ABSTRACT
Background
Epidemiological data on the role of overall dietary patterns in nonalcoholic fatty liver disease (NAFLD) are limited, especially from population-based prospective studies.
Objectives
We investigated the associations between dietary patterns assessed by predefined diet quality indexes (DQIs) and NAFLD risk by cirrhosis status in African Americans, Japanese Americans, Latinos, Native Hawaiians, and whites from the Multiethnic Cohort (MEC).
Methods
A nested case-control analysis was conducted within the MEC. NAFLD cases were identified by linkage to 1999–2016 Medicare claims. Four DQIs—Healthy Eating Index (HEI)-2015, Alternative Healthy Eating Index-2010, alternate Mediterranean diet score, and Dietary Approaches to Stop Hypertension (DASH) score—were calculated from a validated FFQ administered at baseline. Conditional logistic regression was used to estimate the ORs and 95% CIs with adjustment for multiple covariates.
Results
Analyses included 2959 NAFLD cases (509 with cirrhosis; 2450 without cirrhosis) and 29,292 matched controls. Higher scores for HEI-2015 (i.e., highest compared with lowest quintile OR: 0.83; 95% CI: 0.73, 0.94; P for trend = 0.002) and DASH (OR: 0.78; 95% CI: 0.69, 0.89; P for trend < 0.001), reflecting favorable adherence to a healthful diet, were inversely associated with NAFLD risk. Whereas there were no differences by sex or race/ethnicity, the inverse association was stronger for NAFLD with cirrhosis than for NAFLD without cirrhosis (P for heterogeneity = 0.03 for HEI-2015 and 0.05 for DASH).
Conclusions
Higher HEI-2015 and DASH scores were inversely associated with NAFLD risk in this ethnically diverse population. The findings suggest that having better diet quality may reduce NAFLD risk with more benefit to NAFLD with cirrhosis.
Keywords: steatosis, NAFLD, nutrition, food, cirrhosis
Introduction
Nonalcoholic fatty liver disease (NAFLD) is the most common cause of chronic liver disease in the United States, affecting 80–100 million people, and is increasing nationally and globally (1). Diet may influence NAFLD development by regulating overall adiposity and interacting with genetic predisposition to NAFLD (2). To prevent and manage NAFLD, dietary modification has been suggested in terms of overall dietary patterns as well as single foods or nutrients (3–5). Dietary patterns reflect the complexity of foods consumed and are often assessed by predefined diet quality indexes (DQIs) (6–8). For NAFLD, the Mediterranean diet has been recommended owing to its potential to prevent cardiovascular events or related diseases (4, 9). However, evidence on the efficacy of other theoretical dietary patterns in NAFLD prevention and treatment is still limited. Most previous studies are either cross-sectional or case-control and have generally supported the thesis that recommended dietary patterns are inversely associated with NAFLD (10–14). Prospective studies examining the association between dietary patterns and NAFLD have rarely been conducted, especially in racially/ethnically diverse populations.
In the current study, using data from the Multiethnic Cohort (MEC) study that comprises African-American, Japanese-American, Latino, Native Hawaiian, and white participants, we prospectively investigated the associations of 4 DQIs [Healthy Eating Index (HEI)-2015, Alternative Healthy Eating Index (AHEI)-2010, alternate Mediterranean diet (aMED) score, and Dietary Approaches to Stop Hypertension (DASH) score] with NAFLD risk overall and by sex, race/ethnicity, and cirrhosis status.
Methods
Study population
The MEC was established to investigate the roles of diet and other lifestyle factors in cancer and other chronic diseases, as previously described (15). The cohort was designed to include adults living in Hawaii and California from 5 targeted racial/ethnic groups: African American, Native Hawaiian, Japanese American, Latino, and white. In 1993–1996, >215,000 men and women aged 45–75 y completed a 26-page comprehensive questionnaire on diet, medical history, and lifestyle (available from: https://www.uhcancercenter.org/mec). MEC participants who reached age 65 y were linked to Medicare Services claims (1999–2016) using Social Security numbers, sex, and date of birth; 93% of them were successfully linked (16). For this study, we restricted the analyses to the Medicare fee-for-service (FFS) participants in the MEC (n = 123,196) because the outcome NAFLD was identified using FFS claims (see “Case selection” below). We excluded participants who were not from the 5 major ethnic groups (n = 7511), had invalid dietary data based on implausible macronutrient intakes (n = 4498), and had missing baseline information on the important relevant variables (e.g., BMI, diabetes, vigorous physical activity) (n = 5756). A total of 105,431 eligible participants were available for nested case-control analysis. The excluded and eligible participants had similar duration of Medicare coverage, smoking status, average BMI, and energy intake, but the eligible participants were younger, had a higher education level, and a higher proportion were Japanese Americans.
