Severe Hepatic Steatosis by Controlled Attenuation Parameter Predicts Quality of Life Independent of Fibrosis

Abstract

Background & Aim

Liver fibrosis is associated with poor patient-reported outcomes (PROs), but the impact of steatosis is unknown. We aimed to evaluate the impact of steatosis on PROs independent of liver fibrosis.

Methods

We evaluated the impact of steatosis, measured by Controlled-Attenuation Parameter (CAP) on transient elastography, and PROs using the 2017–2018 National Health and Nutrition Examination Survey (NHANES) database. We used univariate and multivariate logistic and ordinal regression to evaluate categorical CAP score with PROs measuring physical disability, general health and depression.

Results

Of 4,509 participants included, 38% had severe steatosis (> 280 dB/m). Those with severe steatosis were older and more likely to be male (56% vs. 43% and 51%). On univariate analysis, severe steatosis was associated with more difficulty walking (P = 0.01), dressing (P = 0.005), lifting objects (P = 0.02), bending (P < 0.001), and moving large objects (P = 0.0006). After multivariate adjustment, severe steatosis remained associated with difficulty lifting objects (odds ratio [OR]: 1.7, 95% confidence interval [CI]: 1.2–2.4, P = 0.01) and difficulty bending (OR: 1.8, 95% CI: 1.2–2.7, P = 0.006). Severe steatosis increased risk of having any of the disabilities (OR: 1.7, 95% CI: 1.2–2.4, P = 0.008) and had higher ordinal disability index (OR: 1.6, 95% CI: 1.2–2.2, P = 0.007). Lastly, severe steatosis was also associated with worse self-perceived health status (OR: 1.5, 95% CI: 1.2–1.9, P = 0.002), while general health compared to one year ago and depression trended toward significance.

Conclusion

Patients with severe steatosis are at increased risk of physical disability and have worse self-perceived health status independent of liver fibrosis.

Introduction

Chronic liver disease (CLD) remains a rising cause of morbidity and mortality in the United States (US) [1] and around the world [2, 3]. Nonalcoholic fatty liver disease (NAFLD), defined by the presence of hepatic steatosis, is becoming the leading cause of CLD and liver transplantation worldwide [4]. While improving mortality of CLD is imperative, addressing morbidity is increasingly important as it directly impairs patient’s quality of life, increases caregiver burden [5], strains healthcare resource utilization and heightens healthcare costs [6]. Apart from liver-related clinical decompensation, CLD itself is linked to worse health-related quality of life (HRQOL) [7, 8]. Although data on NAFLD is more scarce, accumulating evidence suggests a strong association between NAFLD and reduced HRQOL [9, 10] and increased healthcare costs [6]. Studies evaluating whether poor HRQOL in NAFLD is a result of advanced fibrosis, cirrhosis or steatosis itself are limited.

Patients with NAFLD have worse cognitive impairment [11,12,13], anxiety and depression [14], and increasing disability, particularly in cirrhosis, measured by increasing disability-adjusted life years [15]. Periodic assessment of symptom burden is crucial as treatment may improve patient reported outcomes (PROs) and HRQOL [16], while also informing true burden of NAFLD. Additionally, PRO measurements are sensitive to improvements in underlying conditions and are often considered by the United States Food and Drug Administration when evaluating trials.

We sought to evaluate the association of Controlled Attenuation Parameter (CAP) score, an ultrasound-based measurement of hepatic steatosis, with PROs assessing physical function, general and mental health using National Health and Nutrition Examination Survey (NHANES) questionnaire data.

Materials and Methods

Study Population

This is an analysis of data from the participants of the 2017–2018 cycle of NHANES, which is conducted in the US by the National Center for Health Statistics at the Centers for Disease Control and Prevention. It is a cross-sectional survey program that includes individuals considered representative of the general non-institutionalized US population. To produce a representative national cohort, NHANES uses a stratified multistage probability sampling design with oversampling of specific ethnic, income, and age groups. The survey combines structured interviewing for data relevant to health and diet with physical and laboratory examinations [17].

