Introduction: Chronic liver disease (CLD) is associated with increased morbidity and mortality. Understanding health disparities can inform appropriate interventions. We aimed to study mortality outcomes of those with CLD by the income level (income-to-poverty ratio <5 as lower income and ≥5 as higher income). Methods: In this retrospective cohort study, we analyzed data of adults from the National Health and Nutrition Examination Survey, 1999–2018. CLD included viral hepatitis, nonalcoholic fatty liver disease (NAFLD), and alcohol-associated liver disease (ALD). Results: We analyzed 59,204 adults: 47,224 without CLD and 11,980 with CLD. The CLD group was older, more likely male, racial/ethnic minority groups or foreign-born, and had lower educational and income levels (p < 0.001). Most (80.02%) CLD participants did not have college degrees and had lower income (79.18%). Among CLD participants, similar differences were observed between lower and higher income groups. Lower income participants with CLD had significantly higher 10-year cumulative mortality compared to higher income CLD participants (15.26 vs. 8.00%, p < 0.001), with consistent findings in viral hepatitis and NAFLD subgroups (p < 0.001) but not ALD (p = 0.71). Adjusting for age, sex, race, birthplace, lower income CLD participants were 2.01 (hazard ratio [HR]: 2.01; 95% CI: 1.79–2.26) times more likely to die overall and in viral hepatitis (HR: 2.05; 95% CI: 1.31–3.24) and NAFLD subgroups (HR: 2.32; 95% CI: 1.69–3.18) but not ALD (HR: 1.17; 95% CI: 0.55–2.51). Conclusion: Lower income, foreign-born, and racial/ethnic minority groups were disproportionately represented among those with CLD, with lower income and CLD individuals having double the mortality risk compared to their higher income counterparts. Interventions should be culturally appropriate and address socioeconomic barriers.

Chronic liver disease (CLD) is a significant cause of morbidity and mortality globally, with complications from cirrhosis and hepatocellular carcinoma accounting for approximately 1.75 million deaths every year [1‒3]. As of 2017, 1.5 billion people have CLD worldwide, with the most common conditions being nonalcoholic fatty liver disease (NAFLD), hepatitis B virus (HBV) infection, hepatitis C virus (HCV) infection, and alcohol-associated liver disease (ALD) [4, 5]. A population-based study focusing on the incidence, prevalence, and mortality of CLD and its complications in the USA showed that CLD-related mortality continues to increase, with an age-adjusted mortality rate of 14.2% in 2017 [6].

Prior reports have noted racial and ethnic, gender, and geographic disparities in the risk factors, prevalence, and mortality rates of CLD in the USA [7‒13]. Some of these observed disparities may be related to differences in the socioeconomic levels of the various groups. As income inequality in the USA has increased over the past few decades, socioeconomic gaps in the prevalence of chronic diseases and life expectancy have also expanded [14‒18]. Indeed, between 2001 and 2014, higher income was found to be associated with greater longevity, and differences in life expectancy across income groups have increased over time in the USA [14]. Between 1999 and 2018, people in the USA with college or higher education or higher family income had consistently lower levels of cardiovascular risk factors than those with lower income or educational levels [15].

Additionally, notable socioeconomic differences in the prevalence of liver disease and mortality from cirrhosis have also been reported [19, 20]. From 1968 to 1997, individuals with less than a college degree had approximately twice the risk of dying from cirrhosis compared to college-educated individuals [19] However, nationally representative data on mortality outcomes of individuals with CLD by income levels are sparse. We aimed to evaluate the long-term mortality of people with CLD in the USA by the family income-to-poverty level and to investigate the association between family income and mortality in people with CLD.

Data Source

This was a retrospective analysis of a population-based sample of persons with CLD utilizing the National Health and Nutrition Examination Surveys (NHANES) during 1999–2018. NHANES collects data through comprehensive written questionnaires, physical examinations, and biological samples from a complex multistage, stratified, clustered probability sample that is representative of the noninstitutionalized, civilian population of the USA. The CLD investigated in this study were HBV or HCV infection, NAFLD, and ALD. The study protocols were approved by the Institutional Review Board of the National Center for Health Statistics, and written informed consent was obtained from each participant.

Study Definitions

Viral hepatitis infection was defined as having either current or a history of HBV or HCV infection. Current HBV infection was defined by the presence of positive hepatitis B surface antigen, and a history of HBV or past infection was defined by a positive hepatitis B core antibody in the absence of hepatitis B surface antigen. Current HCV infection was defined by positive HCV RNA, and a history of past infection was defined by positive anti-HCV antibody in the absence of HCV RNA.

NAFLD was defined using the US fatty liver index (US FLI) [21] of >30, a noninvasive test inclusive of age, race, waist circumference, gamma-glutamyl transferase activity, fasting insulin, and fasting glucose levels that has been validated in the US population for the diagnosis of NAFLD [21], in the absence of HBV and HCV infection and significant alcohol use (>2 drinks/day for men or >1 drink/day for women). ALD was defined by a history significant alcohol use (>3 drinks/day for the past 12 months for men and >2 drinks/day for the past 12 months for women) but no HBV or HCV infection in the presence of elevated liver enzymes levels (aspartate aminotransferase or alanine aminotransferase >25 U/L in women and >35 U/L in men) in the absence of a total bilirubin level >3 mg/dL [22, 23].

Individuals in this study were not grouped for coexistence of more than one CLD. Each CLD subgroup contains individuals having only one CLD.

The family income-to-poverty ratio is an index developed by NHANES using the Department of Health and Human Services (HHS) poverty guidelines, which are issued every year and vary by family size and by state. NHANES codes participants with a family income-to-poverty ratio of 5 or greater or less than 5 due to confidentiality and disclosure concerns. These cut points classify people with income levels in approximately the top 25% and those with income levels in the bottom 75% of the US population into higher income or lower income groups, respectively [15].