Selection of cases and controls
For the current nested case-control analysis, NAFLD and cirrhosis cases were identified among eligible participants by using 1 inpatient or ≥2 outpatient/carrier qualifying claims during a 1-y period between 1999 and 2016. NAFLD was identified using International Classification of Diseases, ninth revision (ICD-9) codes (571.8 and 571.9) and tenth revision (ICD-10) codes (K75.81, K760, K7689, K741, and K769) as previously described (17, 18). Using the American Association for the Study of Liver Diseases guidelines (19), NAFLD cases who reported >21 drinks/wk (men) or >14 drinks/wk (women) were reclassified as alcoholic liver disease. The median time between cohort entry (baseline) and the first NAFLD-related claim was 17.4 y. Cirrhosis status among NAFLD patients was identified using the following codes: 571.2, K7030, K7031 (cirrhosis with alcoholism with and without ascites); 571.5, K740, K7460, K7469 (cirrhosis without mention of alcohol); 456.0, 456.1, 456.20, 456.21, I8501, I8500, I8511, I8510 (esophageal varices); 567.23, K652 (spontaneous bacterial peritonitis); 572.2, K7290, K7291 (hepatic encephalopathy); and 572.4, K767 (hepatorenal syndrome). Controls were randomly selected among eligible participants without liver disease and individually matched to cases (with a ratio ≤10:1) by birth year, sex, ethnicity, and length of FFS Medicare enrollment. A total of 2959 cases (509 with cirrhosis and 2450 without cirrhosis) and 29,292 controls were included in analyses. The mean number of controls per case was 9.9.
The Institutional Review Boards for the University of Southern California and the University of Hawaii approved this study.
DQIs
Each participant completed a quantitative food-frequency questionnaire (QFFQ) at baseline designed from 3-d measured food records (15). Importantly, a calibration substudy showed satisfactory correlations between energy-adjusted daily nutrient estimates from the QFFQ and three 24-h recalls (19). As part of the Dietary Patterns Methods Project, diet quality of participants was assessed using the following predefined DQIs based on responses to the QFFQ: HEI-2015, AHEI-2010, aMED, and DASH (20–22). The HEI-2015 (0–100 points with 13 components) reflects adherence to the 2015–2020 Dietary Guidelines for Americans (23). The AHEI-2010 (0–110 points with 11 components) was developed to identify dietary patterns consistently associated with lower risk of chronic disease (24). The aMED (0–9 points with 9 components) was an adaptation of the Mediterranean Diet Score (MDS) (25), which was consistently associated with lower risk of chronic disease (26, 27). The DASH index (8–40 points with 8 components) reflects adherence to the DASH diet that was designed to help treat or prevent high blood pressure (28, 29). For all DQIs, higher scores reflect better quality of diets.
Statistical analysis
To estimate the associations between the DQIs and NAFLD risk, ORs and 95% CIs were calculated with conditional logistic regression, modeling NAFLD status. The DQIs were divided into quintiles, which were determined by the overall distribution of each DQI in both cases and controls. Trend tests were performed by entering a variable assigned the median of the appropriate quintile group as a continuous variable. P values for dose-response were based on the Wald statistics for trend variables.
Matched sets were used as strata in the conditional logistic models, which accounted for the matching criteria: birth year, sex, ethni-city, and length of Medicare FFS enrollment. Multivariate ORs were further adjusted for variables that were related to liver disease in the MEC (17, 30): BMI (in kg/m2), physical activity (hours spent in vigorous works or sports per day), total energy intake (log-transformed kcal/d), and coffee consumption (0, >0‒1, 2‒3, ≥4 cups/d). For the HEI-2015 and DASH, alcohol consumption (g/d) was also adjusted for, because these 2 indexes do not include a component for alcohol intake. A sensitivity analysis excluding the alcohol component from the aMED and AHEI-2010 was performed. Because the associations between DQIs and NAFLD risk did not vary by sex, subgroup analyses by race/ethnicity and cirrhosis status were performed in men and women combined. Tests for heterogeneity across racial/ethnic groups were based on the Wald statistics with 4 df for the cross-product terms of trend variables and race/ethnicity. We tested differences in the DQI trend parameters for the outcomes of NAFLD with and without cirrhosis by using case-only unconditional logistic regression with the event being cirrhosis (yes compared with no) and adjusting for the matching variables. All analyses were performed using SAS version 9.4 (SAS Institute, Inc.). Statistical significance was considered at P < 0.05 and all tests were 2-sided.
Results
Table 1 presents selected baseline characteristics in cases and controls. Compared with controls, cases were more likely to have higher BMI and less likely to be physically active and drink alcoholic beverages and coffee. Mean scores of the 4 DQIs were slightly lower in NAFLD cases than in controls.
TABLE 1.