Measurement of Liver Steatosis

For the 2017–2018 NHANES cycle, vibration-controlled transient elastography (VCTE) using Fibroscan^® model 502 V2 Touch, equipped with medium (M) probe (74% of patients) or extra-large (XL) probe (26% of patients), was performed by trained technicians on all participants aged 12 years and over. Participants deemed ineligible included those who were unable to lie down on the examination table, were pregnant or that could not provide a pregnancy test, had an implanted electronic medical device, wore a bandage, or had lesions on the side of the abdomen where measurements were taken. VCTE results were reported as complete if 10 or more complete stiffness (E) measurements were obtained, fasting time was at least 3 h, and if the liver stiffness interquartile (IQRe) range/median E was less than 30%. NHANES performed CAP measurements on 5948 participants. We included 4509 participants in our analysis that were aged 20 years or older and had a complete VCTE exam. Median CAP cutoffs of 248 dB/m, 268 dB/m, and 280 dB/m were considered as S1, S2, and S3, analogous to a meta-analysis on optimal steatosis cutoffs by Karlas et al. [18].

Background Information

We extracted questionnaire information on age, gender, ethnicity, alcohol consumption in the past 12 months, and smoking history. Previously diagnosed conditions including liver disease, diabetes, and hypercholesterolemia were also retrieved. Body mass index (BMI) and waist circumference were both obtained from the examination data. Laboratory measurements included alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), glycosylated hemoglobin (HbA1c), total bilirubin, and ferritin.

Patient Reported Outcomes (PROs)

We included variables on self-perceived health status and self-reported disabilities which represented more physically demanding activities and PHQ-9 depression questionnaire from NHANES. The individual items measuring physical function and disability included: difficulty walking, difficulty dressing, difficulty lifting objects, difficulty climbing stairs, difficulty bending, and difficulty moving large objects. A binary outcome of having any of these disabilities was also measured. The individual items were then combined to create an ordinal disability scale from 0 to 6. Self-perceived health status was reported on an ordinal scale from excellent, very good, good, fair or poor. Participants also provided their general health status compared to 1 year ago as better, about the same or worse.

Statistical Analyses

We first compared the prevalence of participant characteristics including demographics, PROs, smoking history, alcohol consumption history, anthropometric measurements, and laboratory measurements among individuals with CAP score stages S1, S2, and S3. Pair-wise comparison was employed with Chi-square test for binary and categorical variables and linear regression for continuous variables.

Univariate logistic regression was used to assess the association of median CAP (S1, S2, and S3) as a categorical predictor and the individual PRO items as binary outcomes, including: difficulty walking, dressing, lifting objects with 10 lb. in weight, climbing stairs, bending, and moving large objects. We also analyzed the association between median CAP and the presence of any of these disabilities as a binary variable under a single univariate model. Ordinal regression analysis was used to determine the association of median CAP and perceived health status (excellent, very good, good, fair, and poor), perceived health in a year-span (better, about the same, worse), and the total number of the individual PRO items present. A threshold of 10 out of 27 possible points from the PHQ-9 items of the questionnaire was utilized as a diagnosis of major depression disorder (MDD) to assess as a binary variable under a univariate logistic model with categorical median CAP. We then performed multivariate logistic and ordinal regression analyses to estimate odds ratios (ORs) of predicting these PROs based on median CAP score after adjustment for age, gender, BMI, median liver stiffness measurement (LSM) (kPa), MDD diagnosis, and cancer history. MDD diagnosis was not included as predictor in the multivariate model analyzing MDD as the outcome.

All data analyses were conducted with Stata version 16.1, incorporating NHANES-provided sampling weights.

Results

Prevalence of US Adults with Steatosis and PROs

We analyzed a total of 4509 participants with a complete VCTE exam aged 20 years or older from the total of 9254 participants in the NHANES 2017–2018 cycle. Weighted analysis of these participants represented a total of 207.8 million adults, 64.8% of the total US population. The analysis also indicated that 79.6 million (95% confidence interval [CI] 73.3–86.0 million) had CAP higher than 280 dB/m, indicating that 24.8% of the total US population have severe steatosis (Table 1). The total of participants with both PROs and CAP measurement varied according to the specific questionnaire item. The lowest total of responders was allotted to the “difficulty climbing stairs” item, with a total of 1,759 participants or a weighted representation of 72.1 million adults (95% CI 63.8 – 80.4 million) of the total US population. The presence of individual disabilities in the NHANES 2017–2018 survey increased with CAP score, with a total of 27.1 million adults with scores > 248 dB/m. This translates to a prevalence of 8.5% of the total population. The range of responders was highest in “health status compared to 1 year ago” item, with 4,506 total eligible participants or 207.7 million adults (95% CI 192.0 – 223.0) under weighted analysis. Among these, “fair” and “poor” self-perceived health status increased with CAP score (Table 1).

Table 1.