We also conducted supplemental analyses for people with income levels in the bottom 25% (family income-to-poverty ratio <1), middle 50% (family income-to-poverty ratio 1–5), and top 25% (family income-to-poverty ratio >5) of the US population, grouping them into lower, middle-, and higher income groups, respectively. Further supplemental analyses were conducted for CLD participants according to the educational level (lower vs. higher educational status), with the lower educational level characterized as not having college education (without high school diploma/GED certificate or with high school diploma/GED certificate) and the higher educational level characterized as having college education (some college or associate’s degree or college degree or higher).

Study Outcomes

The primary study outcome was overall mortality. Public-use Linked Mortality Files (LMF) are available for 1986–2018 NHIS, 1999–2018 NHANES, and NHANES III (https://www.cdc.gov/nchs/data-linkage/mortality-public.htm). We also determined factors associated with overall mortality.

Statistical Analysis

We performed weighted analyses using NHANES survey weights, which account for the survey design, survey nonresponse, poststratification, and oversampling. Descriptive statistics for continuous variables were reported as mean ± standard deviation or as proportions. We used χ2 tests to compare the distribution of categorical variables between groups. We used Student t-tests to compare the distribution of continuous variables between groups. The Kaplan-Meier method was used to generate survival curves by the income level, and the log-rank test was used to determine differences between groups. We performed univariable and multivariable Cox’s proportional hazard models to estimate hazard ratios (HRs) relating income level to mortality. Subgroup analysis by specific liver disease etiology of CLD was performed. We used Stata for all analyses (Stata Corp, ver. 17, College Station, TX, USA) with statistical significance considered for p < 0.05, two-tailed.

Participant Characteristics

Total Cohort

In the overall cohort of 59,204 participants, we identified 47,224 participants without CLD and 11,980 participants with CLD (online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000539858). Overall, compared to those without CLD, the CLD group was older (49.59 ± 17.24 vs. 45.21 ± 17.44), more likely male (57.06 vs. 46.01%), a racial and ethnic minority group (non-Hispanic black: 12.26 vs. 11.86%, Hispanic: 17.57 vs. 14.02%, non-Hispanic Asian: 3.67 vs. 2.19%) or foreign-born (14.17 vs. 10.51%), and had lower educational and income levels (all p < 0.001). Most participants with CLD also had lower income (79.18%) and were not college graduates (80.02%) (Table 1).

Table 1.

Characteristics of participants by the presence of CLD

CharacteristicsWithout CLD (N = 47,224)With CLDa (N = 11,980)p value
Overall, N = 59,204 
Mean age, years 45.21±17.44 49.59±17.24 <0.001 
Sex (%) <0.001 
 Female 53.99 42.94 
 Male 46.01 57.06 
Race (%) <0.001 
 Non-Hispanic white 71.93 66.50 
 Non-Hispanic black 11.86 12.26 
 Hispanic 14.02 17.57 
 Non-Hispanic Asian 2.19 3.67 
Educational level (%) <0.001 
 Without high school diploma/GED certificate 16.85 23.6 
 With high school diploma/GED certificate 25.32 26.23 
 Some college or associate’s degree 29.54 30.15 
 College degree or higher 28.3 20.02 
Family income-to-poverty ratio (%) <0.001 
 ≥5 25.89 20.82 
 <5 74.11 79.18 
Marital status (%) 0.39 
 Not married 44.98 44.29 
 Married 55.02 55.71 
Birthplace (%) <0.001 
 Foreign-born 10.51 14.17 
 US-born 89.49 85.83 
CharacteristicsWithout CLD (N = 47,224)With CLDa (N = 11,980)p value
Overall, N = 59,204 
Mean age, years 45.21±17.44 49.59±17.24 <0.001 
Sex (%) <0.001 
 Female 53.99 42.94 
 Male 46.01 57.06 
Race (%) <0.001 
 Non-Hispanic white 71.93 66.50 
 Non-Hispanic black 11.86 12.26 
 Hispanic 14.02 17.57 
 Non-Hispanic Asian 2.19 3.67 
Educational level (%) <0.001 
 Without high school diploma/GED certificate 16.85 23.6 
 With high school diploma/GED certificate 25.32 26.23 
 Some college or associate’s degree 29.54 30.15 
 College degree or higher 28.3 20.02 
Family income-to-poverty ratio (%) <0.001 
 ≥5 25.89 20.82 
 <5 74.11 79.18 
Marital status (%) 0.39 
 Not married 44.98 44.29 
 Married 55.02 55.71 
Birthplace (%) <0.001 
 Foreign-born 10.51 14.17 
 US-born 89.49 85.83 

GED, general education development.

aIncludes persons with prior infection with viral hepatitis B or C, NAFLD, and ALD.

