Quality of life in adults with nonalcoholic fatty liver disease: Baseline data from the nonalcoholic steatohepatitis clinical research network

Authors

  • Kristin David,

    1. Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, University of California, San Diego, La Jolla, CA
    2. Joint Doctoral Program in Public Health, San Diego State University and University of California San Diego, La Jolla, CA
    Search for more papers by this author
  • Kris V. Kowdley,

    1. Center for Liver Disease, Virginia Mason Medical Center, Seattle, WA
    Search for more papers by this author
  • Aynur Unalp,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
    Search for more papers by this author
  • Fasiha Kanwal,

    1. John Cochran VA Medical Center, Saint Louis and Division of Gastroenterology, Saint Louis University School of Medicine, Saint Louis, MO
    Search for more papers by this author
  • Elizabeth M. Brunt,

    1. Washington University School of Medicine, Department of Pathology and Immunology, Saint Louis, MO
    Search for more papers by this author
  • Jeffrey B. Schwimmer,

    Corresponding author
    1. Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, University of California, San Diego, La Jolla, CA
    2. Department of Gastroenterology, Rady Children's Hospital, San Diego, CA
    • Associate Professor of Pediatrics, Division of Gastroenterology, Hepatology, and Nutrition, Department of Pediatrics, University of California, San Diego, 200 West Arbor Drive, MC 8450, San Diego, CA 92103
    Search for more papers by this author
    • fax: 619-543-7537

  • NASH CRN Research Group


  • Potential conflict of interest: Nothing to report.

  • A list of members of the Nonalcoholic Steatohepatitis Clinical Research Network can be found in the appendix.

Abstract

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in the United States. The association between NAFLD and quality of life (QOL) remains unclear. These data are important to estimate the burden of illness in NAFLD. The aim was to report QOL scores of adults with NAFLD and examine the association between NAFLD severity and QOL. QOL data were collected from adults with NAFLD enrolled in the Nonalcoholic Steatohepatitis Clinical Research Network using the Short Form 36 (SF-36) survey, and scores were compared with normative U.S. population scores. Liver biopsy histology was reviewed by a central pathology committee. A total of 713 subjects with NAFLD (male = 269, female = 444) were included. Mean age of subjects was 48.3 years; 61% had definite nonalcoholic steatohepatitis (NASH), and 28% had bridging fibrosis or cirrhosis. Diabetes was present in 27% of subjects. Subjects with NAFLD had worse physical (mean, 45.2) and mental health scores (mean, 47.6) compared with the U.S. population with (mean, 50) and without (physical, 55.8; mental, 52.5) chronic illness. Subjects with NASH reported lower physical health compared with subjects with fatty liver disease without NASH (44.5 versus 47.1, P = 0.02). Subjects with cirrhosis had significantly (P < 0.001) poorer physical health scores (38.4) than subjects with no (47.6), mild (46.2), moderate (44.6), or bridging fibrosis (44.6). Cirrhosis was associated with poorer physical health after adjusting for potential confounders. Mental health scores did not differ between participants with and without NASH or by degree of fibrosis. Conclusion: Adults with NAFLD have a significant decrement in QOL. Treatment of NAFLD should incorporate strategies to improve QOL, especially physical health. (HEPATOLOGY 2009.)

The increasing prevalence of obesity in North America has had significant effects on the prevalence of obesity-related conditions, including nonalcoholic fatty liver disease (NAFLD).1 NAFLD is now the most common chronic liver disease in the United States.2, 3 Epidemiological surveys estimate that 3%-34% of the general population have NAFLD2, 4–6 and that 2%-5% have nonalcoholic steatohepatitis (NASH).6–8

Because of the increasing importance of NAFLD, many studies have focused on understanding the epidemiology, natural history, and associated comorbidities of NAFLD. Quality of life (QOL) has been less well evaluated in patients with NAFLD, but it is a key element in assessing the burden of a disease and the personal impact of the disease on daily living. QOL measurement also allows for a common metric to compare the burden of different diseases. Previous studies of QOL in NAFLD have been characterized by mixed populations of liver disease, with relatively small numbers of patients with NAFLD.9, 10 The one study with a larger number of patients with NAFLD used solely a disease-specific, rather than generic, QOL measure.11 Notwithstanding the limitations, these previous studies suggest that NAFLD is associated with impaired QOL. Patients with multiple other forms of chronic liver disease, including hepatitis C virus,12 symptomatic hepatitis B virus,13 and cholestatic liver disease,14 have also been shown to have decrements in QOL.

