Global epidemiology of metabolic dysfunction-associated steatotic liver disease: 2026 update

Abstract

Metabolic dysfunction-associated steatotic liver disease (MASLD) now affects approximately one-third of the global adult population and is the most common chronic liver disease worldwide. An epidemiological update is needed because recent developments have implications for clinical care and treatment of MASLD as a multisystem disease. The epidemic of MASLD is growing. Global Burden of Disease 2023 analyses estimate approximately 1.3 billion prevalent cases, with projections approaching 1.8 billion by 2050, and a disproportionate rise in metabolic dysfunction-associated steatohepatitis (MASH) and advanced fibrosis. MASLD is increasingly recognized as a systemic metabolic disorder with hepatic, cardiovascular, renal, and other extrahepatic manifestations. Recently, the 79^th World Health Assembly adopted a resolution recognizing steatotic liver disease as a major global noncommunicable disease challenge. Concurrently, cardiovascular-kidney-metabolic frameworks and non-hepatology guidelines have incorporated liver fibrosis assessment into evaluations of selected high-risk populations. Advances in non-invasive diagnostics have expanded case-finding, while the availability of MASH-directed pharmacotherapies has shifted the clinical and economic rationale for earlier detection. This narrative review synthesizes the latest epidemiological data on global and regional prevalence and incidence, comorbidities, hepatic and extrahepatic outcomes, advances in screening and prevention strategies, and future directions.

Key Words: Metabolic dysfunction-associated steatotic liver disease; Metabolic dysfunction-associated steatohepatitis; Epidemiology; Comorbidity; Hepatic and extrahepatic outcomes; Screening; Prevention

Core Tip: The epidemiology of metabolic dysfunction-associated steatotic liver disease (MASLD) is rapidly evolving, driven by a rising global burden and shifting patterns across populations. MASLD is no longer viewed as a simple liver-specific disease but as a condition associated with a broad spectrum of hepatic and extrahepatic complications. Advances in non-invasive diagnostics and new pharmacotherapies are reshaping approaches to disease detection and management. This narrative review summarizes recent updates on MASLD epidemiology and discusses their implications for clinical practice.

INTRODUCTION

Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly termed non-alcoholic fatty liver disease (NAFLD), has emerged as the most prevalent chronic liver disease worldwide and a major manifestation of systemic metabolic dysfunction[1,2]. The prognostic impact of MASLD extends far beyond the liver, as it is associated with a broad range of hepatic and extrahepatic complications, including metabolic dysfunction-associated steatohepatitis (MASH), cirrhosis, hepatocellular carcinoma (HCC), cardiovascular disease (CVD), chronic kidney disease (CKD), and certain extrahepatic malignancies, all of which contribute to increased mortality and healthcare costs[3-5].

The epidemiological landscape for MASLD has changed markedly in recent years. The global burden of MASLD continues to rise rapidly. The Global Burden of Disease (GBD) 2023 study estimated approximately 1.3 billion prevalent cases of MASLD worldwide in 2023, a 142.7% increase in crude case numbers since 1990, with projections approaching ~1.8 billion by 2050[6]. Although absolute case numbers and disability-adjusted life-years (DALYs) have risen markedly, age-standardized rates have increased more modestly, underscoring the combined effects of population growth, aging, and changing metabolic risk profiles[7,8]. The global policy environment has shifted accordingly. In May 2026, the 79^th World Health Assembly approved a resolution to that effect, marking the first time steatotic liver disease was explicitly included in the global noncommunicable disease and universal health coverage agenda[9]. MASLD is increasingly recognized within a multi-organ, multisystem framework: The 2026 AHA/ACC/ADA/ASN guideline on cardiovascular-kidney-metabolic (CKM) syndrome underscores the importance of MASLD across the broader spectrum of systemic metabolic dysfunction[10]. MASLD, obesity, type 2 diabetes mellitus (T2DM), and CVD are increasingly viewed as a cardiovascular-liver-metabolic (CLM) syndemic, a synergy of interrelated epidemics that share pathophysiological pathways and socioeconomic determinants leading to an even greater health care burden[11,12]. Furthermore, the cardiovascular-kidney-liver-metabolic (CKLM) framework has gradually emerged, emphasizing shared metabolic risk factors and the interconnected nature of hepatic, cardiovascular, and renal diseases[13-16]. A multidisciplinary primary-care consensus has integrated non-invasive liver fibrosis assessment into routine cardiovascular and metabolic risk stratification[17]. A revolution in non-invasive diagnostic and screening tools has reshaped what epidemiologists can measure at scale[18,19]. Additionally, the introduction of disease-modifying pharmacotherapies for MASH and liver fibrosis, such as resmetirom and semaglutide, has improved the potential cost-effectiveness of targeted screening and strengthened the rationale for earlier detection in selected high-risk groups[20,21]. This narrative review summarizes current evidence on the global and regional epidemiology of MASLD, major comorbidities, hepatic and extrahepatic outcomes, advances in screening, prevention strategies, and priorities for future research.

SEARCH STRATEGY AND SELECTION OF EVIDENCE

We searched PubMed/MEDLINE and EMBASE from January 2015 through June 2026, supplemented by the Cochrane Database of Systematic Reviews and the World Health Organization and GBD online repositories. The principal search terms combined “metabolic dysfunction-associated steatotic liver disease”, “MASLD”, “metabolic dysfunction-associated fatty liver disease”, “MAFLD”, “non-alcoholic fatty liver disease”, and “NAFLD” with “epidemiology”, “prevalence”, “incidence”, “burden”, “fibrosis”, “screening”, and “cost-effectiveness”. Priority was given to large systematic reviews and meta-analyses, GBD analyses, nationally representative studies, major clinical practice guidelines, and policy documents. Reference lists of key articles were hand-searched. We selected studies based on methodological quality, sample size, population representativeness, and relevance to global or regional epidemiology. Given the rapidly evolving nomenclature, searches included both current and historical terminology. Most of the core quantitative evidence was generated during the NAFLD era, so estimates are reported using the terminology of the original studies. Although NAFLD and MASLD populations substantially overlap, they are not identical, and this distinction should be considered when interpreting the estimates.

EPIDEMIOLOGY OF MASLD

Prevalence and incidence of MASLD

Large-scale meta-analyses consistently indicate a high global prevalence of MASLD, though estimates vary substantially across populations and diagnostic methods. A recent analysis reported a pooled prevalence of 33.6% [95% confidence interval (CI): 28.1%-39.5%], with extremely high between-study heterogeneity[22]. Earlier meta-analyses generally estimated the prevalence of MASLD at approximately 29%-33%[23-25]. Differences in age structure, geographic region, diagnostic modality, study period, metabolic risk profile, and sampling strategy limit direct comparisons. A comprehensive meta-analysis of 63 cohort studies encompassing over 4.1 million person-years estimated a global MASLD incidence of 46.2 cases per 1000 person-years (95%CI: 43.2-49.3)[26]. MASLD incidence has risen sharply from approximately 3.7% annually in 1994-2006 to 5.9% in 2010-2014, a 60% increase before plateauing in the most recent period[25].

