Cortisol and testosterone levels on a weekend and a work day in threee mountain villages in Selska Valley of northwest Slovenia

Background: In contemporary western populations, a week commonly involves 5 days of paid work (work days) and two non-working days (weekend). Work days are usually perceived as being more stressful than non-work days and this hypothesis has been tested in several studies, most of which selected subjects with jobs that are perceived to have high stress.Aim: The study measured salivary cortisol and testosterone on a work day and a weekend in a community-based sample of people going about their everyday lives and tested the hypothesis that hormone levels will be higher on a work day.Subjects and methods: Slovenian alpine villagers (30 females and 25 males) were sampled without reference to their occupation. Each individual was measured on two occasions, a day on a weekend and a work day as they went about their usual activities in the afternoon.Results: Cortisol (mean = 3.32 ng ml(-1), range 0.4-27.9) and testosterone (mean = 121 pg ml(-1), range 17-424) values were similar to other populations. Neither the age of subjects nor the time in the afternoon of sample collection were associated with hormone concentrations. On each day of collection, cortisol and testosterone values were correlated for each sex, with the estimate of the correlation coefficient ranging from 0.57 to 0.88. For females, testosterone values were higher on the weekend than the work day (102 pg ml(-1) and 60 pg ml(-1), respectively) but not for males (mean across both days 134 pg ml(-1)). Independent of this effect, the presence of a spouse or other adult in the house was significantly associated with lower testosterone levels in both sexes. Husband and wife testosterone values are correlated on the weekend (r = 0.67, p = 0.02) but not on the work day. Mean cortisol values for the weekend and work day were not different and there was no correlation between levels on these two days.Conclusions: These results, although based upon a small sample size, reveal potential relationships between testosterone, work-rest activities, and the presence-absence of a social partner that warrant further investigation.