Testosterone in menopause: a review of the evidence and prescribing practice

Abstract

Testosterone treatment for postmenopausal women with hypoactive sexual desire disorder (HSDD) is controversial. The absence of a strong recommendation from the National Institute for Health and Care Excellence (NICE), as well as the lack of a UK licensed product for women, has led to national variation on local formulary positions, resulting in a postcode lottery of prescribing practice and provision. However, the demand for testosterone for this indication has increased 15-fold over the past 10 years and NHS spend in England has increased from £150,000 in 2015 to £2.5m in 2023. Consensus is urgently needed on national pathways and prescribing. 

Keywords: menopause, testosterone, sexual dysfunction

Introduction

Testosterone is an important hormone for sexual and metabolic function in women^​1​. Testosterone contributes to libido, sexual arousal and orgasm by increasing dopamine levels in the central nervous system and maintains normal metabolic function, muscle and bone strength, urogenital health, mood and cognitive function^​2,3​.  

Testosterone circulates in high concentrations relative to estradiol in women and it is estimated that healthy, young, premenopausal women naturally produce 200–250 micrograms of testosterone a day^​4,5​. Although testosterone plays a significant role and is present in higher concentrations than estradiol in women, research on its effects — as well as the consequences of testosterone deficiency and replacement in women — remains limited^​4​.   

Testosterone levels naturally decline throughout a woman’s life, with levels falling by 25% between the ages of 20 and 40 years, with a further 10–20% decline over the following 10 years, plateauing around the age of 55 years^​6,7​. Clinical practice suggests that this gradual decline in testosterone levels over decades may go unnoticed by some women, while others may be more sensitive to the changing levels of testosterone. There is limited research on individual differences in hormonal sensitivity and symptom expression during menopause. 

Declining testosterone can cause hypoactive sexual desire disorder (HSDD). Testosterone is used to treat HSDD but there are no licensed testosterone products for women in the UK and NICE is yet to make a strong recommendation for its use. 

In its guidance ‘Menopause: diagnosis and management’, published in November 2015, NICE advises to “consider testosterone supplementation for menopausal women with low sexual desire if HRT alone is not effective”^​8​.

NICE words its recommendations carefully, based on the quality of the underpinning evidence. Where there is clear and strong evidence of benefit, NICE will use the word ‘offer’, but the word ‘consider’ will be used where the benefit is less certain^​9​. NICE has a ‘consider’ recommendation for testosterone for low libido. 

This article explores the evidence base around testosterone prescribing, the patchwork nature of formulary positions across England, and the surge in demand and NHS spend in particular patient groups. We advise primary care pharmacists and GPs to only consider testosterone replacement for postmenopausal women with HSDD when oestrogen replacement is optimised. We advocate for an improved evidence base and cost-effectiveness analyses of a testosterone treatment using women-specific formulations.

Hypoactive sexual desire disorder 

Circulating available androgens have been implicated in several domains of sexual response and are related to symptoms such as low sexual desire, poor sexual arousal, orgasm and diminished wellbeing in postmenopausal women^​3​. Reduced levels of circulating testosterone may lead to HSDD, which is defined as recurrent or persistent deficiency or absence of sexual desire and fantasies for sexual activity that results in marked distress or interpersonal difficulty. HSDD is the most common female sexual dysfunction and peaks during menopause years^​10,11​. HSDD is present in 12.3% of women aged 45–64 years, compared with 8.9% of women aged 18–44 years and 7.4% of women aged over 65 years^​11​. 

There is no single standardised test to diagnose HSDD, but guidelines issued by the International Society for the Study of Women’s Sexual Health recommend diagnostic strategies, including the use of the Decreased Sexual Desire Screener (DSDS), a validated tool for use in general practice, and/or a sexual history^​10,12​. HSDD is diagnosed on presentation of persistent lack of sexual desire that has lasted for at least six months, accompanied by clinically significant personal distress, such as frustration, grief, guilt or worry. It may involve decreased or absent spontaneous or responsive desire, a loss of motivation to engage in sexual activity and avoidance of sexual situations, provided these symptoms are not secondary to another condition, such as sexual pain disorders.

Given the biopsychological nature of HSDD, consultations with women should include an assessment of symptom severity, underlying contributing factors and the impact on the woman’s quality of life. Low libido is a frequently reported concern but does not necessarily indicate the presence of a disorder in the absence of marked distress or interpersonal difficulty. The treatment approach begins with patient education, the adjustment of modifiable risk factors, and — when required — the addition of further therapeutic interventions^​13​.

