ISSN 0439-755X
CN 11-1911/B

Acta Psychologica Sinica ›› 2023, Vol. 55 ›› Issue (9): 1489-1500.doi: 10.3724/SP.J.1041.2023.01489

• Reports of Empirical Studies • Previous Articles     Next Articles

Moderating effects of CSF3R and action control between stress and healthy eating: Preliminary evidence for an individual health action against stress model

HU Yueqin1, WANG Lizhong2, CHEN Gang2, GAN Yiqun3()   

  1. 1School of Psychology, Beijing Normal University, Beijing 100875, China
    2WeGene, Shenzhen Zaozhidao Technology Co. Ltd., Shenzhen 518042, China
    3School of Psychological and Cognitive Sciences and Beijing Key Laboratory of Behavior and Mental Health, Peking University, Beijing 100871, China
  • Published:2023-09-25 Online:2023-06-09
  • Contact: GAN Yiqun


A healthy diet is essential to maintaining a healthy lifestyle, yet many factors may hinder a healthy eating plan. Some studies have found that people tend to consume more unhealthy foods to cope with stress. But the opposite findings also exist. Some people under stress may still maintain a healthy diet. The mechanisms by which people make different dietary choices under stress are not yet clear.
This study combined health behavior theories and vagal tone theory and proposed a gene stress interaction model of dietary choice, arguing that different dietary choices under stress may be related to two types of vagal tone: trophotropic or ergotrophic, representing the tendency to regenerate and recover or the tendency to prepare for fight or flight. CSF3R is a granulocyte colony-stimulating factor receptor that is closely associated with chronic psychological stress responses, and its different genotypes may lead to different stress responses, exhibiting the aforementioned tendency toward or away from nutrition. In addition, action control suggested by Health Action Process Approach may also play a critical role in the implementation of healthy eating intentions into healthy eating behaviors. Figure 1 shows the conceptual model proposed in this study, the Individual Health Action Against Stress Model.
A sample of 14,675 Chinese adults completed genotyping of their saliva samples, as well as questionnaires on perceived stress, action control, and healthy dietary intentions and behaviors. Their mean age was 28.17 years (SD = 7.09); 6363 (43.4%) were male, 8267 (56.3%) were female, and 45 (0.3%) did not report gender; 3601 (24.5%) had a master's degree or higher, 8614 (58.7%) had a bachelor's degree, 1597 (10.9%) had a college degree, 817 (5.6%) had a high school education, and the remaining 46 (0.3%) had less than high school education. Perceived stress was measured by the Perceived Stress Scale (Cohen et al., 1983). Action control was measured by a questionnaire developed by Sniehotta et al. (2005). And healthy dietary intentions and behaviors were measured by the same items used in Schwarzer & Luszczynska (2008). CSF3R was used as a candidate gene reflecting vagal tone due to its proven impact on stress response. Table 1 shows the descriptive statistics and correlations.
Regression analysis was used to test the effect of stress on healthy dietary intentions and behaviors, and the moderating effect of the CSF3R gene and action control. Table 2 shows the regression results. Specifically, Model 1-2 tested the mediating effect of intention between stress and behavior, Model 3-5 tested the moderating effect of gene between stress and intention, and Model 6 tested the moderating effect of action control between intention and behavior. Regression results found that stress predicted a significant reduction in healthy dietary intentions, B = -0.034, SE = 0.002, p <.001, 95% CI = [-0.038, -0.030], and a subsequent reduction in healthy dietary behaviors, with a direct effect of B = -0.008, SE = 0.001, p <.001, 95% CI = [-0.010, -0.006], and an indirect effect of B = -0.022, SE = 0.002, 95% CI = [-0.025, -0.019]. The CSF3R gene rs4076431 and its linked loci rs4498771, rs10752589, rs9660229 moderated the relationship between stress and healthy eating intentions, B = 0.013, SE = 0.004, p <.01, 95% CI = [0.004, 0.021], while action control moderated the relationship between healthy eating intentions and behaviors, B = -0.067, SE = 0.003, p <.001, 95% CI = [-0.073, -0.060]. In rs4076431, for example, the negative relationship between stress and healthy eating intentions was stronger in AA genotype (trophotropic) than the G carriers (ergotrophic). And higher levels of action control were associated with more healthy eating behaviors and less indirect impact of stress on behavior. The moderating effect of gene and action control is illustrated in Figure 2.
These results supported the individual health action against stress model, which may be applied to healthy diet education and stress management. Individual's stress levels reflecting environmental influence, dominance of vagal tone (or genotype) and action control reflecting individual physiological and psychological factors may all have an impact on dietary choices.