The expanding impact of MASLD is reflected in global disease burden estimates. The 2023 GBD study estimated that approximately 1.3 billion individuals were living with MASLD worldwide, with a 142.7% increase in crude numbers since 1990 and a 28.6% increase in the age-standardized prevalence rate[6]. Projections indicate that the burden of MASLD will reach ~1.8 billion by 2050. An estimated ~3.6 million DALYs were attributable to MASLD worldwide in 2023, corresponding to an age-standardized DALY rate of 39.6 per 100000 population[6].

Regional epidemiological characteristics

Substantial geographic heterogeneity exists, reflecting differences in metabolic risk factors, population characteristics, and healthcare access (Figure 1)[27]. The prevalence of MASLD is particularly high in Latin America (44.4%), consistent with the region’s high burden of obesity, T2DM, and dietary transitions[25,28]. Similarly, recent MASLD estimates from the Middle East and North Africa (MENA) have exceeded 40%, with high prevalence of obesity, T2DM, and other cardiometabolic risk factors[29,30]. In contrast, the Asia-Pacific region shows a lower overall MASLD prevalence (approximately 28.0%) but distinct epidemiological features, including a higher proportion of lean MASLD, a greater prevalence of genetic susceptibility variants such as the I148M variant of patatin-like phospholipase-3, and a relatively greater contribution of liver-related complications in selected populations[25,31,32]. In Europe, pooled prevalence estimates of MASLD generally fall within 25%-30%, though marked variation exists across countries and population subgroups[25,33]. The reported prevalence of MASLD in Africa among the general population is about 13.5%, but this likely underestimates the true burden because of scarce data, limited diagnostic capacity, low awareness, and challenges in healthcare access[23,34]. In Oceania, Australia has an age-standardized adult MASLD prevalence of 35.9%, and Indigenous and Pacific Island populations face far higher risks of metabolic liver disease than White populations, with a projected 40% increase in MASH cases over the next decade[35].

Figure 1 Global prevalence rates of metabolic dysfunction-associated steatotic liver disease in the general adult population. Data are synthesized from several meta-analyses[23,25,29]. MENA: Middle East and North Africa.

MASLD in children and adolescents

Pediatric MASLD is an emerging public health concern because early-life disease onset may result in prolonged exposure to metabolic and hepatic injury throughout the life course[36]. A recent meta-analysis estimated a global prevalence of MASLD of 7.0% (95%CI: 4.1%-11.7%) among 5-24-year-old, with prevalence increasing with age and calendar year and higher in boys[37].

Among children with overweight or obesity, the pooled prevalence of MASLD was 41.2%[38]. A meta-analysis of children with T2DM reported a comorbidity rate of 36.6%, whereas magnetic resonance-based studies found rates as high as 55.0%[39]. A GBD-based analysis confirmed that global MASLD prevalence among adolescents aged 10-19 years rose from 3.73% in 1990 to 4.71% in 2019, a relative increase of 26.3%[40]. Long-term outcomes of childhood MASLD are cause for concern. The LIVERS study (involving 1096 children with MASLD aged 2-18 years, with an average follow-up of 8.5 years) found that 3.4% of children died, the cumulative incidence of cirrhosis was 4.7%, and nearly half of deaths were liver-related[41]. Childhood MASLD also significantly increases the risk of developing incident T2DM and CVD[42]. A recent cohort study found that adolescents with MASLD had greater carotid intima-media thickness and left ventricular mass and an elevated burden of cardiometabolic disorders, underscoring the importance of early intervention[43]. Historically, MASLD was regarded as a condition predominantly affecting adolescents; however, accumulating evidence now shows that its onset is not confined to this age group, and a substantial burden is increasingly recognized in preschool-aged children[44]. A 2026 Finnish population-based study further linked maternal cardiometabolic factors, childhood dietary patterns, and metabolic profiles to higher alanine aminotransferase trajectories in childhood and adolescence[45]. These findings underscore the developmental origins of MASLD and support early, life-course prevention strategies.

Lean MASLD

Lean MASLD is a significant yet often overlooked subgroup, accounting for approximately 5% to 20% of all MASLD cases globally, with a notably higher prevalence in Asian populations[46,47]. Despite a normal body mass index (BMI), these patients with lean MASLD paradoxically exhibit a high metabolic burden; the prevalence of T2DM and prediabetes is 15.6% and 22.9%, respectively[48,49].

A 2026 study using the Lancet Commission’s new multidimensional obesity criteria found that 84.2% of BMI-defined lean MASLD patients meet obesity criteria when body fat percentage and anthropometric measures are included[50]. These findings suggest that a substantial proportion of BMI-defined lean MASLD may reflect ethnic differences in the relationships among BMI, adiposity, and fat distribution; however, external validation across populations is needed before reclassifying the phenotype more broadly. Nevertheless, a subset of patients with MASLD remains genuinely lean by all anthropometric measures, and understanding the genetic, epigenetic, and gut microbial drivers of this phenotype remains a research priority.

COMORBIDITY BURDEN

Obesity

Obesity is the strongest and most prevalent risk factor for MASLD. A meta-analysis of 151 studies found that the prevalence of MASLD was approximately 70% among adults with overweight, and approximately 75% among those with obesity[51]. Among these groups, approximately 35% had MASH, and severe fibrosis (stages F3-F4) accounted for about 7%[51]. Prevalence of MASLD generally increases with obesity severity, although estimates vary considerably by population setting, diagnostic method, and the distribution of metabolic risk factors[52]. In individuals with severe obesity, the prevalence of MASLD can be as high as 95%[52].

Type 2 diabetes

T2DM is among the highest-risk clinical settings for MASLD. Updated meta-analyses indicate that approximately 65% of adults with T2DM have MASLD, and prevalence has increased over the last decades[53]. Importantly, the burden of progressive liver disease is substantial: Approximately one-third have MASH, around one-third have clinically significant fibrosis, and approximately 15% have advanced fibrosis[54]. These estimates are substantially higher than those in the general adult population, supporting routine fibrosis risk assessment for patients with T2DM. Patients with MASLD and T2DM have a worse prognosis, with a significantly increased risk of all-cause mortality, CVD, CKD, liver-related events, and HCC[55-57].