Characteristics of nonalcoholic fatty liver disease cases and controls in the Multiethnic Cohort1
| Cases | Controls | |
|---|---|---|
| Participants, n | 2959 | 29,292 |
| Age at cohort entry, y2 | 57.7 ± 7.8 | 57.8 ± 7.8 |
| Sex2 | ||
| Male | 1111 (37.5) | 11,088 (37.9) |
| Female | 1848 (62.5) | 18,204 (62.1) |
| Race/ethnicity2 | ||
| African American | 202 (6.8) | 2013 (6.9) |
| Native Hawaiian | 185 (6.3) | 1838 (6.3) |
| Japanese American | 1490 (50.4) | 14,685 (50.1) |
| Latino | 611 (20.6) | 6053 (20.7) |
| White | 471 (15.9) | 4703 (16.1) |
| BMI, kg/m2 | 27.2 ± 4.9 | 25.7 ± 4.6 |
| Physical activity,3 h/d | 0.31 ± 0.69 | 0.36 ± 0.78 |
| Daily intake | ||
| Alcohol, g | 2.8 ± 6.3 | 6.8 ± 19.3 |
| Energy, kcal | 2117.2 ± 994.2 | 2126.4 ± 983.2 |
| Coffee, cup | 1.0 ± 1.1 | 1.1 ± 1.1 |
| Coffee, g | 242.1 ± 268.7 | 254.9 ± 272.9 |
| ≥1 cup of coffee | 1503 (50.8) | 15,668 (53.5) |
| ≥2 cups of coffee | 736 (24.9) | 7630 (26.0) |
| ≥4 cups of coffee | 128 (4.3) | 1452 (5.0) |
| Diet quality index | ||
| HEI-2015 | 66.3 ± 10.5 | 67.2 ± 10.5 |
| AHEI-2010 | 64.8 ± 9.4 | 65.2 ± 9.5 |
| aMED | 4.1 ± 1.8 | 4.2 ± 1.8 |
| DASH | 23.4 ± 4.4 | 23.8 ± 4.4 |
Values are n (%) or mean ± SD unless indicated otherwise. AHEI, Alternative Healthy Eating Index; aMED, alternate Mediterranean diet; DASH, Dietary Approaches to Stop Hypertension; HEI, Healthy Eating Index.
Matching factors.
Hours spent in vigorous works or sports per day.
Table 2 presents the ORs and 95% CIs for associations of DQIs with NAFLD risk. HEI-2015, AHEI-2010, and DASH, but not aMED showed an inverse association with NAFLD in the basic model. With multivariate adjustment, the inverse association remained statistically significant for HEI-2015 (for the highest compared with the lowest quintile, OR: 0.83; 95% CI: 0.73, 0.94; P for trend = 0.002) and DASH (OR: 0.78; 95% CI: 0.69, 0.89; P for trend < 0.001), but not for AHEI-2010 (OR: 0.91; 95% CI: 0.80, 1.03; P for trend = 0.34). Similar results were obtained when excluding the alcohol component from the aMED and AHEI-2010 analyses (data not shown).
TABLE 2.
The association between dietary quality indexes and nonalcoholic fatty liver disease in the Multiethnic Cohort1
| Cases, n | Controls, n | OR2 (95% CI) | OR3 (95% CI) | |
|---|---|---|---|---|
| HEI-2015 | ||||
| Q1 (17.9–58.0) | 664 | 5778 | 1.00 (ref.) | 1.00 (ref.) |
| Q2 (58.1–64.3) | 630 | 5812 | 0.93 (0.83, 1.05) | 0.95 (0.84, 1.07) |
| Q3 (64.4–69.9) | 564 | 5884 | 0.82 (0.73, 0.92) | 0.85 (0.75, 0.96) |
| Q4 (70.0–76.4) | 577 | 5878 | 0.84 (0.74, 0.94) | 0.89 (0.79, 1.00) |
| Q5 (76.5–99.6) | 524 | 5940 | 0.75 (0.66, 0.85) | 0.83 (0.73, 0.94) |
| P for trend | <0.0001 | 0.0022 | ||
| AHEI-2010 | ||||
| Q1 (25.1–57.1) | 623 | 5814 | 1.00 (ref.) | 1.00 (ref.) |
| Q2 (57.2–62.8) | 588 | 5857 | 0.94 (0.83, 1.06) | 0.94 (0.84, 1.06) |
| Q3 (62.9–67.6) | 601 | 5846 | 0.96 (0.85, 1.08) | 0.98 (0.87, 1.10) |
| Q4 (67.7–73.1) | 606 | 5856 | 0.96 (0.85, 1.08) | 1.01 (0.89, 1.14) |
| Q5 (73.2–104.5) | 541 | 5919 | 0.85 (0.75, 0.96) | 0.91 (0.80, 1.03) |
| P for trend | 0.024 | 0.3366 | ||
| aMED | ||||
| Q1 (0–2) | 590 | 5592 | 1.00 (ref.) | 1.00 (ref.) |
| Q2 (3) | 531 | 5282 | 0.95 (0.84, 1.08) | 0.98 (0.86, 1.11) |
| Q3 (4) | 561 | 5584 | 0.95 (0.84, 1.08) | 0.99 (0.88, 1.13) |
| Q4 (5) | 565 | 5355 | 1.00 (0.88, 1.13) | 1.06 (0.93, 1.21) |
| Q5 (6–9) | 712 | 7479 | 0.90 (0.80, 1.01) | 1.00 (0.88, 1.15) |
| P for trend | 0.1762 | 0.6136 | ||
| DASH | ||||
| Q1 (9–19) | 603 | 5138 | 1.00 (ref.) | 1.00 (ref.) |
| Q2 (20–22) | 631 | 6269 | 0.85 (0.75, 0.96) | 0.85 (0.76, 0.96) |
| Q3 (23–24) | 510 | 4898 | 0.88 (0.77, 0.99) | 0.89 (0.78, 1.01) |
| Q4 (25–27) | 650 | 6749 | 0.81 (0.72, 0.91) | 0.81 (0.71, 0.91) |
| Q5 (28–39) | 565 | 6238 | 0.76 (0.67, 0.86) | 0.78 (0.69, 0.89) |
| P for trend | <0.0001 | 0.0003 | ||
AHEI, Alternative Healthy Eating Index; aMED, alternate Mediterranean diet; DASH, Dietary Approaches to Stop Hypertension; HEI, Healthy Eating Index.
