Unhealthy diet has been associated with obesity. Evening type has been associated with unhealthier food and nutrient intake that could predict a higher risk of obesity among them as compared to morning type. However, thus far no study has examined the interrelationships between chronotype, a healthy diet and obesity. We examined whether a healthy diet mediates the association between chronotype and obesity and whether chronotype modifies the association between a healthy and obesity. The National FINRISK 2007 Study included 4421 subjects aged 25-74 years. Diet was assessed using a validated food frequency questionnaire. Baltic Sea diet score (BSDS), including nine dietary components, was used as a measure of adherence to a healthy Nordic diet. Weight, height, body fat percentage and waist circumference were measured, and body mass index values were calculated. Chronotype was assessed using a shortened version of Horne and Östberg's morningness-eveningness questionnaire (MEQ). The sum score calculated from MEQ was either used as a continuous variable or divided into tertiles of which the lowest tertile demonstrated evening preference and the highest tertile demonstrated morning preference. A series of regression analyses were conducted to determine whether the BSDS mediates the association between chronotype and obesity. Likelihood ratio test was used to determine whether chronotype modifies the association between the BSDS and the obesity measures. After testing the interaction, chronotype-stratified analysis for the association between the BSDS and obesity measures was determined by linear regression. Generally, the evening types had lower adherence to the BSDS and were more often smokers (men), physically inactive and had lower perceived health than the other chronotypes (p < 0.05). The poorer health behavior of this group, however, was not manifested in their obesity measures, and no evidence that the BSDS would mediate the association between chronotype and obesity was found (p > 0.05). No evidence that chronotype would modify the association between the BSDS and obesity was found either (p > 0.05).
