Disparities in Metabolic Dysfunction-Associated Steatotic Liver Disease Prevalence, Diagnosis, Treatment, and Outcomes: A Narrative Review

Abstract

Background

Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as non-alcoholic fatty liver disease (NAFLD), is a leading cause of morbidity and mortality, and health disparities have been shown to influence disease burden.

Aim

In this review, we aim to characterize disparities in prevalence, diagnosis, treatment, and outcomes of MASLD, and to make recommendations for next steps to minimize these disparities.

Methods

Literature search on PubMed and Scopus databases was conducted to identify relevant articles published before September 2, 2024.

Results

Relative to women and White populations, MASLD is more common in men and Hispanic populations and less common in Black populations. It is also more prevalent among those with lower SES. Noninvasive clinical scores may perform differently across groups, and screening practices vary both for initial disease and for progression to metabolic dysfunctionassociated steatohepatitis (MASH), formerly called non-alcoholic steatohepatitis (NASH). Women and Black and Hispanic patients suffer worse outcomes including rates of progression to MASH and mortality.

Conclusions

Health disparities related to race, ethnicity, gender, and socioeconomic factors impact multiple stages of care for patients with MASLD.

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) is an extremely common disease associated with significant morbidity and mortality [1]. Rates of MASLD along with other components of metabolic syndrome have been rising over the last 2 decades, especially among young adults [2, 3]. Disparities exist in MASLD incidence, prevalence, screening, diagnosis, and outcomes along lines of race, ethnicity, sex, and socioeconomic status (SES) [4–6]. These disparities are attributable to a variety of biological factors including genetics, epigenetics, and cellular metabolism, individual factors including lifestyle and environment, and structural factors including provider practice, systemic bias, and access to and quality of healthcare [7–11]. The aim of the current review is to characterize these disparities with the dual aim of informing providers of factors that may contribute to MASLD disparities in their own patient population and identifying areas for future research that may be helpful in their mitigation.

Methods

Nomenclature

The majority of this review was conducted prior to June 2023, when a change in nomenclature from NAFLD to MASLD was announced [12]. Therefore, while the term MASLD is used throughout this manuscript, the analysis was based on search terms and diagnostic criteria for NAFLD. Steatotic liver disease (SLD) refers to a spectrum of disease involving fatty changes to the liver associated with metabolic pathology including early-stage disease represented by steatosis without cellular injury as well as later stage steatohepatitis and cirrhosis.

Search Strategy

This narrative review was conducted with a quasi-systematic approach, in that a single rather than double-reviewer system was utilized. A search string of “NAFLD disparities” was determined to have the optimal balance of sensitivity and specificity for the scope of the present review. PubMed and Scopus were searched concurrently; these results are as of August 16, 2023. Strict inclusion/exclusion criteria were not applied; rather, articles were broadly evaluated for reported findings related to the subject of disparities in MASLD. The PubMed search returned 143 results, of which 54 were eliminated following title and abstract review. The Scopus search returned 90 results, of which 22 were eliminated following title and abstract review. The full text of 92 unique results was reviewed and an additional 38 were eliminated based on a lack of relevance. In total, this primary search returned data from 54 articles to inform this review, 39 of which were present in both database searches, 12 of which were unique to PubMed, and 3 of which were unique to Scopus. A diagram reflecting this search and inclusion process is included in Fig. 1.

Fig. 1.

A diagram displaying the distribution of included articles and the process by which they were evaluated

One final search was done on September 2, 2024 with the purpose of including new evidence that may have been published since the initial search was performed as well as to include articles suggested based on reviewer feedback. This secondary search identified a total of 18 additional articles. All included articles are listed in Online Resource 1.

Review and Integration of Findings

Each included article was evaluated by a single reviewer (BG or KCM), who identified relevant component findings. Findings were then collated by subject matter and integrated. In areas where data specific to MASLD were lacking, supporting articles that discuss themes and populations that relate to health disparities in MASLD (e.g., diabetes, obesity, etc.) have also been cited.

Results

Differences exist in the incidence and prevalence of MASLD by race and ethnicity, sex, and SES, and intersectional relationships across these axes are also present. Along with epidemiological differences in disease prevalence, differences exist in the accuracy of and rates at which patients undergo screening for and diagnosis of MASLD. Disparities in screening and diagnosis may be related to social factors such as provider biases, differences in provider practice, and resource availability, but also may be related to the varied performance of biochemical markers and clinical tools used to screen for and monitor MASLD across different populations [13–16]. In screening for the progression of disease, practices may differ according to the resources available within the healthcare setting, individual provider practices, and patient race/ethnicity, but research in this area is sparse [15, 17]. Further, approaches to treatment vary widely across providers, which has been attributed in part to lack of adherence to published guidelines, provider bias, and healthcare access; however, further research is needed regarding management practices, particularly as viable treatment options continue to emerge [17, 18].