Using a large, multicenter sample, this study tested the hypothesis that individuals with NAFLD have lower QOL compared with individuals without NAFLD. QOL scores of individuals with NAFLD were compared with both a general reference U.S. population that included individuals with chronic illnesses and a healthy subset of the general population that excluded individuals with chronic illnesses. Because liver biopsy specimens were available, a special focus of this study was the association between histological severity of NAFLD and QOL to assess whether increasing severity of disease was associated with lower QOL.

Abbreviations

BMI, body mass index; CRN, Clinical research Network; MCS, mental component score; NAFLD, nonalcoholic fatty liver disease; PCS, physical component score; QOL, quality of life; NASH, nonalcoholic steatohepatitis; quality of life; SF-36, Short Form 36.

Patients and Methods

Subjects

The Nonalcoholic Steatohepatitis Clinical Research Network (NASH CRN) was established by the National Institute of Diabetes and Digestive and Kidney Diseases in 2002 to assess the natural history, pathogenesis, and therapy of NAFLD in the United States.15 The baseline QOL data were obtained from adult subjects enrolled in two NASH CRN studies: (1) the NAFLD Database, an observational cohort study; and (2) a randomized, placebo-controlled clinical trial, PIVENS (Pioglitazone versus Vitamin E versus Placebo for the Treatment of Nondiabetic Patients with Nonalcoholic Steatohepatitis; Clinical Trial number NCT00063622). These studies were conducted by eight clinical centers and a central data coordinating center (see Acknowledgment for roster). Both study protocols were approved by all participating center institutional review boards and an independent data and safety monitoring board. Each participant provided written informed consent.

Measures

Demographics.

Demographic information collected during screening interviews as part of the registration process included age, sex, self-reported ethnic and racial affiliation, highest educational level achieved, marital status, employment status, and annual income.

Anthropometrics.

Weight and height were measured with participants standing wearing light clothing. Body mass index (BMI) was calculated as weight (kg) divided by the height (meters) squared. Weight status was defined as normal (BMI 18.5-24.9), overweight (BMI 25-29.9), mild obesity (BMI 30-34.9), moderate obesity (BMI 35-39.9), and severe obesity (BMI > 40).

Histology.

The NASH CRN Pathology Committee developed and validated a feature-based histological scoring system that encompasses the spectrum of lesions of NAFLD.16 Liver biopsy slides from subjects were read centrally by the Pathology Committee, during which biopsies were rigorously evaluated according to the published scoring system.16 Steatosis was scored according to amount (%) of biopsy occupied using a four-point scale. A diagnosis of NAFLD required the presence of at least 5% steatosis. Fibrosis was staged from 0 to 4, with 0 = none; 1a = mild zone 3 (central) perisinusoidal fibrosis, 1b = moderate zone 3 perisinusoidal fibrosis, 1c = periportal and portal fibrosis (zone 1) only; 2 = both perisinusoidal and periportal or portal fibrosis; 3 = bridging fibrosis, and 4 = cirrhosis. Diagnostic determinations of each biopsy were also assigned. Categories used were: definite steatohepatitis, definitely not steatohepatitis, borderline steatohepatitis (zone 3 pattern), and borderline steatohepatitis (zone 1 pattern).

Quality of Life.

The Short Form 36 (SF-36) is a generic QOL assessment survey that is validated, widely used, and has shown good psychometric properties in diverse disease states including patients with advanced liver disease.17–21 The SF-36 consists of 36 questions that make up eight subscales (physical function, physical role limitation, bodily pain, general health, vitality, social function, emotional role limitation, and mental health). Within each dimension, 0 is the worst and 100 is the best possible score.22 An overall physical health score [physical component scale (PCS)] and mental health score [mental component scale (MCS)] are derived from the subscale scores. Norm-based scoring of the SF-36 was done according to the instructions provided in the SF-36 users manual.23 Norm-based scoring involves a linear transformation to transform scores to a mean of 50 and standard deviation of 10. Norm-based scoring results in a straightforward interpretation of the QOL scores; that is, because the subscales have different ceilings and floors each score is affected by not only the disease being measured but also by arbitrary differences in the ceilings and floors of the particular scale. Therefore, familiarity with the mean of each scale is necessary when interpreting a particular scale score. With norm-based scoring, because each scale is scored to have a mean of 50 and the same standard deviation of 10, interpretation of a particular scale is clear-cut, greater than 50 equals better health than the U.S. population, and less than 50 equals poorer health; additionally, one can easily see how many standard deviations a score is above or below the mean.

