Women’s testosterone levels decline with age – so why are we still using one ‘normal’ range?

A new Hertility study reveals that testosterone levels in women change continuously throughout adult life, challenging the idea that one broad reference range can accurately define what is “normal” for every woman at every age (Wainwright et al., 2026).

Testosterone is often thought of as a male hormone but women produce it too. It plays an important role in sexual function, mood, muscle mass, bone health, metabolism and wider reproductive physiology (Davis and Wahlin-Jacobsen, 2015).

It is also commonly tested when investigating symptoms such as excess facial or body hair, persistent acne, irregular periods, scalp hair loss and possible polyendocrine metabolic ovarian syndrome (PMOS), formerly known as polycystic ovary syndrome (PCOS).

But interpreting the result is not always straightforward.

Testosterone results are generally compared with a laboratory reference range and labelled as either inside or outside that range. However, many existing ranges are based on relatively small groups of women and use broad age categories.

Published in the peer-reviewed Journal of Endocrinological Investigation, Hertility’s study suggests that age should become a much bigger part of how testosterone results are understood.

What did the study find?

Researchers analysed testosterone levels from 5,323 carefully screened women aged 19-59 from across England, Scotland and Wales.

The main finding was clear: average testosterone levels were approximately 1.4% lower with each additional year of age.

The amount of variation between women also increased with age. This means testosterone did not simply decline at one defined life stage, such as menopause. Instead, levels changed gradually throughout adult life.

In practical terms, the same testosterone result may not mean exactly the same thing for a 22-year-old and a 52-year-old.

The researchers therefore developed an age-continuous reference model. Rather than placing women into broad groups such as ages 20-49, which is what one of the current reference ranges does, the model shows how a result compares with the testosterone distribution expected at each individual age.

About the research

The study retrospectively analysed anonymised data collected through Hertility’s reproductive health service between September 2020 and August 2025.

More than 30,000 records were initially available. Strict criteria were then applied to establish a reliable ‘healthy’ reference population.

The final reference group included 5,323 women who had regular menstrual cycles, a normal BMI and no reported history of reproductive health conditions or hirsutism (excess hair growth).

Samples were collected during the early part of the menstrual cycle, following a fast, and analysed using the same laboratory platform. This helped reduce some of the variation caused by sample timing and different testing methods.

A separate group of 2,338 women who reported hirsutism was also assessed.

What did the study find about excess facial or body hair?

Hirsutism describes the growth of thicker, darker hair in areas such as the face, chest, abdomen or back.

It can be a clinical sign of androgen excess and is commonly associated with PMOS, although not everyone with hirsutism will have raised testosterone or PMOS (Escobar-Morreale et al., 2012).

The study found that women reporting hirsutism had significantly higher testosterone levels on average than women in the reference population.

The difference was most noticeable at younger ages. At age 18, the most typical testosterone level in the hirsutism group was around 24% higher than in the reference group. This difference became smaller with age but remained present throughout adult life.

However, there was considerable overlap between the two groups.

This is important because it shows why symptoms should not be dismissed simply because a testosterone result is labelled normal. Equally, a raised testosterone result cannot diagnose the cause of someone’s symptoms on its own.

Blood results and symptoms need to be interpreted together.

What is a normal testosterone level in women?

There is no single testosterone level that is universally normal for every woman.

The reference range shown on a blood test can vary depending on the laboratory, the testing method and the population used to create it.

Testosterone levels can also be influenced by menstrual-cycle timing, time of day, body mass index, stress, hormonal contraception, pregnancy and menopausal stage (Schiffer et al., 2023).

Hertility’s findings add age as another essential consideration.

A result may fall inside a broad laboratory range while still being relatively high or low compared with what is typically expected for someone of that specific age.

It is also important to understand that a reference range is not the same as a diagnostic threshold. A reference range describes the values found in a selected population; it does not create a clear dividing line between health and disease.

Why current testosterone reference ranges may be limited

One widely used testosterone assay manufacturer established its female reference range using just 149 women, divided into two broad categories: ages 20-49 and ages 50 and over.

By comparison, Hertility’s new model was developed using 5,323 carefully selected women and estimated expected testosterone distributions continuously across ages 19-59.

Reference ranges may also differ between NHS trusts, private providers and laboratories because testing sites use different platforms or locally established ranges.

This means the same testosterone level may be interpreted differently depending on where someone is tested.

The study does not propose that clinical decisions should be made from age alone. Instead, it provides greater context to help clinicians understand how common or unusual an individual result is compared with other women of the same age.

What does this mean for PMOS?

Higher testosterone is commonly associated with PMOS, a complex hormonal and metabolic condition that can affect menstrual cycles, ovulation, skin, hair and fertility.

Androgen excess may be identified through visible symptoms, such as hirsutism, or through hormone testing. International guidance recognises both clinical and biochemical signs of androgen excess when assessing someone for PMOS (Teede et al., 2023).

The Hertility study does not create a new diagnostic cut-off for PMOS.

