Study protocol was described previously in details (4, 9). Briefly, 143 patients with NAFLD and 471 controls were included in the present case-control study. The cases were patients with NAFLD diagnosed by a gastroenterologist for presence of hepatic steatosis in ultrasound exam during previous month who were referred to a tertiary hepatology clinic to be examined by Fibroscan. Controlled attenuation parameter (CAP) score of more than 263 and fibrosis score> 7 in the Fibroscan result were two criteria for NAFLD diagnosis. Controls were randomly selected age-matched subjects from the same clinic among patients who had no evidence of hepatic steatosis on the ultrasound exam. All participants were recruited using convenience sampling method based on inclusion criteria and written consent was obtained. The local Ethics Review Committee approved the study protocol.
At baseline, participants were asked about their age, employment, marital status, education, smoking, past medical history, current use of medications and other factors during a 45-min in-person interview. Physical activity level as metabolic equivalent hours per week (METs h/wk) was evaluated using questionnaire. Also, a valid and reliable 168-item semi quantitative food frequency questionnaire (FFQ) was used to assess dietary intakes (10).The data from the FFQ was used to determine foods consumption frequency on a daily, weekly or monthly basis during the past year with standard portion sizes, as commonly consumed by Iranians. Data obtained from FFQ were converted to gram intake per day for each food item in order to assess the nutrient and total energy intakes using the Nutritionist 4 software (First Data bank) (11).
Energy-dense nutrient-poor snacks, in this study, were divided into four categories as follow: biscuits and cakes (biscuits, crackers, cakes, cookies and traditional Iranian confectioneries such as gaz, sohan, noghl, halva, Yazdi cakes), candies and chocolates, salty snacks (potato chips, puff snacks) and soft drinks. Fruits, dairy products, and cereals with low or medium energy density and high nutritional value were not considered as snacks. As a whole, these four groups formed total snacks. All these practices were done by a trained dietitian.
Statistical analysis: All statistical analyses were carried out using SPSS (Version 21.0; Chicago, IL, USA), and P-values at < 0.05 were considered significant. Comparison of baseline characteristics and dietary intakes between two study groups were performed using student t-test for continuous variables and chi-square for categorical variables. To evaluate associations between energy-dense nutrient-poor snacks and NAFLD risk, the study participants were divided into 4 categories on the basis of quartiles of total snacks intake and the lowest quartile was set as the reference category. ANOVA test was used to compare the variables between quartiles and P for trend was calculated using linear regression. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multiple logistic regression analysis. Analyses were adjusted for age, sex, body mass index (kg/m2), physical activity (MET-h/wk), alcohol and energy intake (kcal/d). Also, to perform the linear trend tests, quartile-specific medians were assigned to each quartile.