Night work has been suggested as a risk factor for diabetes. Individuals with high triglyceride levels, high LDL cholesterol, low HDL cholesterol and obesity, especially abdominal obesity, have a greater chance of developing diabetes. The aim of this study was to analyze glycemic levels, total cholesterol, HDL-C, LDL-C, triglycerides and the anthropometric alterations that precede diabetes, considering their possible association with nigh work among a non-diabetic population. Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) comprises of 15,105 civil servants (35-74 years old) at baseline (2008-2010). The following parameters were analyzed: serum cholesterol (total cholesterol, HDL-C, LDL-C), triglycerides and glucose drawn from 12-hour fasting blood sample, glycated hemoglobin and 2-hour plasma glucose obtained after a 75 g oral glucose tolerance test, BMI, hip and waist measurements using standard equipment and techniques. Participants with diabetes, retired workers and day workers with previous experience of night work were excluded. Generalized linear models, a gamma regression model with an identity link function, were performed to test the association of night work with metabolic and anthropometric variables. The study sample consisted of 3918 men and 4935 women; 305 (7.8%) and 379 (7.7%) of the participants were men and women who worked at night, respectively. Among the men, the exposure to night work was associated with an increase in BMI (b-value = 0.542; p = 0.032) and waist circumference (b-value = 1.66; p = 0.014). For women, increased fasting plasma glucose (b-value = 2.278; p < 0.001), glycated hemoglobin (b-value = 0.099, p < 0.001) and 2 hour plasma glucose (b-value = 5.479, p = 0.001) were associated with night work after adjustments. No significant associations between night work and triglycerides, LDL-C, HDL-C, total cholesterol levels or waist-hip ratio were found. The influences of night work on metabolic and anthropometric factors suggest night work as a potential risk factor for type 2 diabetes. Further studies are needed to investigate the inconclusive data on gender differences in the associations.