Data Analysis.

Descriptive statistics (mean, median, percentiles, standard deviation, range) were used to characterize the sample population. A Student t test (continuous data) or chi-squared test (categorical data) was used to compare characteristics between individuals with and without NASH. A Student t test was used to compare SF-36 scores between individuals with and without NASH. Univariable regression analyses were conducted to examine the unadjusted association between subjects' characteristics and SF-36 scores. Multivariable regression analyses were conducted to identify factors independently associated with PCS and MCS SF-36 scores. Analysis of variance was used to compare SF-36 scores by degree of fibrosis; Tukey's post-hoc test was used to identify where specific differences, identified globally in the analysis of variance, occurred. Dummy variables were created for categorical variables with more than two levels. For analyses, age was categorized as 18 to 30 years, 31 to 40, 41 to 50, 51 to 60, and 61 to 76 years. Data were assessed, as appropriate, for meeting regression assumptions before running analyses [that is, linearity (assessed by partial regression plot), lack of multicollinearity (multicollinearity defined as tolerance < 0.20 or variance-inflation factor ≥ 4), homoscedasticity (assessed by simple regression plot), and absence of outliers (defined as standardized residuals greater than 3.3)]. A P-value of less than 0.05 was considered statistically significant.

Results

Subjects' Baseline Characteristics

There were 713 adults enrolled in the database or PIVENS trial of the NASH CRN between October 27, 2004 and October 30, 2007 who had complete biopsy results available at the time of analysis, at least 5% steatosis, and complete QOL data. The mean age of the sample was 48 ± 12 years, and 62% were female. Most participants were non-Hispanic white (76%), followed by Hispanic (13%), and non-Hispanic, nonwhite (8%). Most participants had some college education (71%), and one third had completed a bachelor's degree or higher. More than half (58%) of the sample had a household income of at least $50,000 per year, and 70% were currently employed. Most adults were married (69%). The mean BMI was 34.3 ± 6.5 kg/m2, 30% of subjects were mildly obese, 25% were moderately obese, and 17% were severely obese. Type 2 diabetes was present in 27%. Sixty-one percent (436/713) had definite NASH, 20% (141/713) had borderline NASH, and 19% (136/713) had NAFLD but definitely not NASH. Twenty-three (167/713) percent had no fibrosis, whereas 28% had bridging fibrosis or cirrhosis (197/713) (Table 1).

Table 1. Baseline Characteristics
CharacteristicAll Subjects N = 713NASH N = 436No NASH N = 136P Value*
  • *

    P value is from comparison (t test or chi-squared test) between participants with definite NASH and participants with NAFLD but without NASH.

