Low Testosterone in Men: A Complete Evidence-Based Guide

Half.

That is the share of men with an initial low testosterone value who come back normal on a repeat draw without any treatment. The finding comes from a long-running study of more than 1,700 men in Boston who had their blood drawn at multiple visits over years. It is also consistent with what is known about how much testosterone varies day to day.

But you won’t find this in the marketing copy of the testosterone clinics that have nearly quadrupled their U.S. prescription volume over the past two decades. It does not appear in the supplement category that takes in billions of dollars a year on the premise that more testosterone is the answer to a man’s complaints. And you don’t see it reported on the automated results that direct-to-consumer lab companies send to their customers.

If a single low value normalizes half the time without any intervention, then the test result isn’t quite as cut and dried as most people think. 

Why is low testosterone both over- and under-diagnosed?

Testosterone prescriptions have quadrupled over the last 20 years though population-level testosterone has remained roughly the same.^1,2 Considering that a quarter of men starting Testosterone Replacement Therapy (TRT) have no testosterone lab drawn in the year before getting their prescription, you might think that there’s an overtreatment problem.¹ And you’d be right. 

The HIM study found that 38.7% of middle-aged men in U.S. primary care had a total testosterone below 300 ng/dL.¹² Tighter definitions that require both lab evidence and clinical symptoms yield prevalence figures in the low single digits, in the range of 2 to 6% depending on the study and the symptom criteria used.¹¹ The nearly six-fold spread between those two estimates is the space in which the current overprescribing happens.

But there’s also an undertreatment problem.

Men with classical or persistent symptomatic testosterone deficiency (“low T”) get turned away by primary care physicians uncomfortable with hormone therapy, dismissed for values technically inside the reference range, or scared off treatment by outdated cardiovascular concerns the field has since updated.¹⁸

Within the same study where 38.7% of men had low testosterone, only one in ten with low testosterone were receiving treatment.¹² Specific populations get missed routinely. Chronic opioid users suppress their own testosterone production at very high rates, and the population is rarely evaluated for it in primary care.¹⁷ After the 2013 to 2015 cardiovascular fear period, U.S. testosterone prescribing fell sharply, and the trial that has since substantially walked back the cardiovascular concern, TRAVERSE, did not arrive until 2023.¹⁸ A meaningful share of men who would have qualified for treatment under the existing guidelines spent that decade without it.

Both problems are real, and a careful evaluation should protect against both. A man who needs replacement and does not get it suffers in his function, his bones, his mood, and his metabolism. A man started on replacement without a diagnosis pays a different price. The diagnostic step is what separates the two, and the current system is short-circuiting it in both directions.

A quarter of new testosterone prescriptions are written without a blood test

A study looking at the U.S. insurance claims database covering 2001 through 2011 found that approximately 25% of new testosterone prescriptions were written without a testosterone measurement in the year prior.¹ A quarter of the time, the decision to commit a man to a hormone for a potentially indefinite course of treatment was being made without confirming the hormone was actually low. 

Follow-up work showed that even among men who did get tested, nearly half had no follow-up draw in the year after starting therapy to confirm the dose was bringing them into range.¹ Even among men who did get tested before starting, 19.5% did not meet Endocrine Society guidelines for low testosterone but were prescribed it anyway.¹

The 2001 to 2011 window predates the boom in direct-to-consumer wellness clinics. Testosterone prescriptions in the U.S. have exploded across that interval, and a growing share of the prescribing has moved to telehealth platforms that do not run their evaluations the way a traditional endocrinology clinic does. The current numbers are almost surely worse.

Do testosterone booster supplements actually work?

In a study reviewing the 50 most popular testosterone booster products on the U.S. market, 62% had no published data supporting the claims on the label. Of the products that did have data, 10% had evidence showing the active ingredients decreased testosterone.³ 

There is also a safety concern: contamination. An analysis of FDA enforcement records found that 12% of the muscle-building supplements that had drawn enforcement action contained undisclosed synthetic steroids.⁴ 

A man who takes a product labeled as a natural booster and ends up with an actual androgen in his system will suppress his own production. He will come off the product with lower testosterone than he started with.

Why is a single low testosterone test not a diagnosis?