Hypertension

Hypertension and MASLD are closely intertwined. A meta-analysis of 561 studies found that the prevalence of MASLD is 44.3% among individuals with hypertension[58]. The association between hypertension and MASLD is likely bidirectional[59]. A study of three large cohorts found that hypertension increases the risk of adverse clinical outcomes by 30% [adjusted hazard ratio (HR) = 1.30], liver stiffness progression by 57% (adjusted HR = 1.57), and histologic liver fibrosis progression by 41% (adjusted HR = 1.41)[60]. Population-based studies show increased risk of both prevalent hypertension [odds ratio (OR) = 1.22-1.59] and incident hypertension (HR = 1.26-1.54) with greater steatosis severity[61]. Among patients with MASLD, fibrosis stage ≥ F2 independently predicted new-onset hypertension (HR = 2.39), with incidence rates of 4.6 vs 1.1 per 100 person-years in those without liver fibrosis[61].

Hyperlipidemia

Hyperlipidemia is highly prevalent in MASLD and serves as both a marker of metabolic dysfunction and a major determinant of cardiovascular risk[62]. A meta-analysis found that the prevalence of MASLD among individuals with hyperlipidemia was 59.9% (95%CI: 52.6%-67.0%), the highest among all comorbidities examined[58]. In the Netherlands Epidemiology of Obesity study, MASLD prevalence was 57% among participants with hypertriglyceridemia, rising to 81% with concurrent obesity and to 86% when T2DM and hypertriglyceridemia coexisted[63]. In a large United States cohort, the prevalence of hyperlipidemia was 59.8% in patients with MASLD vs 37.3% in healthy controls[64].

HEPATIC AND EXTRAHEPATIC OUTCOMES OF MASLD

The clinical burden of MASLD stems from both progressive liver disease and extrahepatic complications (Figure 2). Characterizing the frequency and determinants of these outcomes is essential to understanding disease burden and improving risk stratification. Table 1 summarizes meta-analytic estimates of MASLD-related hepatic and extrahepatic outcomes.

Figure 2 Hepatic and extrahepatic complications associated with metabolic dysfunction-associated steatotic liver disease. Metabolic dysfunction-associated liver disease represents a multisystem metabolic disorder in which hepatic disease progression and extrahepatic manifestations arise through overlapping but distinct biological pathways. Hepatic outcomes reflect the progressive spectrum from steatosis to metabolic dysfunction-associated steatohepatitis, fibrosis, cirrhosis, and hepatocellular carcinoma, whereas extrahepatic outcomes represent clinically relevant associations involving cardiovascular, renal, metabolic, oncological, and other organ systems. The magnitude and mechanisms underlying these associations vary across outcomes and remain an active area of investigation. MASH: Metabolic dysfunction-associated steatohepatitis; HCC: Hepatocellular carcinoma; T2DM: Type 2 diabetes mellitus; PCOS: Polycystic ovary syndrome.

Table 1 Meta-analytic estimates of hepatic and extrahepatic outcomes of metabolic dysfunction-associated steatotic liver disease.

Ref.	Study characteristics	MASLD definition	Outcomes	Pooled estimates (95%CI)
Hepatic outcomes
Zamani et al[65], 2025	46 studies; ~8 million adults from general-population cohorts	Liver imaging or blood-based scores	Advanced fibrosis (F3-F4)	Prevalence 3.3% (2.4-4.2)
			Cirrhosis	Prevalence 1.3% (0.9-1.7)
Owrangi et al[66], 2025	35 studies; 513742 MASLD patients	Liver imaging, ICD codes, or histology	Liver-specific mortality	Incidence 2.12 per 100 PY
			Decompensation events	Incidence 7.17 per 100 PY
			HCC	Incidence 4.45 per 100 PY
	General outpatient/primary-care settings (4 studies)		Cirrhosis	Prevalence 3.26% (2.47-4.31)
	High-risk hospitalized or biopsy-referred subgroup (31 studies)		Cirrhosis	Prevalence 14.5% (11.2-18.6)
Le et al[67], 2024	79 studies; 1377466 patients	Liver imaging, ICD codes, or histology	Cirrhosis	Incidence 10.9 per 1000 PY
			HCC	Incidence 3.39 per 1000 PY
Extrahepatic outcomes
Cardiovascular disease
Yang et al[75], 2026	9 studies; 3202447 MASLD patients	Liver imaging, blood-based scores, ICD codes, or histology	Incident CVD	HR 1.33 (1.21-1.46)
Mantovani et al[76], 2021	36 cohort studies; ~5.8 million middle-aged adults across several countries	Liver imaging, ICD codes, or histology	Any fatal or non-fatal CVD event	HR 1.45 (1.31-1.61)
			Fatal CVD events	HR 1.30 (1.08-1.56)
			Nonfatal CVD events	HR 1.40 (1.20-1.64)
			Combined fatal and nonfatal CVD endpoint	HR 1.81 (1.39-2.36)
	Subgroup analyses in patients with advanced MASLD		Fatal CVD events	HR 2.03 (1.17-3.54)
			Nonfatal CVD events	HR 2.30 (1.20-4.42)
			Combined fatal and nonfatal CVD endpoint	HR 2.54 (1.73-3.73)
Myocardial infarction
Lv and Zhao[77], 2024	7 articles comprising 8 prospective and 2 retrospective cohort studies	Liver imaging, blood-based scores, or ICD codes	Incident myocardial infarction	HR 1.26 (1.08-1.47)
Heart failure
Mantovani et al[79], 2022	11 studies; ~12.2 million middle-aged adults across several countries	Liver imaging, blood-based scores, ICD codes, or histology	Incident heart failure	HR 1.50 (1.34-1.67)
Atrial fibrillation
Mantovani et al[80], 2025	16 studies; ~19.5 million middle-aged adults across several counties	Liver imaging, blood-based scores, ICD codes, or histology	Incident atrial fibrillation	HR 1.20 (1.10-1.32)
Chronic kidney disease
Mantovani et al[84], 2022	13 cohort studies; 1222032 individuals	Liver imaging, blood-based scores, ICD codes, or histology	Incident CKD stage ≥ 3	HR 1.43 (1.33-1.54)
Kueh et al[85], 2026	36 studies; 23 prevalence studies (575615 participants; 56248 CKD cases) and 13 incidence studies (27996 incident CKD events)	Liver imaging, blood-based scores, or ICD codes	Prevalence of CKD	15.22% (9.69-23.12)
			Incidence of CKD	22.17 per 1000 PY (11.44-32.89)
Extrahepatic malignancies
Mantovani et al[88], 2022	10 cohort studies; 182202 middle-aged individuals	Liver imaging or ICD codes	Incident oesophagus, stomach, pancreas, colorectal cancers	RR 1.5-2.0
			Incident lung, breast, gynaecological, and urinary system cancers	RR 1.2-1.5
Zhou et al[89], 2024	18 studies; ~16.7 million individuals	Liver imaging, blood-based scores, ICD codes, or histology	Incident gastric, pancreatic, thyroid, urinary cancers	RR 1.4-1.5
			Incident colorectal and female genital cancers	RR 1.3-1.4
Duan et al[90], 2026	29 studies	Liver imaging, ICD codes, or histology	Colorectal polyps	HR 1.42 (1.22-1.66)
			Colorectal cancer	HR 1.22 (1.16-1.29)
			Gastric cancer	HR 1.18 (1.10-1.26)
			Cholangiocarcinoma	HR 1.26 (1.17-1.36)
Musculoskeletal disorders
Hussaini et al[95], 2025	16 studies; individual study sample sizes 57-18815	Liver imaging or ICD codes	Sarcopenia	OR 2.24 (1.74-2.89)
Su et al[96], 2023	19 studies: 9 studies (21294 participants) for bone mineral density, 6 studies (133319) for osteoporosis, 5 studies (227901) for osteoporotic fractures	Liver imaging or histology	Incident osteoporosis	RR 1.28 (1.08-1.52)
			Incident osteoporotic fractures	RR 1.17 (1.00-1.37)
Endocrine disorders
Manzano-Nunez et al[97], 2023	36 studies; 7374 participants	Liver imaging or blood-based scores	MASLD prevalence in PCOS	43% (35-52)
Mantovani et al[98], 2021	33 studies; 501022 individuals	Liver imaging, or histology	Incident T2DM	HR 2.19 (1.93-2.48)
	Subgroup with more severe MASLD			HR 2.69 (2.08-3.49)
	Subgroup with advanced fibrosis			HR 3.42 (2.29-5.11)
Chan et al[99], 2024	129 studies; ~6 million individuals	Liver imaging, blood-based scores, ICD codes, or histology	Incident T2DM	HR 2.56 (2.10-3.13)
	Subgroup with advanced liver disease			HR 3.60 (2.10-6.18)
Neurological and psychiatric
Mou and Yao[101], 2025	16 studies; ~2 million patients	Liver imaging, blood-based scores, ICD codes, or histology	Prevalent depression	OR 1.28 (1.06-1.55)
Lu et al[102], 2022	6 studies; 2345929 patients	Liver imaging, blood-based scores, or ICD codes	Incident dementia	HR 1.04 (1.00-1.08)
Huang et al[103], 2023	7 studies; 879749 patients	Liver imaging or ICD codes	Incident dementia	HR 1.01 (0.94-1.08)
Respiratory diseases
Mantovani et al[104], 2019	6 cohort studies; 133707 individuals	Liver imaging or blood-based scores	Predicted FEV1	WMD -2.43% (-3.28 to -1.58)
			Predicted FVC	WMD -2.96% (-4.75 to -1.17)
Anisa and Adrianison[105], 2025	6 studies; 185432 patients	Liver imaging or ICD codes	Incident OSA	HR 1.65 (1.39-1.96)