Age    
Mean ± SD (range) years48.2 ± 12.0 (18–76)49.3 ± 11.947.3 ± 12.10.076
Sex[n (%)]    
 Female444 (62.3%)292 (67.0%)73 (53.7%)0.006
 Male269 (37.7%)144 (33.0%)63 (46.3%) 
Race[n (%)]    
 White, non-Hispanic543 (76.2%)342 (78.4%)97 (71.3%)0.324
 Nonwhite, non-Hispanic58 (8.1%)26 (6.0%)13 (9.6%) 
 Multiracial or missing, non-Hispanic21 (2.9%)15 (3.4%)3 (2.2%) 
 Hispanic91 (12.8%)53 (12.2%)23 (16.9%) 
Education[n (%)]    
Kindergarten-8th grade15 (2.1%)10 (2.3%)2 (1.5%)0.003
Grades 9–1141 (5.8%)21 (4.8%)13 (9.6%) 
Completed high school152 (21.3%)102 (23.4%)23 (16.9%) 
Some college272 (38.1%)182 (41.7%)41 (30.1%) 
Bachelor's degree or higher232 (32.5%)120 (27.5%)57 (41.9%) 
Income[n (%)]    
$15,00055 (7.7%)31 (7.1%)15 (11.2%)0.123
$15,000–$29,99989 (12.5%)57 (13.1%)16 (11.9%) 
$30,000–$49,000149 (20.9%)103 (23.6%)21 (15.7%) 
$50,000±409(57.4%)241 (55.3%)81 (61.2%) 
Missing11 (1.5%)4 (0.9%)3 (2.2%) 
Weight (kg)    
 Mean ± SD (range)96.9 ± 20.8 (48.4–169.8)97.0 ± 20.396.2 ± 20.90.689
Height (m)    
 Mean ± SD (range)1.68 ± .10 (1.4–1.9)1.67 ± 0.101.68 ± 0.100.744
BMI    
 Mean ± SD (range)34.3 ± 6.5 (20.0–59.5)34.5 ± 6.334.1 ± 6.70.558
Fibrosis[n (%)]    
None167 (23.4%)28 (6.4%)94 (69.1%)<0.001
1a, mild, zone 3 perisinusoidal,109 (15.3%)68 (15.6%)9 (6.6%) 
1b, moderate, zone 3, perisinusoidal73 (10.2%)64 (14.7%)1 (0.7%) 
1c, portal/periportal only29 (4.1%)2 (0.5%)15 (11.0%) 
Zone 3 and periportal138 (19.4%)107 (24.5%)4 (2.9%) 
Bridging131 (18.4%)118 (27.1%)2 (1.5%) 
Cirrhosis66 (9.3%)49 (11.2%)11 (8.1) 

Baseline Characteristics, No NASH Versus NASH

There was a significantly (P = 0.006) greater proportion of women in the group with NASH (67%) than in those with NAFLD but without NASH (54%). Type 2 diabetes was more prevalent among adults with NASH than in those without NASH (32% versus 19%, P = 0.005). There were no significant differences between adults with and without NASH with regard to age, race, marital status, employment status, or BMI (Table 1).

Baseline QOL Scores

Individuals with NAFLD had significantly (P < 0.001) lower QOL for both physical (45.2) and mental health (47.6) than the general healthy U.S. population (mean PCS score = 55.8, MCS score = 52.5). Individuals with NAFLD also had significantly (P < 0.001) lower QOL scores compared with the general US population (including individuals with chronic illnesses; mean PCS and MCS scores = 50) (Fig. 1). Age and sex adjustment of the U.S. normative sample did not change these results. All SF-36 subscale scores were also lower for adults with NAFLD compared with the U.S. general and healthy populations. Among adults with NAFLD, the lowest (worst) subscale mean scores were in general health (42.4), vitality (44.8), and physical functioning (45.6) (Fig. 2).

Figure 1.

Mean SF-36 PCS and MCS scores of adults enrolled in NASH CRN (n = 713) and US normative sample with (n = 1982) and without (n = 571) chronic illness. Error bars display standard deviations. Black bar = mean scores of Individuals with NAFLD. Gray bar: Mean scores of general U.S. reference population. White bar with black diagonal lines: Mean scores of healthy (that is, no chronic illness) reference sample. Mean PCS and MCS scores of NAFLD sample are significantly (P < 0.001) lower than PCS and MCS scores of general U.S. population and healthy sample.

Figure 2.

Mean SF-36 subscale scores of adults enrolled in NASH CRN (n = 713) and US normative sample with (n = 1982) and without (n = 571) chronic illness. Error bars display standard deviations. Black bar = mean scores of Individuals with NAFLD. Gray bar: Mean scores of general U.S. reference population. White bar with black diagonal lines: Mean scores of healthy (that is, no chronic illness) reference sample. PF, physical function; RP, role-physical; RE, role-emotional; VT, vitality; MH, mental health; SF, social function; BP, bodily pain; GH, general health. Mean subscale scores of NAFLD sample are significantly (P ≤ 0.001) lower than subscales scores of general U.S. population and healthy sample.

