Context: Previous studies of seasonal variation of testosterone and other hormones in men have produced mixed results regarding the number and timing of peaks and nadirs, and whether hormones vary seasonally at all. Wide variation in study designs, sample sizes, analytical methods and characteristics of the study populations may account for the heterogeneity of results.

Objective: To determine if serum total, free, and bioavailable testosterone, DHT, sex hormone binding globulin, LH, DHEA, DHEAS, estrone, estradiol and cortisol vary seasonally in men.

Design: Two blood samples were drawn 1–3 days apart at study entry and again 3 and 6 months later (maximum: six samples per subject). Hormone levels 1–3 days apart were averaged to reduce short-term intra-subject variation.

Setting: The community-dwelling population of Boston, Massachusetts.

Study Participants: 134 men 30–79 years old were randomly selected from the respondents to the Boston Area Community Health Survey. 121 men who completed all six visits were included in the analysis.

Main Outcome Measures: In a repeated measures analysis, 3-month change in hormone levels over 3 months, measured twice per subject and, in a sinusoidal nonlinear regression with random subject effects, average hormone level is samples 1–3 days apart.

Results: Aside from cortisol, no evidence of seasonal variation in hormone levels was found. The amplitude of seasonal variation was much smaller than total intra-individual variation for all hormones considered.

Conclusions: Seasonal variation is likely an unimportant source of variation clinically and in epidemiological studies of hormone levels.