Stratified by matching factors (birth year, sex, race/ethnicity, and length of Medicare enrollment).
Further adjusted for BMI (in kg/m2), physical activity (h/d), total energy intake (log-transformed kcal/d), and coffee consumption (0, >0‒1, 2‒3, ≥4 cups/d). For HEI-2015 and DASH, further adjusted for alcohol intake (g/d).
The associations between DQIs and NAFLD risk did not vary by sex (Ps for heterogeneity > 0.26), although the inverse trends for HEI-2015 and DASH appeared stronger in men than in women (Supplemental Table 1). In racial/ethnic-specific analyses, no differences in the associations across the 5 racial/ethnic groups were observed (Ps for heterogeneity > 0.19) (Supplemental Table 2).
In analyses stratified by cirrhosis status (Table 3), the inverse association was stronger for NAFLD with cirrhosis than for NAFLD without cirrhosis (P for heterogeneity = 0.03 for HEI-2015 and 0.05 for DASH). In addition, higher AHEI-2010 scores were associated with a lower risk of NAFLD with cirrhosis (OR: 0.64; 95% CI: 0.46, 0.89; P for trend = 0.04), whereas no significant association was found for NAFLD without cirrhosis (P for heterogeneity = 0.15).
TABLE 3.
Association between dietary quality indexes and nonalcoholic fatty liver disease by cirrhosis status in the Multiethnic Cohort1
| Cirrhosis | No cirrhosis | ||||||
|---|---|---|---|---|---|---|---|
| Cases | Controls | OR (95% CI)2 | Cases | Controls | OR (95% CI)2 | P-heterogeneity | |
| HEI-2015 | |||||||
| Q1 (17.9–58.0) | 113 | 986 | 1.00 (ref.) | 551 | 4792 | 1.00 (ref.) | |
| Q2 (58.1–64.3) | 134 | 1017 | 1.14 (0.87, 1.51) | 496 | 4795 | 0.91 (0.80, 1.03) | |
| Q3 (64.4–69.9) | 106 | 1009 | 0.96 (0.72, 1.28) | 458 | 4875 | 0.83 (0.73, 0.95) | |
| Q4 (70.0–76.4) | 83 | 1036 | 0.75 (0.55, 1.02) | 494 | 4842 | 0.92 (0.80, 1.05) | |
| Q5 (76.5–99.6) | 73 | 1018 | 0.74 (0.54, 1.03) | 451 | 4922 | 0.84 (0.74, 0.97) | |
| P for trend | 0.0088 | 0.0315 | 0.0341 | ||||
| AHEI-2010 | |||||||
| Q1 (25.1–57.1) | 120 | 1032 | 1.00 (ref.) | 503 | 4782 | 1.00 (ref.) | |
| Q2 (57.2–62.8) | 105 | 1052 | 0.82 (0.61, 1.09) | 483 | 4805 | 0.97 (0.85, 1.10) | |
| Q3 (62.9–67.6) | 113 | 1040 | 0.91 (0.68, 1.20) | 488 | 4806 | 0.99 (0.86, 1.13) | |
| Q4 (67.7–73.1) | 102 | 979 | 0.93 (0.69, 1.24) | 504 | 4877 | 1.02 (0.89, 1.17) | |
| Q5 (73.2–104.5) | 69 | 963 | 0.64 (0.46, 0.89) | 472 | 4956 | 0.96 (0.84, 1.10) | |
| P for trend | 0.0395 | 0.8272 | 0.1450 | ||||
| aMED | |||||||
| Q1 (0–2) | 100 | 998 | 1.00 (ref.) | 490 | 4594 | 1.00 (ref.) | |
| Q2 (3) | 100 | 929 | 1.10 (0.81, 1.48) | 431 | 4353 | 0.95 (0.83, 1.09) | |
| Q3 (4) | 105 | 1010 | 1.09 (0.80, 1.49) | 456 | 4574 | 0.97 (0.85, 1.12) | |
| Q4 (5) | 94 | 902 | 1.11 (0.80, 1.55) | 471 | 4453 | 1.05 (0.91, 1.21) | |
| Q5 (6–9) | 110 | 1227 | 1.02 (0.72, 1.42) | 602 | 6252 | 1.00 (0.86, 1.15) | |
| P for trend | 0.9491 | 0.6397 | 0.5319 | ||||
| DASH | |||||||
| Q1 (9–19) | 95 | 768 | 1.00 (ref.) | 508 | 4370 | 1.00 (ref.) | |
| Q2 (20–22) | 117 | 1025 | 0.90 (0.67, 1.21) | 514 | 5244 | 0.84 (0.74, 0.96) | |
| Q3 (23–24) | 88 | 857 | 0.78 (0.56, 1.07) | 422 | 4041 | 0.90 (0.78, 1.03) | |
| Q4 (25–27) | 110 | 1236 | 0.69 (0.51, 0.94) | 540 | 5513 | 0.83 (0.72, 0.95) | |
| Q5 (28–39) | 99 | 1180 | 0.68 (0.49, 0.94) | 466 | 5058 | 0.80 (0.69, 0.92) | |
| P for trend | 0.0058 | 0.0054 | 0.0493 | ||||
AHEI, Alternative Healthy Eating Index; aMED, alternate Mediterranean diet; DASH, Dietary Approaches to Stop Hypertension; HEI, Healthy Eating Index.