Participants with CLD by Family Income-to-Poverty Ratio

CLD in general is more common in participants with a family income-to-poverty ratio <5, regardless of race or etiology of CLD. Among participants with CLD, we found similar disparities between participants with lower income and those with higher income with participants with CLD and lower income being older (49.09 ± 17.94 vs. 50.96 ± 11.58), more likely male (55.17 vs. 64.75%), more likely non-Hispanic black (13.61 vs. 5.69%) or Hispanic (10.59 vs. 5.33%) or foreign-born (15.16 vs. 6.81%), and having lower educational levels (without high school diploma: 27.14 vs. 6.19%) (all p < 0.001) (Table 2). The findings were also mostly similar in subgroup analysis comparing participants with higher versus lower income by individual liver disease etiology with 4,422 participants with viral hepatitis (including subgroups with current and past infection), 5,718 participants with NAFLD, and 2,212 participants with ALD (online suppl. Table 1). Findings for lower, middle-, and higher income levels demonstrate similar disparities, with lower income CLD participants being more likely non-Hispanic black (20.29 vs. 10.21 vs. 16.29%) or Hispanic (32.14 vs. 13.79 vs. 25.63%) or foreign-born (24.37 vs. 11.15 vs. 24.64%) and having lower educational levels (without high school diploma: 44.17 vs. 18.33 vs. 33.79%) (all p < 0.001) (online suppl. Table 3). Results were mostly similar in subgroup analysis as well when comparing participants in lower, middle-, and higher income levels by individual liver disease etiology, namely, by viral hepatitis, NAFLD, and ALD (online suppl. Table 7).

Table 2.

Characteristics of participants with CLD by income level

CharacteristicsFamily income-to-poverty ratio ≥5 (N = 2,494)Family income-to-poverty ratio <5 (N = 9,486)p value
With CLDa, N = 11,980 
Mean age, years 50.96±11.58 49.09±17.94 0.003 
Sex (%) <0.001 
 Female 35.25 44.83 
 Male 64.75 55.17 
Race (%) <0.001 
 Non-Hispanic white 85.22 62.98 
 Non-Hispanic black 5.69 13.61 
 Hispanic 5.33 10.59 
 Non-Hispanic Asian 3.76 3.37 
Educational level (%) <0.001 
 Without high school diploma/GED certificate 6.19 27.14 
 With high school diploma/GED certificate 16.18 28.99 
 Some college or associate’s degree 31.53 30.13 
 College degree or higher 46.10 13.73 
Marital status (%) <0.001 
 Not married 27.23 48.41 
 Married 72.77 51.59 
Birthplace (%) <0.001 
 Foreign-born 6.81 15.16 
 US-born 93.19 84.84 
CharacteristicsFamily income-to-poverty ratio ≥5 (N = 2,494)Family income-to-poverty ratio <5 (N = 9,486)p value
With CLDa, N = 11,980 
Mean age, years 50.96±11.58 49.09±17.94 0.003 
Sex (%) <0.001 
 Female 35.25 44.83 
 Male 64.75 55.17 
Race (%) <0.001 
 Non-Hispanic white 85.22 62.98 
 Non-Hispanic black 5.69 13.61 
 Hispanic 5.33 10.59 
 Non-Hispanic Asian 3.76 3.37 
Educational level (%) <0.001 
 Without high school diploma/GED certificate 6.19 27.14 
 With high school diploma/GED certificate 16.18 28.99 
 Some college or associate’s degree 31.53 30.13 
 College degree or higher 46.10 13.73 
Marital status (%) <0.001 
 Not married 27.23 48.41 
 Married 72.77 51.59 
Birthplace (%) <0.001 
 Foreign-born 6.81 15.16 
 US-born 93.19 84.84 

GED, general education development.

aIncludes persons with prior infection with viral hepatitis B or C.

Within each racial and ethnic group, about 2 in 3 (73–76%) participants in the non-Hispanic white and non-Hispanic Asian groups had lower income, while about 9 in 10 (87–92%) of Hispanic and non-Hispanic black participants had lower income (Fig. 1a), with consistent findings in subgroup analyses of participants with viral hepatitis, or NAFLD, or ALD (Fig. 1b, c). Disparities remain in further sub-analysis within the viral hepatitis subgroup, with the lowest percentage of lower income participants observed among non-Hispanic white participants (82.87%) compared to 87.02% among non-Hispanic Asian and 95–96% among Hispanic and non-Hispanic black participants (online suppl. Fig. 2A); and again, with similar disparities observed in further subgroups of participants with HBV infection overall or current HBV infection, as well as in subgroups of participants with HCV infection overall and current HCV infection (online suppl. Fig. 2B–D). By lower, middle-, and higher income levels for the overall cohort, about 1 in 10 (10–15%) participants in the non-Hispanic white and Asian groups had lower income, while about 1 in 4 (26–29%) of Hispanic and non-Hispanic black participants had lower income (online suppl. Fig. 4A), with similar disparities found in further sub-analyses of participants with viral hepatitis, NAFLD, ALD (online suppl. Fig. 4B–D), current viral hepatitis infection, HBV infection overall, current HBV infection, HCV infection overall, but not for participants with current HCV infection (online suppl. Fig. 6A–E).

Fig. 1.

Distribution of race by income levels in the overall CLD cohort (a) and in subgroups: viral hepatitis history (b), nonalcoholic fatty liver disease (NAFLD) (c), and alcohol-associated liver disease (ALD) (d).

Fig. 1.

Distribution of race by income levels in the overall CLD cohort (a) and in subgroups: viral hepatitis history (b), nonalcoholic fatty liver disease (NAFLD) (c), and alcohol-associated liver disease (ALD) (d).

Close modal

Participants with CLD by Educational Level

CLD is more common in participants without college education. Among participants with CLD, significant disparities were noted between participants with lower education levels and those with higher educational levels, with CLD participants lacking college education being more likely non-Hispanic black (14.36 vs. 10.12%) or Hispanic (24.88 vs. 10.23%) or foreign-born (18.64 vs. 9.77%) and having lower income (family income-to-poverty ratio <5: 90.51 vs. 68.24%) (all p < 0.001) (online suppl. Table 5).