Background characteristics by median controlled attenuation parameter categories

	Median CAP (dB/m)			P-value
	S1 (< 248)	S2 (248–280)	S3 (> 280)
Population estimate1	90.4	37.7	79.6
Age2	43.9 (42.5–45.4)	50.8 (49.0–52.5)	51.2 (49.8–52.6)	< 0.001
Gender
Male	39.0 (43.2%)	19.2 (50.9%)	44.3 (55.6%)	0.001
Female	51.4 (56.8%)	18.5 (49.1%)	35.4 (44.4%)
Race
White	56.8 (62.9%)	23.4 (61.9%)	49.6 (62.3%)	0.003
Black	12.2 (13.5%)	4.2 (11.1%)	6.8 (8.6%)
Hispanic	12.1 (13.4%)	5.7 (15.1%)	15.0 (18.9%)
Asian	5.2 (5.8%)	2.5 (6.6%)	4.4 (5.5%)
Others	4.0 (4.5%)	2.0 (5.2%)	3.8 (4.8%)
General health condition
Excellent	11.8 (13.1%)	4.3 (11.4%)	4.0 (5.0%)	< 0.001
Very Good	33.1 (36.5%)	11.0 (29.2%)	19.5 (24.5%)
Good	31.2 (34.5%)	15.2 (40.3%)	34.2 (42.9%)
Fair	9.8 (10.9%)	4.6 (12.1%)	17.2 (21.5%)
Poor	1.3 (13.9%)	0.5 (1.4%)	2.1 (2.7%)
Health compared to 1year ago
Better	20.2 (22.4%)	6.8 (17.9%)	14.6 (18.3%)	0.1
About the same	7.5 (8.3%)	3.5 (9.4%)	8.9 (11.1%)
Worse	62.7 (69.4%)	27.4 (72.7%)	56.2 (70.5%)
Difficulty walking quarter of a mile	5.3 (5.8%)	3.4 (8.9%)	8.3 (10.4%)	0.06
Difficulty dressing	4.5 (5.0%)	2.4 (6.2%)	7.7 (9.7%)	0.01
Difficulty climbing stairs	3.7 (4.1%)	2.6 (6.8%)	5.3 (6.6%)	0.20
Difficulty lifting objects	8.3 (9.2%)	4.7 (12.5%)	11.9 (14.5%)	0.10
Difficulty bending/kneeling	14.0 (15.5%)	9.1 (24.1%)	25.2 (31.7%)	< 0.001
Difficulty moving large objects	12.8 (14.1%)	7.3 (19.4%)	17.4 (21.8%)	0.06
Liver Disease
Fatty Liver	0.5 (0.6%)	0.3 (0.9%)	3.0 (3.7%)	0.04
Viral Hepatitis	0.9 (1.0%)	0.4 (1.2%)	0.7 (0.8%)
Autoimmune	0.1 (0.1%)	0.04 (0.1%)	0.1 (0.1%)
Other causes	1.1 (1.3%)	0.5 (1.3%)	1.5 (1.8%)
BMI
Underweight	2.7 (3.0%)	0.2 (0.6%)	0.1 (0.1%)	< 0.001
Normal weight	42.2 (46.6%)	6.5 (17.3%)	4.6 (5.8%)
Overweight	28.8 (31.9%)	15.0 (39.7%)	21.2 (26.6%)
Obese	16.3 (18.0%)	15.7 (41.6%)	53.3 (66.9%)
History of diabetes	3.2 (3.6%)	3.3 (8.8%)	15.4 (19.4%)	< 0.001
Hypercholesterolemia	23.8 (26.3%)	12.8 (33.9%)	31.7 (39.8%)	< 0.001
Depression diagnosis (PHQ-9)	6.9 (7.6%)	2.3 (6.0%)	6.2 (7.8%)	0.5
Number of drinks a day in the past 12months
Never or none in the last year	16.8 (18.6%)	7.1 (18.9%)	16.2 (20.4%)	0.1
Former heavy drinker	2.7 (3.0%)	1.5 (3.9%)	3.3 (4.1%)
Drinker	61.5 (68.0%)	23.5 (62.2%)	49.5 (62.2%)
Excessive Drinker	6.0 (6.7%)	3.5 (9.3%)	7.6 (9.5%)
Non-responders	3.4 (3.7%)	2.2 (5.7%)	3.0 (3.8%)
Smoking history
Never smoker	55.4 (61.3%)	21.4 (56.8%)	44.0 (55.2%)	0.01
Past smoker	18.4 (20.3%)	10.5 (27.9%)	22.8 (28.6%)
Current smoker	16.6 (18.4%)	5.8 (15.3%)	12.9 (16.2%)
AST (IU/L)	21.3 (20.0–22.6)	22.0 (20.4–23.6)	23.6 (22.6–24.6)	0.003
ALT (IU/L)	19.6 (18.2–21.0)	21.7 (20.3–23.1)	28.4 (26.8–29.9)	< 0.001
Alkaline phosphatase (IU/L)	72.4 (69.8–75.0)	77.4 (74.7–80.2)	80 (77.8–82.1)	0.001
Total bilirubin (mg/dL)	0.48 (0.46–0.51)	0.50 (0.46–0.54)	0.45 (0.43–0.47)	0.1
Ferritin (ng/mL)	124.8 (114.0–135.5)	158.3 (134.2–182.4)	176.0 (162.0–190.0)	< 0.001
HbA1c (%)	5.40 (5.36–5.43)	5.61 (5.51–5.70)	6.0 (5.92–6.06)	< 0.001

