Testosterone Use in Men
March 2014
Volume 5, Issue 1

Many Men Appear to Start Testosterone Therapy Without Being Properly Tested

J Clin Endocrinol Metab. 2014;Jan 1:jc20133570 [Epub ahead of print]

Introduction: Current clinical guidelines from The Endocrine Society recommend use of testosterone supplementation in patients with low testosterone levels who are symptomatic and have low levels confirmed by repeated laboratory tests. The guidelines discourage initiation of treatment based on one low testosterone measurement. This study was designed to compare trends in testosterone testing and supplementation in the United States (US) and the United Kingdom (UK).

Methods: The retrospective, cohort study analyzed commercial and Medicare insurance claims from the US and general practitioner health-care records from the UK between 2000 and 2011. The study identified 410,019 men in the US and 6,858 men in the UK who began taking testosterone during this period. The analysis also identified 1,114,329 men in the US and 66,140 men in the UK having new testosterone laboratory measurements taken during this time.

Results: Testosterone testing rates increased over the study period in both countries—from 13.0 to 46.4 per 10,000 person years in the UK and from 39.6 to 170.0 per 10,000 person years in the US.

The yearly initiation rate of testosterone therapy use increased dramatically in the United States from 20.2 to 75.7 per 10,000 person years; the rate spiked the most in 2008. The rate was markedly lower in the UK baseline and with little change over the study period—from 3.4 to 4.5 per 10,000 person years.

In the US, 40.2% of men did not have a testosterone test in the 180 days before beginning therapy and 50% had only one test. In the UK these rates were 53.8% and 32.7%, respectively.

In both countries, transdermal gels became the most commonly used initial choice of treatment formulation over injections and patches over the study period.

Conclusion: Testosterone testing and use has increased over the past decade, with a markedly greater increase in the US compared with the UK. High rates of inadequate laboratory testing prior to initiating therapy was found in both countries.


Tamara L. Wexler, MD, PhD, is an endocrinologist specializing in reproductive and neuroendocrinology, and Attending in Medicine, Massachusetts General Hospital, Boston, MA.

In this study, Layton and his colleagues describe the increase in testosterone testing and use of replacement in adults in the US and UK. Of note, the study is purely descriptive, using information from health databases to identify large populations. The authors identified two large cohorts: one identified men who had a testosterone measurement, and investigated the characteristics of those tested and the frequency of testosterone initiation in subsequent 90 days; the second identified men who initiated testosterone treatment, and looked at the rate of testing conducted in the preceding 180 days. As all data was based on coding, any incomplete or inaccurate coding, or differences in coding practice between the US and UK, obfuscate the results.

During the time period studied (2000 to 2011), testosterone testing rates increased, most markedly in the US, where both a higher baseline rate and a steeper increase in testing were seen—nearly 4-fold in the US, compared to an increase of approximately one-third in the UK. A dramatic increase in slope for both testing and testosterone use was observed in approximately 2008 in US, while rates were far more steady in the UK.

The most striking finding in this study is the dearth of testosterone testing in the 180 days prior to initiation of testosterone therapy. In the UK, 53.8% of men starting testosterone had no total testosterone level measured in the preceding 180 days, and 32.7% had only one test in the database. In the US cohort, 40.2% had no testosterone level measured, and 50% had only one test—ie, <10% had repeat testosterone tests in the 180 days prior to testosterone initiation.

The suggestion that testosterone is being prescribed to patients who do not meet criteria for low testosterone levels and, particularly, without any testosterone measurements in evidence, is concerning. Even accounting for a degree of testing not captured by study due to limitations of design or database information, the results point to a large number of men who are being started on testosterone therapy without appropriate supporting evidence. Particularly given the uncertainty of risk and benefit in testosterone use, we should not initiate testing without knowing a patient’s testosterone level.

Next Article:
The Role of Testosterone and Estradiol in Changes Attributed to Male Hypogonadism