MASLD: Metabolic dysfunction-associated steatotic liver disease; CI: Confidence interval; HR: Hazard ratio; ICD: International Classification of Diseases; HCC: Hepatocellular carcinoma; PY: Person-year; CVD: Cardiovascular disease; CKD: Chronic kidney disease; RR: Relative risk; OR: Odds ratio; PCOS: Polycystic ovary syndrome; T2DM: Type 2 diabetes mellitus; FEV1: Forced expiratory volume in 1 second; FVC: Forced vital capacity; OSA: Obstructive sleep apnea; WMD: Weighted mean difference.

MASH, advanced fibrosis, cirrhosis, decompensation and HCC

Liver fibrosis stage is the strongest determinant of liver-related prognosis in MASLD, and the global burden of advanced fibrosis and cirrhosis is rising in parallel with the underlying MASLD epidemic. Before considering MASLD prevalence and progression estimates, it is important to note that without liver histology, MASH prevalence is difficult to assess in the general population.

Liver biopsy remains the reference standard for diagnosing and staging MASH, but it is invasive, potentially subject to sampling variability, and unsuitable for surveillance. No simple, non-invasive gold standard currently exists for MASH diagnosis. This diagnostic limitation complicates estimates of MASH prevalence and progression and contributes to the wide between-study heterogeneity observed in meta-analyses.

The global prevalence of MASH among the general adult population is estimated at approximately 5.3%, with regional variation: Latin America 7.1%, MENA 5.9%, South Asia 5.4%, and Western Europe 4.0%[25]. A meta-analysis of 46 studies (involving approximately 8 million individuals) found that the pooled prevalence of advanced liver fibrosis (F3-F4) in the general adult population was 3.3% (95%CI: 2.4%-4.2%), and the prevalence of cirrhosis was 1.3% (95%CI: 0.9%-1.7%)[65]. Notably, the prevalence has increased significantly since 2016. The prevalence of advanced fibrosis rose from 2.0% to 4.7% (P = 0.004), and the prevalence of cirrhosis rose from 0.8% to 1.8% (P = 0.034), suggesting that the disease burden is accumulating at an accelerated rate[65].

When focusing on the MASLD population, prevalence estimates depend heavily on the study population’s baseline risk. A meta-analysis of the MASLD population (35 studies involving 513742 patients with MASLD) found that in general outpatient/primary care settings, the pooled prevalence of cirrhosis was 3.26% (95%CI: 2.47%-4.31%); among high-risk MASLD patients who were hospitalized or referred for liver biopsy, the prevalence was 14.51% (95%CI: 11.22%-18.57%)[66]. The all-cause mortality rate among patients with MASLD-related cirrhosis was 7.91 per 100 person-years (95%CI: 4.44-13.71), of which the liver-specific mortality rate was 2.12 per 100 person-years[66]. Stratified analysis showed that the mortality rate among patients with compensated cirrhosis was 4.85 per 100 person-years, whereas once they entered the decompensated stage, the mortality rate rose to 12.47 per 100 person-years[66]. The incidence of decompensation events (ascites, variceal bleeding, hepatic encephalopathy, etc.) was 7.17 per 100 person-years, the annual incidence of HCC was 4.45 per 100 person-years, and the liver transplantation rate was 1.84 per 100 person-years[66]. A meta-analysis of the MASLD population (79 studies, 1377466 patients) reported an incidence of cirrhosis of 10.9 per 1000 person-years and of HCC of 3.39 per 1000 person-years[67].