^1.Population estimates are shown in units of million population

^2.Continuous variables reported as mean ± (SD) if normally distributed and median (IQR) if not normally distributed

^3.CAP Controlled attenuation parameter, dB/m Decibels per meter, BMI Body mass index, PHQ-9 Patient Health Questionnaire-9, AST Aspartate aminotransferase, ALT Alanine aminotransferase, IU International units, mg/dL Milligrams per deciliter, ng/mL Nanograms per milliliter, HbA1c Hemoglobin A1c

Participants’ Characteristics with Steatosis

Demographics, anthropometric measurements, self-perceived health condition, PROs, depression diagnosis according to PHQ-9 questionnaire, alcohol consumption, and smoking history across the three CAP score categories are reported in Table 1. Individuals with CAP category S3 were more likely to be older, male, white, obese, and had metabolic comorbidities. Of note, no significant increase in prevalence is evident in alcohol consumption patterns, highlighting the variable association between alcohol and steatosis.

Liver Steatosis Associated with Patient Reported Outcomes (PROs)

Functional status, represented as PROs, reflects the quality of life and living independence of the patients with CLD. We further investigated if the three CAP score categories were associated with participants’ functional status. Among the PROs, a significant difference within the median CAP scores was present for difficulty bending/kneeling and difficulty dressing. The prevalence for these PRO items in patients with CAP above 280 dB/m was 9.7% and 31.7%, respectively. Increase in prevalence was evident across all other individual PRO items as CAP score increases. This prevalence disparity among PROs highlights the difficulty of accurately predetermining the specific functional component affected by the physiologic consequences of steatosis in these patients.

Factors Associated with Patient Reported Outcomes (PROs)

In our unadjusted logistic and ordinal regression analyses, CAP category S3 was positively associated with individual PRO items excluding “difficulty climbing stairs”, the presence of any of these disabilities, and with their accumulated presence in participants (Table 2). General perceived health status and diagnosis of depression according to PHQ-9 questionnaire were also positively associated with S3 in the ordinal model (Table 3). Notably, perceived health in comparison with a year prior was not significantly associated despite the relationship with actual self-perceived general health status.

Table 2.

The association of controlled attenuation parameter with physical disability

NHANES item	Univariate Analysis	P-value	Multivariate analysis1	P value
Difficulty walking	S1: Ref S2: 1.3 (0.7–2.4) S3: 1.7 (1.2–2.6)	0.3	S1: Ref S2: 1.4 (0.7–2.7) S3: 1.3 (0.8–2.0)	0.3
		0.01		0.3
Difficulty dressing	S1: Ref S2: 1.0 (0.6–1.6) S3: 1.6 (1.2–2.2)	0.8	S1: Ref S2: 1.0 (0.5–3.1) S3: 1.5 (0.9–2.5)	0.9
		0.005		0.09
Difficulty lifting objects	S1: Ref S2: 1.1 (0.7–1.7) S3: 1.4 (1.0–1.8)	0.7	S1: Ref S2: 1.4 (0.8–2.3) S3: 1.7 (1.2–2.4)	0.2
		0.02		0.01
Difficulty climbing stairs	S1: Ref S2: 1.4 (0.7–3.1) S3: 1.5 (0.9–2.3)	0.3	S1: Ref S2: 1.6 (0.7–3.7) S3: 1.4 (0.9–2.3)	0.2
		0.1		0.1
Difficulty bending	S1: Ref S2: 1.4 (0.8–2.3) S3: 2.7 (2.0–3.7)	0.1	S1: Ref S2: 1.3 (0.7–2.3) S3: 1.8 (1.2–2.7)	0.5
		< 0.001		0.006
Difficulty moving large objects	S1: Ref S2: 1.1 (0.7–1.5) S3: 1.4 (1.1–1.7)	0.7	S1: Ref S2: 1.2 (0.8–1.7) S3: 1.3 (0.9–1.7)	0.3
		0.006		0.1
Any disability2	S1: Ref S2: 1.1 (0.7–1.5) S3: 1.4 (1.1–1.7)	0.5	S1: Ref S2: 1.2 (0.7–2.0) S3: 1.7 (1.2–2.4)	0.4
		< 0.001		0.008
Ordinal disabilities3	S1: Ref S2: 1.3 (0.8–2.0) S3: 1.9 (1.4–2.6)	0.3	S1: Ref S2: 1.4 (0.8–2.4) S3: 1.6 (1.2–2.2)	0.2
		0.001		0.007