The Massachusetts Male Aging Study followed more than 1,700 men in the Boston area over time, drawing blood at multiple visits. Among men who met criteria for testosterone deficiency at an initial visit, more than 50% had normal values on repeat testing without any treatment.⁶ 

The men who normalized tended to have mild deficiency at the first draw, with values near the threshold rather than deeply suppressed, and were leaner and healthier on average than the men who remained suppressed. Severe persistent hypogonadism rarely normalizes without treatment.

This is why the correct workup requires two morning draws, performed while fasted, and not during an acute illness.¹⁰ One bad night of sleep, a recent illness, a stressful week, a meal beforehand, or even an afternoon draw in place of a morning one can pull the number down enough to cross a diagnostic threshold. The clinical implication is that a single low value is never a diagnosis. 

What is testosterone, and what does the number actually mean?

Testosterone is a steroid hormone produced mainly in the Leydig cells of the testes, with small contributions from the adrenal glands. Its production is controlled by a feedback loop called the hypothalamic-pituitary-gonadal axis (HPG axis). The hypothalamus releases gonadotropin-releasing hormone (GnRH), which signals the pituitary to release luteinizing hormone (LH) and follicle-stimulating hormone (FSH). LH tells the Leydig cells to make testosterone. FSH supports sperm production. Testosterone itself feeds back on both the hypothalamus and the pituitary to regulate its own production. The pattern of values across the HPG axis tells you where a problem lives. A low testosterone with a high LH points to a testicular problem (primary testosterone deficiency). A low testosterone with a low or inappropriately normal LH points to a problem in the brain (secondary testosterone deficiency).

Free vs total testosterone: what your lab is actually reporting

Most of the testosterone in your bloodstream is bound to a protein called sex hormone-binding globulin, or SHBG, and to albumin. Only a small fraction circulates free. Total testosterone is the sum of all three fractions. 

The free and albumin-bound fractions together make up what is sometimes called “bioavailable testosterone”. Bioavailable testosterone is useful to know when SHBG is abnormal. For example, SHBG tends to run high in older lean men and low in men with obesity or insulin resistance.⁹ In a man with a normal SHBG, the total testosterone usually tracks with the bioavailable fraction, making a directly measured free testosterone less useful.

Why higher testosterone is not better

Androgen receptors are the cellular machinery that translates a testosterone level into a biological effect. They are present in finite numbers in each tissue, and once enough of them are bound by testosterone, the tissue is producing its full response. Adding more testosterone past that point does not increase the response. The receptors are already running at capacity.

The best evidence for this “Saturation Model” comes from the prostate, which appears to saturate at a testosterone concentration of roughly 250 ng/dL.²⁶ Below that threshold, prostate tissue responds to changes in testosterone concentration. Above it, additional testosterone produces no further prostate effect. This is the finding that has dismantled the decades-old fear that testosterone replacement causes prostate cancer in men with normal testosterone. At the levels most men sit at, prostate androgen receptors are already saturated, so adding more testosterone does not produce more growth.

The response to exercise tells the same story. Men and women carry a roughly fifteen to twenty-fold difference in circulating testosterone, but when they complete matched resistance training programs, they gain similar relative muscle and similar relative strength. If testosterone within the normal range were the rate-limiting factor for training-induced muscle gain, that finding would not exist. Muscle adapts to the mechanical signal of training; testosterone is permissive across most of its physiologic range rather than dose-dependent within it.²⁷ Other tissues have different thresholds, and the data is less clean. Skeletal muscle appears to operate on a broader plateau within the normal range; libido, mood, energy, bone, and red blood cell production each have their own kinetics, and the saturation framework for those endpoints is more inferred than directly measured.¹⁰

If your testosterone is in the middle of the normal range, pushing it to the top of the normal range will not deliver the energy, libido, or muscle gains the supplement and clinic marketing promise. The receptors that respond at your level are already responding. Pushing past the physiological range, which is what supraphysiologic dosing in wellness-clinic protocols actually does, adds risk faster than it adds benefit.

Why does the response to a given testosterone level vary so much?

The saturation model explains why trying to “optimize” testosterone levels within the normal range is rarely worth it, but it does not explain why two men at the same testosterone value can present with completely different symptom pictures. The best mechanistic answer for that variation is the androgen receptor itself. 

