The Epidemiology of Nonalcoholic Fatty Liver Disease: A Global Perspective

Mariana Lazo, M.D., M.Sc.; Jeanne M. Clark, M.D., M.P.H.


Semin Liver Dis. 2008;28(4):339-350. 

In This Article

Risk Factors for NAFLD

Insulin Resistance

Insulin resistance is considered the pathophysiological cause of NAFLD and several cross-sectional studies have found strong associations between direct measures of insulin resistance and NAFLD. In addition, several studies have looked at the association between NAFLD and metabolic syndrome, which is considered the clinical construct of insulin resistance.[104,105] A long-term prospective study demonstrated that the presence of metabolic syndrome is associated with increased odds for US-defined NAFLD, adjusted OR = 4.0 (95% CI, 2.6-6.1) and 11.2 (95% CI, 4.9-25.9) for men and women respectively. The same study found that NAFLD regression was less likely among those with metabolic syndrome.[37]

Consistent with the likelihood that insulin resistance plays a primary role in NAFLD pathogenesis, mild insulin resistance is very common at the earliest stages of NAFLD, and more severe insulin resistance (i.e., metabolic syndrome or type 2 diabetes) correlates with more advanced stages of NAFLD.[106] Finally, a cross-sectional study found that the presence of metabolic syndrome was associated with higher odds of NASH (OR = 3.2, 95% CI, 1.2-8.9) and fibrosis (OR = 3.5; 95% CI, 1.1-11.2), even adjusting for sex and BMI.[107] The mechanisms of liver disease in insulin resistance states have recently been reviewed.[4]


Worldwide, obesity remains the most important and well-described risk factor for NAFLD (Fig. [1]). The results of several cross-sectional studies and case-control studies have shown that people with NAFLD have higher waist circumference (WC) or BMI than those without NAFLD, and have reported significant associations between abdominal obesity (OR = 1.10-1.13; 95% CI, 1.02-1.22 per 1-cm increase in WC) and NAFLD.[108] In the Dionysos study, NAFLD was present in 94%, 67%, and 24.5% of the obese, overweight, and normal weight populations, respectively.[109] Evidence from a prospective study published recently from Japan reinforced this association. This study found that during 5 years of follow-up among people without NAFLD, significant alcohol consumption or hepatitis B or C infection at baseline, the presence of a BMI > 25 kg/m2 and high body fat (estimated by impedance method, men: ≥ 20%, women: ≥ 25%) was associated with an increased risk of developing elevated aminotransferases, HR = 3.6 (95% CI, 2.2-5.9), and HR = 2.8 (95% CI, 1.7-4.8), respectively.[48] Finally, this association has also been supported by studies showing a dose-response relationship between levels of obesity and degrees of NAFLD, with more advanced stages of the disease among those with severe obesity.[15,90,91]

Figure 1.

Prevalence of nonalcoholic fatty liver disease (NAFLD) versus prevalence of obesity in the general population.

Although the mechanisms responsible for the increased risk of NAFLD with abdominal obesity have not been fully elucidated, current findings suggest that obesity leads to insulin resistance, which, in turns, leads to NAFLD.[110] It is also postulated that obesity increases the risk of developing NAFLD after exposure to certain other insults. For instance, obesity appears to potentiate and lower the threshold for alcohol-related liver problems. Bellentani et al found US evidence of fatty liver in 46% of nonobese and 95% of obese heavy drinkers, demonstrating that obesity doubles the prevalence of alcohol-induced fatty liver disease,[111] which has been confirmed by another group.[112]

Obstructive Sleep Apnea

Several studies have reported that patients with obstructive sleep apnea (OSA) have a higher prevalence of NAFLD.[4,113,114,115,116] This association may be mediated, in part, through increased BMI and insulin resistance, but also independently of it. Recent evidence suggests that OSA could be an independent risk factor for liver injury through intrahepatic hypoxia. Furthermore, experimental evidence has suggested that transient intermittent hypoxia occurring in OSA may be the "2nd hit" for the progression of NAFLD to NASH and fibrosis.[4,117,118,119] However, this explanation remains hypothetical and most of the epidemiological evidence suggests that OSA is associated with NAFLD through insulin resistance.

Genetic Factors

A genetic susceptibility to NAFLD, NASH, and its complications has been postulated. Environmental factors such as obesity cause simple steatosis in the vast majority of people; however, more advanced forms of NAFLD and hepatocellular cancer are found in only a minority of the exposed, leading to the hypothesis that susceptibility is inherently determined. This novel hypothesis is now being actively addressed.

Epidemiologic studies showing familial aggregation of NAFLD and insulin resistance further support a genetic basis for NAFLD.[120,121] Additionally, studies have found interethnic variations in the prevalence of NAFLD and NAFLD-related "cryptogenic" cirrhosis, suggesting that susceptibility to NAFLD has a genetic component.[122,123]

In a general formulation of determinants of different stages of NAFLD, the following are elements where genetically determined host characteristics might be important: (1) insulin resistance, (2) magnitude and pattern of fat deposition, (3) hepatic lipid metabolism (synthesis, storage, utilization-oxidation, and export), (4) general and NAFLD-related fibrosis, (5) cytokines and adipokine production, and (6) oxidative stress. With the application of the new and powerful tools of modern molecular and cellular epidemiology and biology, a more complete understanding of the genetic and epigenetic changes involved in NAFLD can be anticipated.[124,125]


Based on studies in animals demonstrating strong associations between diet composition and balance and the development of fatty liver, it has been postulated that dietary habits may promote NAFLD in humans. The mechanisms by which diet may play a role include modulation of HT accumulation and regulation of the antioxidant activity, as well as through changes in insulin sensitivity and postprandial triglyceride metabolism.[126,127] The results of epidemiologic studies have been inconsistent and it has been difficult to distinguish the effect of different components of diet per se. Although some studies have highlighted the positive association between carbohydrate intake and NAFLD occurrence and severity,[128] others have underlined the effect of different types of fat.[129] A recent population-based study conducted in Israel found that after controlling for BMI, age, gender, and total calories, the consumption of soft drinks (OR = 1.45; 95% CI, 1.13-1.85) and meat (OR = 1.37; 95% CI, 1.04-1.83) were significantly associated with an increased odds of NAFLD. In addition, a higher intake of fish rich in omega-3 may be associated with decreased odds of NAFLD (OR = 0.73; 95% CI, 0.52-1.04).[130]

Gut-derived Bacteria

Discussion of the role of gut-derived bacteria and their products in liver injury has been lingering in the NAFLD literature for several years and its mechanisms has been reviewed in detail elsewhere.[131,132] Both in vivo and human studies have suggested that the steatotic liver is particularly predisposed to damage by additional insults such as exposure to gut bacteria-derived lipopolysaccharide or the endotoxin-mediated inflammatory response, which could promote more advanced stages of the disease such as NASH or cirrhosis. Thus, conditions that alter the gut flora and permeability might confer higher risk for disease.

Importantly, patients with NAFLD who undergo jejunal-ileal bypass surgery have been shown to have a higher risk of developing intestinal bacterial overgrowth, which, in turn, would potentially place them at risk for developing more advanced disease, such as NASH. Consistent with this hypothesis, a study conducted in Australia found that the presence of small intestinal bacterial overgrowth was significantly higher among patients with NASH compared with controls.[133]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: