Higher Intake of Vegetable Protein and Lower Intake of Animal Fats Reduce the Incidence of Diabetes in Non-Drinking Males: A Prospective Epidemiological Analysis of the Shika Study

Abstract

Although nutrient intake and alcohol consumption are both closely associated with the incidence of diabetes, their interrelationships remain unclear. Therefore, we herein have investigated the interrelationships among nutrient intake, alcohol consumption, and the incidence of diabetes using longitudinal data. This study included 969 residents ≥40 years living in Japan. In 2011 and 2012, a baseline study was conducted using questionnaires on basic demographics, diabetes, nutrient intake, and lifestyle habits. In 2018 and 2019, a follow-up study was performed using questionnaires and medical records on diabetes. Two-way analysis of covariance (two-way ANCOVA) was used to test the interactions of drinking habits and diabetes incidence on nutrients intake. The prospective relationship between nutrient intake at baseline and the incidence of diabetes in the follow-up stratified by drinkers and non-drinkers was evaluated using multiple logistic regression analysis. Interactions were observed for vegetable protein intake (p = 0.023) and animal fat intake (p = 0.016) in males. Vegetable protein intake negatively correlated with the incidence of diabetes in non-drinkers (odds ratio (OR): 0.208; 95% confidence interval (95% CI): 0.046–0.935; p = 0.041). Furthermore, animal fat intake positively correlated with the incidence of diabetes in non-drinkers (OR: 1.625; 95% CI: 1.020–2.589; p = 0.041). Therefore, vegetable protein and animal fat intakes in combination with drinking habits need to be considered for the prevention of diabetes.