A UK Epidemic of Testosterone Prescribing


Earn H. Gan; Stewart Pattman; Simon H. S. Pearce; Richard Quinton


Clin Endocrinol. 2013;79(4):564-570. 

In This Article


The PCA data from England, Wales and Scotland have clearly shown a progressive rise in the prescribing of testosterone from 2001–2010. This is in keeping with the data available globally, including Switzerland,[19] United States[20] and Australia,[21] suggesting a potential 'pandemic' of testosterone prescribing. There are a number of potential explanations for this trend. Firstly, conditions associated with hypogonadism might be on the rise. For example, the birth prevalence per 100 000 men with Klinefelter syndrome (KS), the most common pathological cause of androgen deficiency, is reported as having risen from 109 in the 1960s–1970s to 223 during the period of 1986–2006.[22,23] There has been a definite increase in the number of childhood and adult cancer survivors, comprising men who have received radiotherapy to the pituitary or testis or have been orchidectomized for seminoma or teratoma of the testis. Finally, opioids such as tramadol and slow-released morphine sulphate tablets are now more widely used for the treatment of chronic noncancer pain. However, the nearly twofold increase in testosterone prescribing is highly unlikely to be solely attributable to an increased prevalence of male hypogonadism, especially as most men with KS, the commonest cause of pathological male hypogonadism, only become hypogonadal later in life.[24]

Another possible explanation for the observed prescribing trend is increased testosterone testing in primary care, which could potentially result in either improved diagnosis of male hypogonadism and/or an unwarranted increase in TRT in men with borderline serum testosterone concentrations. It is entirely appropriate to treat symptomatic men with borderline-low testosterone under defined circumstances, such as gonadotrophin deficiency resulting from pituitary tumours or prescribed opiate analgesia, or elevated gonadotrophins diagnostic of primary testicular dysfunction (e.g. KS). However, with these important exceptions, there is insufficient evidence for initiating TRT in most men with a borderline-low serum testosterone due to FHH. There are likewise no data to support testosterone therapy in the context of age-related frailty.[3,25,26] We simply do not know whether borderline biochemical hypotestosteronaemia related to FHH is maladaptive, neutral or conceivably even adaptive.

Our local survey has demonstrated that the absolute number of men with likely unequivocal hypogonadism (testosterone <6·0 nm) has remained relatively constant despite a surge in testosterone testing in primary care. The rising number of requests has identified significantly more men with testosterone level in the 6–9 nm range, although 9 nm is the lower limit of the adult male normal range in our assay. Admittedly, this local pattern of requests and results might not reflect the overall situation in the UK, but there are currently no other comparable published data available.

Finally, a plausible explanation for the observed trend in testosterone prescribing in the UK is that an increasing number of eugonadal men might be receiving unnecessary testosterone therapy, particularly older men with 'nongonadal' illness. Although the newer testosterone preparations have undoubtedly improved the quality of life for many men with organic hypogonadism, due to ease of use and better pharmacokinetic profile, pharma has promoted TRT to a broader population of older men with sexual dysfunction.[27,28] Over the past 3 years, the advertising tracker Kantar Media has reported an increase of more than 170% in spending on advertising by pharmaceutical companies such as Abbott and Eli Lilly in the USA to promote TRT.[29]

Although direct-to-consumer pharmaceutical marketing is not permitted in the UK, the public has virtually unfettered access to pharma websites as key sources of information relating to ageing, erectile dysfunction and the 'andropause'.[30–33] A general theme from pharma-hosted online sources is that testosterone deficiency is common and particularly so with advancing age. Hence, it should be no surprise if men with nonspecific, physiological, age- or illness-related symptoms should increasingly consider these as possible manifestations of testosterone deficiency, with a consequent belief that TRT might restore their quality of life and virility. The question is whether this information campaign has over-reached the actual evidence?