Overall Morality in Participants with CLD by Family Income Poverty Ratio

The cumulative 10-year mortality incidence for lower income participants was significantly higher than those of higher income participants with CLD (15.26 vs. 8.00%, p < 0.001, Fig. 2a) with consistent findings in subgroups of participants with viral hepatitis (21.92 vs. 11.67%, p = 0.001, Fig. 2b) and NAFLD (20.22 vs. 11.17%, p < 0.001, Fig. 2c) but not for ALD where there was no difference by income-to-poverty ratio (6.91 vs. 5.97%, p = 0.71, Fig. 2d). Among the subgroups with viral hepatitis (online suppl. Fig. 3A–C), we found similar findings in subgroups of participants with hepatitis B overall and current hepatitis B as well as in subgroups of participants with hepatitis C overall and those with current hepatitis C infection. When analyzing by lower, middle-, and higher income participants, the cumulative 10-year mortality incidence for lower income participants was slightly higher than that of their middle-income counterparts but not higher compared to higher income participants with CLD (1.50 vs. 1.46 vs. 1.82%, p < 0.001, online suppl. Fig. 5A). There were no differences between the three income-to-poverty ratio levels in subgroups of participants with viral hepatitis (2.30 vs. 2.22 vs. 2.16% p = 0.92, online suppl. Fig. 5B), NAFLD (1.93 vs. 2.03 vs. 2.67%, p = 0.06, online suppl. Fig. 5C), ALD (0.90 vs. 0.59 vs. 1.05%, p = 0.08, online suppl. Fig. 5D), current viral hepatitis infection (2.00 vs. 2.22 vs. 2.29%, p = 0.87, online suppl. Fig. 7A), HBV infection overall (2.41 vs. 2.26 vs. 2.27%, p = 0.83, online suppl. Fig. 7B), current HBV infection (2.67 vs. 2.34 vs. 1.18%, p = 0.59, online suppl. Fig. 7C), HCV infection overall (1.93 vs. 2.00 vs. 2.41%, p = 0.85, online suppl. Fig. 7D), or current HCV infection (1.84 vs. 2.27 vs. 3.33%, p = 0.42, online suppl. Fig. 7E).

Fig. 2.

Overall mortality in the overall CLD cohort (a) and in subgroups: viral hepatitis history (b), nonalcoholic fatty liver disease (NAFLD) (c), and alcohol-associated liver disease (ALD) (d).

Fig. 2.

Overall mortality in the overall CLD cohort (a) and in subgroups: viral hepatitis history (b), nonalcoholic fatty liver disease (NAFLD) (c), and alcohol-associated liver disease (ALD) (d).

Close modal

In the multivariable Cox regression analyses adjusted for age, sex, race, and birthplace, CLD participants with lower income were about 2 times more likely to die than CLD participants with higher income (HR: 2.07; 95% CI: 0.70–6.14) overall and in subgroup of participants with viral hepatitis (HR: 2.16; 95% CI: 0.73–6.40) with similar findings within subgroup of those with HBV and HCV history (online suppl. Table 2) and in subgroups of participants with NAFLD (HR: 2.32; 95% CI: 1.69–3.18) (Table 3 and online suppl. Table 2). However, among the subgroup with ALD, the income level was not significantly associated with mortality (HR: 1.04; 95% CI: 0.48–2.24) (Table 3).

Table 3.

Factors associated with overall mortality in the overall CLD cohort and in subgroups by liver disease etiology

Income-to-poverty ratioNumber of eventsUnivariable HR (95% CI)p valueMultivariable HRa (95% CI)p value
Overall, N = 53,446 
 ≥5 490  <0.001 
 <5 4,494 2.24 (1.98–2.53) <0.001 2.03 (1.80–2.29) 
Viral hepatitis history, N = 4,422 
 ≥5 36  0.003 
 <5 506 1.92 (1.34–2.75) 0.001 2.03 (1.27–3.23) 
NAFLD, N = 5,718 
 ≥5 67  <0.001 
 <5 639 2.49 (1.73–3.59) <0.001 2.32 (1.69–3.18) 
ALD, N = 2,212 
 ≥5 13  0.65 
 <5 91 1.09 (0.52–2.26) 0.82 1.04 (0.48–2.24) 
Income-to-poverty ratioNumber of eventsUnivariable HR (95% CI)p valueMultivariable HRa (95% CI)p value
Overall, N = 53,446 
 ≥5 490  <0.001 
 <5 4,494 2.24 (1.98–2.53) <0.001 2.03 (1.80–2.29) 
Viral hepatitis history, N = 4,422 
 ≥5 36  0.003 
 <5 506 1.92 (1.34–2.75) 0.001 2.03 (1.27–3.23) 
NAFLD, N = 5,718 
 ≥5 67  <0.001 
 <5 639 2.49 (1.73–3.59) <0.001 2.32 (1.69–3.18) 
ALD, N = 2,212 
 ≥5 13  0.65 
 <5 91 1.09 (0.52–2.26) 0.82 1.04 (0.48–2.24) 

CLD, chronic liver disease; HR, hazard ratio.

aAdjusted for age, sex, race, and birthplace.

In multivariable Cox regression analyses adjusted for age, sex, race, and birthplace for CLD participants by lower, middle-, and higher income, participants with lower income were about one-and-a-half times more likely to die than CLD participants with higher income (HR: 1.58; 95% CI: 1.32–1.89) overall (online suppl. Table 4). However, among subgroups with viral hepatitis infection overall, NAFLD, ALD (online suppl. Table 4), current viral hepatitis infection, HBV infection overall, current HBV infection, HCV infection overall, or current HCV infection (online suppl. Table 8), the income level was not significantly associated with mortality.