MASLD is increasingly recognized as one of the fastest-growing etiologies of HCC, although hepatitis B virus and hepatitis C virus remain the leading global causes[68]. However, the proportion attributable to MASLD is rising rapidly, driven by declining viral hepatitis prevalence and the expanding pool of metabolic risk factors. In studies using administrative databases that more closely reflect real-world scenarios, the 1-, 3-, 5-, and 10-year cumulative incidence of HCC among individuals with MASLD and advanced fibrosis was 0.8%, 2.4%, 3.9%, and 8.8%, respectively, and among those without advanced fibrosis was 0.1%, 0.5%, 0.7%, and 1.3%[69]. In the VALID cohort, the risk of incident HCC increased by 18% for every 5 kPa increase in liver stiffness (measured by transient vibration-controlled elastography); annual incidence per 100 person-years ranged from 0.34 (LSM 10-14.9 kPa) to 0.94 (≥ 25 kPa)[70]. A Japanese multicenter cohort reported HCC incidence per 1000 person-years: 1.7 (F0-F1), 4.5 (F2), 14.2 (F3), and 16.9 (F4)[71]. Non-cirrhotic HCC is a growing concern, with about one-third of MASLD-HCC cases occurring without cirrhosis[72,73].

CVD

Beyond liver-related outcomes, MASLD consistently affects extrahepatic organ systems, with CVD being the leading cause of mortality[74]. CVD is the leading cause of death in MASLD, accounting for approximately 40% of all deaths[62].

In a large pooled analysis, CVD mortality was 4.20 per 1000 person-years[11]. A meta-analysis of 9 cohort studies (involving 3202447 patients with MASLD) found that MASLD was associated with a 33% higher risk of incident CVD (pooled HR = 1.33, 95%CI: 1.21-1.46)[75]. An earlier updated meta-analysis of 36 prospective cohort studies comprising 5802226 individuals and 99668 incident CVD events over a median follow-up of 6.5 years reported a pooled HR of 1.45 (95%CI: 1.31-1.61), with risk rising markedly across fibrosis stages (pooled HR = 2.54 for advanced fibrosis)[76]. Another meta-analysis of 10 cohort studies focusing on myocardial infarction found that MASLD was associated with a 26% higher risk of incident myocardial infarction (pooled HR = 1.26, 95%CI: 1.08-1.47)[77]. The association extends beyond atherosclerosis to incident heart failure. A cohort study found that individuals with MASLD had a higher risk of new-onset heart failure, particularly heart failure with preserved ejection fraction[78]. A meta-analysis of 11 cohort studies (involving 11242231 individuals) found that MASLD was associated with a 50% increase in new-onset heart failure (HR = 1.50, 95%CI: 1.34-1.67)[79]. MASLD also significantly increases the risk of cardiac arrhythmias. In a meta-analysis, MASLD was significantly associated with an increased risk of incident atrial fibrillation (pooled HR = 1.20, 95%CI: 1.10-1.32)[80].

The relationship between MASLD and CVD is bidirectional: Incident CVD was associated with a higher risk of subsequent liver-related events (adjusted HR = 1.93)[81], and higher predicted CVD risk scores were associated with greater liver fibrosis burden and long-term liver-related outcomes[82]. Fibrosis-4 (FIB-4) index is increasingly recognized not only as a noninvasive liver fibrosis test but also as a biomarker of systemic metabolic and CVD risk. FIB-4 ≥ 2.67 is common in patients with coronary artery disease (CAD) and is associated with higher CAD incidence, stent thrombosis, restenosis, and long-term cardiovascular risk[83]. The 2026 CKM guideline places the FIB-4 index within an integrated CLM risk assessment rather than as a liver-only noninvasive test, thereby strengthening the link between liver epidemiology and CVD risk stratification[10].

CKD

CKD is closely intertwined with both MASLD and CVD, forming a pathological triad within the CKM syndrome framework. A meta-analysis of 13 cohort studies including 1222032 individuals and 33840 cases of incident CKD stage ≥ 3 over a median follow-up of 9.7 years reported a 43% higher long-term risk of incident CKD associated with MASLD (HR = 1.43, 95%CI: 1.33-1.54)[84]. A meta-analysis of 36 observational studies reported a pooled global CKD prevalence of 15.22% (95%CI: 9.69%-23.12%) among individuals with MASLD and a pooled CKD incidence of 22.17 per 1000 person-years (95%CI: 11.44-32.89)[85]. The all-cause mortality rate was 18.28 per 1000 person-years (95%CI: 3.45-33.11) among patients with coexisting MASLD and CKD - more than double that of the MASLD-only population (7.26 per 1000 person-years; 95%CI: 2.97-11.56)[85].

Key risk factors for CKD among patients with MASLD include T2DM (OR = 1.94, 95%CI: 1.69-2.22), hypertension (OR = 1.49, 95%CI: 1.29-1.72), and hyperlipidemia (OR = 1.22, 95%CI: 1.20-1.24)[85]. A Swedish biopsy-proven cohort study of 11082 patients with MASLD with a median follow-up of 17.6 years reported a CKD incidence of 53.1 per 10000 person-years, corresponding to an 85% higher risk than a matched control population (adjusted HR = 1.85, 95%CI: 1.68-2.04). Risk of CKD increased progressively across fibrosis stages from simple steatosis (adjusted HR = 1.69) to cirrhosis (adjusted HR = 2.86)[86]. A systematic review of prospective cohort studies confirmed that liver fibrosis severity is consistently identified as a key determinant of renal impairment in MASLD, significantly increasing the risk of CKD and accelerating its progression[87]. Within the broader CKM syndrome framework, both CKD and CVD are manifestations of the same underlying metabolic-vascular dysfunction that drives MASLD progression. These data support evaluating integrated hepatic, cardiovascular, and renal risk-assessment pathways, but well-designed prospective studies are needed to determine whether these strategies improve long-term clinical outcomes and are cost-effective.

Extrahepatic malignancies

Extrahepatic cancers are an increasingly recognized component of the MASLD disease burden. A meta-analysis of 10 cohort studies comprising 182202 individuals and 8485 incident extrahepatic cancers over a median follow-up of 5.8 years reported a nearly 1.5- to 2-fold increased risk of developing gastrointestinal cancers and a 1.2- to 1.5-fold increased risk of lung, breast, gynecological, and urinary system cancers associated with MASLD[88]. A meta-analysis of 18 cohorts (including ~16.7 million individuals) found approximately 1.4-1.5-fold higher risk for gastric, pancreatic, thyroid, and urinary cancers, 1.3-1.4-fold higher risk for colorectal and female genital cancers, and a more modest 1.2-fold increase for breast cancer in MASLD[89]. A 2026 meta-analysis of 29 cohort studies quantified site-specific gastrointestinal cancer risk: Colorectal polyps (HR = 1.42), colorectal cancer (HR = 1.22), gastric cancer (HR = 1.18), cholangiocarcinoma (HR = 1.26), and gallbladder cancer (HR = 1.41)[90]. For colorectal neoplasia specifically, a 2025 meta-analysis of more than 56 million individuals confirmed that MASLD significantly increased the risk of colorectal adenomas, overall neoplasia, and colorectal cancer, with MASH showing an even higher risk[91]. A multinational propensity-matched cohort study also demonstrated that MASLD was associated with a more than twofold risk of colorectal cancer, and MASH with a fivefold increase[92]. In a large-scale health check-up cohort, MASLD was associated with a higher incidence of colorectal cancer (adjusted HR = 2.11, 1.9 vs 1.1 per 1000 person-years) and, in follow-up colonoscopy, with an increased risk of both incident colorectal neoplasia (HR = 1.89, 5.1% vs 2.7%) and colorectal cancer (HR = 1.64, 1.0% vs 0.7%)[93]. A 2026 Australian cohort with a 19.7-year follow-up found that MASLD was associated with incident colorectal cancer (HR = 2.90), an association that was primarily observed in women[94].