Component Studies

Roughly half of included studies were observational (40, 55.6%), while review articles (23, 31.2%), experimental studies (6, 8.3%), and editorials (3, 4.2%) comprised the remaining papers (Fig. 2). Reviews and editorials were not utilized as a significant source of primary data; however, their aggregation of multiple studies on a topic facilitated drawing broader and more widely supported conclusions. The clinical focus of component articles as well as the axes of disparities addressed by each is found in Figs. 3 and 4, respectively.

Fig. 2.

Component studies by type

Fig. 3.

Number of articles by subject

Fig. 4.

Number of articles by axis of disparity

The majority of studies were observational; numerous review studies were also included. Few component studies were experimental in nature. Of the 72 studies included in this review, the majority addressed disparities in incidence and prevalence and many articles discussed outcomes. Relatively fewer papers included information pertaining to screening/diagnosis, management/treatment, and research disparities. Race/ethnicity was the most commonly discussed axis of disparity among the studies included in this review. Sex was also frequently addressed, while socioeconomic status and other axes along which disparities exist, such as the presence of comorbidities or body mass index (BMI), were identified in fewer articles.

Disparities exist across MASLD prevalence, screening, diagnosis, treatment, and outcomes. The factors contributing to disparities in each of these domains are distinct but related and are summarized in the concept map in Fig. 5. Below we discuss these factors organized by the axis of disparity.

Fig. 5.

A concept map illustrating the relationship between various factors that have been demonstrated to contribute to disparities in MASLD. Factors contributing to disparities in incidence and prevalence, screening and diagnosis, and treatment all contribute to disparate outcomes

Disparities by Race and Ethnicity

Incidence and Prevalence

It is well established that among adults in the United States, Hispanic populations have the greatest burden of MASLD, followed by non-Hispanic White populations, and non-Hispanic Black populations [19]. The risk of disease was quantified in a 2018 systematic review and meta-analysis of 34 studies, that found, compared to non-Hispanic White patients, a relative risk of MASLD of 1.36 (95% CI = 1.08–1.73) for Hispanic patients and 0.68 (95% CI = 0.54–0.84) for non-Hispanic Black patients [6]. Mexican Americans also have a higher risk of MASLD than other Hispanic Americans [20]. Limited research suggests that Asian and non-Hispanic White populations have a similar prevalence of disease [11]. While the present review focuses primarily on MASLD in the United States, it is worth noting that prevalence is high globally, and compared to North America, disease burden is even higher in Latin America, the Middle East and North Africa, and Southeast Asia [21].

Additionally, there is substantial overlap in patients with MASLD and metabolic syndrome, and with the shift in nomenclature from NAFLD to MASLD, patients must now have at least one component of cardiometabolic disease to meet criteria for diagnosis of MASLD [12]. However, the relationship between MASLD and metabolic syndrome is not straightforward and is modified by a number of other variables, including race. For example, the higher prevalence of metabolic comorbidities such as obesity and diabetes among Hispanic individuals likely contributes to their higher burden of MASLD [22]. However, given that Black individuals have a higher prevalence of obesity and insulin resistance but lower prevalence of MASLD compared to Whites, and that White patients with MASLD are more likely to have metabolic syndrome than Black patients with MASLD (AHR = 2.05; 95% CI = 1.95–2.15 vs. AHR = 1.76; 95% CI = 1.58–1.96), other factors must also be contributing [23–26]. Socioeconomic and genetic factors have been implicated, as discussed later on, though the relationship between race/ethnicity and MASLD is complex and incompletely understood. Similarly, diabetes on its own is a predictor of MASLD and MASH in White (OR = 3.12; 95% CI = 1.50–6.51) but not Black (OR = 1.80; 95% CI = 0.66–4.96) patients [28]. In a recent study by Shaheen et al. [27], disparities in MASLD severity as measured by transient elastography were noted across racial/ethnic groups with normoglycemia (compared to non-Hispanic White patients, aOR = 3.1; 95% CI = 1.4–6.9 in Mexican American patients; aOR = 0.4; 95% CI = 0.3–0.6 in non-Hispanic Black patients), but not consistently in those with prediabetes or diabetes. The lower prevalence of disease in non-Hispanic Black populations may be related to the existence of an insulin resistance paradox wherein Black patients, on average, have higher insulin resistance, yet are less likely to develop MASLD, than Hispanic patients [28]. This could be due to differences in lipid homeostasis, triglyceride levels, visceral fat accumulation, and/or behavioral risk factors [28]. For example, Black patients tend to have less visceral fat accumulation and lower plasma triglyceride levels than Hispanic patients [28]. Mechanisms for racial/ethnic differences in insulin resistance as it relates to MASLD may be explained by (a) “liver-outward” pathways in which hepatic steatosis triggers insulin resistance, or (b) “periphery inward” cascades in which racial/ethnic differences in insulin-mediated suppression of adipose tissue leads to differential lipid storage and metabolism in the liver relative to skeletal muscle [29].