Stratified by matching factors and adjusted for BMI (in kg/m2), physical activity (h/d), total energy (log-transformed kcal/d), and coffee consumption (0, >0‒1, 2–3, ≥4 cups/d). For HEI-2015 and DASH, further adjusted for alcohol intake (g/d).
Discussion
In the MEC, we found an inverse association between diet quality and NAFLD risk. Higher HEI-2015 and DASH scores were associated with a lower risk of NAFLD. Whereas no difference was found by sex or race/ethnicity, the inverse association was stronger for NAFLD with cirrhosis than for NAFLD without cirrhosis.
Several observational studies have examined overall dietary patterns in relation to NAFLD. Using a factor analysis, which is an a posteri-ori, data-driven method to define dietary patterns, a cross-sectional study in Korea identified 3 dietary patterns: the traditional, Western and high-carbohydrate, and simple meal patterns (10). The traditional dietary pattern was associated with an increased risk of NAFLD, and the simple meal pattern was associated with a decreased risk, whereas the Western and high-carbohydrate pattern was not significantly associated (10). In a cross-sectional study with 170 NAFLD patients in Iran (11), a healthy dietary pattern identified by factor analysis was associated with a lower risk of fibrosis, whereas a Western pattern was associated with a higher risk of fibrosis. In a Greek case-control study (12), a fast food–type pattern derived by factor analysis was associated with a higher risk of NAFLD. Using predefined dietary indexes, a cross-sectional study in a Chinese population reported that Diet Quality Index-International but not MDS was associated with lower prevalence of NAFLD (13). A case-control study in Iran found an inverse association between the DASH score and NAFLD risk (14). In a recent report from the Framingham Heart Study that prospectively assessed changes in diet and liver fat over 6 y (2), increases in the MDS and AHEI scores were associated with reduced liver fat accumulation and lower risk and severity of fatty liver among 1521 participants, especially those with high genetic risk scores for NAFLD. These observational, mostly cross-sectional or case-control, studies generally supported preferable dietary patterns and high-quality diets as being inversely associated with NAFLD development or progression.
A recent comprehensive review including dietary patterns in the management of NAFLD concluded that recommending the Mediterranean diet seems prudent, although more studies are required to determine which components of the DASH diet provide the greatest benefits, and other dietary patterns need further investigation (4). The Mediterranean diet may reduce important cardiovascular disease risk factors such as total cholesterol and LDL cholesterol, based on a systematic review of 11 randomized trials (31). However, randomized trials examining the effects of the Mediterranean diet on liver histology are limited (4). Nevertheless, the European Association for the Study of the Liver, European Association for the Study of Diabetes, and European Association for the Study of Obesity recommended the Mediterranean diet for NAFLD management (32). In a randomized clinical trial addressing weight loss and metabolic status among NAFLD patients, the DASH diet, primarily designed for hypertension, was shown to have beneficial effects on body weight and metabolic profiles including liver enzymes, triglycerides, insulin-metabolism markers, inflammatory markers, and oxidative-stress markers (33).
