Based on a large population of 1113 men aged 30-60 at baseline (mean: 44.1 years, standard deviation: 10.5), we investigated whether intra-individual changes in testosterone (T) and related reproductive hormones during a ten year period were dependent of marital status at baseline and follow-up. The studied men were part of a health survey in Denmark, conducted between 1982 and 1984 with a follow-up examination approximately ten years later. Data on reproductive hormones, measured in serum, and lifestyle and marital status were obtained at both time points. As expected, an age-related decline in testosterone was observed. However, independent of age and lifestyle, we observed that men who went from unmarried to married (n=81) during the study period experienced an accelerated age-related decline in testosterone (-6.6nmol/L) whereas men who went from married to unmarried (n=67) experienced an attenuated age-related decline (-2.3nmol/L). Men who were either married or unmarried at both time points (n=167, n=798, respectively) had a testosterone decline in between (-3.7nmol/L and -4.6nmol/L, respectively). Changes in T/LH ratio did not differ according to marital status indicating that the lowered T level is not compensated by increasing LH levels. This could suggest a modification of the gonadostat due to an adaptation to changing life circumstances.