Genetics also seem to play a role in differences in MASLD prevalence by race and ethnicity. The increased risk of disease in Hispanic populations and lower risk in Black populations is thought to be related in part to a polymorphism of the PNPLA3 gene, which codes for a triacylglycerol lipase active in hepatic adipocytes. The rs738409:G polymorphism of this gene leads to increased hepatic fat accumulation and inflammation and increases the risk of the onset of MASLD, and this polymorphism is more frequent in individuals of Hispanic origin [1, 5, 11, 30, 31]. Among Hispanic individuals in the United States, Mexican Americans have both the highest frequency of the rs738409:G allele (55%) and the highest prevalence of MASLD (22%), and Puerto Rican individuals have both the lowest frequency of the allele (32%) and a lower prevalence of MASLD compared to most other Hispanic populations (15.8%) [11]. One small study also found a high frequency of the rs738409:G allele among individuals with Hmong ethnicity [32]. Compared to non-carriers, homozygotes for this polymorphism had more than twofold higher hepatic fat levels [31]. Conversely, a different variant of the PNPLA3 allele (rs6006460:T) has been found to be associated with lower hepatic fat content and was commonly found in African Americans but rarely found in White or Hispanic individuals [31]. There is debate as to the extent to which genetics impact the prevalence of MASLD. Using participant data from the Dallas Heart Study, it was estimated that these two PNPLA3 alleles accounted for 72% of observed ancestry-related differences in hepatic fat content [31]. Other studies have reported that variation in the PNPLA3 allele may account for 2 to 3% of total variance in MASLD [33–35]. Numerous other genes have also been implicated in the development of MASLD such as TM6SF2, GCKR, PPP1R3B, TMC4/MBOAT7, NCAN, and LYPLAL1, among others [1, 13, 30, 33, 35–39].

Several other factors may also contribute to racial/ethnic differences in MASLD incidence and prevalence. Racial differences in inflammatory cascades and vitamin D metabolism have been described; Black patients demonstrate higher levels of the active metabolite 1,25-dihydroxy vitamin D despite lower levels of precursor 25-hydroxy vitamin D as well as differential binding affinity between vitamin D receptors and their associated ligands [40]. These differences may be related to MASLD pathogenesis due to the role of vitamin D in regulating free fatty acid metabolism and the induction of the hepatic inflammatory response [40]. Further, childhood hepatic fat accumulation may vary by ethnicity, with implications for pre-pregnancy factors such as maternal age, education level, household income, pre-pregnancy BMI, marital status, and parity, as well as pregnancy factors such as maternal calorie intake, smoking status, alcohol use, and pregnancy complications. These factors appear to have a greater impact on subsequent hepatic fat content in certain ethnicities; in a study of school-aged children in the Netherlands, this included those of Cape Verdean, Dutch Antillean, Surinamese-Creole, and Turkish backgrounds [41].

Lifestyle and social factors are important determinants of health and are likewise implicated in the racial/ethnic disparities observed in MASLD prevalence [5, 42]. In terms of lifestyle, several studies report differences in diet by race. Greater carbohydrate consumption, which is associated with MASH, has been observed in Hispanic and Asian populations [28, 43]. As discussed in a later section, socioeconomic factors including healthcare quality and access, food insecurity, cultural and language barriers, and socioeconomic disparities are also likely linked to MASLD disparities by race/ethnicity [1, 5, 11, 42]. Among racial/ethnic groups in the United States, Hispanic populations have the worst access to care and quality of care, experiencing the worst health outcomes as a result [9]. Inadequate access to care and screening could result in lower rates of primary prevention, thereby increasing disease prevalence in this population [11].