Instead, it reinforces that a testosterone result within a broad laboratory range does not automatically exclude androgen excess or PMOS – particularly where someone has relevant symptoms or irregular periods.

Testosterone should be considered alongside menstrual history, symptoms, other hormone results and the wider clinical picture.

What does this mean for patients?

For anyone having their testosterone tested, the key takeaway is that the number should not be interpreted in isolation.

Questions you may want to discuss with your healthcare professional include:

• Is this result typical for someone of my age?
• Does it fit with my symptoms and menstrual-cycle pattern?
• Were other relevant hormones, such as SHBG, measured?
• Could medication or hormonal contraception affect my result?
• Do my symptoms require further assessment despite a result within range?

Rapidly developing symptoms, marked changes in hair growth, voice deepening or other signs of virilisation should always be assessed promptly by a healthcare professional (Elhassan et al., 2025).

What are the study’s limitations?

Although this is a large study, it is important to recognise its limitations.

The research was cross-sectional, meaning it compared different women at different ages rather than following the same women throughout their lives. The estimated 1.4% annual difference therefore describes a population-level pattern, not the precise rate at which every individual woman’s testosterone will decline.

Health information and symptoms were also self-reported, and the study population was predominantly white. Further research is needed to confirm and expand the model in more ethnically diverse populations.

The study measured total testosterone using an immunoassay. Results may differ between laboratory methods, so the new range should be applied specifically within the context of the assay used.

Why better hormone reference ranges matter

A “normal range” can appear objective, but it depends on who was included, how their samples were collected, how their hormones were measured and how age and symptoms were considered.

Large, carefully characterised datasets allow us to move beyond simply asking whether a result falls inside or outside a broad range.

Instead, we can ask:

Is this result typical for this individual, at this age, with these symptoms?

At Hertility, this is why reproductive health research matters: to provide stronger evidence, support more personalised interpretation and help people access clearer answers about their bodies.

Read the peer-reviewed research

Wainwright, E., Getreu, N., Crawford, L. et al. (2026) ‘Redefining testosterone reference ranges for adult females’, Journal of Endocrinological Investigation. https://doi.org/10.1007/s40618-026-02929-w

Frequently asked questions

Do women have testosterone?

Yes. Women naturally produce testosterone in the ovaries and adrenal glands. Although levels are lower than in men, testosterone contributes to sexual function, mood, bone health, muscle mass, metabolism and reproductive physiology.

Does testosterone decrease with age in women?

The Hertility study found that average testosterone was approximately 1.4% lower with each additional year of age across women aged 19-59. This is a population-level estimate and does not mean every individual’s level falls by exactly this amount each year.

What are the symptoms of high testosterone in women?

Possible signs of androgen excess include excess facial or body hair, persistent acne, irregular or absent periods and scalp hair thinning. These symptoms can have several causes and require appropriate clinical assessment.

Can you have androgen-related symptoms with a normal testosterone result?

Yes. The study found considerable overlap between testosterone levels in women with and without hirsutism. A result within the laboratory range does not automatically rule out clinically meaningful symptoms or androgen excess.

Can testosterone diagnose PMOS?

Not on its own. Testosterone can contribute to a PMOS assessment, but diagnosis must also consider symptoms, menstrual or ovulatory patterns and other possible causes.

References

Davis SR, Wahlin-Jacobsen S (2015) Testosterone in women—the clinical significance. Lancet Diabetes Endocrinol 3:980–992. https://doi.org/10.1016/S2213-8587(15)00284-3 

Elhassan YS, Hawley JM, Cussen L, Abbara A, Clarke SA, Kempegowda P et al (2025) Society for endocrinology clinical practice guideline for the evaluation of androgen excess in women. Clin Endocrinol (Oxf) 103:540–566. https://doi.org/10.1111/cen.15265 

Escobar-Morreale HF, Carmina E, Dewailly D, Gambineri A, Kelestimur F, Moghetti P et al (2012) Epidemiology, diagnosis and management of hirsutism: a consensus statement by the Androgen Excess and Polycystic Ovary Syndrome Society. Hum Reprod Update 18:146–170. https://doi.org/10.1093/humupd/dmr042 

Schiffer L, Kempegowda P, Sitch AJ, Adaway JE, Shaheen F, Ebbehoj A et al (2023) Classic and 11-oxygenated androgens in serum and saliva across adulthood: a cross-sectional study analyzing the impact of age, body mass index, and diurnal and menstrual cycle variation. Eur J Endocrinol 188:86–100. https://doi.org/10.1093/ejendo/lvac017 

Teede HJ, Tay CT, Laven JJE, Dokras A, Moran LJ, Piltonen TT et al (2023) Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome. J Clin Endocrinol Metab 108:2447–2469. https://doi.org/10.1210/clinem/dgad463 

Wainwright, E. et al. (2026) ‘Redefining testosterone reference ranges for adult females’, Journal of Endocrinological Investigation. https://doi.org/10.1007/s40618-026-02929-w