Army Testosterone Screening: Study Warns of False Alarms
A JAMA study suggests mandatory Army testosterone screening could flag up to 25% of service members. See why experts oppose blanket testing and what follows.
Why mandatory testosterone screening could create more questions than answers
On paper, a screening programme sounds simple: test a big group, flag the people with low results, and send them for follow-up. In medicine, though, a screening result isn't the same thing as a diagnosis. When you test people without symptoms, even a small mismatch between the test threshold and the condition you're trying to catch can produce a lot of uncertainty - plus a heavy workload for the health system.
That's the concern raised by a new JAMA study about mandatory testosterone screening in the US Army. Researchers estimate that testing could produce a low testosterone result for as many as one in four active-duty soldiers, even though the true rate of clinically significant testosterone deficiency is probably much lower. The reason this matters isn't just the blood test itself. A low result can set off repeat tests, consultations, diagnoses, and treatment - each step with its own costs and risks.
This article breaks down what the study measured, what it can and can't show, why major medical groups have pushed back on blanket testing, and what a more clinically grounded screening policy would need to get right.
What the Army testosterone study found
The study analyzed testosterone measurements from male, active-duty Army service members who took part in a larger military health survey. Researchers at the Pennington Biomedical Research Center and the US Army Research Institute of Environmental Medicine used data collected at five US Army installations.
Blood was drawn in the morning while participants were fasting. Samples were collected between August 2024 and January 2026. Researchers measured total testosterone in 479 men, excluding three men who were already receiving testosterone treatment.
The analysis was weighted statistically to represent active-duty Army personnel. The weighted average age was about 30, and the weighted average body mass index was 28.4 - within the overweight range. The study's headline point is that using a low-testosterone screening threshold across the force could flag up to 25 percent of soldiers tested.
That number needs context. It's an estimate of how many people might receive a low result under a screening approach. It doesn't mean one in four soldiers has a confirmed medical condition, needs treatment, or would benefit from testosterone therapy. The gap between a screening result and a clinical diagnosis is the whole issue here.
What the study can tell us
The study gives a practical sense of the follow-up volume that broad testing could create. It also anchors the policy debate in real measurements from real people, rather than speculation.
But it shouldn't be read as a count of confirmed cases. The researchers measured testosterone in a sample and used statistical weighting to estimate how common low readings might be. Based on what's reported, the study doesn't show that people with low results have persistent deficiency or symptoms. It also doesn't track what happens later under the Pentagon's broader screening policy, which was announced separately.
A screening programme has multiple steps: who gets tested, what threshold triggers attention, whether results are repeated, how clinicians assess symptoms and other causes, and what treatment - if any - is offered. A study of test results can inform the early steps. It can't replace evidence about the entire clinical pathway.
Why the timing matters
Defense Secretary Pete Hegseth announced mandatory testosterone screening in July. The Pentagon later released clinical guidance in September, and the JAMA study adds evidence about how many low results broad testing might generate.
That creates an operational question as much as a medical one: can the military health system reliably tell the difference between results that warrant clinical investigation and results that are temporary, unrelated to symptoms, or not clinically meaningful? And if a lot of people are sent into follow-up, the quality and capacity of that follow-up becomes part of the policy itself.
The source report and discussion are available from Ars Technica.
Why a low test result is not a diagnosis
Testosterone levels aren't a single fixed number for everyone forever. Results can vary with timing and with day-to-day health or circumstances. That's part of why the study's use of morning blood draws matters: when you compare results to a threshold, the conditions of the test matter.
Even so, one low reading doesn't prove a person has a persistent condition. A clinician has to interpret the result in context, including symptoms and the person's medical history. A low result in someone who feels well isn't the same as a diagnosis in someone who has compatible symptoms and repeat findings.
This is a common feature of screening, not something unique to testosterone. Screening tests are meant to identify people who might need further assessment. They aren't built to "settle" every person's diagnosis on their own. The more widely you apply a test, the more important it is to explain that distinction clearly and to run follow-up in a consistent, evidence-based way.
Prevalence changes the meaning of a test
If a condition is uncommon in the group you're testing, broad screening can generate many low readings that still don't turn into a confirmed diagnosis. That can happen even when the test is reasonably useful for people who already have symptoms or other reasons to be evaluated.
A big part of the issue is who gets included. Testing someone because they have symptoms gives clinicians a reason to investigate. Testing everyone - regardless of symptoms - changes the balance: a lot more people are tested, including those whose chance of a clinically significant condition is low. Some of them will fall below a threshold, but that alone doesn't decide whether treatment makes sense.
So, the study's estimate is best treated as a warning about potential screening volume - not as evidence that the Army has uncovered a widespread unmet need. That difference should shape how leaders communicate the results, plan staffing, and decide whether blanket testing is justified.
Thresholds involve trade-offs
A threshold defines what counts as "low," but setting it isn't purely technical. A more inclusive threshold can identify more people for review, and it can also increase the number of repeat tests and further assessments. A more restrictive threshold can reduce follow-up - while risking missed cases where clinical evaluation would have helped.
