The present study found that eating behavior is strongly associated with NAFLD/MAFLD. Obesity can occur due to a false perception of eating habits, actual dietary consumption, and unfavorable eating habits. For instance, some patients may think that they gain body weight even by drinking water, which is a gap in the recognition of constitution and weight, and some patients may insufficiently chew food, which is a bad eating habit13. The EBQ was originally developed to evaluate unaware gaps and habits in eating behaviors, especially in obese patients13. Since behavior therapy can improve adherence to lifestyle interventions for obesity17, several guidelines for obesity treatment have emphasized the importance of behavior therapy in patients with obesity12. However, few attempts have been made to develop behavioral therapies for the treatment of NAFLD/MAFLD. We hypothesized that eating behavior can affect NAFLD/MAFLD and that the EBQ can be an effective modality for evaluating eating behavior among those with NAFLD/MAFLD.
The first intriguing finding of the present study was that eating behavior was significantly worse in participants with NAFLD/MAFLD than in those without NAFLD/MAFLD. The second intriguing finding of the present study was that the influence of eating behavior on MAFLD was stronger than that on NAFLD. The results of the present study expand our knowledge of the pathophysiology of NAFLD/MAFLD and provide an important perspective for the multidisciplinary treatment of patients with NAFLD/MAFLD.
Our study showed a significant difference in eating behaviors between participants with and without NAFLD/MAFLD. In previous studies, the total EBQ score was higher in obese patients than in non-obese patients and was associated with actual food consumption18,19. Our study revealed that young Japanese male adults with NAFLD/MAFLD had significantly higher total EBQ scores than those without NAFLD/MAFLD. As for eating behavior category, decision tree and random forest analyses revealed that “perception of constitution and weight” was the strongest contributor for NAFLD/MAFLD. The robust association between “perception of constitution and weight” and NAFLD/MAFLD can be explained by a prior study that demonstrated that “perception of constitution and weight” was strongly correlated with BMI (r = 0.41)18, which can stratify the risk of NAFLD/MAFLD20. The initial approach to behavioral therapy varies according to each patient’s problematic eating behavior; therefore, physicians should evaluate eating behavior and adopt individualized behavior therapy to improve adherence to lifestyle modifications in patients with NAFLD/MAFLD 11. Since the EBO has good internal reliability (coefficient alpha = 0.81)18, the novelty of our findings should be emphasized in that assessment of eating behavior using the EBQ can be a basic approach to initiating behavior therapy in young adults with NAFLD/MAFLD.
The second finding of our study was that the effect of eating behavior on MAFLD was stronger than that on NAFLD. In our study, the comparison of eating behavior categories between participants with and without MAFLD showed a larger number of statistically significant differences than between those with and without NAFLD (four vs. two categories). In addition, multivariate logistic regression analysis revealed that two categories of eating behavior, “perception of constitution and weight” and “bad eating habits,” were independently associated with MAFLD, whereas only one category, “perception of constitution and weight,” was associated with NAFLD. Since eating behavior can lead to obesity, our results provide a reasonable explanation for the fact that patients with MAFLD have a higher risk of obesity, fibrosis, cardiovascular disease, and mortality than those with NAFLD16,21,22. Therefore, the assessment of eating behavior can be a useful strategy to identify individuals at high risk of MAFLD who require early detection and intervention, especially at a younger age.
An additional finding of our study was that we identified an effective question item to stratify the risk of NAFLD/MAFLD. Participants who had ≥ 2 points in the question asking that “I believe I easily gain weight than others” had significantly higher prevalence of MAFLD and NAFLD than those with < 2 points and the results were confirmed by the data-mining analysis. A recent study has shown that patients with NAFLD have good knowledge of their disease; however, approximately half of patients with NAFLD are not ready for lifestyle modifications, including diet and exercise10. Although participants who had ≥ 2 points in this question may have genetic or constitutional factors that can lead to obesity, our results imply that individuals with NAFLD/MAFLD have unfavorable recognition of their constitution, which can limit adherence to lifestyle interventions. Thoughts can influence mood and behavior, and cognitive restructuring can be a useful method to modify an individual’s mindset to promote favorable effects on lifestyle modifications11. Therefore, modifying unfavorable recognition based on the results of the EBQ can be a useful method to organize effective lifestyle modifications in patients with NAFLD/MAFLD. In addition, since the EBQ has been reported to be useful in monitoring eating behavior23, the questionnaire is also applicable to behavior therapy and monitoring during lifestyle interventions in patients with NAFLD/MAFLD.
However, some limitations of this study must be addressed. First, this was a single-center study that included only young Japanese male adults, and the results may not be applicable to other regions, females, and age groups. Second, unassessed variables in the MAFLD diagnosis may underscore the prevalence of MAFLD in our study. Despite these limitations, our study has several strengths, including the use of a young cohort, originality, and application of data mining analysis and has meaningful implications for the association between eating behavior and NAFLD/MAFLD among young male adults.
In conclusion, our study adds to the growing evidence that eating behavior is significantly worse in individuals with NAFLD/MAFLD than in those without. Furthermore, our study provides evidence that the EBQ is useful for behavioral therapy in patients with NAFLD/MAFLD. Our study encourages better application of a multidisciplinary approach to maximize the benefit of lifestyle interventions in young adults and reduce the future burden of NAFLD/MAFLD.