In the current study, the inverse association between diet quality and NAFLD risk was evident for HEI-2015 and DASH but not for AHEI-2010 and aMED overall. The major difference between HEI-2015 and DASH compared with AHEI-2010 and aMED is that the former 2 indexes do not include alcohol consumption as a component, whereas the latter 2 indexes do. However, the associations with these indexes remained similar, whether or not they were also adjusted for alcohol intake. For NAFLD with cirrhosis, a significant inverse association was found for AHEI-2010, in addition to HEI-2015 and DASH, but not for aMED. The aMED may be less sensitive to the variance in diet quality owing to its smaller range of total score (0–9 points) than the other indexes (HEI-2015: 0–100; AHEI-2010: 0–110; and DASH: 8–40). Because aMED scores are based on the median intakes of the populations, the high score in the MEC could be lower in other populations. Also, the Mediterranean diet may not be practical in some non-Mediterranean countries or populations. Nevertheless, in previous studies in the MEC, the aMED was associated with lower risk of liver cancer (34), colorectal cancer (35), colorectal cancer–specific mortality among women diagnosed with colorectal cancer (36), and mortality from all causes, cardiovascular disease, and cancer (21). In a subgroup of the MEC (n ∼ 2000), high-quality diets at baseline were associated with a lower risk of NAFLD 20 y later, which was defined as an MRI-based measure of >5.5% liver fat (37). The risk reduction in the highest tertile of DQIs was statistically significant for the HEI-2010, AHEI-2010, and DASH, and suggestive for the aMED. The weaker or lack of association between the aMED and NAFLD in the MEC warrants further examination.
The etiology of NAFLD is multifactorial and likely involves interactions between genetic and environmental factors (38). There are several proposed mechanisms linking diet quality to NAFLD risk including anti-inflammatory and antioxidant properties of certain diet components such as fiber, monounsaturated and omega-3 fatty acids, and phytosterols (39, 40). Diet can also influence NAFLD development by its effect on modulating body weight and metabolic syndrome (39). Potential interactions between diet quality and genetic predisposition to NAFLD have also been reported (2).
The strengths in our study include the large sample size with 5 racial/ethnic groups, the populations-based design, and comprehensive information on diet and covariates related to NAFLD. Especially, the prospective design of the study minimizes the likelihood that diet quality at baseline was affected by the disease process. However, we cannot rule out the possibility that NAFLD status was misclassified if it was already present before cohort entry or was present but was never reported on a Medicare claim. Measurement error is inevitable in dietary assessment based on a self-administered QFFQ, but tends to be uncorrelated with disease in a prospective study, resulting in attenuated risk estimates. In the current analysis, we used dietary data measured only once at baseline, although dietary habits might change over time. Indeed, diet quality slightly improved over 10 y among participants in the MEC. Among the controls in the current study, 16,795 completed both baseline and 10-y follow-up QFFQs. The DQIs for these individuals were moderately correlated between the 2 surveys; Spearman rank correlation coefficients were 0.57 for HEI-2015 and DASH, 0.50 for AHEI-2010, and 0.47 for aMED. Because the current study was performed in the FFS participants, the results may not be generalizable to the entire MEC cohort. However, the FFS subcohort was large and still incorporated racial/ethnic and socioeconomic diversity (16).
In conclusion, our findings suggest having better diet quality may reduce NAFLD risk in this ethnically diverse population. The benefit of a high-quality diet to NAFLD prevention may be larger for NAFLD with cirrhosis than for NAFLD without cirrhosis.
Supplementary Material
ACKNOWLEDGEMENTS
The authors’ responsibilities were as follows—VWS: designed the research; VWS and S-YP: conducted the research and had primary responsibility for the final content; S-YP: analyzed the data; and all authors: wrote the paper and read and approved the final manuscript.
Notes
Supported by NIH/National Cancer Institute grants R01CA228589 (to VWS), U01CA164973 (to LRW), and R03CA223890 (to SY-P).
Author disclosures: The authors report no conflicts of interest.
The funding agency had no role in the design, implementation, analysis, and interpretation of the data.
Supplemental Tables 1 and 2 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/cdn/.
Abbreviations used: AHEI, Alternative Healthy Eating Index; aMED, alternate Mediterranean diet; DASH, Dietary Approaches to Stop Hypertension; DQI, diet quality index; FFS, fee-for-service; HEI, Healthy Eating Index; ICD, International Classification of Diseases; MDS, Mediterranean Diet Score; MEC, Multiethnic Cohort; NAFLD, nonalcoholic fatty liver disease; QFFQ, quantitative food-frequency questionnaire.