A polymorphism in the AR gene, the number of CAG trinucleotide repeats, varies between men. A shorter repeat sequence produces a more sensitive receptor; a longer sequence produces a less sensitive one.²⁸ Two men with identical total testosterone, identical SHBG, and identical free testosterone can sit on opposite sides of this polymorphism and have different lived experiences of the same hormonal level. 

Are testosterone levels really declining across generations?

You have seen the headline. Testosterone levels are falling. Every generation has less than the one before. There is usually a thumbnail attached that implies civilization is ending. The decline is real at a population level, but the magnitude of it has been overstated and the cause is not exactly a mystery. 

Why do people say testosterone levels are declining?

The original finding is from the Massachusetts Male Aging Study. Researchers found that men born about fifteen years apart had testosterone levels approximately 15 to 20% lower at the same age, even after adjusting for health factors including obesity.⁵ Similar patterns showed up in Finnish and Israeli cohorts. Three independent groups in three countries, using the same type of lab test, found the same thing. That is the origin story of the often-quoted “1% per year decline”.

How much are men’s testosterone levels actually decreasing?

A recent meta-analysis looked at testosterone levels reported in 1,256 studies and more than one million subjects across 53 years of data, from 1971 through 2024. After adjusting for age and BMI, the decline was approximately 0.56% per year, roughly half the magnitude of the commonly cited figure.⁷ The same analysis also found that LH declined in parallel with testosterone while FSH did not. That pattern points to the hypothalamus (brain) as the site of the change and away from the testes themselves.

When the authors broke the data down by testing method, however, studies that used the most accurate available method — liquid chromatography-tandem mass spectrometry (LC-MS/MS, often called “mass spec”) — showed no decline in testosterone levels. The authors attributed this to a shorter surveillance window for mass spectrometry, about thirty years of data versus fifty. Running the numbers though, if the true rate were 0.3% per year, a high-quality dataset covering more than 100,000 subjects across thirty years should detect it. The most accurate test, in the largest available sample, found nothing.

Why the testosterone decline figure is overstated

The headline figure is a mixture of three things: 1) an increase in body fat that BMI misses, 2) an error comparing two different types of testosterone tests across eras, and 3) perhaps a small biological decline.

BMI has problems. It does not distinguish muscle from fat or provide any information about where the fat is located. Roughly half of individuals classified as non-obese by BMI are obese by DXA-measured body fat.²⁹ As a result, most obesity guidelines recommend obtaining a waist circumference in addition to BMI. The waist circumference serves double duty: it identifies individuals who are carrying too much body fat that BMI misses, and it tells you how much of the fat is visceral fat that drives metabolic disease. 

Over the years, the waist circumference at any given BMI has gone up across birth cohorts. Men at a BMI of 27 today carry more visceral fat than men at the same BMI in 1985. All three of the major secular decline studies adjusted for BMI and reported the remaining decline as unexplained. They were all making the same measurement error.

The lab tests used have problems too: immunoassays do not measure testosterone directly. They infer the concentration by measuring how much a labeled decoy molecule gets displaced from an antibody. Other steroids, including estradiol, dihydrotestosterone, and dehydroepiandrosterone, displace the decoy. This is called cross-reactivity, and it means immunoassays systematically overestimate testosterone, especially at low concentrations.⁹ When researchers retested stored blood samples using the more accurate LC-MS/MS test, the proportion of samples falling below 300 ng/dL nearly doubled. The Endocrine Society subsequently revised the lower limit of normal from 300 ng/dL to 264 ng/dL.¹⁰ Comparing a test that routinely reads high (immunoassay) to a test that reads accurately, but lower, creates a decline that isn’t really there. 

Why the testosterone decline mostly disappears when you fix both errors

In 2012, researchers compared archived blood samples collected from American men in 1988-1991 to those collected in 1999-2004. They eliminated the testing-method problem by running every sample through the same laboratory on the same day using the same assay. They eliminated the BMI problem by adding waist circumference to their regression model. After adjusting for age, race, BMI, waist circumference, and lifestyle factors, the decline disappeared. There was no statistically significant change in total or free testosterone in a nationally representative U.S. sample.⁸ 

The practical implication is that the real population-level decline, once both errors are corrected, is at most very small. Whatever portion of the trend would require an environmental or generational cause beyond rising metabolic disease is below what current datasets can reliably detect.