Overall Morality in Participants with CLD by Educational Level

In multivariable Cox regression analyses adjusted for age, sex, race, and birthplace for CLD participants by the educational level, participants without college education were about one-and-a-half times more likely to die than CLD participants with college education (HR: 1.52; 95% CI: 1.41–1.64) overall and in subgroups of participants with viral hepatitis (HR: 1.38; 95% CI: 1.05–1.81) and NAFLD (HR: 1.38; 95% CI: 1.13–1.67) (online suppl. Table 6).

In this population-based study of almost 60,000 US people, we aimed to discern the income levels and outcomes of individuals with CLD and found that lower income, foreign-born, and racial and ethnic minority groups were disproportionately represented among those with CLD. Importantly, almost one-fourth of individuals in the overall cohort had CLD. This proportion is consistent with a previous finding in a study on the prevalence of NAFLD and risk factors mortality in the USA [24]. People with CLD and lower income also had double the mortality risk compared to their higher income counterparts. However, while long-term mortality was significantly and independently associated with the income level among people with viral hepatitis (current or past infection) and people with NAFLD, there was no significant impact by income on mortality of people with ALD. These findings highlight severe socioeconomic disparities among people with CLD, with the greatest burden of disease falling on lower income, racial and ethnic minority groups, and foreign-born individuals.

The higher prevalence of lower income, racial and ethnic minority groups, and foreign-born status among people with CLD is consistent with prior reports. For viral hepatitis B, much of this infection in the USA is among foreign-born people [25]. The prevalence of hepatitis B is also higher among racial and ethnic minority groups with the highest prevalence in Asian individuals followed by black and Hispanic individuals [25]. For HCV infection, prior studies demonstrate a significant association between lower income among racial and ethnic minority groups and injection drug use, consequently contributing to greater prevalence of HCV infection among non-Hispanic black and Hispanic individuals, with lower income persons experiencing greater burden of disease [26]. Our results are also consistent with a prior report of only 78% of the overall US Hispanic population compared to 93% of Hispanic individuals with CLD observed in our study [7] having lower income, which can be due to various cultural as well as socioeconomic barriers predisposing lower income individuals to health disparities.

It is well known that many liver diseases are more common in male. Although males were most adversely affected by CLD regardless of the income level compared to females, in the lower income group, females comprised 45% of this cohort. In fact, across all the CLDs studied, 30% to almost 50% of the lower income group was comprised of females. Therefore, practitioners need to have a high index of suspicion that lower income females are also at risk for CLD and should undergo similar screening for risk factors as males.

The disproportionate CLD burden among the lower income group is also apparent within the population with CLD in which those with CLD and lower income had 2 times the risk of mortality compared to those with CLD and higher income overall, after adjustment for age, sex, race, and foreign-born versus US-born status. These findings further highlight the impact of lower income as a socioeconomic factor contributing to health disparities in that it impacts not only the CLD disease burden but also mortality outcome among those with CLD.

Another important finding of the current study is the differences in the association between lower income status and mortality outcome by liver disease etiology. While the association between lower income and higher mortality was consistent for the group with viral hepatitis and NAFLD, there was no significant association between income and mortality among those with ALD. While previous studies indicate that individuals with lower socioeconomic status are disproportionately affected by alcohol-attributable harm, including ALD, which is likely due to risk factors commonly found in lower income individuals, smoking, obesity, access to health services, and drinking cultures [27‒29], the income level did not seem to impact mortality outcome among those with ALD. Potential reasons for this may be due to other nonfinancial factors that affect health outcomes among people with alcohol use disorders such as poor social isolation or nonadherence to follow-up care [30‒32]. There is also no known effective medical treatment for ALD, while antiviral therapies that may be more accessible to higher income people with viral hepatitis are well-known to decrease the risk of liver cancer, liver decompensation, and premature deaths [33].

There is also disproportionate CLD burden among the lower educational-level group in which those with CLD and without college education had about 1.5 times the risk of mortality compared to those with CLD and with college education overall, after adjustment for age, sex, race, and foreign-born versus US-born status. The greater prevalence of CLD, lower income, racial and ethnic minority groups, and foreign-born status among people with lower education levels is consistent with prior reports. It is already well noted that there is a strong association between education and CLD prevalence and mortality [34‒38]. Thus, our findings corroborate and underscore the impact of lower educational status as a socioeconomic factor contributing to health disparities in that it impacts not only CLD disease burden but also mortality outcomes among those with CLD.

Our results are consistent with prior studies indicating greater burden of CLD among Hispanic and non-Hispanic black participants relative to other groups [8, 9], and with previous studies indicating socioeconomic disparity in CLD mortality in lower income individuals [19]. Few prior studies have included a large enough sample of non-Hispanic black participants to draw conclusions about racial differences in NAFLD [39], while our study is representative of the US population and highlighted racial disparity in viral hepatitis and ALD as well. Additionally, foreign-born status was heavily represented among CLD persons, which is consistent with multiple prior reports of greater prevalence of viral hepatitis [40, 41]. The linkage between foreign-born status and CLD can be due to lack of preventive measures such as vaccination, early detection techniques, medications, and education in different countries as well as language, financial, and cultural barriers immigrants face in seeking care in the USA to prevent CLD incidence and complications.

Nonetheless, our study highlights the disparate burden of CLD among the different races and ethnicities, and although studies have individually highlighted some of these findings, our study places these findings together using a representative sample of the population of the USA. With these results, policymakers can now focus on interventions that may be most appropriate for each group. For instance, between 60% and 70% of those with an HBV infection were not born in the USA regardless of their income status, which highlights the need to make sure the outreach and interventions to treat this group are culturally focused. A recent study presented a machine learning-generated algorithm which includes the place of birth to help practitioners easily identify those at risk so that culturally developed interventions can be provided to the patient [42‒44]. Also, since over half the individuals with CLD in the lower income group lacked higher education, participants’ level of health literacy must also be considered when developing any interventions [45, 46]. One suggestion shown to be effective is the use of visual aids when working with those without a college degree [47].

