The phrase “female androgen insufficiency” sounds like something invented by pharmaceutical marketing departments, and depending on who you ask, the debate about whether it should be a formal diagnosis is ongoing. But the symptoms are not controversial. The clinical evidence that low testosterone causes specific, measurable problems in women is real, documented, and increasingly difficult to ignore — even if the medical establishment’s official response has ranged from “not sure” to “yes, but we still won’t approve a product for it.”
Let’s work through what the research actually shows.
The Symptom Picture
The cardinal symptoms of low testosterone in women are:
Reduced libido and sexual motivation. This is the most consistently documented effect. Not just reduced interest in sex — reduced spontaneous desire, reduced responsiveness, reduced capacity for sexual arousal. The distinction between “libido” (the intrinsic drive) and “responsiveness” (arousal in context) matters clinically, because women with low testosterone often report that they no longer experience the unprompted interest in sex that they previously had, even when circumstances are favorable and relationship quality is high.
Fatigue and reduced sense of wellbeing. This is harder to isolate because fatigue has many causes, but women with documented low testosterone levels report a qualitatively specific kind of tiredness — low motivation, reduced sense of vitality, a general dimming of energy that precedes any actual exertion. In the Boston University TDSM trial (Huang, Basaria, et al., 2014), testosterone administration in hysterectomized women with low testosterone was associated with dose-dependent improvements in psychological wellbeing scores in addition to sexual function scores.
Loss of muscle mass and physical performance. Testosterone has anabolic effects in women at physiologically relevant concentrations. In the Huang et al. (2014) dose-response trial — which used graded weekly injections of testosterone enanthate in 71 surgically menopausal women with baseline total testosterone below 31 ng/dL — the highest dose group (25 mg weekly, which brought testosterone levels to around 210 ng/dL) showed significant increases in lean body mass, chest-press power, and loaded stair-climb power compared to placebo. Changes were dose-dependent, correlated with free testosterone concentrations, and measurable within 24 weeks.
Cognitive symptoms. Brain fog, reduced verbal fluency, and difficulty with concentration are reported by women with low testosterone, and androgen receptors in the prefrontal cortex and hippocampus suggest a plausible mechanism. The evidence here is thinner and more contested than the libido and muscle data, but it is not absent.
Menopause: Natural vs. Surgical
Here is a distinction that does not get made clearly enough in most clinical discussions.
Natural menopause — the cessation of menstruation due to ovarian follicle depletion, typically around age 51 — causes a dramatic drop in estrogen, a substantial drop in progesterone, but a relatively modest drop in testosterone. The climacteric ovary continues to produce androgens even after it stops producing estrogen. Women in natural menopause retain roughly 50–60% of their premenopausal testosterone production, with the remainder coming from adrenal androgens and peripheral conversion.
Surgical menopause — bilateral oophorectomy, the removal of both ovaries — is a different hormonal event entirely. When both ovaries are removed, testosterone production drops precipitously and immediately, because the ovaries were contributing a substantial fraction of the total. Studies document a roughly 50% reduction in circulating testosterone within weeks of surgical oophorectomy. Because oophorectomy is often performed alongside hysterectomy, particularly for cancer prevention or endometriosis management, this surgical testosterone deficiency is common.
The clinical consequence is measurable. Women who undergo surgical oophorectomy report significantly worse sexual function outcomes than naturally menopausal women — even when both groups receive estrogen replacement therapy. The estrogen restores what the estrogen deficiency took away. The testosterone deficiency remains, untreated.
This disparity was the clinical rationale behind the TDSM (Testosterone Dose Response in Surgically Menopausal women) trial at Boston University, which specifically enrolled hysterectomized women with and without oophorectomy, with baseline testosterone below 31 ng/dL or free testosterone below 3.5 pg/mL. The trial ran from 2005 to 2011 and is one of the more rigorous dose-response datasets available for female testosterone.
The Decline Curve
The timeline of testosterone decline in women is also worth understanding, because it undermines the common assumption that testosterone is only a concern at menopause.
Female testosterone peaks in the early-to-mid 20s and then begins a gradual decline that continues through the fourth and fifth decades. By age 40, many women have testosterone levels roughly half those of women in their 20s. By the time of natural menopause, the decline has been underway for 20–25 years.
This means that symptoms of low testosterone can emerge well before menopause — and frequently do. Premenopausal women in their 30s and early 40s who report declining libido, persistent fatigue, and gradual muscle loss often have testosterone levels that are quantifiably lower than age-matched peers. Whether to treat premenopausal androgen insufficiency is a contested clinical question, partly because the evidence base is thinner for that population. But the hormone physiology is not complicated: the decline does not wait for the hot flashes to start.
What the Intervention Evidence Shows
Several well-designed placebo-controlled trials have tested testosterone in women and found consistent improvements in sexual function.
The most extensively studied formulation was the Intrinsa testosterone patch (300 mcg/day), manufactured by Procter & Gamble. Phase III trials published between 2003 and 2006 found significant improvements in “satisfying sexual events” and sexual desire in both surgically and naturally menopausal women. The Buster et al. (2005) trial in surgically menopausal women showed a statistically significant increase in satisfying sexual events (roughly 1 additional event per 4-week period compared to placebo — which, before you roll your eyes, was a clinically meaningful effect to the participants).
The drug received approval in Europe in 2006 under the name Intrinsa. It was never approved in the United States. The FDA declined approval in 2004, citing concerns about long-term cardiovascular and breast cancer safety data. Procter & Gamble subsequently withdrew the application worldwide in 2012, citing commercial reasons. The clinical evidence did not disappear — only the commercial vehicle for accessing it did.
That is, essentially, the current situation. The evidence for benefit in postmenopausal women is robust, particularly for libido. The evidence for harm at physiologic doses in women is much weaker than the FDA’s caution implied. But without an approved product, access remains dependent on compounding pharmacies, off-label use of male-formulated products at dramatically reduced doses, and clinicians who are willing to work outside the standard playbook.
A Note on Assay Accuracy
One more factor that contributes to underdiagnosis: the tests used to measure testosterone in women are often unreliable at female ranges.
Most standard immunoassay-based testosterone tests were designed and calibrated for the male range of 300–1,000 ng/dL. At female concentrations in the 10–50 ng/dL range, measurement error is proportionally larger and cross-reactivity with other steroids creates additional noise. The gold standard for female testosterone measurement is liquid chromatography-tandem mass spectrometry (LC-MS/MS), which the Huang et al. (2014) trial used specifically because of this limitation.
In a typical outpatient lab setting, a woman’s testosterone might be reported as 18 ng/dL or “within normal range” — when the “normal range” was derived from a general population including women on oral contraceptives (which suppress testosterone), women of very different ages, and data collected with assays that cannot reliably distinguish 18 from 25 ng/dL at these concentrations. Whether 18 ng/dL is symptomatic depends on the woman, her baseline, and her receptor sensitivity — none of which a single lab value captures.
This is one reason clinical assessment of female androgen insufficiency remains symptom-anchored rather than purely biochemical. The numbers are a starting point. They are not the whole picture.
Next up: Post 3 — The ISSWSH and FDA Standoff: why the professional society that studies female sexual medicine recommends testosterone for HSDD, and why the FDA has never approved a product to deliver it.