Comparison of QOL Between Subjects With and Without NASH

The mean PCS of participants with NAFLD but not NASH was 47.1, which was significantly (P = 0.001) lower than the normative U.S. population. Notably, adults with definite NASH (PCS 44.5) reported significantly (P = 0.02) poorer physical health compared with adults with no NASH (PCS 47.1). For SF-36 subscale scores, adults with NASH reported significantly poorer vitality (44.3 versus 46.6 for subjects with and without NASH, respectively, P = 0.04), general health (41.8 versus 44.2 for subjects with and without NASH, respectively; P = 0.02), bodily pain (47.7 versus 50.0 for subjects with and without NASH, respectively; P = 0.04) and role limitations caused by physical health (45.9 versus 48.3 for subjects with and without NASH, respectively, P = 0.04). Individuals with and without NASH did not significantly differ in mental health (Table 2).

Table 2. QOL of Participants in the NASH CRN With and Without NASH
 All Subjects (N = 713)NASH (N = 436)No NASH (N = 136)P -Value *
  • *

    P value is from comparison (t test) between participants with definite NASH and participants with NAFLD but without NASH.

SF-36 PCS (standardized)    
 Mean ± SD (range)45.2 ± 10.9 (15.7–63.8)44.5 ± 11.047.1 ± 10.40.018
SF-36 MCS (standardized)    
 Mean ± SD (range)47.6 ± 11.0 (9.2–68.7)47.5 ± 10.948.6 ± 11.30.342
SF-36 Sub-Scales (standardized)Mean ± SDMean ± SDMean ± SD 
 Physical functioning45.6 ± 11.344.9 ± 11.747.0 ± 11.00.066
 Role limitations due to physical health46.5 ± 11.645.8 ± 11.648.3 ± 11.40.036
 Role limitations due to emotional problems47.1 ± 12.246.9 ± 12.148.6 ± 12.00.154
 Vitality44.8 ± 11.244.4 ± 11.146.6 ± 11.50.043
 Mental health48.3 ± 10.848.0 ± 10.749.1 ± 11.70.336
 Social functioning46.9 ± 11.646.9 ± 11.348.0 ± 12.00.328
 Bodily pain48.0 ± 11.247.7 ± 11.250.0 ± 11.40.043
 General health42.4 ± 10.841.8 ± 10.944.2 ± 10.90.023

Factors Associated with QOL in Subjects with NAFLD

In univariable analyses, the following characteristics were significantly (P < 0.05) associated with poorer physical health: age greater than 40 years, female sex, type 2 diabetes, whether NASH was present, income less than $15,000, divorced, separated or widowed marital status, and BMI of at least 40. Race and level of education were not associated with physical health score.

In multivariable analysis, characteristics significantly associated with poorer physical health, after adjusting for all other variables in the model, included older age (versus age 18-30 years: age 41-50 years, B: −3.34, P = 0.031; age 51-60 years, B: −4.69, P = 0.003; age 61-76 years, B: −4.85, P = 0.004), female sex (B: −4.67, P < 0.001), type 2 diabetes (B: −4.11, P < 0.001), and lower income ($30,000-$49,400 versus <$15,000, B: 4.65, P = 0.002; ≥$50,000 versus <$15,000, B: 7.24, P < 0.001). The presence of NASH as determined by liver histology was not independently associated with poorer physical health.

Characteristics significantly (P < 0.05) associated with poorer mental health in univariate analysis included female sex, type 2 diabetes, income of less than $15,000, less than high school diploma, and divorced, separated, or widowed marital status. Age, presence of NASH, race, and BMI were not associated with the mental health score.

In multivariable analysis, characteristics significantly associated with poorer mental health, after adjusting for all other variables in the model, included female sex (B: −2.40, P = 0.007), age 18-30 years (versus 61-76 years, B: −4.91, P = 0.010), type 2 diabetes (B: −2.12, P = 0.028), less than high school diploma (B: −3.94, P = 0.020), and annual household lower income (≥$50,000 versus <$15,000, B: 5.36, P = 0.001).

Gender-Stratified Multivariable Analyses

Because sex was a significant independent predictor of both MCS and PCS, multivariable models were re-run stratified by sex. Older age, type 2 diabetes, and low household income were associated with poorer physical health in both men and women. Divorced, separated, or widowed marital status (versus married) was associated with worse mental health in men; low household income and less than high school diploma were associated with worse mental health in women.