Treatment of HSDD

Sexual disorders in women are complex and require an individualised approach to address modifiable biopsychosocial factors — such as personal or relationship difficulties — through counselling or referral for individual/couples therapy and/or pharmacological intervention to address any menopause-related hormonal deficiencies^{​3,10,13​}.  The association of testosterone levels and sexual desire disorders is variable, and there is no specific serum testosterone concentration that reliably correlates with the presence, absence or severity of HSDD^​11​.

HRT

The lack of oestrogen production in menopausal women may induce sexual symptoms related to vulvovaginal atrophy and other climacteric symptoms, such as sleep disturbances and mood changes, which can negatively impact sexual wellbeing. Vulvovaginal atrophy is caused by decreased oestrogenisation of the vaginal tissue and can lead to symptoms such as vaginal dryness, irritation, soreness and dyspareunia, all of which could affect a woman’s libido^​14​. Vaginal dryness tends to increase in severity with time since menopause, which can be relieved with vaginal moisturisers, systemic and/or local oestrogen treatment^​15,16​.

The British Menopause Society (BMS) recommends that oestrogen-based HRT, with or without progestogen, is optimised in the management of menopause symptoms, such as vulvovaginal atrophy and low libido^​13​. The BMS recommends that HRT is started with a low-dose preparation and reviewed after three months to evaluate its effectiveness and tolerability, and at least once a year thereafter, unless there are clinical reasons for an earlier review (such as inadequate treatment response, side effects or adverse events). Each review should also include a discussion of any bleeding pattern, a review of HRT type and dose, and a discussion to assess the ongoing risk and benefit of HRT^​17​.

Optimisation of HRT may involve increasing oestrogen content of HRT up to the maximum licensed dose, as tolerated, switching to an alternative preparation, changing formulation or adding vaginal oestrogen therapy, which can be used in conjunction with systemic HRT^​13​.  

Testosterone

If HRT with adequate oestrogen intake has not been effective, then testosterone supplementation can be considered in menopausal women with low sexual desire. Testosterone supplementation is not a third component of HRT; however, it is an option if standard, well-established hormone replacement with oestrogen-based HRT has been ineffective alone. This is an unlicensed indication, but supported by the National Institute for Health and Care Excellence and the British Menopause Society in guidance on management of menopausal symptoms. While NICE recognises the potential benefits of testosterone, it does not issue a strong recommendation for its use^​2,8​.

Testosterone gels for HSDD

There are several transdermal testosterone gels available in the UK that are licensed for hypogonadism owing to androgen deficiency in men; however, there are currently no licensed products for women in the UK. Male formulations are used off-label for use in postmenopausal women with low sexual desire^​18​. The dose required for treatment in women for low sexual desire is much lower than the licensed indication used for hypogonadism owing to androgen deficiency in men, and dose directions to achieve this dosing can be confusing. For example, the recommended dose of Tostran gel (testosterone; ADVANZ Pharma) for male use in hypogonadism (60mg daily) is approximately one twelfth of the recommended dose for female use (10mg of testosterone on alternate days)^​19​.  For this reason, patients must be advised not to follow the directions in the patient information leaflet provided with the medication.

There are two testosterone brands commonly used off-label for low sexual desire in postmenopausal women, including Tostran 2% transdermal gel pump and Testogel 40.5mg/2.5g transdermal gel sachets (Besins Healthcare). Prescribers must be aware of the differences between the brands of testosterone gel used in women, see Table 1^{​2,19,20​}. 

In January 2023, the Medicines and Healthcare products Regulatory Agency (MHRA) published a drug safety update that warned of the risk of harm to children following accidental exposure to Testogel^​21​. Premature puberty and genital enlargement have been reported in children who were in close physical contact with an adult using topical testosterone and who were repeatedly accidentally exposed to this medicine^​21​.

Other strengths/formulations are available, such as Testogel 16.2mg/g transdermal gel (20.25mg per actuation)^​22​ and Testavan 20mg/g transdermal gel (23mg per actuation) with applicator^​23​; however, these are not suitable for female use owing to the pump mechanisms, making it too difficult to deliver the required dose.

Previously, there were two other testosterone gel options; however, Testogel 50mg/5g gel sachets were discontinued in March 2022, and Testim 50mg/5g transdermal gel unit dose tubes (Besins Healthcare (UK) Ltd) were discontinued in December 2023^​24​.