Other systemic manifestations

Beyond cardiovascular, renal, and oncological outcomes, MASLD is associated with an increasing number of underrecognized systemic complications. A 2025 meta-analysis of 16 studies found that patients with MASLD had a higher prevalence of sarcopenia than controls (14.86% vs 6.49%), with a pooled OR of 2.24 (95%CI: 1.74-2.89, P < 0.001; I² = 95%)[95]. Osteoporosis and frailty are underrecognized musculoskeletal burdens in MASLD. A meta-analysis of 19 studies (281136 individuals) found that MASLD was associated with lower bone mineral density (mean difference -0.019 g/cm²) and with increased risk of osteoporosis (HR = 1.28, 95%CI: 1.08-1.52) and osteoporotic fractures (HR = 1.17, 95%CI: 1.00-1.37)[96].

Reproductive and endocrine disorders represent another important aspect of the MASLD burden. A 2023 meta-analysis of 36 studies (7374 participants) found a pooled MASLD prevalence of 43% (95%CI: 35%-52%; I² = 97.2%) among young women with polycystic ovary syndrome (recently renamed polyendocrine metabolic ovarian syndrome), substantially higher than that of age-matched control women[97]. A meta-analysis of 33 observational studies comprising 501022 individuals and 27953 cases of incident T2DM over a median follow-up of 5 years found that MASLD was associated with a twofold increased risk of developing new-onset T2DM (HR = 2.19, 95%CI: 1.93-2.48)[98]. This risk increased progressively with MASLD severity: Individuals with more severe MASLD had an HR of 2.69 (95%CI: 2.08-3.49) for incident T2DM, and those with advanced fibrosis had an HR of 3.42 (95%CI: 2.29-5.11)[98]. A more recent umbrella meta-analysis of 129 Longitudinal studies corroborated and extended these findings, reporting a pooled HR of 2.56 (95%CI: 2.10-3.13) for incident T2DM among patients with MASLD compared with non-MASLD individuals. Subgroup analyses confirmed that individuals with advanced MASLD had a substantially greater risk of incident T2DM (HR = 3.60, 95%CI: 2.10-6.18) than those with less severe MASLD (HR = 1.63, 95%CI: 1.00-2.45)[99].

Neurological and psychiatric manifestations are increasingly recognized. Emerging evidence supports a liver-brain axis in MASLD, linking hepatic steatosis and fibrosis to cognitive impairment, depression, and an increased risk of neurodegenerative diseases[100]. However, current evidence does not support a consistent independent association between MASLD and dementia. A 2025 meta-analysis of 16 studies (~2 million individuals) confirmed that MASLD increased the risk of depression (OR = 1.28, 95%CI: 1.06-1.55)[101]. However, the association between MASLD and dementia remains controversial. A 2022 meta-analysis of 6 cohort studies (2345929 participants) reported a marginal association between MASLD and dementia (pooled HR = 1.04, 95%CI: 1.00-1.08)[102], whereas a 2023 meta-analysis of cohort studies including 879749 individuals found no significant associations of MASLD with all-cause dementia (pooled HR = 1.01, 95%CI: 0.94-1.08) and vascular dementia (pooled HR = 0.99, 95%CI: 0.86-1.13)[103].

Furthermore, MASLD is associated with respiratory diseases. A meta-analysis of 6 observational studies (133707 individuals, predominantly of Asian ethnicity) found that MASLD was associated with significant reductions in predicted forced expiratory volume in 1 second (weighted mean difference: -2.43%, 95%CI: -3.28 to -1.58) and forced vital capacity (-2.96%, 95%CI: -4.75 to -1.17)[104]. A 2025 meta-analysis of 6 Longitudinal cohort studies (185432 participants) provided evidence of a bidirectional relationship between MASLD and obstructive sleep apnea (OSA): Baseline OSA was associated with more than a twofold increased risk of incident MASLD (pooled HR = 2.29, 95%CI: 1.93-2.71), whereas baseline MASLD was associated with a 65% increased risk of developing incident OSA (pooled HR = 1.65, 95%CI: 1.39-1.96). However, the latter estimate was based on only two cohort studies[105].

SCREENING ADVANCES AND COST-EFFECTIVENESS

From liver fibrosis assessment to risk stratification

The epidemiological focus of MASLD has evolved from detecting steatosis to identifying clinically significant fibrosis, and more recently to predicting future liver-related events. Early population studies relied primarily on imaging or biochemical surrogates to estimate steatosis prevalence. The wider adoption of non-invasive fibrosis tests (NITs), including FIB-4 and vibration-controlled transient elastography (VCTE), enabled fibrosis case-finding outside specialist clinics. However, most individuals with MASLD and liver fibrosis will never progress to cirrhosis, while a small subset accounts for most long-term liver-related events. This is where the next technological evolution comes into play: Serial monitoring with NITs and algorithm-derived prognostic scores. Rather than a new device, the advance was methodological: Using large longitudinal cohorts to track NIT trajectories over time and to develop multivariable risk models. The current question is: Which patients will progress to cirrhosis or experience long-term liver-related events?

Serum-based tests

The epidemiological significance of serum-based biomarkers lies not in their diagnostic accuracy alone but in their ability to extend case-finding beyond specialist clinics into primary care and community settings (Figure 3). FIB-4 index, derived from routinely available laboratory values (age, serum transaminases, and platelet count), is the most widely used first-line tool. In low-prevalence settings, the FIB-4 index has a high negative predictive value for advanced fibrosis and is therefore particularly useful as a first-line rule-out test. Its simplicity supports large-scale population screening. The more consequential shift is the emergence of direct fibrogenesis biomarkers and multi-omic panels that may redefine what epidemiologists can measure at scale. Plasma levels of insulin-like growth factor-binding protein 7 have emerged as a promising predictor of long-term liver-related events, but its incremental value, calibration, and generalizability require further validation in independent population-based cohorts before routine epidemiological use[106]. The epidemiological implications are significant: If validated in population-based cohorts, such tools could shift the field from prevalence estimation to incidence tracking and dynamic risk monitoring - the foundation of a true population screening infrastructure.