Men who stay lean, active, and free of chronic illness maintain testosterone levels that approach those of younger men into their seventies.³⁰ There is a survivorship caveat: men we observe at 75 with normal testosterone are not a random sample of 75-year-olds, they are the ones who got there, and their trajectory does not by itself prove that any 50-year-old can replicate it. What the longitudinal data does support is the weaker but still useful claim: men whose body composition, sleep, and chronic disease burden stay favorable have a substantially shallower testosterone trajectory than men whose health declines.

Is the age-related testosterone decline inevitable?

Generational decline and age-related decline are different questions. Generational decline asks whether a 50-year-old in 2024 has lower testosterone than a 50-year-old in 1980. Age-related decline asks what happens to the same man as he moves from 30 to 50 to 70.

The Baltimore Longitudinal Study of Aging followed healthy men over decades and found a slow age-related decline in total testosterone of approximately 1 to 2% per year after age 30.³¹ This is not due to aging per se, as the vast majority of that decline tracks with body composition, chronic illness, and medication burden as men age. 

Active, healthy older men maintain testosterone values in the same range as middle-aged men.³⁰ The framing of an inevitable hormonal collapse is not supported by evidence.

So what actually drives low testosterone in a healthy adult man?

Most men who present with a low testosterone value in adult primary care or a wellness clinic have a signaling problem at the level of the brain rather than an issue with the testes. The good news is much of the signaling problems are driven by something modifiable. The bad news is, these common signaling problems are often undertreated, or managed incorrectly. 

Why visceral fat is the most common cause of low testosterone

In men, obesity more than doubles the odds of developing testosterone deficiency and the relationship is dose dependent. Overweight men show an average testosterone reduction of about 66 ng/dL, and men with obesity show an average reduction of about 147 ng/dL.²⁰ The clinical pattern is common enough to have a name: male obesity-related secondary hypogonadism or MOSH.²¹

There are multiple reinforcing mechanisms driving testosterone down as a result of excess body fat. Adipose tissue contains aromatase, an enzyme that converts testosterone into estradiol. About 80% of the estradiol circulating in men is produced this way in peripheral fat. The more visceral fat a man carries, the more of his testosterone gets pulled into estradiol production, and the accumulating estradiol then feeds back on the hypothalamus to lower the signal from the brain to the testes. The same fat tissue produces leptin and inflammatory adipokines (TNF-alpha, interleukin-6, and others), both of which appear to contribute independently to GnRH and Leydig-cell suppression, though the relative weight of each is still being worked out. Insulin resistance, which travels with visceral adiposity, lowers SHBG and disturbs the HPG axis through additional routes. A man with significant visceral adiposity is being suppressed at multiple points along the axis simultaneously.

But wait, there’s more. There is also a feedback loop. Low testosterone itself drives the accumulation of visceral fat, which then suppresses testosterone further. The system runs itself into the ground unless something interrupts it.

The good news is that the same loop runs in reverse. Weight loss reverses MOSH at a rate proportional to the visceral fat lost, and the mechanism by which the fat is lost (lifestyle, GLP-1 medication, bariatric surgery) appears to matter less than the fat loss itself.²⁰ The European Male Aging Study followed men prospectively and found that secondary testosterone deficiency is, in many cases, a transient state: a substantial proportion of men recovered normal testosterone, with the recovery driven primarily by improvements in body weight and chronic illness burden.²⁵ The diagnosis is not necessarily permanent if something is done about it.

How sleep deprivation lowers testosterone

Testosterone production depends on sleep duration more than it depends on the time of day. The hormone follows a daily rhythm, peaking in the early morning and falling through the afternoon, but the production itself is driven by sleep itself rather than by the clock. Testosterone begins rising at sleep onset and rises most steeply across the first three hours, and the morning value is more strongly predicted by measured sleep duration than by the precise time of the draw.^14,15

One study showed that just one week of sleep restriction (5 hours per night) dropped testosterone in healthy young men by 10 to 15%.¹⁴ Men whose sleep is chronically fragmented, whether by shift work, parenting, or untreated obstructive sleep apnea, often have low testosterone as a downstream consequence.