There are also limitations of our study. The first is that ALD is probably vastly underestimated given the reliance on self-report of alcohol consumption, which has been shown to underreport the true prevalence of alcohol consumption. Second, though NHANES utilizes complex multistage, stratified, clustered probability sampling to represent the noninstitutionalized, civilian population of the USA, it is likely that there is still undersampling of people with limited English ability and marginalized populations such as those with substance use disorders. However, both of these limitations likely underestimate the observed disparities but not change the direction of our results.

In this study, we aimed to highlight socioeconomic disparities among those with CLD and found those whose income-to-poverty ratio was <5 were disproportionately affected. Outside of the non-Hispanic white population, the Hispanic group carried the greater burden for NAFLD and ALD compared to the other ethnicities, while the non-Hispanic black group carried a greater burden for viral hepatitis. Those with a lower education level and those born outside of the USA were also disproportionally impacted with lower income women carried a greater burden of CLD. Among those with CLD, those with NAFLD and viral hepatitis and lower income had 2 times the risk of mortality compared to their higher income counterparts, but income status did not have a significant impact on mortality among those with ALD. Given these findings, interventions must not only be culturally appropriate but also require attention to the potential lack of health literacy and social support and be sex-specific.

NHANES collects data through comprehensive written questionnaires, physical examinations, and biological samples from complex multistage, stratified, clustered probability samples that are representative of the noninstitutionalized, civilian population of the USA. Written informed consent was obtained from each participant. The study protocols for the multiple NHANES between 1999 and 2018 were reviewed and approved by the National Center for Health Statistics (NCHS) Ethics Review Board (ERB) of the US Centers for Disease Control and Prevention (CDC); Approval No. #98–12, #2005–06, #2011–17, #2018–01. Further information can be found at https://www.cdc.gov/nchs/nhanes/irba98.htm.

Mindie Nguyen: research funding: Pfizer, Enanta, AstraZeneca, Delfi Technologies, GSK, Gilead, CurveBio, Exact Sciences, Helio Health, Glycotest, and Vir Biotech; consulting: Exelixis, Gilead, Intercept, GSK, Exact Science. Other authors have no conflict of interest to declare.

Brian Thanh Nguyen was supported by the AGA-Dr. Harvey Young Education and Development Foundation’s Young Guts Scholar Program from September 2023 to May 2024, during which this study was conducted. However, this grant was not specifically dedicated to this study’s design, execution, analysis, manuscript conception, planning, writing, and decision to publish.

Conceptualization: Vy Hoang Nguyen and Mindie H Nguyen. Methodology: Brian Thanh Nguyen, Vy Hoang Nguyen, Ramsey Cheung, and Mindie H Nguyen. Formal analysis: Brian Thanh Nguyen, Vy Hoang Nguyen, and Mindie H. Nguyen. Writing – original draft: Brian Thanh Nguyen, Linda Henry, and Mindie H. Nguyen. Data interpretation and writing – review and editing: Brian Thanh Nguyen, Vy Hoang Nguyen, Michael Le, Linda Henry, Ramsey Cheung, and Mindie H. Nguyen. Supervision and guarantor of the manuscript: Mindie H. Nguyen. All the authors approved the final version of the article, including the authorship list.

Additional Information

Brian Thanh Nguyen and Vy Hoang Nguyen should be considered joint first authors.

All data are available publicly at https://wwwn.cdc.gov/nchs/nhanes/. Further inquiries can be directed to the corresponding author.

