Methods. aEuro integral Total testosterone and sex hormone-binding globulin levels were analysed and free testosterone levels was calculated in 3413 men participating in the fifth Troms circle divide o study in 2001. Self-administered questionnaires including information about education, marital status, smoking habits and the Hopkins Symptom Checklist-10 (SCL-10, a 10-item psychological health questionnaire) were completed. The cross-sectional data were analysed with partial association and analysis of variance and covariance. Results. aEuro integral The complete SCL-10 was not associated with total or free testosterone, but symptoms of anxiety were negatively associated with both total and free testosterone (p < 0.05). Men presumed to be testosterone deficient, with testosterone levels in the lowest 10th percentile, had increased SCL-10 score compared to men with higher testosterone levels (p == 0.021), before and after adjusting for age, waist circumference, marital status, education and smoking. There was an even stronger association between men presumed to be testosterone deficient and symptoms of anxiety (p < 0.001). However, men with more pronounced symptoms indicating mental disorder did not have lower testosterone levels. Conclusions. aEuro integral Men presumed being testosterone deficient had a higher symptom score, in particularly regarding anxiety, but they did not have pathological symptoms. Thus, lower testosterone levels was only associated with subthreshold symptoms of anxiety and depression.