Sleep apnea: an under-diagnosed cause of low testosterone

Obstructive sleep apnea (OSA) is the under-recognized driver of low testosterone in men with obesity, loud snoring, morning headaches, treatment-resistant hypertension, or daytime sleepiness. It affects approximately 15 to 30% of men in North America and is substantially underdiagnosed.¹⁵ During an apneic event, the soft tissue of the upper airway collapses and seals the windpipe shut. Repeated cycles produce low blood oxygen levels (hypoxia), which causes the person to wake up briefly, disrupting and fragmenting their sleep. The fragmentation is what suppresses testosterone production: the sleep-driven signal that normally lifts testosterone across the night never gets to run uninterrupted. The hypothalamus lowers GnRH output, the pituitary lowers LH, and the testes go quiet. The low testosterone on the lab report is real, but its proximate cause is the untreated sleep apnea. 

Untreated moderate-to-severe OSA is an independent risk factor for hypertension, coronary artery disease, atrial fibrillation, congestive heart failure, and stroke. Exogenous testosterone can slightly relax upper airway musculature and worsen apnea events.¹⁰ A man started on TRT for low testosterone driven by undiagnosed apnea is having his actual problem made worse by the treatment of the apparent problem. The downstream cardiovascular and metabolic risk is the part of this picture that does not show up on the testosterone panel.

A reasonable screening tool is the STOP-BANG questionnaire. Score one point for each yes: 

• Do you Snore loudly?
• Do you feel Tired during the day? 
• Has anyone Observed you stop breathing? 
• Do you have or are you treated for high blood Pressure? 
• BMI greater than 35? 
• Age over 50? 
• Neck circumference over 40 cm? 
• Are you a man (Gender)? 

A score of three or more raises risk and warrants formal evaluation by in-lab polysomnography or a physician-interpreted home sleep test.¹⁶ Any man presenting with fatigue and a low testosterone value who has not been screened for sleep apnea has not yet been evaluated.

Other reversible causes of low testosterone

Beyond visceral adiposity and sleep apnea, several conditions suppress testosterone through the HPG axis and respond to treating the underlying problem rather than to exogenous hormone. They are easy to miss when the starting point is a wellness clinic drawing a number.

Chronic opioid therapy

Opioids inhibit GnRH and LH release at the hypothalamus and pituitary, producing a state called opioid-induced androgen deficiency, or OPIAD. The suppression is consistent across opioid classes, dose-dependent, and reverses on discontinuation.¹⁷ Any man on long-term opioid therapy with low testosterone should not be evaluated as if his lifestyle is the driver.

Hyperprolactinemia

Elevated prolactin from a pituitary adenoma or from a long list of medications (some antipsychotics, some antiemetics, some antidepressants) suppresses GnRH and produces a hypogonadotropic pattern with low testosterone, low LH, low FSH, and a high prolactin. The workup is a serum prolactin. The treatment is targeting the cause rather than starting exogenous testosterone.

Hemochromatosis

Iron overload from undiagnosed hemochromatosis deposits in both the pituitary and the testes, producing a mixed pattern with variable LH and FSH. Ferritin and transferrin saturation are the screening tests. The treatment is therapeutic phlebotomy, not hormones.

Other medical conditions

Many other medical conditions can suppress the HPG axis centrally and produce testosterone deficiency: untreated depression, advanced kidney or liver disease, uncontrolled type 2 diabetes, advanced HIV, and a small number of other systemic illnesses.

The fatigue and low libido that bring the man to the clinic can come from the underlying illness, the secondary testosterone suppression, or both. 

Endocrine-disrupting chemicals

Phthalates, bisphenol A, and several pesticide classes are structurally similar enough to endogenous hormones to interfere with hormone signaling in vitro and in animal models. The challenge with the human population data is dietary confounding: the same ultra-processed food and beverage environment that delivers most chemical exposure also delivers excess Calories and the downstream metabolic disease that suppresses testosterone directly. 