Androfeme cream (Lawley Pharmaceuticals Pty Ltd) is a brand of testosterone licensed in Australia for use in women^​2​. While it can be imported to the UK under a special-order license, it is not approved for use on any formulary in England and should not be prescribed on NHS prescription. Private providers may opt to prescribe Androfeme, which can be an expensive option for women, costing approximately £100 for a 100-day supply. There have been unofficial reports that Androfeme cream is being reviewed by the MHRA. The authors contacted the MHRA, who were unable to comment on pending regulatory approvals to third parties. At the time of writing, there were no active UK licences for Androfeme cream. 

A testosterone patch authorised by the European Medicines Agency in 2006 for use in women with surgical menopause was withdrawn from the market in 2012 for commercial reasons^​25​.

Testosterone monitoring 

Testosterone prescribing for women requires careful consideration to ensure both efficacy and safety. It is essential to follow a structured monitoring schedule to assess the treatment’s effectiveness and identify any potential side effects, such as excess hair growth, acne and weight gain. According to the BMS guidelines, testosterone treatment should be regularly reviewed, with blood tests conducted before initiating therapy, 2–3 months after starting and subsequently every 6–12 months^​2​.

Limitations to accessing testosterone for menopause

The lack of a UK licensed product is one of many reasons why women experience difficulties accessing testosterone, and because GPs may be reluctant to prescribe current licensed formulations off-label for women. Prescribers will look to national and local guidance to support decisions on whether to prescribe testosterone to women for the management of menopausal symptoms. 

Formularies within individual integrated care systems will guide clinicians on whether testosterone is supported locally. The absence of a strong recommendation from NICE has led to national variation in the formulary positions of testosterone for low libido in menopause.  

There are 42 integrated care boards (ICBs) in England^​26​, each of which has at least one of its own formularies. There are 46 ICB formularies in the public domain and each was manually checked by one of the authors (KNA) for the inclusion of testosterone gel for low libido in postmenopausal women; the outcome of this review is summarised in the following section.

Formulary traffic light system

Formularies use a traffic light system (TLS) to provide a framework to define where clinical, and therefore prescribing, responsibility should lie through the categorisation of individual medicines. Individual formularies have subtle differences in the definitions of the TLS.

Generally, the following definitions reflect if and when testosterone prescribing for menopause can take place. The variation in formulary position means women across the country have very different experiences of accessing treatment. The position of the formularies can be seen in Figure 1. 

There are subtypes of amber under the TLS that dictate where testosterone can be initiated and at which point transfer of care from the specialist to GP can occur. For three-quarters (75%) of formularies that classify testosterone as amber, treatment should be provided under specialist initiation and continuation of prescribing until the patient is stabilised and reviewed. In these cases, a specialist monitors the patient’s response, adjusts the dose and treats side effects, as required. The GP may then be asked to take over prescribing responsibility once the patient is reviewed and stabilised on a maintenance dose. For the remaining 25%, the decision to initiate testosterone must be taken by a specialist; however, the first prescription may be written by the specialist, or the GP may be asked to provide the first and ongoing prescriptions.

A total of 13 formularies did not include testosterone within the TLS (see Figure 1)^​18​. One formulary that did not include testosterone had appraised the evidence and made a clear ‘Do not prescribe’ recommendation on the basis of lack of evidence. Ten formularies did not include testosterone on the formulary and it is unclear whether this was an intentional decision or if it had not yet been considered for inclusion. The remaining two formularies were not accessible to determine formulary position.

For medicines without a NICE technology appraisal — and in cases such as testosterone for low libido, where NICE is less decisive in its recommendation — local appraisal of the evidence is required. Local formulary decision-making groups will search for, appraise, interpret and contextualise the available evidence. It is likely that the same evidence base for testosterone for low libido is being considered in the decision-making process, yet formularies may reach different conclusions.

While clinical effectiveness and strength of evidence is crucial to deciding whether a medicine should be included on a formulary (see below), other decision-making criteria need to be applied, such as cost-effectiveness and resource impact, patient safety, place in therapy relative to available treatments, national guidance and priorities, local health priorities, equity of access and stakeholder views. 

The evidence of the effectiveness and safety of testosterone in menopausal women

In 2019, Islam et al. published a comprehensive summary of randomised trial data on the safety and efficacy of testosterone^​27​. They performed a systematic review and meta-analysis of trials of testosterone treatment (of at least 12 weeks duration) and recorded multiple outcomes of interest, including sexual function, cardiometabolic and cognitive measures. Between 1990 and 2018, they identified 36 randomised controlled trials (RCTs) involving 8,480 participants. The majority of trials were based in higher income countries, mainly in the United States (n=17) and Australia (n=12), with just one UK trial.