Lastly, data regarding prevalence come from a mix of retrospective studies, which may be biased by underdiagnosis, and prospective studies involving universal screening, which are more likely to reflect true prevalence [17]. This phenomenon may contribute to the reported lower relative prevalence of MASLD among Black populations.

Screening and Diagnosis

While alanine aminotransferase (ALT) is a nonspecific marker of liver injury and inflammation, it is often an early indicator of liver damage in MASLD and other chronic liver disease that prompts further workup [44]. Differences in testing patterns for liver enzymes may therefore subsequently contribute to differing rates of timely diagnosis of MASLD. Kim and Khalili (2023) found that Black (OR = 1.3, p < 0.001) and Latinx (OR = 1.1, p = 0.005) patients and patients with excess body weight (OR = 1.6, p < 0.001) and diabetes (OR = 4.3, p < 0.001) were more likely to have ALT testing than non-Latinx White patients and patients without excess body weight or diabetes, respectively [44]. Conversely, Asian/Pacific Islander patients (OR = 0.6, p < 0.001) were less likely to have ALT testing [44].

In interpreting results of screening, results suggest both that providers may interpret similar results differently according to patient characteristics, and that non-invasive measures used to risk-stratify patients with concern for MASLD may perform differently across racial/ethnic groups. Regarding the former, in one study, Latinx and Asian/Pacific Islander patients were significantly more likely to have an undiagnosed abnormal ALT than non-Latinx White or Black patients [44]. In terms of the latter, age at liver biopsy, sex, presence of diabetes, total bilirubin level, albumin, and NAFLD Activity Score (NAS) predict advanced fibrosis in White but not Black patients [13, 15].

Further, Balakrishnan et al. (2021) evaluated the performance of several commonly used clinical indices in a predominantly Hispanic population: the NAFLD fibrosis score (NFS), BARD score (Body mass index, Aspartate aminotransferase [AST]/ALT Ratio, Diabetes), Fibrosis-4 (FIB-4), and AST-to-platelet ratio index (APRI) and found that all of these scores have similar discriminatory ability (i.e., area under the receiver operating characteristic) for MASLD-related fibrosis when used in Hispanic relative to White populations [45]. However, because disease prevalence differs between these populations, such measures have different positive and negative predictive values in the respective populations. Namely, when used in Hispanic populations, all of these non-invasive risk scores have higher positive predictive value (PPV) and lower negative predictive value (NPV); the FIB-4 has the highest PPV at 100% and the NFS has the highest NPV at 85.4% [45]. Authors therefore suggest that the NFS should be used preferentially for screening as the initial score to rule out advanced disease in Hispanic populations [45]. As performing non-invasive testing is more likely to detect positive cases in Hispanic populations, ensuring appropriate testing in this group would likely lessen the burden of undiagnosed MASLD even more so than in other racial and ethnic groups.

While screening is not recommended in all patients, the 2023 AASLD guidance document for diagnosis and management of MASLD recommends regular non-invasive testing via the FIB-4 index for likelihood of advanced fibrosis in patients with suspected MASLD based on metabolic risk factors or incidental imaging findings [46]. However, these guidelines note that the reliability of performance of non-invasive testing including FIB-4 has not been well studied in patients with diabetes, and may be less sensitive in this population [47]. Given that there is a higher prevalence of diabetes in Black and Hispanic populations compared to White and non-Hispanic populations, the decreased reliability of non-invasive testing in patients with diabetes has the potential to widen racial and ethnic disparities in diagnosis of MASLD [48, 49].

Management and Outcomes

In terms of race, current research suggests that although Black patients are less likely to develop MASLD, those with disease have the same risk of developing MASH compared to White patients [25]. By contrast, Asian Americans with MASLD tend to have less severe and less progressive disease [43].

When looking at hospitalizations, Hispanic individuals with MASLD are more likely to be hospitalized compared to other racial groups, and the rate of hospitalization has been increasing in Hispanic individuals with MASLD compared to other races, possibly driven in part by the fact that Hispanic individuals are more likely to present to emergency departments than to preventative care settings [50, 51]. Black race is associated with increased risk of discharge to another health facility rather than to home relative to White race, which in turn has higher risk than Hispanic and Asian populations. Black race is also associated with longer average length of stay and higher comorbidity burden [50]. However, race was not associated with differences in inpatient mortality.

