Dietary habits and genetic susceptibility: correlations between nutritional intake and genetic risks for schizophrenia and bipolar disorder

Abstract

Dietary habits may impact the prevention and management of schizophrenia (SCZ) and bipolar disorder (BD), and genetic and environmental factors can influence both these habits and these disorders. This study investigated the effects of genetic predispositions to SCZ and BD on current dietary habits among older adults with lifestyle-related diseases, potentially offering insights for preventive mental health strategies. A cohort of 730 older patients who were diagnosed with or suspected of having lifestyle-related diseases was assessed for eight current dietary categories: miso soup, Japanese tea, green and yellow vegetables, light-colored vegetables, fruits, pickles, meats, and soybeans. Polygenic risk scores (PRSs) for the risk of SCZ and BD, including BD types I and II, the shared risk of SCZ and BD, and the differentiation of SCZ from BD, were calculated utilizing data from large-scale genome-wide association studies (GWASs). Our findings revealed that PRSs for SCZ and BD risk significantly influenced specific dietary habits, particularly decreased consumption of nutrient-rich foods such as light-colored vegetables (SCZ, R² = 0.0096, p = 3.54 × 10^-3; BD, R² = 0.0074, p = 9.09 × 10^-3) and soybeans (SCZ, R² = 0.0061, p = 0.019; BD, R² = 0.014, p = 8.38 × 10^-4). Notable differences in dietary effects were observed between PRSs for BD I and BD II, with a more pronounced impact associated with BD I (e.g., light-colored vegetables, BD I, R² = 0.015, p = 3.11 × 10^-4; BD II, p > 0.05). Moreover, shared genetic factors for SCZ and BD were correlated with lower intakes of miso soup (R² = 0.013, p = 1.21 × 10^-3), Japanese tea (R² = 0.0092, p = 5.59 × 10^-3), light-colored vegetables (R² = 0.010, p = 2.92 × 10^-3), and soybeans (R² = 0.014, p = 3.13 × 10^-4). No significant correlations were found between PRSs for differentiating SCZ from BD and any dietary patterns (p > 6.25 × 10^-3). Genetic risks shared by individuals with SCZ and BD may influence dietary choices in older adults, emphasizing the potential for dietary modifications as part of comprehensive strategies for the prevention of the SCZ and BD onset, as well as for the treatment of individuals at risk of or diagnosed with SCZ and BD.

Fig. 1. Partial correlations across eight dietary habits, with age, sex, and cohort as covariates.

*p < 6.25 × 10⁻³.

Fig. 2. Influences of polygenic risk scores for the risk of SCZ and BD, including BD types I and II shared SCZ and BD risk (SCZ + BD), and differentiating SCZ from BD (SCZ vs. BD) at different P_T levels on eight dietary habits.

SCZ schizophrenia, BD bipolar disorder. *p < 0.05, **p < 0.01, ***p < 6.25 × 10⁻³.

Fig. 3. Effects of PRSs for the shared risk of SCZ and BD and the risk of BD I on consumption frequencies of miso soup, Japanese tea, light-colored vegetables, and soybeans.

Box plots among food consumption frequencies based on a questionnaire with a Likert-scale indicate the individual PRSs for the shared risk of SCZ and BD and the risk of BD I in older patients with lifestyle-related diseases. PRSs polygenic risk scores, SCZ schizophrenia, BD bipolar disorder, BD I BD type I. PRSs corrected for cohort were z-standardized.