Replicating the protocol from this systematic review, we have identified one further small, randomised trial that has been published since 2018, in which 70 participants (35 participants in each arm) were randomised to weekly transdermal testosterone (50mg) plus daily oral 1mg estradiol valerate or only oestrogen for eight weeks^​28​. However, given the small size and shorter duration of treatment, it would not have affected the conclusions of Islam et al.’s systematic review (see main findings below).

In terms of ongoing research, in 2024, the National Institute for Health and Care Research (NIHR) funded the ‘Evaluating the clinical and cost-effectivenesS of TEstosteronE to improve Menopause-related quality of life (ESTEEM) trial. This RCT aims to recruit 416 participants to 12 months of testosterone gel or placebo to measure menopause-related quality of life beyond altered sexual function, with an expected end date of October 2028^​29​.

Efficacy

The 36 randomised trials identified by Islam et al. were heterogeneous in terms of route of administration, oral (n=15) vs transdermal (n=18) vs other routes (sublingual, subcutaneous, etc). Trials included women who had undergone natural menopause (n =793) or surgical menopause (n=755). Where the primary outcome of the trial was satisfying sexual events and sexual desire, there were enough comparisons to draw conclusions on the pooled effects of testosterone. For other sexual outcomes (e.g. orgasms), the pooled data were more limited owing to variable use of questionnaire types and timing. 

A total of 18 studies provided data for the effects of testosterone on sexual function. There were 8 studies that showed mean change in the number of satisfying sexual events per month, with testosterone modestly increasing the number of events compared with placebo/comparator (mean difference per month was 0.85, 95% confidence interval [CI] 0.52–1.18). Other outcomes reported by the other 10 trials included testosterone therapy improving sexual desire, pleasure, arousal, orgasm, responsiveness and self-image, all while reducing sexual distress.

Limitations

It is of note that the majority of trials (n=21) excluded women with “identifiable causes of sexual concerns”, such as dyspareunia, vaginal dryness or depression, reducing the generalisability of the findings and emphasising the need for a holistic approach to this population. Another limitation of many of these studies is the lack of an adequate control arm. As discussed, the mainstay of menopause management is adequate HRT. Of the 36 students, 14 students were placebo-controlled without standardisation of HRT prescribing. The data were too limited to make conclusions on other outcomes, such as cognitive performance (3 studies, n=159), musculoskeletal impact (7 studies, n=500) or psychological wellbeing (4 studies, n=605). While the major finding from the pooled analysis was that testosterone can increase the number of sexual events per month (mean difference per month = 0.85), without reference to a minimally important clinical difference it is uncertain whether this increase is important/noticeable for patients. 

Safety

In a pooled analysis of 10 studies (n=1,774 participants) where testosterone was administered topically (rather than orally), there was no significant rise in LDL cholesterol or reduction in total cholesterol, HDL-cholesterol and triglycerides. An overall increase in weight was recorded with testosterone treatment (mean difference 0.48kg, 95% CI 0.16–0.79) and up to 10% of women can experience an increase in acne (risk ratio [RR] 1.46, 95% CI 1.11–1.92) and hirsutism (RR 1.69, 95% CI 1.33–2.14).

There was no evidence of endometrial proliferation with testosterone use^​30​.  No randomised trials investigating breast cancer risks with transdermal testosterone replacement have been performed. 

Cost of testosterone gels for female use

No studies have been completed that assess the cost-effectiveness of testosterone gel for female use, but Table 2 below shows the estimated cost of treatment^​2,31​.

Approximately £45 per patient per year does not seem like a significant cost implication; however, the demand of testosterone has risen significantly in recent years and the population level demand for testosterone has a significant impact on NHS resources^​32​. In addition to the prescription costs, it is important to consider the associated costs of blood testing and the time clinicians spend on patient reviews. Owing to the differing TLS classifications of testosterone across regions, blood testing may occur in both primary and secondary care settings and may transition between them during treatment. Given the potential variability in contracts, access to locally enhanced services and tariffs, a cost analysis has not been included in this work. These monitoring costs should be factored into any future studies conducting cost-effectiveness analyses. 

Figure 2 shows the upward trend in testosterone gel prescription items prescribed in general practice in England for females aged 40 years and over (where demographics are identifiable; data from NHS Business Services Authority).