References
- 1. Rinella ME. Nonalcoholic fatty liver disease: a systematic review. JAMA. 2015;313:2263–73. [DOI] [PubMed] [Google Scholar]
- 2. Ma J, Hennein R, Liu C, Long MT, Hoffmann U, Jacques PF, Lichtenstein AH, Hu FB, Levy D. Improved diet quality associates with reduction in liver fat, particularly in individuals with high genetic risk scores for nonalcoholic fatty liver disease. Gastroenterology. 2018;155:107–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. George ES, Forsyth A, Itsiopoulos C, Nicoll AJ, Ryan M, Sood S, Roberts SK, Tierney AC. Practical dietary recommendations for the prevention and management of nonalcoholic fatty liver disease in adults. Adv Nutr. 2018;9:30–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Perdomo CM, Frühbeck G, Escalada J. Impact of nutritional changes on nonalcoholic fatty liver disease. Nutrients. 2019;11:677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Mirmiran P, Teymoori F, Asghari G, Azizi F. Diet quality and nonalcoholic fatty liver disease. Hepatobiliary Surg Nutr. 2019;8:262–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13:3–9. [DOI] [PubMed] [Google Scholar]
- 7. Kant AK. Dietary patterns and health outcomes. J Am Diet Assoc. 2004;104:615–35. [DOI] [PubMed] [Google Scholar]
- 8. Potter J, Brown L, Williams RL, Byles J, Collins CE. Diet quality and cancer outcomes in adults: a systematic review of epidemiological studies. Int J Mol Sci. 2016;17:1052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Trovato FM, Castrogiovanni P, Malatino L, Musumeci G. Nonalcoholic fatty liver disease (NAFLD) prevention: role of Mediterranean diet and physical activity. Hepatobiliary Surg Nutr. 2019;8:167–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Chung GE, Youn J, Kim YS, Lee JE, Yang SY, Lim JH, Song JH, Doo EY, Kim JS. Dietary patterns are associated with the prevalence of nonalcoholic fatty liver disease in Korean adults. Nutrition. 2019;62:32–8. [DOI] [PubMed] [Google Scholar]
- 11. Soleimani D, Ranjbar G, Rezvani R, Goshayeshi L, Razmpour F, Nematy M. Dietary patterns in relation to hepatic fibrosis among patients with nonalcoholic fatty liver disease. Diabetes Metab Syndr Obes. 2019;12:315–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kalafati IP, Borsa D, Dimitriou M, Revenas K, Kokkinos A, Dedoussis GV. Dietary patterns and non-alcoholic fatty liver disease in a Greek case-control study. Nutrition. 2019;61:105–10. [DOI] [PubMed] [Google Scholar]
- 13. Chan R, Wong VW, Chu WC, Wong GL, Li LS, Leung J, Chim AM, Yeung DK, Sea MM, Woo J et al.. Diet-quality scores and prevalence of nonalcoholic fatty liver disease: a population study using proton-magnetic resonance spectroscopy. PLoS One. 2015;10:e0139310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hekmatdoost A, Shamsipour A, Meibodi M, Gheibizadeh N, Eslamparast T, Poustchi H. Adherence to the Dietary Approaches to Stop Hypertension (DASH) and risk of nonalcoholic fatty liver disease. Int J Food Sci Nutr. 2016;67:1024–9. [DOI] [PubMed] [Google Scholar]
- 15. Kolonel LN, Henderson BE, Hankin JH, Nomura AM, Wilkens LR, Pike MC, Stram DO, Monroe KR, Earle ME, Nagamine FS. A multiethnic cohort in Hawaii and Los Angeles: baseline characteristics. Am J Epidemiol. 2000;151:346–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Setiawan VW, Virnig BA, Porcel J, Henderson BE, Le Marchand L, Wilkens LR, Monroe KR. Linking data from the Multiethnic Cohort Study to Medicare data: linkage results and application to chronic disease research. Am J Epidemiol. 2015;181:917–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Setiawan VW, Stram DO, Porcel J, Lu SC, Le Marchand L, Noureddin M. Prevalence of chronic liver disease and cirrhosis by underlying cause in understudied ethnic groups: the multiethnic cohort. Hepatology. 2016;64:1969–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Noureddin M, Zelber-Sagi S, Wilkens LR, Porcel J, Boushey CJ, Le Marchand L, Rosen HR, Setiawan VW. Diet associations with nonalcoholic fatty liver disease in an ethnically diverse population: the Multiethnic Cohort. Hepatology. 2019, in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Stram DO, Hankin JH, Wilkens LR, Pike MC, Monroe KR, Park S, Henderson BE, Nomura AM, Earle ME, Nagamine FS et al.. Calibration of the dietary questionnaire for a multiethnic cohort in Hawaii and Los Angeles. Am J Epidemiol. 