^1.Adjusted for age, sex, body mass index, median liver stiffness measurement, depression diagnosis, cancer history

^2.Binary variable produced by the presence of any of the six disabilities from the questionnaire data

^3.Ordinal variable produced by the number of disabilities present in each participant from the questionnaire data

^4.NHANES National Health and Nutrition Examination Survey, S1 Controlled Attenuation Parameter score < 248 dB/m, S2 Controlled Attenuation Parameter score 248–280 dB/m, S3 Controlled Attenuation Parameter score > 280 dB/m

Table 3.

The association of controlled attenuation parameter with depression and general health

NHANES item	Definition of outcome	Univariate analysis	p-value	Multivariate analysis1	p-value
Depression questionnaire (PHQ-9)2	Binary variable (0/1) produced by scoring > 10 points out of a total of 27 points from 9 items of questionnaire	S1: Ref S2: 1.2 (0.6–2.4) S3: 1.8 (1.0–3.1)	0.5	S1: Ref S2: 0.7 (0.3–1.5) S3: 1.6 (0.9–2.8)	0.2
			0.04		0.1
Health status	Ordinal variable with the following possible statuses: Excellent Very Good Good Fair Poor	S1: Ref S2: 1.3 (1.0–1.7) S3: 2.4 (2.0–2.9)	0.05	S1: Ref S2: 1.0 (0.8–1.5) S3: 1.5 (1.2–1.9)	0.8
			< 0.001		0.002
General health compared to 1year ago	Ordinal variable with the following possible statuses: Better About the same Worse	S1: Ref S2: 1.2 (1.0–1.5) S3: 1.1 (0.9–1.4)	0.08	S1: Ref S2: 1.3 (1.0–1.8) S3: 1.3 (1.0–1.6)	0.04
			0.3		0.06

^1.Adjusted for age, sex, body mass index, median liver stiffness measurement, depression diagnosis, cancer history

^2.Depression questionnaire variable not included as predictor in multivariate analysis

^3.NHANES National Health and Nutrition Examination Survey, PHQ-9 Patient Health Questionnaire-9, S1 Controlled Attenuation Parameter Score < 248 dB/m, S2 Controlled Attenuation Parameter score 248–280 dB/m, S3 Controlled Attenuation Parameter score > 280 dB/m

We examined these self-perceived variables in multivariate logistic and ordinal regression models to adjust for covariates related to PROs reported in the literature (Table 2). Positive associations remained significant across items that represented more physically demanding PROs, including difficulty lifting objects and bending. Disabilities in both binary and ordinal forms maintained their positive association with S3 CAP category, underlining their independent predictive value with steatosis. Perceived health status compared to 1 year ago trended toward significance after multivariate adjustment in the S3 CAP category (Tables 2, 3). A sensitivity analysis after excluding excessive alcohol use and adjusting for diabetes shows similar direction and magnitude of effect of steatosis on outcomes (Supplemental Tables 1 and 2).

Discussion

Although typically considered a silent, asymptomatic condition, these data demonstrate that hepatic steatosis is associated with poor PROs. Complementing data on self-reported NAFLD [9] and other PRO studies in patients with estimated (by Fatty Liver Index) or ultrasound-defined NAFLD [10, 12], our data provides estimation of the impact of hepatic steatosis on PROs in a nationally representative sample using CAP, the most sensitive metric of liver fat.

