimates are drawn from the referenced external literature. NNT/NNH: not applicable. Adverse events: not applicable.
Practice implications
This correspondence does not present original data and should be interpreted solely as expert methodological commentary. The central clinical point is valid and epidemiologically well-grounded: tobacco smoking is a high-magnitude competing risk for premature death (approximately 10–12 years of lost life expectancy) via cancer, COPD, and cardiovascular/cerebrovascular disease—all independent of liver disease. If SLD cohorts contain higher rates of current or former smokers than the general reference population, and if smoking status is not adequately measured and adjusted for in multivariable models, premature mortality attributable to SLD (and specifically to diabetes and binge drinking) may be overestimated or miscategorized by etiology. The findings of the index study (Younossi et al., 2026) that type 2 diabetes and binge drinking are leading mortality risk factors in SLD cannot be dismissed by this commentary, but the correspondence raises a legitimate question about residual confounding by tobacco exposure. Clinicians should note that this critique is common to observational studies of liver disease mortality broadly. Only the abstract of the index study and only the text of this correspondence are available for review; neither the full multivariable model specifications of Younossi et al. nor the degree to which smoking was controlled can be assessed from the information provided. It is therefore not possible to determine from available data whether the index study did or did not adequately adjust for tobacco exposure. __LIMITATIONS__ 1. This is a correspondence, not a primary study; it generates no new evidence and cannot be critically appraised as a clinical trial or cohort study. 2. The correspondence does not provide patient-level data, effect estimates, or a formal statistical argument—the claim about tobacco confounding is qualitative. 3. The referenced '12 years of premature death' figure is a population-level average derived from external cohorts that may not reflect the SLD patient population specifically. 4. It is unknown whether the index study (Younossi et al., 2026) already adjusted for smoking status in its multivariable models, as the full text of that study was not available for review. 5. The correspondence does not quantify the magnitude of potential confounding, making it difficult to assess its practical impact on the index study's conclusions. 6. Generalizability of the smoking-mortality estimate (derived from general population studies) to the SLD population—who may have distinct smoking patterns, competing metabolic comorbidities, and different background mortality rates—is assumed but not established. 7. The commentary conflates average population smoking effects with the SLD subgroup without accounting for disease-specific interaction effects. 8. Funding source not reported. __BOTTOM_LINE_CATEGORY__ low_certainty __BOTTOM_LINE__ This correspondence highlights tobacco smoking as a potentially underappreciated competing cause of premature death that should be rigorously controlled for in observational analyses of SLD mortality; clinicians and researchers should verify that smoking status is adequately measured and adjusted for in any SLD outcomes study before attributing excess mortality exclusively to hepatic risk factors such as type 2 diabetes or binge drinking. The concern is methodologically legitimate but does not invalidate the index study's findings absent evidence that smoking was omitted from multivariable adjustment. No change in clinical practice is warranted based on this correspondence alone.