1.
Younossi
ZM
,
Stepanova
M
,
Younossi
Y
,
Golabi
P
,
Mishra
A
,
Rafiq
N
, et al
.
Epidemiology of chronic liver diseases in the USA in the past three decades
.
Gut
.
2020
;
69
(
3
):
564
8
.
2.
Asrani
SK
,
Devarbhavi
H
,
Eaton
J
,
Kamath
PS
.
Burden of liver diseases in the world
.
J Hepatol
.
2019
;
70
(
1
):
151
71
.
3.
Mokdad
AA
,
Lopez
AD
,
Shahraz
S
,
Lozano
R
,
Mokdad
AH
,
Stanaway
J
, et al
.
Liver cirrhosis mortality in 187 countries between 1980 and 2010: a systematic analysis
.
BMC Med
.
2014
;
12
:
145
Published 2014 Sep. 18.
4.
Cheemerla
S
,
Balakrishnan
M
.
Global epidemiology of chronic liver disease
.
Clin Liver Dis
.
2021
;
17
(
5
):
365
70
. Published 2021 Jun 4.
5.
GBD 2017 Disease and Injury Incidence and Prevalence Collaborators
.
Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017
.
Lancet
.
2018
;
392
(
10159
):
1789
858
.
6.
Moon
AM
,
Singal
AG
,
Tapper
EB
.
Contemporary epidemiology of chronic liver disease and cirrhosis
.
Clin Gastroenterol Hepatol
.
2020
;
18
(
12
):
2650
66
.
7.
Flores
YN
,
Yee
HF
Jr
,
Leng
M
,
Escarce
JJ
,
Bastani
R
,
Salmerón
J
, et al
.
Risk factors for chronic liver disease in Blacks, Mexican Americans, and Whites in the United States: results from NHANES IV, 1999-2004
.
Am J Gastroenterol
.
2008
;
103
(
9
):
2231
8
.
8.
Shaheen
M
,
Schrode
KM
,
Pan
D
,
Kermah
D
,
Puri
V
,
Zarrinpar
A
, et al
.
Sex-specific differences in the association between race/ethnicity and NAFLD among US population
.
Front Med
.
2021
;
8
:
795421
. Published 2021 Dec 2.
9.
Schneider
AL
,
Lazo
M
,
Selvin
E
,
Clark
JM
.
Racial differences in nonalcoholic fatty liver disease in the U.S. population
.
Obesity
.
2014
;
22
(
1
):
292
9
.
10.
Emmanuel
B
,
Shardell
MD
,
Tracy
L
,
Kottilil
S
,
El-Kamary
SS
.
Racial disparity in all-cause mortality among hepatitis C virus-infected individuals in a general US population, NHANES III
.
J Viral Hepat
.
2017
;
24
(
5
):
380
8
.
11.
Liu
G
,
Holmberg
SD
,
Kamili
S
,
Xu
F
.
Racial disparities in the proportion of current, unresolved hepatitis C virus infections in the United States, 2003-2010
.
Dig Dis Sci
.
2014
;
59
(
8
):
1950
7
.
12.
Moore
KJ
,
Gauri
A
,
Koru-Sengul
T
.
Prevalence and sociodemographic disparities of Hepatitis C in Baby Boomers and the US adult population
.
J Infect Public Health
.
2019
;
12
(
1
):
32
6
.
13.
Desai
AP
,
Mohan
P
,
Roubal
AM
,
Bettencourt
R
,
Loomba
R
.
Geographic variability in liver disease-related mortality rates in the United States
.
Am J Med
.
2018
;
131
(
7
):
728
34
.
14.
Chetty
R
,
Stepner
M
,
Abraham
S
,
Lin
S
,
Scuderi
B
,
Turner
N
, et al
.
The association between income and life expectancy in the United States, 2001-2014 [published correction appears in JAMA. 2017 Jan 3;317(1):90]
.
JAMA
.
2016
;
315
(
16
):
1750
66
.
15.
Abdalla
SM
,
Yu
S
,
Galea
S
.
Trends in cardiovascular disease prevalence by income level in the United States
.
JAMA Netw Open
.
2020
;
3
(
9
):
e2018150
.
16.
Bor
J
,
Cohen
GH
,
Galea
S
.
Population health in an era of rising income inequality: USA, 1980-2015
.
Lancet
.
2017
;
389
(
10077
):
1475
90
.
17.
Truesdale
BC
,
Jencks
C
.
The health effects of income inequality: averages and disparities
.
Annu Rev Public Health
.
2016
;
37
(
1
):
413
30
.
18.
Pickett
KE
,
Wilkinson
RG
.
Income inequality and health: a causal review
.
Soc Sci Med
.
2015
;
128
:
316
26
.
19.
Singh
GK
,
Hoyert
DL
.
Social epidemiology of chronic liver disease and cirrhosis mortality in the United States, 1935-1997: trends and differentials by ethnicity, socioeconomic status, and alcohol consumption
.
Hum Biol
.
2000
;
72
(
5
):
801
20
.
20.
Roesch-Dietlen
F
,
González-Santes
M
,
Sánchez-Maza
YJ
,
Díaz-Roesch
F
,
Cano-Contreras
AD
,
Amieva-Balmori
M
, et al
.
Influence of socioeconomic and cultural factors in the etiology of cirrhosis of the liver
.
Rev Gastroenterol Mex Engl Ed
.
2021
;
86
(
1
):
28
35
.
21.
Ruhl
CE
,
Everhart
JE
.
Fatty liver indices in the multiethnic United States national health and nutrition examination survey
.
Aliment Pharmacol Ther
.
2015
;
41
(
1
):
65
76
.
22.
Singal
AK
,
Bataller
R
,
Ahn
J
,
Kamath
PS
,
Shah
VH
.
ACG clinical guideline: alcoholic liver disease
.
Am J Gastroenterol
.
2018
;
113
(
2
):
175
94
.
23.
Wong
T
,
Dang
K
,
Ladhani
S
,
Singal
AK
,
Wong
RJ
.
Prevalence of alcoholic fatty liver disease among adults in the United States, 2001-2016
.
JAMA
.
2019
;
321
(
17
):
1723
5
.
24.
Le
MH
,
Devaki
P
,
Ha
NB
,
Jun
DW
,
Te
HS
,
Cheung
RC
, et al
.
Prevalence of non-alcoholic fatty liver disease and risk factors for advanced fibrosis and mortality in the United States
.
PLoS one
.
2017
;
12
(
3
):
e0173499
.
25.
Le
MH
,
Yeo
YH
,
Cheung
R
,
Henry
L
,
Lok
AS
,
Nguyen
MH
.
Chronic hepatitis B prevalence among foreign-born and U.S.-Born adults in the United States, 1999-2016
