Abstract
This national survey study uses National Health and Nutrition Examination Survey (NHANES) data to estimate the prevalence of alcoholic fatty liver disease (AFLD) overall and with stage 2 or greater and stage 3 or greater fibrosis among US adults from 2001 to 2016.
Alcoholic liver disease (ALD), comprising a spectrum of diseases ranging from alcoholic fatty liver disease (AFLD) to advanced ALD (including alcoholic hepatitis, cirrhosis, and cirrhosis complications),1 is a leading cause of mortality in the United States, with nearly 250 000 deaths attributed to ALD in 2010.2,3 Overall US clinical burden of ALD remains unclear, perhaps because of lack of a definitive standard for identifying ALD. This study focused on the specific, more well-defined subset of AFLD to estimate national prevalence among US adults.
Methods
We used the 2001-2016 National Health and Nutrition Examination Survey (NHANES) data set, a series of cross-sectional, nationally representative surveys of the noninstitutionalized US population, including in-person interviews and health examinations in mobile examination centers. The NHANES 2001-2016 overall response rate was 70.3% (range, 60.7%-78.3%) for interviews and 67.2% (range, 58.1%-72.7%) for examinations. AFLD was identified based on alcohol use (>28 g/d in women and >42 g/d for men in the past 12 months) and elevated liver enzyme levels (aspartate aminotransferase [AST] or alanine aminotransferase [ALT] >25 U/L [0.42 μkat/L] in women and >35 U/L [0.58 μkat/L] in men), in the absence of elevated total bilirubin level (<3 mg/dL [51.3 μmol/L]) and after excluding hepatitis C and hepatitis B infections.1 Individuals with metabolic syndrome (National Cholesterol Education Program Adult Treatment Panel III4 criteria) were excluded given associated increased risks of nonalcoholic fatty liver disease. Hepatic fibrosis was assessed using an AST-to-platelet ratio index ([AST/upper limit of normal for AST]/platelet count) greater than 0.7 (stage ≥2 fibrosis) and Fibrosis-4 score ([age × AST]/[platelet count × ALT½]) greater than 2.67 (stage ≥3 fibrosis), which represent standard definition cutoffs of well-validated tools for assessing hepatic fibrosis.
Multivariable logistic regression with appropriate sample weights was used to derive AFLD prevalence. Linear trends in prevalence were analyzed using orthogonal polynomial contrasts. Interaction terms were included in logistic regression models to assess whether prevalence changes differed by sex, age, and race/ethnicity. Missing data, ranging from less than 1% to 15%, were addressed using multivariable imputation by chained equations, which included relevant demographic and clinical covariates. Two-tailed P < .05 indicated statistical significance (Stata version 14.0 [StataCorp]). The Alameda Health System institutional review board granted exempt status and waiver of informed consent.
Results
Among 34 423 respondents included, 4.3% were identified with AFLD (60.6% men; 63.0% non-Hispanic white; mean age, 40.2 years). From 2001-2002 to 2015-2016, AFLD prevalence remained stable from 4.3% (95% CI, 3.5%-5.0%) to 4.7% (95% CI, 4.2%-5.1%) (P = .69), AFLD with stage 2 or greater fibrosis increased from 0.6% (95% CI, 0.5%-0.8%) to 1.5% (95% CI, 1.3%-1.8%) (P < .001), and AFLD with stage 3 or greater fibrosis increased from 0.1% (95% CI, 0.02%-0.10%) to 0.2% (95% CI, 0.2%-0.4%) (P = .045) (Figure). Changes in prevalence over time did not differ by sex, age, or race/ethnicity, and no significant interactions were observed across all variables (Table).
Figure. Trends in Prevalence of Total Alcoholic Fatty Liver Disease (AFLD), AFLD With Stage 2 or Greater Fibrosis, and AFLD With Stage 3 or Greater Fibrosis Among US Adults From 2001-2002 to 2015-2016.
Shaded areas adjacent to curves indicate 95% CIs. Data are based on 34 423 individuals in NHANES surveys.