QOL by Degree of Fibrosis

As shown in Fig. 3, there was a significant (P < 0.001) overall difference in physical health score between the fibrosis groups; scores tended to worsen as the amount of fibrosis increased. Tukey's multiple comparison tests demonstrated that participants with cirrhosis reported significantly worse physical health compared with each of the other groups (all P-values ≤ 0.001). After adjusting for age, sex, race, marital status, education, annual household income, BMI, and type 2 diabetes, presence of cirrhosis was independently associated with significantly worse physical health score compared with no fibrosis (B = −5.06, P < 0.001). If the participants with cirrhosis, however, were removed from analysis, those participants with NAFLD but not cirrhosis had a PCS score of 45.9, which remains significantly (P < 0.001) lower than the normative reference population. The difference in mental health scores did not significantly differ between the fibrosis categories.

Figure 3.

Median PCS scores and interquartile range by degree of fibrosis. ANOVA and Tukey's post-hoc analysis demonstrated that PCS score for individuals with cirrhosis was significantly (P < 0.001) lower than that of each of the other groups.

Discussion

These data demonstrate that individuals with the full spectrum of NAFLD have reduced QOL. Impairment in QOL is most evident in physical health, and mental health is affected to a lesser degree. Furthermore, there was an inverse relationship between severity of NAFLD and lower QOL, and increased fibrosis portended a decrease in physical QOL. Factors independently associated with lower QOL included older age, female sex, presence of type 2 diabetes, and poverty.

QOL was particularly low in individuals with cirrhosis; after adjustment for other characteristics cirrhosis remained a significant independent predictor of lower physical health score. Previous studies have reported lower QOL in patients with cirrhosis compared with normal controls and liver disease patients without cirrhosis. However, most of those studies were limited by small sample sizes,19, 24, 25 use of single-center samples,13, 26–30 disease-specific measures,31, 32 or the inclusion of univariable analysis only.9, 26–28, 33 Studies that have used surrogate measures of cirrhosis, such as presence of splenomegaly or patient self-report of cirrhosis, suggested that cirrhosis may be independently associated with lower QOL, in a variety of other liver diseases.34, 35 In this study there was a strong association between presence of cirrhosis and lower QOL in individuals with NAFLD.

Subjects with NAFLD reported physical QOL scores that were an average of 4.9 points lower than the general U.S. reference population while mental health scores were on average 2.4 points lower. The reference population includes individuals with chronic diseases, possibly including NAFLD—22% of the sample reported the presence of type 2 diabetes and 27% reported hypertension, two conditions associated with NAFLD. Comparing the NAFLD sample with a subset of the reference population that excluded individuals with chronic disease,23 the physical scores of the NAFLD sample were an average of 10.7 points lower, and mental health scores were an average of 4.9 points lower. Discussions surrounding what constitutes a clinically significant difference in QOL scores are longstanding.36 A difference in scores of at least half of a standard deviation (that is, 5 points) is generally believed to be a conservative estimate of a clinically significant effect size.37 In this study, physical health scores were a half standard deviation lower than the U.S. general population and a full standard deviation lower than individuals without chronic health conditions. Conversely, the mental health scores of individuals with NAFLD are likely clinically significantly lower than in individuals without chronic health conditions, but not significantly different than in the U.S. general population.

Although QOL is a relatively new concept to the study of NAFLD, QOL research itself is not new. Standard methods in QOL research are to detect whether QOL differences exist, and, if so, to then identify what factors underlie the difference. By identifying and intervening on modifiable factors that are found to be associated with decreased QOL, improvements in QOL can be realized. Whereas NAFLD is generally asymptomatic, fatigue is one symptom shown to be common in individuals with NAFLD38 and to negatively affect QOL of individuals with other types of liver disease.9, 24, 39–42 Thus, NAFLD-associated symptoms, in particular fatigue, may contribute to decreased QOL. A recent study demonstrated that fatigue was significantly associated with a decrease in all domains of the Chronic Liver Disease Questionnaire in individuals with NAFLD.11 In the current study, vitality was one of the QOL dimensions on which individuals with NAFLD reported the lowest scores.

The current data suggest that individuals with NAFLD have a greater degree of limitation in physical health than patients with other types of liver disease. Using a liver disease–specific instrument, patients with NAFLD reported greater systemic symptoms and greater degrees of limitations in activity compared with patients with hepatitis B virus or hepatitis C virus.11 It is notable that the PCS scores of individuals with NAFLD in this study are comparable to the PCS scores reported for patients with hepatitis B virus who had decompensated cirrhosis.13 From cross-sectional studies, one cannot determine whether poor physical function contributes to NAFLD and NAFLD severity, or conversely, whether presence of NAFLD results in poor physical function.