Figure 3 Non-invasive tests for metabolic dysfunction-associated steatotic liver disease screening and surveillance. This framework summarizes the clinical applications, limitations, and future directions of non-invasive tests for metabolic dysfunction-associated liver disease. Established approaches, including serum-based fibrosis scores and imaging-based assessments, have been incorporated into current clinical pathways for risk stratification and fibrosis assessment. Emerging technologies, such as artificial intelligence-assisted ultrasound, three-dimensional volumetric imaging, and electronic health record-integrated risk assessment, show potential for improving case identification and personalized management but require further validation before broader clinical implementation. The diagnostic performance and clinical utility of each approach depend on the target condition, patient population, technical characteristics, and clinical setting. VCTE: Vibration-controlled transient elastography; MRE: Magnetic resonance elastography; EHR: Electronic health record; BMI: Body mass index; AI: Artificial intelligence; 3D: Three-dimensional.

Imaging-based tests

Imaging has reshaped MASLD epidemiology. VCTE is portable, enabling population-level estimates of fibrosis prevalence and making the liver fibrosis burden a measurable public health parameter. However, measurement feasibility and accuracy may be reduced in individuals with severe obesity or unfavorable body habitus, potentially introducing differential ascertainment bias in population studies[107]. The epidemiological consequence is an inverse detection problem: The populations most likely to have disease are the least reliably captured by the most scalable tool.

Magnetic resonance elastography (MRE) has demonstrated high accuracy, with an area under the receiver operating characteristic curve of 0.92-0.94 for advanced liver fibrosis in MASLD[108]. MRE is considered one of the most accurate non-invasive approaches for liver fibrosis assessment. However, its high cost, need for specialized hardware, and limited availability confine it to research and tertiary centers. MRE’s epidemiological contribution is therefore not to generate population prevalence data, but to anchor our understanding of what VCTE-based estimates mean and how much they may undercount. Head-to-head studies comparing the two imaging modalities suggest that VCTE-derived prevalence estimates are systematically lower than MRE-based estimates in the same populations, implying that current epidemiological data represent a floor rather than a ceiling.

Cost-effectiveness

A major recent development in MASLD epidemiology is the United States Food and Drug Administration (FDA) approval of effective pharmacotherapies for adults with noncirrhotic MASH and moderate-to-severe fibrosis. Resmetirom, an oral, liver-directed, thyroid hormone receptor beta-selective agonist, approved in March 2024, achieved MASH resolution in 26%-30% of patients vs 10% on placebo[109]. Subcutaneous semaglutide 2.4 mg/week, approved in August 2025, achieved MASH resolution in 62.9% vs 34.3% on placebo[110]. These multiple therapeutic mechanisms offer a realistic prospect for personalized therapy tailored to disease stage and phenotype. These FDA approvals significantly alter the screening landscape, as the identification of clinically significant liver fibrosis may now lead to disease-specific pharmacological treatment. This provides targeted screening for MASLD-related fibrosis, offering a clear therapeutic target and potentially improving its cost-effectiveness profile. Thus, MASLD is increasingly a condition that, beyond lifestyle modifications and comorbidity management, may be amenable to targeted pharmacological intervention.

The cost-effectiveness of screening depends on three variables: Test performance, disease prevalence and rates of disease progression in the target population, and, most importantly for epidemiology, the cost of available pharmacotherapies. When effective treatment is expensive, screening must be narrowly targeted to high-prevalence subgroups at high risk of disease progression to be economically justified. Importantly, if treatment costs decline sufficiently, the cost-efficient screening population may broaden beyond the highest-risk subgroups, although expanding to the general population remains unproven. This relationship means that drug pricing policy is, in effect, epidemiological policy: It determines how many people are screened, how much disease is detected, and ultimately the measured prevalence.

A decision-analytic model showed that sequential FIB-4/enhanced liver fibrosis (ELF) screening yielded 8.610 disability-adjusted life-years at a $100000/disability-adjusted life-year threshold in the context of resmetirom ($47400/year)[20]. If drug costs fell below $11570 per year, broader ELF-alone screening became preferred, highlighting a threshold effect in which a single price point determines whether screening is limited to high-risk patients or extended to all individuals with metabolic risk factors[20]. If multiple pharmacotherapies enter the market and competition drives prices down, model-based analyses suggest that the prevalence threshold at which screening could be cost-effective may gradually decline.

The same dynamic applies to pediatric MASLD screening, where waist-to-height ratio-based two-stage strategies were cost-effective across multiple willingness-to-pay thresholds[111]. A possible future development, contingent on lower pharmacotherapy costs, is a gradual shift from opportunistic, targeted case-finding in high-risk groups to broader population-level screening; whether this occurs is uncertain. If screening broadens, measured fibrosis prevalence may rise - primarily because detection improves rather than because the underlying disease becomes more common. Importantly, no randomized clinical-outcome data demonstrate that general-population screening reduces overall or liver-related mortality; current evidence supports targeted case-finding in established high-risk groups, while broader population screening remains a model-based prospect.

PREVENTIVE STRATEGIES

Population-level prevention

The rise in MASLD cannot be addressed solely through individual-level behavioral interventions because obesity and metabolic risk are strongly shaped by food environments, socioeconomic conditions, urban design, and commercial determinants of health[112,113]. The global tripling of obesity rates since 1975 is not a failure of individual willpower but a consequence of food systems and the environment that make ultra-processed, calorie-dense, nutrient-poor foods the default for billions of people[114]. Effective prevention therefore requires structural interventions: Taxes on sugar-sweetened beverages, mandatory front-of-pack food labeling, restrictions on marketing of unhealthy foods to children, and urban planning that promotes active transportation. These policies have proven effective in reducing obesity and T2DM incidence in several countries, and the same mechanisms apply to preventing MASLD[115-117]. In China, the national ‘Weight Management Year’ initiative, launched in 2024, is a major public health effort that addresses conditions, including MASLD, through population-wide awareness campaigns and integrated prevention strategies[118].

The projected growth of MASLD is concentrated in low- and middle-income countries, particularly in sub-Saharan Africa and North Africa and the Middle East, where a rapid nutritional transition is occurring without protective public health infrastructure[6]. These countries face a double burden of undernutrition and overnutrition, with the weakest healthcare infrastructure and the fewest hepatology specialists[119,120]. International organizations and professional societies must prioritize MASLD on global health agendas commensurate with its burden.

Integrated healthcare approaches

The current siloed healthcare organization, with separate departments for hepatology, endocrinology, cardiology, and nephrology, is misaligned with the multisystem nature of MASLD[121,122]. The CKM syndrome guideline and the primary-care CKLM consensus both point toward integrated care pathways that assess hepatic, cardiovascular, renal, and metabolic risks simultaneously[10,17]. Primary care is the natural home for this integration because most patients with MASLD are managed outside specialist hepatology[123]. Embedding fibrosis risk assessment (starting with the FIB-4 index) into routine diabetes and obesity management could dramatically increase case detection at minimal cost.