The Impact of Severe Steatosis

Liver fibrosis independently predicts important liver-related outcomes [19], but the impact of steatosis is less clear. Whereas it has been argued that fibrosis is the most important factor to discern for risk stratification [20], this may obscure the risk of important deficits in quality of life and PROs among patients with steatosis. In our study, we highlight the presence of severe steatosis (CAP score > 280 dB/m) to be associated with poor physical function and general health. First off, the prevalence of functional disability increases as CAP score worsens. On univariate assessment, those with severe steatosis were more likely to report any disability at all, have a higher ordinal disability score, and express worse functional status based on individual items with the exception of “difficulty climbing stairs”. After adjusting for LSM and other predictors, the more physically demanding items, such as difficulty lifting objects (OR: 1.7, 95% CI: 1.2–2.4, P = 0.01) and difficulty bending (OR: 1.8, 95% CI: 1.2–2.7, P = 0.006), remained significantly worse, while the remaining items trended toward significance. Those with severe steatosis had an OR of 1.7 (1.2–2.4, P = 0.008) in reporting any disability and had 1.6 times more disability on the ordinal scale (95% CI: 1.2–2.2, P = 0.007).

General health perception and depression were also impacted by the severity of steatosis on CAP score. On univariate analysis, severe steatosis predicted worse perception of one’s general health as well as depression by PHQ-9. After multivariate adjustment, the association of severe steatosis self-perceived health status remained significant (OR 1.5, 95% CI: 1.2–1.9, P = 0.002), while self-perceived general health compared to 1 year ago trended toward significance. Depression, on the other hand, lost significant difference in the multivariate model. Overall, severe steatosis consistently shows worse disability and self-perceived general health and serves as a predictor of worse HRQOL, independent of liver fibrosis measured by LSM.

PROs Independent of Fibrosis

Whether worse HRQOL measures are a result of advanced fibrosis, cirrhosis, or underlying steatosis itself remains unclear. A cohort study with of 156 patients with biopsy-proven NAFLD found fatigue to be severe in patients with NAFLD reported as reduced physical activity and increased daytime somnolence, independent of biochemical and histological markers of liver disease severity [21]. Other studies showing association of PROs independent of liver disease severity or fibrosis are lacking. Our study demonstrates severe steatosis by CAP score is predictive of PROs independent of LSM, suggesting a more direct effect of steatosis on physical function and quality of life independent of liver fibrosis.

Strengths and Limitations

The large sample sizes, an inherent strength of the NHANES database, optimize the generalizability of our results to national population level estimates. Additionally, CAP score is a more reliable and sensitive method for steatosis compared to previous measures used in other studies. Despite the strength and additional clinical insight this study provides, there are some limitations. First and foremost, the observational nature makes it prone to several biases and confounding. Recall bias, for instance, is an inherent concern when using survey data and there is no exception here. Components of metabolic syndrome, such as BMI, may also influence our results. However, adjusting for BMI in our multivariate model continued to yield significant results. Additionally, unmeasured confounders not accounted for in our analysis may influence the results of the study. However, this was offset by using multivariate logistic and ordinal regression models to minimize the effect of confounding. Prospective data is required to confirm our findings.

CAP score assessment of steatosis has robust clinical application, but optimal cut-offs are yet to be determined. Current cut-offs were devised from a heterogeneous patient population. As the optimal cut-off for severe steatosis requires elucidation among patients with NAFLD, the cut-offs used in our study can be applied to the general population given the generalizability of our results.

Conclusion

Severe steatosis is associated with a meaningful, deleterious impact on PROs. These nationally representative data underscore the potentially hidden aspects of the true burden of NAFLD. Furthermore, these data highlight the role of PRO measurement in clinical trials of therapeutics that address hepatic steatosis.

Abbreviations

CLD:

Chronic liver disease

US:

United States

NAFLD:

Nonalcoholic fatty liver disease

HRQOL:

Health-related quality of life

PRO:

Patient-reported outcomes

CAP:

Controlled attenuation parameter

NHANES:

National health and nutrition examination survey

VCTE:

Vibration-controlled transient elastography

IQRe:

Interquartile range/median E

dB/m:

Decibels per meter

BMI:

Body mass index

ALT:

Alanine aminotransferase

AST:

Aspartate aminotransferase

ALP:

Alkaline phosphatase

HbA1c:

Hemoglobin A1c

PHQ-9:

Patient Health Questionnaire-9

MDD:

Major depression disorder

OR:

Odds ratio

LSM:

Liver stiffness measurement

kPa:

Kilopascals

CI:

Confidence interval