Methods: This was a cross-sectional study conducted in male and female adults living in the urban area of Yogyakarta, Indonesia. Adiposity was determined based on body weight, body mass index (BMI), percent body fat, and waist and hip circumference. Data on coffee consumption and other dietary components were collected using a semiquantitative food frequency questionnaire along with other caffeine-containing beverages such as tea, chocolate, and other beverages. The -866 G/A UCP2 gene variation was analyzed using polymerase chain reaction-restriction fragment length polymorphism. The correlation between coffee intake and adiposity was tested using linear regression test with adjustment for sex, age, energy intake, table sugar intake, and total caffeine intake.

Results: In all subjects, coffee intake was inversely correlated with body weight (β = -0.122, p=0.028), BMI (β = -0.157, p=0.005), and body fat (β = -0.135, p=0.009). In subjects with AA + GA genotypes, coffee intake was inversely correlated with body weight (β = -0.155, p=0.027), BMI (β = -0.179, p=0.010), and body fat (β = -0.148, p=0.021). By contrast, in subjects with GG genotype, coffee intake was not correlated with body weight (β = -0.017, p=0.822), BMI (β = -0.068, p=0.377), and body fat (β = -0.047, p=0.504).