Differences in race and ethnicity have also been described relating to treatment of MASLD. Patients recruited for lifestyle programs are predominantly White, and ethnic minorities are harder to recruit, with socioeconomic factors also contributing to this discrepancy [52]. However, a meta-analysis examining lifestyle programs for prevention of weight gain found that ethnicity does not moderate success of such interventions, the majority of which were conducted via mobile platform [53]. Regarding drug therapies for MASLD, Resmetirom was recently the first medication to be approved by the Food and Drug Administration (FDA) specifically for the treatment of MASH and fibrosis; however, data surrounding long-term clinical outcomes and generalizability across racial/ethnic groups are not yet available [54]. While not currently FDA-approved for treatment of MASLD, recent studies have also shown that glucagon-like peptide-1 receptor agonists (GLP-1RAs) may reduce hepatic steatosis and fibrosis, though no mortality benefit has been observed and larger studies are still ongoing [46, 55, 56]. Nonetheless, use of GLP-1RAs in diabetes has been shown to reduce mortality and other clinically significant adverse outcomes, and their use in obesity even among patients without diabetes has been shown to reduce adverse cardiometabolic outcomes [57, 58]. Given that diabetes and obesity are two major risk factors and comorbidities seen in MASLD, these drugs will likely be increasingly used in patients with MASLD. However, emerging data have revealed disparities in access to these medications. Despite there being a higher burden of diabetes, obesity, and cardiovascular disease among people of color, use of GLP-1RAs for treatment of diabetes is lower in Black (aOR = 0.81; 95% CI = 0.79–0.83), Asian (aOR = 0.59; 95% CI = 0.56–0.62), and Hispanic (aOR = 0.91; 95% CI = 0.88–0.93) populations than in non-Hispanic Whites, independent of socioeconomic and health insurance status [59]. The global shortage of medications in this class likely contributes to disparities in their use as well [60]. However, a single-center study of patients with concurrent diabetes and MASH found that while use of GLP-1RAs is low, there is no difference in the rate of use across racial/ethnic groups [61].

In examining outcomes across race/ethnicity, there remains a paucity of knowledge on whether disparities exist. A 2018 systematic review and meta-analysis from Rich et al. was unable to identify clear differences, citing contradictory study findings related to disparities across racial and ethnic groups in all-cause mortality and risk of progression to cirrhosis [6]. When examining disease severity, Rich et al. also concluded that among patients with MASLD, there were no significant differences in the proportion of patients with fibrosis by race or ethnicity, despite finding higher risk of steatohepatitis in Hispanic populations compared to non-Hispanic Black and White populations [6]. They suggest a multitude of factors contributing to these contradictory results, including variation across studies in method of MASLD diagnosis confirmation, differences across study populations in lifestyle habits, and prevalence of metabolic syndrome or high-risk genetic polymorphisms.

Lastly, both explicit and implicit provider bias can impact provider decisions related to screening and treatment, thus further compounding disparities in outcome by race and ethnicity [17]. For example, despite similar risk of progression among Black patients, a pediatric study determined that African Americans with obesity and abnormal liver enzymes are much less likely to be referred to hepatology, endocrinology, or weight management clinics than White or Hispanic patients [62]. Another study examining management of patients with both diabetes and MASLD concluded that non-Hispanic Black patients were less likely than non-Hispanic White patients to receive evidence-based management, as defined by liver ultrasound, transient elastography, or hepatology evaluation [63]. Differences in socioeconomic status and access to healthcare also likely influence racial and ethnic discrepancies in disease outcomes, as discussed later in this review.

Research

Non-Hispanic white patients are overrepresented in genetic studies of MASLD and MASH, thus caution must be employed in generalizing these results to other races and ethnicities [36]. In clinical trials of MASLD treatment, Black and Hispanic patients are underrepresented relative to the burden of disease experienced by these respective populations [64]. Of note, nearly 90% of patients in the trial for Resmetirom were White, making it difficult to determine whether this therapy will show the same benefit in other racial and ethnic groups [54]. Additionally, many of the non-invasive screening tests for MASLD were validated in studies that did not evaluate their reliability among those of Asian descent [65].