2000;151:358–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Liese AD, Krebs-Smith SM, Subar AF, George SM, Harmon BE, Neuhouser ML, Boushey CJ, Schap TE, Reedy J. The Dietary Patterns Methods Project: synthesis of findings across cohorts and relevance to dietary guidance. J Nutr. 2015;145:393–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Harmon BE, Boushey CJ, Shvetsov YB, Ettienne R, Reedy J, Wilkens LR, Le Marchand L, Henderson BE, Kolonel LN. Associations of key diet-quality indexes with mortality in the Multiethnic Cohort: the Dietary Patterns Methods Project. Am J Clin Nutr. 2015;101:587–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Panizza CE, Shvetsov YB, Harmon BE, Wilkens LR, Le Marchand L, Haiman C, Reedy J, Boushey CJ. Testing the predictive validity of the Healthy Eating Index-2015 in the Multiethnic Cohort: is the score associated with a reduced risk of all-cause and cause-specific mortality?. Nutrients. 2018;10:452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Reedy J, Lerman JL, Krebs-Smith SM, Kirkpatrick SI, Pannucci TE, Wilson MM, Subar AF, Kahle LL, Tooze JA. Evaluation of the Healthy Eating Index-2015. J Acad Nutr Diet. 2018;118:1622–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Chiuve SE, Fung TT, Rimm EB, Hu FB, McCullough ML, Wang M, Stampfer MJ, Willett WC. Alternative dietary indices both strongly predict risk of chronic disease. J Nutr. 2012;142:1009–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Fung TT, McCullough ML, Newby PK, Manson JE, Meigs JB, Rifai N, Willett WC, Hu FB. Diet-quality scores and plasma concentrations of markers of inflammation and endothelial dysfunction. Am J Clin Nutr. 2005;82:163–73. [DOI] [PubMed] [Google Scholar]
- 26. Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, Gnardellis C, Lagiou P, Polychronopoulos E, Vassilakou T, Lipworth L, Trichopoulos D. Diet and overall survival in elderly people. BMJ. 1995;311:1457–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med. 2003;348:2599–608. [DOI] [PubMed] [Google Scholar]
- 28. US Department of Health and Human Services. Your guide to lowering your blood pressure with DASH. NIH Publication No. 06-4082 Bethesda, MD: National Heart, Lung, and Blood Institute; 2006. [Google Scholar]
- 29. Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women. Arch Intern Med. 2008;168:713–20. [DOI] [PubMed] [Google Scholar]
- 30. Setiawan VW, Wilkens LR, Lu SC, Hernandez BY, Le Marchand L, Henderson BE. Association of coffee intake with reduced incidence of liver cancer and death from chronic liver disease in the US Multiethnic Cohort. Gastroenterology. 2015;148:118–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Rees K, Hartley L, Clarke A, Thorogood M, Stranges S. ‘Mediterranean’ dietary pattern for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev. 2013;(8):CD009825. [DOI] [PubMed] [Google Scholar]
- 32. European Association for the Study of the Liver (EASL), European Association for the Study of Diabetes (EASD), European Association for the Study of Obesity (EASO). EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease. J Hepatol. 2016;64:1388–402. [DOI] [PubMed] [Google Scholar]
- 33. Razavi Zade M, Telkabadi MH, Bahmani F, Salehi B, Farshbaf S, Asemi Z. The effects of DASH diet on weight loss and metabolic status in adults with non-alcoholic fatty liver disease: a randomized clinical trial. Liver Int. 2016;36:563–71. [DOI] [PubMed] [Google Scholar]
- 34. Bogumil D, Park SY, Le Marchand L, Haiman CA, Wilkens LR, Boushey CJ, Setiawan VW. High-quality diets are associated with reduced risk of hepatocellular carcinoma and chronic liver disease: the Multiethnic Cohort. Hepatol Commun. 2019;3:437–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Park SY, Boushey CJ, Wilkens LR, Haiman CA, Le Marchand L. High-quality diets associate with reduced risk of colorectal cancer: analyses of diet quality indexes in the Multiethnic Cohort. Gastroenterology. 2017;153:386–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Jacobs S, Harmon BE, Ollberding NJ, Wilkens LR, Monroe KR, Kolonel LN, Le Marchand L, Boushey CJ, Maskarinec G. Among 4 diet quality indexes, only the Alternate Mediterranean Diet Score is associated with better colorectal cancer survival and only in African American women in the Multiethnic Cohort. J Nutr. 2016;146:1746–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Maskarinec G, Lim U, Jacobs S, Monroe KR, Ernst T, Buchthal SD, Shepherd JA, Wilkens LR, Le Marchand L, Boushey CJ. Diet quality in midadulthood predicts visceral adiposity and liver fatness in older ages: the Multiethnic Cohort Study. Obesity. 2017;25:1442–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Buzzetti E, Pinzani M, Tsochatzis EA. The multiple-hit pathogenesis of non-alcoholic fatty liver disease (NAFLD). Metabolism. 2016;65:1038–48. [DOI] [PubMed] [Google Scholar]
- 39. Zelber-Sagi S, Ratziu V, Oren R. Nutrition and physical activity in NAFLD: an overview of the epidemiological evidence. World J Gastroenterol. 2011;17:3377–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Anania C, Perla FM, Olivero F, Pacifico L, Chiesa C. Mediterranean diet and nonalcoholic fatty liver disease. World J Gastroenterol. 2018;24:2083–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.