The recent World Health Assembly resolution further calls for integrating steatotic liver disease into primary health care and universal health coverage packages, with an emphasis on equity and access[9]. Implementing these integrated approaches requires systematic investment in physician training and patient education. Primary care physicians need structured training programs to recognize MASLD as a multisystem disease rather than a liver-specific disease, and to interpret NITs (such as FIB-4) competently in clinical practice. Patient education initiatives should emphasize the multisystem nature of MASLD and the critical role of lifestyle modification, moving beyond the misconception that hepatic steatosis is benign or reversible. Breaking down disciplinary barriers among hepatology, cardiology, endocrinology, and nephrology is essential. Joint clinical pathways, shared electronic health records, and multidisciplinary case conferences can facilitate coordinated care.

Individual-level prevention strategies

At the individual level, lifestyle modifications remain the cornerstone of MASLD prevention and management and are the first-line intervention across disease stages. In clinical trials, structured lifestyle interventions show a consistent dose-response relationship: Even modest weight loss reduces hepatic steatosis; 5%-7% weight loss achieves meaningful hepatic benefits, while 7%-10% loss resolves MASH and regresses early liver fibrosis in most patients[124]. The critical issue is what happens beyond the trial setting. Real-world adherence to lifestyle modifications declines rapidly, and most individuals regain weight within 1 year to 2 years, making the clinically demonstrable hepatic benefits largely unsustainable at the population level. These limitations underscore the role of novel pharmacotherapies as complementary strategies rather than replacements for lifestyle modification[125]. Pharmacotherapies that support weight loss and metabolic improvement, including glucagon-like peptide-1 (GLP-1) receptor agonists (semaglutide) and dual glucose-dependent insulinotropic polypeptide/GLP-1 agonists (tirzepatide), may interrupt the cycle of weight regain by providing sustained appetite suppression and metabolic benefits, independent of the behavioral challenges associated with long-term lifestyle adherence.

The combination of lifestyle modifications and pharmacotherapy is a realistic therapeutic strategy for achieving meaningful hepatic outcomes at the population level, particularly for patients with MASH and advanced fibrosis, who are at the highest risk of disease progression and mortality[126,127]. With the availability of metabolism-based and liver-directed agents, including GLP-1 receptor agonists and MASH-targeted therapies, pharmacological interventions may offer additional opportunities for sustained disease modification, particularly among individuals with MASLD/MASH unable to achieve or maintain sufficient weight loss[128].

FUTURE DIRECTIONS

Improving global epidemiological surveillance

Despite substantial advances, critical gaps still remain in our understanding of MASLD epidemiology. Addressing these gaps will be essential to developing effective prevention and control strategies and to reducing the global disease burden. Current estimates of the global burden of MASLD rely heavily on modeling and extrapolation, which introduce substantial uncertainty. The GBD database, while invaluable, has inherent limitations when applied to MASLD: Many cases remain undiagnosed; diagnostic criteria have changed repeatedly (from NAFLD to MAFLD to MASLD); and population-based prevalence studies using MASLD-specific criteria are scarce in most low- and middle-income countries[6]. The wide confidence intervals in meta-analytic estimates reflect both true heterogeneity across populations and limitations in study quality and methodology. Future efforts should focus on standardizing diagnostic and reporting criteria, conducting population-based prevalence studies in underserved regions, and establishing national and international disease registries to track MASLD incidence, progression, and long-term liver-related and extrahepatic clinical outcomes.

Defining high-risk and understudied populations

Two areas of epidemiological uncertainty warrant attention. First, lean MASLD remains poorly characterized. Although much of what was historically classified as “lean MASLD” may reflect ethnic differences in BMI thresholds and body fat distribution, as suggested by the 84.2% obesity prevalence under the new Lancet Commission criteria[50], a genuine subset of patients with lean MASLD exists. Their natural history, genetic architecture, and optimal management remain unclear. Second, pediatric/adolescent MASLD has seen dramatic advances in prevalence estimation and risk score development, yet it remains constrained by a lack of long-term outcome data. The natural history of MASLD diagnosed in childhood - including progression rates to cirrhosis, HCC, and liver-related mortality in adulthood - remains poorly understood. This represents a critical knowledge gap because pediatric-onset MASLD may follow a more aggressive course than adult-onset disease.

Addressing geographic disparities in MASLD research

There is a marked geographical imbalance in MASLD research. Most epidemiological data come from high-income countries and a handful of middle-income countries. Data from Africa, where the pediatric MASLD burden is projected to rise fastest, are extremely limited. Similar gaps persist across Latin America, the Middle East, and South and Southeast Asia. These regions are experiencing the most rapid increases in obesity and T2DM prevalence and have the weakest healthcare infrastructure to respond to them.

Leveraging digital health and artificial intelligence for MASLD epidemiology

Artificial intelligence (AI) may create new opportunities for MASLD epidemiology and clinical practice across multiple domains. A systematic review of large language models (LLMs) in chronic liver disease identified 18 studies that applied LLMs for diagnostics (HCC detection, MASLD identification, fibrosis staging), prognosis (cirrhosis phenotyping, fibrosis progression), clinical decision support (guideline-aligned prescriptions), and patient education (achieving 70%-90% accuracy on MASLD queries)[129]. Beyond LLMs, machine learning applied to electronic health records, imaging data, and multi-omic datasets promises to: (1) Improve population-level case detection by using routinely collected data; (2) Improve risk stratification beyond current fibrosis scores; (3) Enable precision medicine by identifying MASLD subphenotypes with distinct prognostic and therapeutic profiles; and (4) Optimize healthcare resource allocation through predictive modeling.

The FibroX explainable AI tool for advanced fibrosis detection, currently in pilot testing, represents one of the first efforts to prospectively validate AI in real-world clinical workflows[130,131]. However, most available studies remain retrospective or only internally validated. Before population-level deployment, AI models require external validation across healthcare systems and demographic groups, prospective assessment of clinical utility, calibration and fairness analyses, and evaluation of workflow, privacy, and cost implications. Reported performance on patient-information queries should not be generalized to diagnostic or prognostic accuracy.

CONCLUSION

MASLD has emerged as a multisystem disease in an era in which its epidemiological boundaries extend far beyond the liver to include CVD (the leading cause of death in MASLD), type 2 diabetes, CKD, certain extrahepatic malignancies, sarcopenia, reproductive and endocrine disorders, and emerging neurological manifestations. The central implementation challenge is determining how to integrate risk-stratified case-finding and prevention equitably across healthcare systems while generating evidence on clinical benefit, cost-effectiveness, and feasibility across diverse populations. Bridging the gap between evidence and implementation will determine whether the next decade yields a measurable reduction in MASLD-related morbidity and mortality or allows an already escalating epidemic to continue unchecked.