Obesity43–46 and diabetes23 are two common co-morbidities in individuals with NAFLD41 and are also associated with decreased QOL. Moreover, the presence of co-morbid illness has been associated with decreased QOL in patients with other chronic liver diseases.47, 48 In the current study, patients with severe obesity had significantly lower QOL than patients who were normal weight, although there was not an independent relationship between BMI and QOL score after adjustment for other covariates. Conversely, the presence of type 2 diabetes had a significant independent negative association with both physical and mental QOL scores. However, cirrhosis was associated with lower QOL physical health score after adjusting for presence of type 2 diabetes. Thus, type 2 diabetes does not fully explain the decreased QOL in NAFLD.

Assessment of QOL is useful because this measure provides information about the overall burden of NAFLD from the individual's perspective. In addition to the negative impact poor physical health may have on an individual personally, there may be other important implications of poor self-reported physical health. A recent study found that NAFLD is associated with significant increases in medical costs and health care utilization over time.49 The degree that health perception influences health care–seeking behavior is unknown, but it can be hypothesized that poor physical health perception may impact health care utilization in individuals with NAFLD. Productivity also may be affected by poor physical health perception. One study found that lower SF-36 physical health score significantly differentiated between individuals who returned to work after liver transplantation and those who did not, despite the fact that all participants reported good health status as measured by a performance status scale.50 Poor QOL also has implications for mortality. In numerous studies, including those in patients with a variety of chronic and acute disease states (such as type 2 diabetes,51 arthritis,52 chronic obstructive pulmonary disease,53 and hemodialysis patients54), low QOL scores have been associated with increased mortality risk.55 Thus, poor subjective physical health has important implications in terms of increased health care utilization, productivity losses, and mortality.

The multicenter design of the NASH CRN and the recruitment of subjects from a variety of settings make these results generalizable to adults in the United States with a clinical diagnosis of NAFLD. However, these data may not fully reflect the larger population of all individuals with NAFLD, especially those with unrecognized NAFLD. A particular strength of this study was the inclusion of liver histology on all patients with a rigorous, standardized, central biopsy review. A limitation was the cross-sectional nature of the study; thus, we cannot determine cause and effect from these data. Longitudinal studies of NAFLD are warranted to clarify the nature and direction of associations identified from cross-sectional data. In addition, there are domains not captured by the SF-36 that may also contribute to QOL (such as sleep, cognition, sex, family, self-esteem, eating, hobbies, and communication); thus, a broader examination of such factors should be incorporated into future studies.

In summary, individuals with NAFLD have lower QOL than the U.S. general population, with and without chronic illness. Decrement in physical health is particularly apparent. Future research should focus on identifying modifiable factors that affect QOL and can be targeted for improvement. The independent inverse relationship between cirrhosis and physical health is novel and merits further evaluation. The associations between diabetes and NAFLD severity and between diabetes and QOL in adults with NAFLD warrant additional attention because these patients will increasingly be co-managed by internists, endocrinologists, and hepatologists.

Acknowledgements

The following members of the Nonalcoholic Steatohepatitis Clinical Research Network have been instrumental in the design and conduct of the NAFLD Database and PIVENS trial:

Baylor College of Medicine, Houston, TX: Stephanie Abrams, M.D.; Diana Arceo; Denise Espinosa; Leanel Fairly.

Case Western Reserve University Cleveland, OH: MetroHealth Medical Center: Arthur McCullough, M.D. (principal investigator); Diane Bringman, R.N., B.S.N.; Srinivasan Dasarathy, M.D.; Carol Hawkins, R.N.; Yao-Chang Liu, M.D.; Nicholette Rogers, Ph.D., P.A.-C.; Margaret Stager, M.D.

Cleveland Clinic Foundation, Cleveland, OH: Arthur McCullough, M.D., (principal investigator); Srinivasan Dasarathy, M.D.; Kevin Edwards, N.P.; Ruth Sargent, L.P.N.

