A team at Michigan Medicine pulled the charts of 200 men, mean age 52, who had been given a hypogonadism diagnosis and started on testosterone between 2020 and 2025. The question they asked was the simplest one a chart audit can ask: how many of those men got the diagnostic workup their own specialty’s guideline requires before the prescription pad came out? The answer, presented this week at ENDO 2026 in Chicago and echoed by News Medical, was 12 percent.
To see why that number is not a surprise, you have to remember how the testosterone market got here. Twenty years ago “Low T” was a marketing slogan looking for a disease. AbbVie’s AndroGel commercials handed middle-aged men a pamphlet’s worth of vague symptoms (low energy, low libido, low mood) and a clinic to call, and U.S. testosterone prescriptions roughly quadrupled across the following decade. The FDA finally responded in March 2015 with required labeling changes warning of possible cardiovascular risk. Cardiology then spent the better part of ten years arguing about whether testosterone gel was quietly killing people. The 2023 TRAVERSE trial answered the cardiac question (mostly no, for men with documented hypogonadism), and the market took it as the all-clear.
The workup the guideline asks for is not exotic. The 2018 Endocrine Society clinical practice guideline tells clinicians to diagnose hypogonadism only in men with symptoms plus two unequivocally low morning fasting total testosterone levels, drawn on separate days, plus measurements of LH and FSH to figure out whether the problem is in the testicles or upstream in the brain. Two morning draws matter because testosterone is a peak-in-the-morning hormone that swings substantially across the day, drops during acute illness or sleep deprivation, and is suppressed by obesity and opioids. One low afternoon reading from a tired 50-year-old is not a diagnosis. It is a coin flip. The LH/FSH step matters because the right downstream care for a man whose testes have failed (primary hypogonadism) is not the same as for a man whose pituitary is the problem (secondary hypogonadism), and skipping that step can mean missing things you genuinely want to find, including pituitary lesions and hemochromatosis.
Most of these 200 men got none of that. Some got a single low reading and a prescription. Plenty did not have the basic safety labs in the year before they started: 38 percent had no pre-treatment PSA on the chart, and 23 percent had no complete blood count, which is the test you use to catch the predictable polycythemia (thickened blood) that testosterone can drive into stroke territory. Roughly two-thirds of the prescriptions were topical gels, the formulation most aggressively marketed since the AndroGel era.
Then the contraindication numbers, which are the part of this paper that should make a reader stop. The Endocrine Society guideline and the American Urological Association’s parallel document list specific conditions that contraindicate testosterone therapy or require careful evaluation first: untreated severe obstructive sleep apnea, a history of prostate or breast cancer, a PSA above 4 ng/mL without urologic workup, a baseline hematocrit above 50 percent. In the Michigan cohort, 55 percent of the men starting testosterone had obstructive sleep apnea on the problem list, 4 percent had a prior prostate cancer diagnosis, and 1.5 percent had a pre-treatment PSA above 4 ng/mL. The conference coverage did not break out OSA severity or treatment status, and the prior-cancer cases were not all necessarily active, but these are the patient profiles the guideline specifically tells you to slow down with, and the Michigan team’s read is that the slowdown was not consistently happening.
Specialty did not save anyone. Primary care wrote 45 percent of these prescriptions, urologists 35.5 percent, and endocrinologists themselves 18 percent. You can read that two ways. The generous read is that primary care is overstretched and TRT belongs with subspecialists. The honest read is that even the endocrinologists, the doctors whose own society wrote the guideline, were not consistently following it.
What the field will say this week is that prescribers need better point-of-care decision support, an EHR nudge, a checklist. That is true and small. It locates the problem in physician forgetfulness, which is the comfortable answer, rather than in the system the prescriptions actually flow through now. In 2024 the United States wrote roughly 11 million testosterone prescriptions, with the sharpest growth among men aged 35 to 44 and direct-to-consumer telehealth spending exceeding $400 million. The 35-to-44 age band sits well below the ages where age-related decline is even a credible explanation. A growing share of those prescriptions are initiated by nurse practitioners and physician assistants on telehealth platforms whose business model is the recurring monthly cartridge, not the second 8 a.m. lab draw two weeks later. A platform that books a single screening visit and ships gel from a mail-order pharmacy is not negligently skipping the second morning draw. It is selling a product whose unit economics depend on not having a second draw rule out the diagnosis.
It is also fair to ask why so many men are walking in the door for this in the first place. Population testosterone levels in American men have been drifting down for decades, faster than aging alone explains. Obesity, type 2 diabetes, opioid exposure, chronic sleep deprivation, and untreated sleep apnea are all well-documented suppressors of endogenous testosterone, and the Michigan cohort was thick with exactly those conditions: 63 percent obese, 55 percent with sleep apnea, 28 percent diabetic, 40 percent with depression. For most of those men the guideline’s preferred first move is to treat the reversible cause: the apnea, the metabolic disease, the sleep. The body knows how to make testosterone when you stop standing on the hose. Mainstream medicine has not solved the underlying epidemic of cardiometabolic disease that is driving a lot of these symptoms, and the gap has been filled by a subscription industry that treats a number rather than the man holding it.
None of this means TRT is fraudulent. For a man with a real pituitary lesion, a real testicular injury, or properly confirmed primary hypogonadism, testosterone replacement is good medicine, and the TRAVERSE evidence (5,204 men randomized, 7.0 percent cardiovascular events on testosterone versus 7.3 percent on placebo over roughly three years) is reassuring on the cardiac question that haunted the field for a decade. TRAVERSE also flagged signals the field had not seen this clearly before: higher rates of atrial fibrillation (3.5 versus 2.4 percent), pulmonary embolism (0.9 versus 0.5 percent), and acute kidney injury (2.3 versus 1.5 percent) on testosterone. Those are the harms you most want to catch upstream by, for instance, screening for sleep apnea and checking a hematocrit before you start the gel. Which the Michigan study tells you a quarter of these patients did not.
| Event | Testosterone | Placebo |
|---|---|---|
| Cardiovascular events | 7.0% | 7.3% |
| Atrial fibrillation | 3.5% | 2.4% |
| Pulmonary embolism | 0.9% | 0.5% |
| Acute kidney injury | 2.3% | 1.5% |
What would actually close the loop is not a checklist. Payers, including telehealth platforms, should not pay for a new testosterone prescription without two morning total-T values on file, the LH and FSH measurements, and a documented review of the contraindication list. The Endocrine Society wrote that rule. The AUA seconded it. The FDA holds labeling authority over every testosterone product on the market. None of them have produced the operational version, the one the subscription industry actually has to obey. Writing it would mean naming the business model that has grown up in the gap between guideline and prescription, and that is the part the field has so far refused to do.
Sources
- MedPage Today – Not Enough Men On Testosterone Get Guideline-Based Hypogonadism Work-Up (ENDO 2026 coverage)
- News Medical – Study finds gaps in guideline-based testosterone prescribing practices
- Bhasin et al., 2018 – Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline (JCEM)
- American Urological Association – Testosterone Deficiency Guideline
- Lincoff et al., 2023 – TRAVERSE trial, cardiovascular safety of testosterone therapy (PubMed listing)
- Layton et al., 2017 – Marketing and Testosterone Treatment in the USA: A Systematic Review (PubMed)
- U.S. Pharmacist – Nationwide Patterns in Testosterone Replacement Therapy