Table. Prevalence and Number of Adults Aged 20 Years or Older With Alcoholic Fatty Liver Disease (AFLD) and Fibrosis in the United States by Sex, Age, and Race.
| 2001-2002 | 2015-2016 | P Value for Interactionc | |||
|---|---|---|---|---|---|
| No.a | Prevalence, % (95% CI)b | No.a | Prevalence, % (95% CI)b | ||
| AFLD Among Total Population | |||||
| Sex | |||||
| Men | 108 | 5.2 (4.2-6.2) | 116 | 5.8 (3.5-6.3) | .45 |
| Women | 56 | 3.4 (2.8-4.0) | 64 | 3.7 (2.6-4.2) | Reference |
| Age, y | |||||
| 20-44 | 117 | 5.6 (4.6-6.6) | 108 | 6.5 (5.9-7.1) | Reference |
| 45-64 | 40 | 3.6 (2.9-4.4) | 61 | 4.3 (3.7-7.1) | .86 |
| ≥65 | 7 | 1.0 (0.7-1.3) | 11 | 4.2 (3.7-4.9) | .88 |
| Race/ethnicityd | |||||
| White | 71 | 3.8 (2.9-5.2) | 61 | 4.1 (3.2-5.8) | Reference |
| Black/African American | 21 | 3.1 (1.6-4.8) | 22 | 3.4 (1.7-4.0) | .36 |
| Hispanic | 69 | 8.3 (4.9-10.0) | 87 | 9.3 (8.7-10.0) | .91 |
| Other | 3 | 2.5 (1.6-3.8) | 10 | 2.7 (1.0-3.4) | .97 |
| Total | 164 | 4.3 (3.5-5.0) | 180 | 4.7 (4.2-5.1) | |
| AFLD With Stage ≥2 Fibrosis Among Total Population | |||||
| Sex | |||||
| Men | 17 | 0.7 (0.6-0.9) | 38 | 1.8 (1.5-3.0) | .34 |
| Women | 9 | 0.5 (0.4-0.6) | 19 | 1.3 (1.1-1.6) | Reference |
| Age, y | |||||
| 20-44 | 15 | 0.6 (0.5-0.8) | 24 | 1.6 (1.3-2.2) | Reference |
| 45-64 | 10 | 0.8 (0.6-0.9) | 26 | 2.0 (1.6-2.3) | .42 |
| ≥65 | 1 | 0.3 (0.1-0.6) | 7 | 0.6 (0.5-0.8) | .94 |
| Race/ethnicityd | |||||
| White | 12 | 0.6 (0.1-1.0) | 23 | 1.4 (1.1-1.8) | Reference |
| Black/African American | 5 | 0.5 (0.3-0.7) | 7 | 1.2 (0.4-2.0) | .09 |
| Hispanic | 8 | 1.1 (0.8-1.4) | 23 | 2.8 (2.5-3.2) | .53 |
| Other | 1 | 0.3 (0.1-0.4) | 4 | 0.7 (0.5-0.9) | .73 |
| Total | 26 | 0.6 (0.5-0.8) | 57 | 1.5 (1.3-1.8) | |
| AFLD With Stage ≥3 Fibrosis Among Total Population | |||||
| Sex | |||||
| Men | 5 | 0.1 (0.02-0.2) | 12 | 0.4 (0.2-0.7) | .81 |
| Women | 1 | 0.05 (0.02-0.1) | 3 | 0.1 (0.1-0.2) | Reference |
| Age, y | |||||
| 20-44 | 0 | NA | 2 | 0.1 (0.03-0.2) | Reference |
| 45-64 | 5 | 0.2 (0.04-0.3) | 7 | 0.4 (0.1-0.5) | .25 |
| ≥65 | 1 | 0.1 (0.03-0.2) | 6 | 0.3 (0.1-0.8) | NA |
| Race/ethnicityd | |||||
| White | 1 | 0.07 (0.05-0.2) | 4 | 0.2 (0.1-0.4) | Reference |
| Black/African American | 2 | 0.08 (0.03-0.1) | 3 | 0.3 (0.1-0.7) | .70 |
| Hispanic | 2 | 0.1 (0.04-0.2) | 8 | 0.4 (0.1-0.8) | .31 |
| Other | 1 | 0.03 (0.01-0.05) | 0 | NAe | NA |
| Total | 6 | 0.1 (0.02-0.1) | 15 | 0.2 (0.2-0.4) | |
Abbreviation: NA, not analyzable.