.
Hepatology
.
2020
;
71
(
2
):
431
43
.
26.
Zou
B
,
Yeo
YH
,
Le
MH
,
Henry
L
,
Chang
ET
,
Lok
AS
, et al
.
Prevalence of viremic hepatitis C virus infection by age, race/ethnicity, and birthplace and disease awareness among viremic persons in the United States, 1999-2016
.
J Infect Dis
.
2020
;
221
(
3
):
408
18
.
27.
Addolorato
G
,
Russell
M
,
Albano
E
,
Haber
PS
,
Wands
JR
,
Leggio
L
.
Understanding and treating patients with alcoholic cirrhosis: an update
.
Alcohol Clin Exp Res
.
2009
;
33
(
7
):
1136
44
.
28.
Liu
SY
,
Tsai
IT
,
Hsu
YC
.
Alcohol-related liver disease: basic mechanisms and clinical perspectives
.
Int J Mol Sci
.
2021
;
22
(
10
):
5170
Published 2021 May 13.
29.
Stickel
F
,
Datz
C
,
Hampe
J
,
Bataller
R
.
Pathophysiology and management of alcoholic liver disease: update 2016
.
Gut Liver
.
2017
;
11
(
2
):
173
88
[published correction appears in Gut Liver. 2017 May 15;11(3):447].
30.
Farmer
AY
,
Wang
Y
,
Peterson
NA
,
Borys
S
,
Hallcom
DK
.
Social isolation profiles and older adult substance use: a latent profile analysis
.
J Gerontol B Psychol Sci Soc Sci
.
2022
;
77
(
5
):
919
29
.
31.
Engler
PA
,
Ramsey
SE
,
Smith
RJ
.
Alcohol use of diabetes patients: the need for assessment and intervention
.
Acta Diabetol
.
2013
;
50
(
2
):
93
9
.
32.
Hepner
KA
,
Paddock
SM
,
Watkins
KE
,
Hoggatt
KJ
,
Rubenstein
LV
,
Bogart
A
, et al
.
Rates and impact of adherence to recommended care for unhealthy alcohol use
.
J Gen Intern Med
.
2019
;
34
(
2
):
256
63
.
33.
Ogawa
E
,
Chien
N
,
Kam
L
,
Yeo
YH
,
Ji
F
,
Huang
DQ
, et al
.
Association of direct-acting antiviral therapy with liver and nonliver complications and long-term mortality in patients with chronic hepatitis C
.
JAMA Intern Med
.
2023
;
183
(
2
):
97
105
.
34.
Fedeli
U
,
Avossa
F
,
Goldoni
CA
,
Caranci
N
,
Zambon
F
,
Saugo
M
.
Education level and chronic liver disease by aetiology: a proportional mortality study
.
Dig Liver Dis
.
2015
;
47
(
12
):
1082
5
.
35.
Song
Q
,
Liu
C
,
Cheng
W
,
Lin
L
,
Li
T
,
Li
X
, et al
.
Clinical characteristics and risk of all-cause mortality in low education patients with chronic obstructive pulmonary disease in the Chinese population
.
J Glob Health
.
2023
;
13
:
04163
.
36.
Zhao
Q
,
Yin
Y
,
Deng
Y
.
Metabolic associated fatty liver disease and sarcopenia additively increase mortality: a real-world study
.
Nutr Diabetes
.
2023
;
13
(
1
):
21
.
37.
Mackenbach
JP
,
Stirbu
I
,
Roskam
AJ
,
Schaap
MM
,
Menvielle
G
,
Leinsalu
M
, et al,
European Union Working Group on Socioeconomic Inequalities in Health
.
Socioeconomic inequalities in health in 22 European countries
.
N Engl J Med
.
2008
;
358
(
23
):
2468
81
.
38.
IHME-CHAIN Collaborators
.
Effects of education on adult mortality: a global systematic review and meta-analysis
.
Lancet Public Health
.
2024
;
9
(
3
):
e155
65
.
39.
Bambha
K
,
Belt
P
,
Abraham
M
,
Wilson
LA
,
Pabst
M
,
Ferrell
L
, et al
.
Ethnicity and nonalcoholic fatty liver disease
.
Hepatology
.
2012
;
55
(
3
):
769
80
.
40.
Juon
HS
,
Ha
E
,
Kim
F
,
Trang
A
,
Pan
J
,
Blanchard
J
.
Prevalence of viral hepatitis in foreign-born populations in the baltimore-Washington metropolitan area, 2009-2015
.
J Community Health
.
2019
;
44
(
2
):
203
7
.
41.
Le
MH
,
Yeo
YH
,
Cheung
R
,
Henry
L
,
Lok
AS
,
Nguyen
MH
.
Chronic hepatitis B prevalence among foreign-born and U.S.-Born adults in the United States, 1999-2016
.
Hepatology
.
2020
;
71
(
2
):
431
43
.
42.
Ramrakhiani
NS
,
Chen
VL
,
Le
M
,
Yeo
YH
,
Barnett
SD
,
Waljee
AK
, et al
.
Optimizing hepatitis B virus screening in the United States using a simple demographics-based model
.
Hepatology
.
2022
;
75
(
2
):
430
7
.
43.
Freeland
C
,
Bodor
S
,
Perera
U
,
Cohen
C
.
Barriers to hepatitis B screening and prevention for african immigrant populations in the United States: a qualitative study
.
Viruses
.
2020
;
12
(
3
):
305
. Published 2020 Mar 11.
44.
Shankar
H
,
Blanas
D
,
Bichoupan
K
,
Ndiaye
D
,
Carmody
E
,
Martel-Laferriere
V
, et al
.
A novel collaborative community-based hepatitis B screening and linkage to care Program for african immigrants
.
Clin Infect Dis
.
2016
;
62
(
Suppl 4
):
S289
97
.
45.
Perrenoud
B
,
Velonaki
VS
,
Bodenmann
P
,
Ramelet
AS
.
The effectiveness of health literacy interventions on the informed consent process of health care users: a systematic review protocol
.
JBI Database Syst Rev Implement Rep
.
2015
;
13
(
10
):
82
94
.
46.
Nutbeam
D
,
Lloyd
JE
.
Understanding and responding to health literacy as a social determinant of health
.
Annu Rev Public Health
.
2021
;
42
:
159
73
.
47.
Mbanda
N
,
Dada
S
,
Bastable
K
,
Ingalill
GB
,
Ralf W
S
.
A scoping review of the use of visual aids in health education materials for persons with low-literacy levels
.
Patient Educ Couns
.
2021
;
104
(
5
):
998
1017
.