Children's Hospital & Regional Medical Center, Seattle, WA: Melissa Coffey; Karen Murray, M.D.; Melissa Young.

Children's National Medical Center, Washington, DC: Parvathi Mohan, M.D.; Kavita Nair.

Duke University Medical Center, Durham, NC: Manal Abdelmalek, M.D.; Anna Mae Diehl, M.D.; Marcia Gottfried, M.D.; Cynthia Guy, M.D.; Paul Killenberg, M.D.; Samantha Kwan, Yi-Ping Pan, Dawn Piercy, F.N.P.; Melissa Smith.

Indiana University School of Medicine, Indianapolis, IN: Prajakta Bhimalli; Naga Chalasani, M.D.; Oscar W. Cummings, M.D.; Lydia Lee; Linda Ragozzino; Raj Vuppalanchi, M.D.

Johns Hopkins Hospital, Baltimore, MD: Ann Scheimann, M.D.; Michael Torbenson, M.D.

Riley Hospital for Children, Indianapolis, IN: Ann Klipsch, R.N.; Jean Molleston, M.D.; Girish Subbarao, M.D.

St. Louis University, St. Louis, MO: Sarah Barlow, M.D.; Jose Derdoy, M.D.; Joyce Hoffmann; Debra King, R.N.; Joan Siegner, R.N.; Susan Stewart, R.N.; Brent A. Tetri, M.D.; Judy Thompson, R.N.

University of California San Diego, San Diego, CA: Cynthia Behling, M.D.; Manual Celedon; Lisa Clark; Janis Durelle; Tarek Hassanein, M.D.; Joel E. Lavine, M.D., Ph.D.; Susana Mendoza; Jeffrey B. Schwimmer, M.D.; Claude Sirlin, M.D.; Tanya Stein; Allison Tobin.

University of California San Francisco, San Francisco, CA: Kiran Bambha, M.D.; Nathan M. Bass, M.D., Ph.D.; Linda D. Ferrell, M.D.; Danuta Filipowski; Raphael Merriman, M.D.; Mark Pabst; Monique Rosenthal; Philip Rosenthal, M.D.; Tessa Steel.

Virginia Commonwealth University, Richmond, VA: Sherry Boyett, R.N.; Daphne Bryan, M.D.; Melissa J. Contos, M.D.; Michael Fuchs, M.D.; Martin Graham, M.D.; Amy Jones; Velimir A. C. Luketic, M.D.; Bimalijit Sandhu, M.D.; Arun J. Sanyal, M.D.; Carol Sargeant, R.N., M.P.H.; Kimberly Selph; Melanie White, R.N.

Virginia Mason Medical Center, Seattle, WA: Grace Gyurkey; Kris V. Kowdley, M.D.; Jody Mooney, M.S.; James Nelson, Ph.D.; Sarah Roberts; Cheryl Saunders, M.P.H.; Alice Stead; Chia Wang, M.D.; Matthew Yeh, M.D., Ph.D.

Washington University, St. Louis, MO: Elizabeth Brunt, M.D.

Resource Centers

National Cancer Institute, Bethesda, MD: David Kleiner, M.D., Ph.D.

National Institute of Child Health and Human Development, Bethesda, MD: Gilman D. Grave, M.D.; Terry T.K. Huang, Ph.D., M.P.H.

National Institute of Diabetes, Digestive and Kidney Diseases, Bethesda, MD: Edward Doo, M.D.; James E. Everhart, M.D., M.P.H.; Jay H. Hoofnagle, M.D.; Patricia R. Robuck, Ph.D. (Project Scientist); Leonard Seeff, M.D.

Johns Hopkins University, Bloomberg School of Public Health (Data Coordinating Center), Baltimore, MD: Patricia Belt, B.S.; Fred Brancati, M.D., M.H.S.; Jeanne Clark, M.D., M.P.H.; Ryan Colvin, M.P.H.; Michele Donithan, M.H.S.; Mika Green, M.A.; Rosemary Hollick; Milana Isaacson; Wana Kim; Alison Lydecker; Laura Miriel; Alice Sternberg, Sc.M.; James Tonascia, Ph.D.; Aynur Ünalp-Arida, M.D., Ph.D.; Mark Van Natta, M.H.S.; Laura Wilson, Sc.M.; Katherine Yates, Sc.M.

Ancillary