Unweighted number of respondents from National Health and Nutrition Examination Survey study sample.
Derived from multivariable logistic regression with sample weights.
Derived from the interaction term between the time variable (reference = 2001-2002 cycle) and covariate for prevalence.
Self-reported by survey participants in response to interview questions and included the following options: Mexican American, other Hispanic, non-Hispanic white, non-Hispanic black, other race. In this study we categorized non-Hispanic white as “white,” non-Hispanic black as “black/African American,” and other race as “other.” Mexican American and other Hispanic individuals were combined into a single category, “Hispanic.” Race/ethnicity were analyzed because existing data demonstrate racial differences in epidemiology of other chronic liver diseases, alcohol use disorders, and alcoholic-related diseases.
Not analyzable because of insufficient number of respondents.
Discussion
The prevalence of AFLD among US adults remained stable from 2001 to 2016, affecting 4.7% of adults in 2015-2016. However, the prevalence of AFLD with stage 2 or greater fibrosis and AFLD with stage 3 or greater fibrosis increased significantly, affecting 1.5% and 0.2% of adults, respectively, in 2015-2016. This is a particularly concerning observation given that developing fibrosis is the strongest predictor of progression to cirrhosis, liver cancer, and death.5 While studies evaluating national ALD prevalence are lacking, existing studies have demonstrated increasing burden of ALD in population subsets. For example, among patients with end-stage liver disease, ALD became the leading indication for US liver transplants in 2016,6 and a 2018 US population-based study demonstrated increasing cirrhosis death rates largely driven by alcoholic cirrhosis, particularly among individuals aged 25 to 34 years.3
Limitations include the challenges in accurately identifying AFLD using observational data, which may have been affected by misclassification. Noninvasive serologic markers of fibrosis used in this study are well validated, but not specifically for AFLD. The increasing prevalence of US adults with AFLD with stage 2 or greater fibrosis and AFLD with stage 3 or greater fibrosis is concerning and emphasizes the need for greater awareness of unhealthy alcohol use and need for early prevention and intervention efforts.
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Singal AK, Bataller R, Ahn J, Kamath PS, Shah VH. ACG clinical guideline: alcoholic liver disease. Am J Gastroenterol. 2018;113(2):175-194. doi: 10.1038/ajg.2017.469 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Rehm J, Samokhvalov AV, Shield KD. Global burden of alcoholic liver diseases. J Hepatol. 2013;59(1):160-168. doi: 10.1016/j.jhep.2013.03.007 [DOI] [PubMed] [Google Scholar]
- 3.Tapper EB, Parikh ND. Mortality due to cirrhosis and liver cancer in the United States, 1999-2016: observational study. BMJ. 2018;362:k2817. doi: 10.1136/bmj.k2817 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Aguilar M, Bhuket T, Torres S, Liu B, Wong RJ. Prevalence of the metabolic syndrome in the United States, 2003-2012. JAMA. 2015;313(19):1973-1974. doi: 10.1001/jama.2015.4260 [DOI] [PubMed] [Google Scholar]
- 5.Unalp-Arida A, Ruhl CE. Liver fibrosis scores predict liver disease mortality in the United States population. Hepatology. 2017;66(1):84-95. doi: 10.1002/hep.29113 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cholankeril G, Ahmed A. Alcoholic liver disease replaces hepatitis C virus infection as the leading indication for liver transplantation in the United States. Clin Gastroenterol Hepatol. 2018;16(8):1356-1358. [DOI] [PMC free article] [PubMed] [Google Scholar]