The trade-off is easier to manage when the programme is designed around a clear clinical purpose. It's harder when you include people without symptoms and the follow-up pathway isn't well defined. In that situation, the threshold can end up generating medical work rather than finding people who truly need care.
The practical risks of screening everyone
Medical experts - including the Endocrine Society and the American Urological Association - have recommended against blanket testosterone screening. Their concern isn't that testosterone testing has no legitimate role. It's that testing men without symptoms can trigger a chain of evaluations and treatments without a clear chance of benefit.
And the chain matters. A low result can lead to repeat blood tests, appointments, additional investigations, and ultimately a diagnostic label. Each step uses patient time and clinical capacity. If the number of people flagged is large, the programme can also crowd out timely care for service members who already have symptoms or other urgent needs.
Follow-up is part of the intervention
It's easy to treat the blood test as the whole story. In practice, the policy includes everything that happens after the sample is drawn. A programme has to spell out who reviews results, how clinicians decide when to repeat a test, what symptoms or history are relevant, and when treatment is actually considered.
Without that, low results can get handled inconsistently across clinics. One clinician may pursue a full workup; another may decide the same result is insufficient without symptoms or confirmation. Inconsistency is what you'd expect when a screening initiative grows faster than the clinical process designed to handle the results.
The Army's stated clinical guidance is therefore important - but guidance isn't the same as evidence that the programme improves health outcomes. The key question isn't just how many low results the screening finds. It's whether the process identifies people who need care, provides appropriate follow-up, avoids preventable harm, and uses clinical resources effectively.
Treatment can have consequences
The source report notes that unnecessary evaluation can contribute to inaccurate results and diagnoses, overtreatment, and harm - including infertility. That's a major reason not to treat testosterone therapy as a harmless response to a single number.
Treatment decisions should come from a proper clinical assessment, not an automatic pipeline from screening result to therapy. If a programme creates pressure to act on borderline or unconfirmed results, the risk isn't only extra testing. It's that people may end up on treatment without a solid clinical basis - and then face consequences they wouldn't have needed to deal with.
There's also a trust issue. Service members need to understand why the test is being done, what a result can and can't mean, and whether follow-up is voluntary or tied to readiness decisions. If those explanations are unclear, routine screening can feel more like surveillance than care. That perception can shape how openly people talk about symptoms and whether they seek help later.
What a responsible screening policy should consider
A screening policy is easier to judge when you treat it like a complete system, not just a test-ordering decision. The system has to connect clinical rationale, test performance, follow-up capacity, and patient outcomes. If one part is missing, the apparent simplicity of "test everyone" can hide operational and medical costs.
Define the clinical purpose
Leaders should be clear about what problem the programme is trying to solve. Is it aimed at identifying service members who have symptoms but haven't sought care? Improving diagnosis for a particular group? Or measuring testosterone for some other operational reason?
Those objectives aren't interchangeable, and they require different evidence.
If the goal is to improve care for symptomatic patients, it should be easier to access evaluation - not automatically a reason to test people without symptoms. The purpose needs to match both the population and the intervention.
Set the follow-up pathway before expanding testing
Before scaling a screening programme, the health system should be able to answer practical questions:
- What happens after an initial low result?
- When is repeat testing appropriate, and under what conditions?
- How are symptoms and medical history used in the assessment?
- Who is responsible for explaining results and coordinating follow-up?
- What safeguards stop a screening result from turning into an automatic treatment decision?
- How will the programme track unnecessary tests, diagnoses, treatment, and adverse outcomes?
These aren't details to "figure out later." They determine whether screening leads to useful care or a backlog of uncertain cases.
Measure outcomes, not just positive screens
A programme can report high numbers for "tested" and "low results" and still fail to improve health. Evaluation has to follow the full pathway: how many low results get confirmed, how many people actually have relevant symptoms, how many receive treatment, and whether that treatment is appropriate and beneficial.
It should also track costs that are easy to miss: repeat appointments, additional investigations, time away from duties, clinical workload, and adverse effects. Those measures help distinguish a programme that truly detects problems from one that generates activity without clear benefit.
The JAMA study is a prompt to ask these questions before screening becomes routine. Its estimate suggests low results could be common at scale. The next decision should depend on whether the clinical pathway can produce more benefit than harm - not on the idea that a bigger testing number automatically means better care.
Conclusion
The Army screening debate isn't only about whether testosterone can be measured. It's about what happens when you apply a test to a large population - including people without symptoms - and how the health system responds to the results.
The JAMA study estimates that a broad programme could flag up to a quarter of active-duty soldiers with a low result. That's a reason to scrutinize the follow-up process, not a reason to treat low readings as confirmed disease or automatic treatment need. A responsible policy should define its clinical purpose, require careful assessment, and measure patient outcomes - not just screening volume.
If you're reviewing a screening policy in your own organisation, start with what comes after the test: who is likely to be flagged, what happens next, and what evidence will show people are better off. That's where the real value - or the avoidable cost - shows up.