Disparities by Sex

Incidence and Prevalence

A greater prevalence of MASLD in men relative to women has been consistently documented; a 2021 systematic review and meta-analysis found a relative risk (RR) of MASLD of 0.81 for women relative to men (95% CI = 0.68–0.97) [66]. A number of mechanisms for this finding have been proposed. There is substantial evidence that sex hormones (i.e., androgens and estrogens) act as primary mediators of this process [5, 67–70]. Androgens promote pro-inflammatory and cirrhotic pathways, and estrogens appear to be protective against the same, possibly via the TLR-MyD88/IL-6 pathway and/or downstream effects of the ERα receptor including regulation of plasma lipoprotein levels and the inflammatory response [67]. Prevalence rates of MASLD in postmenopausal women are similar to those of men, further demonstrating that the protective effect of estrogen plays a significant role [5]. Postmenopausal decrease in estrogen levels alters visceral fat distribution in a manner that promotes metabolic syndrome [70]. Furthermore, there are limited data to suggest that estrogen-based hormone replacement therapy may mitigate the increase in MASLD prevalence among postmenopausal women [71, 72].

Disruption in gut microbiota has been associated with risk of MASLD via complex microbiota–host interactions related to altered gut permeability, and differences in these pathways between men and women have been noted, possibly also related to differences in sex hormones [70, 73]. Additionally, sex chromosomes themselves may play a role in MASLD by influencing insulin resistance, as boys with Klinefelter syndrome have higher rates of metabolic syndrome than boys with XY sex chromosomes [74, 75]. Other proposed factors contributing to higher prevalence of MASLD in men include differences in the formyl peptide receptor 2, phosphatidylethanolamine N-methyltransferase, hepatic very low density lipoproteins (VLDLs), and differential expression of inflammatory cytokines by hepatic Kupffer cells between men and women [7, 69, 76].

In contrast to the protective factors against hepatic steatosis in women, numerous studies have suggested that high fructose intake may play a greater role in promoting steatosis in women than men, even in the absence of worse obesity [77–81]. A proposed explanation for this is the marked induction of the GLUT8 hepatic fructose transporter enzyme with fructose ingestion that was seen in females but not in males [82].

Management and Outcomes

There is evidence that some treatments, such as lifestyle-based weight loss, are equally effective in men and women, while other treatments, including lipase inhibitors such as Orlistat, are more effective in women [70]. Women with diabetes are more likely than men to receive treatment with GLP-1RAs, which have shown promise in treatment of MASLD as well [46, 59].

In addition to impacting risk of MASLD, sex hormones may also play a role in disease progression, with age-associated differences likely related to female menopausal status as previously reported. Despite similar overall prevalence of MASH between sexes (pooled RR = 1.00; 95% CI = 0.88–1.14), MASH is more common in older women than older men (RR = 1.17; 95% CI = 1.01–1.36), possibly related to the aforementioned anti-inflammatory effect of estrogen in the liver [66, 70]. Regarding disease progression, women are more likely overall to develop advanced liver disease and associated complications [5]. However, some studies have found more advanced hepatic fibrosis in men than women prior to the age of menopause, which has been attributed to the protective effect of estrogen against hepatic fibrosis [70]. Thyroid hormone may also contribute to differential rates of disease progression by sex. Hypothyroidism has been associated with MASLD, and one study has shown that specifically among women there is a strong association between TSH and significant fibrosis [83].

There are also differences in clinically significant outcomes by sex. Adejumo et al. (2019) report that patients hospitalized with MASH are more likely to be women than men (60.2% vs. 39.8%, p = 0.03), with a shorter average length of stay and lower total hospital costs in women but with similar in-hospital mortality rates between the sexes [50]. However, women with MASLD have higher cardiovascular and all-cause mortality than men with MASLD overall [84]. After controlling for other factors, women with MASH are 19% less likely to receive a transplant than men and are more likely to have poor outcomes while on the waitlist (death on the waiting list 17.1% vs 11.4%, p < 0.001); this disparity is further exacerbated by the fact that MASH is the leading indication for liver transplant in women but not men [85].

Disparities by Socioeconomic Status

Incidence and Prevalence

Differences by SES in MASLD prevalence are observed, suggesting that SES may be one of the drivers of disparities. Lower SES is associated with greater MASLD prevalence, an association thought to be mediated in part by nutrition-related factors such as food insecurity [5, 9, 86]. Among low-income adults in the US, food insecurity is independently associated with MASLD (odds ratio [OR] = 1.38; 95% CI = 1.08–1.77) [42]. Food insecurity contributes to MASLD from a young age, as in a cohort of Latinx children in the US, exposure to household food insecurity at age 4 independently associated with a 3.7-fold higher odds of MASLD later in childhood (95% CI: 1.5–9.0, p < 0.01) [87].