It is well established that a significant portion of older men have testosterone levels below the lower limit for healthy, young men, with an average decline in serum testosterone levels of 1–2% per year.[34] Although this has historically been attributed to physiological decline in testosterone production with age, a large 5-year longitudinal study involving 1 382 community-dwelling men reported that the testosterone decline seen in older men was actually largely caused by a combination of an increased burden of nonendocrine disease and modifiable factors, such as obesity and lifestyle, rather than being an inevitable fact of ageing per se.[35] Indeed, the Odense Androgen Study demonstrated similar serum testosterone reference intervals for both young and elderly healthy men.[36] In keeping with these studies, the European Male Aging Study (EMAS), a large cross-sectional study involving 3 369 participants, highlighted that many predisposing lifestyle and health factors contributing to the age-related decline in testosterone levels are modifiable and preventable.[26,37] Overall, these studies do not support the concept of a widespread, inevitable syndrome of age-related sex hormone deficiency in men. The EMAS found a prevalence of late-onset hypogonadism (LOH) of only 2·1% in the general male population and 5·1% in men aged 70–79 years,[26] contrasting with a prevalence for LOH of up to 12% quoted by some pharma promotional material.[32]

Overall, testosterone levels do not correlate well with the symptoms of hypogonadism in older men with LOH,[26] and there is a lack of compelling evidence from large-scale RCTs on the benefits of TRT in ageing men.[15–18] While reduced muscle mass and bone mineral density are clearly established features of organic/syndromic hypogonadism, there is inadequate evidence for a causal relationship between the similar physiological changes observed in older men with FHH. Studies of TRT performed in frail older men have given conflicting results in relation to muscle strength, mood, sexual function and quality of life.[15–18] Moreover, restoring testosterone level in older men to the 'normal range' risks rendering them 'supraphysiologically' replaced. Older men exhibit slower clearance of steroid hormones and are more susceptible to androgen-induced polycythaemia and consequent thrombosis risk, as shown by a meta-analysis including 19 randomized, placebo-controlled studies.[38] Finally, the potential risks of converting a clinically inapparent focus of prostate carcinoma into a more aggressive lesion remain unquantified.

There are good data to suggest maintaining thyroid hormone levels at a lower 'normal range' is beneficial for older people,[39] and the same principle might conceivably also hold true for testosterone. For instance, it was shown that aggressive testosterone replacement in frail men aged 65 years and above was associated with a significantly increased rate of cardiovascular events.[25]

Therefore, GPs should recognize that many of the proposed beneficial effects ascribed to TRT in older men with putative LOH remain unproven and that the potential long-term risks remain uncharacterized at present. Additionally, there is quantifiable health burden and cost relating to testosterone administration and monitoring such as digital rectal examinations, prostate-specific antigen and haematocrit measurement.

The various guidelines and consensus statements currently available[40–43] are mostly based on expert committee reports, opinions or small, nonrandomized controlled studies, representing the very lowest category of the evidence-grading scale. They tend not to distinguish very clearly between organic/syndromic hypogonadism (for which there is evidence of benefit from TRT) and age-/comorbidity-related hypotestosteronaemia (FHH, for which there is generally no evidence). Proposed treatment thresholds are therefore arbitrary and necessarily vary between the different guidelines. For instance, the Endocrine Society Clinical Practice Guideline is unprescriptive in relation to testosterone treatment for older men with low serum testosterone concentration and to the exact treatment threshold.[40] However, it does caution against a general policy of offering testosterone therapy to all older men with low testosterone levels, with the decision to treat needing to be made by clinicians on an individual patient basis.

Although there is no UK national guideline for testosterone prescribing in men, the Society for Endocrinology has recently issued a valuable position statement on male hypogonadism and ageing.[43] This usefully states that, although TRT has been used effectively for many years in younger patients with classical hypogonadism without major adverse events, the same experience should not be extrapolated to the area of LOH, which is still lacking in longer-term studies of sufficient power to establish clinical outcomes.