J Endocrinol Invest. 2026 Jun 24. doi: 10.1007/s40618-026-02929-w. Online ahead of print.
ABSTRACT
PURPOSE: To define an age-continuous reference range for testosterone (T) in females.
METHODS: A large-scale retrospective analysis of anonymised serum T collected via capillary sampling from a private healthcare service. 5,323 female individuals (19-59 years) with no history of reproductive conditions, normal BMI and regular cycles were analysed to establish a robust, age-continuous reference range for T, including probability density distributions and specific quantiles. Competing models of log T values as a function of age were statistically assessed, and the best model selected. Additionally, a cohort with self-reported hirsutism was compared.
RESULTS: A weakly skew-Normal distribution describes the best-fit distribution of log T values at any specific age, such that the mean decreases and the variance increases linearly with age. Analysis of the hirsutism cohort showed significantly higher T levels compared to the reference cohort.
CONCLUSION: Establishing a reliable reference range for female T enables clinicians to differentiate between physiological and pathological states (e.g. hyperandrogenism). Here, we establish a robust, age-continuous reference range for T in a large female population under strict inclusion criteria. The findings underscore the importance of age-specific reference ranges and highlight a substantial decrease in T over adult life. This study provides a valuable tool for clinicians and researchers seeking to compare patient hormone levels to a reliable reference distribution of T levels in women.
PMID:42340622 | DOI:10.1007/s40618-026-02929-w