Diet quality likely plays a major role in the relationship between food insecurity and MASLD, as there is an inverse association between food insecurity and diet quality among adults with MASLD [9]. As measured by the Healthy Eating Index (HEI), which accounts both for adequate intake of nutrients and moderation of intake of refined foods, sodium, sugars, and saturated fats, patients with MASLD tend to have poorer diet quality than patients without MASLD [9]. Socioeconomically disadvantaged groups often bear the brunt of living in an obesogenic environment, and ultra-processed foods, which are calorie-dense and nutrient poor, are usually cheaper than unprocessed or minimally processed foods [88, 89]. Compounding this is the finding made by the Coronary Artery Risk Development in Young Adults study, which revealed fast food consumption to be most strongly related to the availability of fast food close to homes in individuals from low socioeconomic backgrounds [90]. These facts likely explain why low income is a leading risk factor for food insecurity, poor diet quality, and subsequent development of MASLD [89].

Notably, however, while evidence supports a direct correlation between food insecurity and MASLD,the effect of food insecurity on diet quality is not different between MASLD and non-MASLD groups, suggesting that diet quality alone does not mediate the observed relationship between food insecurity and MASLD [5, 9, 17, 42, 91]. Thus, food insecurity and MASLD are associated, but this relationship is not fully explained by diet quality and is likely much more multifactorial. For example, BMI and waist circumference significantly mediate the effect of diet quality on MASLD risk [92]. Other social factors such as proximity and access to healthy food, cultural norms related to diet, utilization of the Supplemental Nutrition Assistance Program (SNAP), and coping responses to stress may also contribute to the association between food insecurity and MASLD, and these factors likely also vary by race and ethnicity [93].

In addition to impacting nutrition, SES influences MASLD prevalence through neighborhood-level factors such as access to green space and crime rate, which have been observed to impact rates of cardiometabolic disease among low-SES populations [17, 94]. Low education level also increases risk of having MASLD among adolescents in the US, most notably in Hispanic males with obesity and hypertension [95]. Higher education level (college or above) is associated with reduced MASLD risk, partially mediated by high-quality diet and physical activity [92]. Physical activity itself reduces risk of MASLD, and the combination of high-quality diet and physical activity reduces risk more so than either individual factor [92]. There is likely an association between physical activity and SES among US adults; however, this association has not been well characterized [96].

Screening and Diagnosis

Disparities in diagnosis related to access to care have not been well studied in patients with MASLD [97]. Moreover, there are little data on in-clinic screening for social determinants of health among patients with risk factors for MASLD, and none among patients with diagnosed MASLD [98]. However, given that screening is predominantly managed in primary care settings, populations with lower primary care access would presumably be less likely to receive appropriate screening for MASLD. There are known racial and ethnic inequities in primary care access and outcomes related to structural racism [99].

Management and Outcomes

Socioeconomic factors have been implicated in MASLD progression and disease outcomes. A study conducted in a tertiary healthcare network in New York identified an association of low SES with higher prevalence of both MASLD and MASH; however, no association was found with fibrosis [100]. As with development of MASLD, food insecurity has also been implicated in disease progression, and is independently associated with advanced fibrosis among low-income adults in the US with MASLD (OR: 1.40, 95% CI: 1.04–1.88) [42, 101]. More advanced fibrosis is also associated with poorer diet quality as measured by HEI [9]. Furthermore, in US adults with MASLD and advanced fibrosis, food insecurity is also associated with greater all-cause mortality [102].

Socioeconomic factors can impact in-hospital outcomes in MASLD as well. Lack of private insurance among hospitalized patients with MASLD associates with significantly increased rate of hospitalization, in-hospital mortality, length of stay, and risk of discharge to another health facility rather than to home [50, 51]. Additionally, lower income has been associated with increased rate of hospitalization and longer length of stay [50, 51].

Disparities in treatment have also been described. Patients recruited for lifestyle programs are primarily highly educated, and patients from socially disadvantaged backgrounds or with lower education levels are harder to recruit [52]. As above, racial and ethnic differences often exist among participants. However, in a meta-analysis looking at lifestyle programs focused on preventing weight gain, SES was found not to moderate program outcomes [53]. High satisfaction with tele-hepatology has been reported among patients in a safety-net clinical setting, though satisfaction was lower among Hispanic patients and those without cell phone or internet access [103]. Data are scare regarding socio-economic disparities in access to hepatology care among patients with MASLD; however, Americans living in rural areas generally have poorer access to specialist care [104].