NHS spend by general practice in England on testosterone gels in women ≥40 years old has increased 15-fold in less than 10 years, see Figure 2^​32​.  Based on the cost of testosterone gel items, we estimate the general practice NHS spend in England has increased from £150,000 in 2015 to £2.5m in 2023. 

There are limitations to the data as it is not possible to differentiate between diagnoses for which testosterone would be prescribed. Indications other than low sexual desire in postmenopausal women may include gender dysphoria, for which numbers over the age of 40 years are small. There will also be some prescription data excluded from the analysis where the demographic information of females aged 40 years and over could not be determined.

The steepest rise has been since 2021, which follows the airing of UK TV personality Davina McCall’s documentaries about the menopause, which is believed to have led to an increase in the number of women presenting to their GP for treatment for the management of menopausal symptoms, including testosterone for low sexual desire^​32,33​.

Testosterone and socioeconomic status

While there is an overall increase in testosterone gels prescribed in women aged 40 years and over nationally, further scrutiny of national prescribing data suggests a correlation between the rate of testosterone prescribing and socioeconomic deprivation. Data on practice-level socioeconomic status were obtained from Public Health England’s National General Practice Profiles^​34​. Socioeconomic status was quantified using the Index of Multiple Deprivation (IMD) score 2019, see Figure 3^​35​.

The IMD score combines information from seven domains, including income, education and health, to produce an overall relative measure of socioeconomic status^​35​.

There are 15.9 million women aged 40 years and over in England and prescribing data showed 116,565 items for testosterone gels were issued on the NHS in 2023/2024^​36​. Prescribing in each GP practice was calculated as number of testosterone gel items per 1,000 female patients aged ≥40 years in 2023/2024. Practice-level prescribing was then categorised by IMD decile (2019), showing the mean with interquartile ranges, where decile 10 represents the practices with the highest IMD score (lowest socioeconomic status)^​34,36​. The overall rate of testosterone prescriptions (per 1,000 women aged ≥40 years) was three times higher in the most affluent practices compared with the most deprived.

The difference in prescribing of testosterone to women may not only be attributed to the decision-making behaviour of the clinician managing a woman’s menopause, but also to the consulting behaviour of the patient. It is likely that educated women, who are better informed on menopause symptomatology and management, are more likely to present to their GP for treatment of low sexual desire linked to menopause or to ask their GP to prescribe testosterone. This is an assumption that aligns with clinical observations, but further research would be necessary to substantiate this claim.

The challenge of private healthcare influencing NHS prescribing practices is an observed phenomenon in primary care. Women in more affluent areas who have the financial means to access private healthcare are within their rights to seek private menopause care; however, private physicians are not subject to the same prescribing restrictions as NHS clinicians who follow locally formulary guidelines. As a result, testosterone may be initiated earlier than recommended, possibly driven by patients who are paying for treatment and may have higher expectations than those treated within the NHS. Once a prescription has been advised or even started in the private sector, it is difficult for GPs to disagree with the advice of a specialist and refuse to prescribe. 

What should be done about this inequity of prescribing?

There is a tendency to interpret inequitable prescribing data as a need to ‘level the playing field’ by encouraging prescribing to more deprived groups. We would not advocate this approach based on the current evidence. This analysis has been presented in an attempt to explain some of the factors that have led to the rise in testosterone prescribing. In keeping with national guidance, the focus should be on optimising HRT prescribing first line. This is not to say that there is not work to do on menopause care in more deprived groups. In 2020, Hillman et al. published results of a cross-sectional study of primary care prescribing data in England in 2018, which showed that the overall prescribing rate of HRT was 29% lower in practices in areas with the highest IMD compared with the most affluent (incidence rate ratio [IRR] = 0.71; 95% CI = 0.68–0.73)^​37​.

Conclusion

Testosterone falls naturally with age and low levels may contribute to HSDD; however, the evidence base for its replacement is lacking and NICE does not strongly recommend its use. As a result, the formulary position of testosterone gels varies nationally. Regardless, testosterone prescribing has increased exponentially since 2020, with a significant variation between practices depending on deprivation scores. Our advice for primary care pharmacists and GPs is to only consider testosterone replacement for postmenopausal women with HSDD when oestrogen replacement is optimised. The major barrier to full NICE support and formulary consensus is the lack of a randomised trial and cost-effectiveness analysis of a testosterone treatment designed for women with an adequate control group. Treatment with off-label male formulations should not remain the status quo.