Lastly, in addition to racial and ethnic disparities in use of GLP-1RAs, which have shown promise in MASLD as described in a previous section, patients with low household income and those with Medicare Advantage rather than commercial health insurance are less likely to receive treatment for diabetes with these medications [59].

Discussion

Disparities exist in MASLD prevalence, screening, diagnosis, treatment, and outcomes; these disparities exist with respect to sex, race, and socioeconomic status, among others. Awareness of these disparities is critical for providers and health systems to promote the delivery of equitable care. The present findings support several conclusions that are relevant to current clinicians. First, significant disparities exist in the prevalence of MASLD across populations, but numerous, variable pathophysiologies underlie these population-level differences. Genetic, cellular, metabolic, endocrine, social, and economic factors all contribute to observed differences in disease prevalence. Therefore, attention should be paid to the individual, intersectional circumstances of the patient when determining disease risk. Maintaining a high clinical index of suspicion for disease across diverse populations may help combat underdiagnosis, both for initial disease and for disease progression to MASH and cirrhosis. Increased vigilance will be required to properly diagnose initial disease and disease progression in order to improve outcomes.

Second, significant disparities exist in outcomes across MASLD patient populations; evidence suggests these are related to observed disparities in prevalence, screening, diagnosis, and treatment, in addition to other as-yet unidentified factors. Disparities at each of these stages may at least partially be related to social drivers of health such as food access and healthcare quality and access. Identifying and addressing social drivers of health is essential in reducing disparities in MASLD outcomes. Increased attention is warranted to promote equitable healthy lifestyle behaviors and access to healthy diet, as these factors are both driving factors in MASLD pathogenesis as well as cornerstones of treatment. Given the discrepancies in referrals for treatment, it is also important to lower barriers to care, such as using telemedicine and mobile-based lifestyle intervention programs, which have shown promise in diverse populations.

Third, the present review reveals a need for further research examining the MASLD disease process and associated disparities, and research that is designed with equity in mind to ensure appropriate representation of affected populations. Further studies are likewise needed on process measures and practices such as screening, diagnosis, and treatment in order to understand the proximal causes of observed disparities. Research into effective treatments should utilize study populations that reflect the diversity of patient populations of interest. While much of this review focused on the MASLD population in the United States and other high-income countries, disparities are also present in low- and middle-income countries, and additional research is warranted to better characterize this phenomenon [105].

Conclusion

MASLD is a growing cause of morbidity and mortality that is associated with significant disparities in prevalence, diagnosis, and outcomes. Further research is needed into these disparities, particularly in areas pertaining to the treatment and outcomes of disease across groups. Providers can take steps to help limit the impact of disparities in their own patient populations by having a high index of suspicion for disease and its progression, and by screening for and addressing social drivers of health such as food security and healthcare access. As metabolic syndrome and its associated sequelae become increasingly common, maintaining the health of patients and populations will rely heavily on the ability of providers and healthcare systems to effectively prevent, diagnose, and treat these diseases. Doing so equitably will be essential to reducing disparities and lessening the overall burden of MASLD-related morbidity and mortality.

Supplementary Material

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10620-024-08722-0.

Abbreviations

AASLD

American Association for the Study of Liver Diseases

AHR

Absolute hazard ratio

ALT

Alanine aminotransferase

AOR

Adjusted odds ratio

APRI

Aspartate aminotransferase-to-platelet ratio index

BMI

Body mass index

CI

Confidence interval

FDA

Food and Drug Administration

FIB-4

Fibrosis-4

GLP-1RA

Glucagon like peptide-1 receptor agonist

HCC

Hepatocellular carcinoma

HEI

Healthy Eating Index

HR

Hazard ratio

MASH

Metabolic dysfunction-associated steatohepatitis

MASLD

Metabolic dysfunction-associated steatotic liver disease

NAFLD

Non-alcoholic fatty liver disease

NAS

NAFLD Activity Score

NASH

Non-alcoholic steatohepatitis

NCEP ATP III

National Cholesterol Education Program Adult Treatment Panel III

NFS

NAFLD fibrosis score

NPV

Negative predictive value

OR

Odds ratio

PPV

Positive predictive value

RR

Relative risk

SES

Socioeconomic status

VLDL

Very low density lipoprotein

Data Availability

No datasets were generated or analysed during the current study.