Whether endocrine-disrupting chemicals contribute independently to the population-level testosterone trend is an open question. It is reasonable to recommend limited intake of ultra-processed foods and other known exposures where the personal cost of doing so is low (food storage choices), without expecting that change alone to change testosterone levels.

How can you tell if your testosterone is actually low?

A diagnosis of testosterone deficiency (TD) requires both laboratory evidence and related symptoms. Neither of these are rigidly defined, as some symptoms are more specific for TD than others, and there’s no single testosterone level that earns the diagnosis. We have a longer self-assessment piece for men trying to decide whether the workup is warranted. The framework here is the abbreviated version.

What symptoms point to testosterone deficiency?

The European Male Aging Study clarified what actually distinguishes testosterone deficiency from everything else: of the long list of complaints men carry into clinics, only three sexual symptoms (decreased morning erections, low libido, and erectile dysfunction) cluster reliably with biochemical hypogonadism.¹¹ The other symptoms tested either did not correlate with testosterone levels or showed weaker, less specific relationships.

Fatigue, mood changes, and difficulty concentrating are common, non-specific, and overlap with sleep deprivation, depression, untreated metabolic disease, and many other conditions unrelated to testosterone, each requiring their own workup. If these conditions are overlooked, a man with severe fatigue caused by anemia may be inappropriately treated with testosterone, which can be a fatal miss if that anemia is caused by colon cancer. 

What sort of testing is required for low testosterone?

An individual whose clinical picture points towards low testosterone should initially have their total testosterone drawn alongside the additional workup for other potential causes of their symptoms. This lab test must be done fasted, first thing in the morning, ideally between 8 and 10 a.m., because of the diurnal rhythm. It should also be done when the individual is otherwise well (not sick) and well rested (no sleep deprivation). If the lab offers it, a mass spectrometry-based assay is preferred over the immunoassay. 

If this initial value is low and the other workup does not identify a more likely cause, a second lab test is warranted. This should be done in the same manner as the first, i.e. in the morning, fasted, no acute illness, and not after a night of poor sleep. At this point, however, additional labs are added on to not only confirm testosterone deficiency, but also to identify the location of the issue in the HPG axis. The additional labs include: SHBG, LH, and FSH at minimum. A calculated or directly measured free testosterone is useful when SHBG is abnormal.¹⁰

How to differentiate between primary and secondary testosterone deficiency?

Total testosterone tells you whether there is a problem. LH and FSH tell you where it lives. Three patterns matter.

Primary testosterone deficiency

Primary deficiency localizes the problem to the testes themselves. The laboratory pattern is low testosterone with high LH and FSH. The brain is sending a maximal signal and the testes are not responding. This pattern is more often genetic or structural, e.g. Klinefelter syndrome, prior chemotherapy, mumps orchitis, testicular trauma, or chronic alcohol exposure. Lifestyle change does not fix this category, and replacement therapy is more often the appropriate treatment when the diagnosis is correct.

Secondary testosterone deficiency

Secondary deficiency localizes the problem to the brain and is much more common than primary deficiency. On laboratory assessment, secondary deficiency shows low testosterone with low or inappropriately normal LH and FSH. The testes are intact and idle. If testosterone levels are low, the normal response is that the brain should be sending more and more LH and FSH. When these signaling hormones are low or inappropriately normal (they should be high), that means something is happening in the hypothalamus and/or the pituitary gland. 

This is the most common pattern in adult clinical practice and the category that most often responds to addressing the upstream driver, i.e. obesity, sleep apnea, opioids, hyperprolactinemia, severe systemic illness, or prior anabolic steroid use.¹³ This pattern, in a man without a structural problem, is the one most likely to respond to addressing the root cause rather than to exogenous hormone.

Mixed testosterone deficiency

Mixed testosterone deficiency often presents with low testosterone with variable or inconsistent LH and FSH. In this case, there are problems at the level of the brain and the testes, each failing simultaneously, usually in the setting of systemic disease such as hemochromatosis, advanced liver or kidney failure, or advanced HIV. These patients need the underlying disease managed first, with hormone replacement considered as a secondary question.

Locating the failure determines the treatment and is a necessary part of the workup. A clinic that draws one number and writes a prescription has skipped this entire step. The pattern across LH, FSH, and total testosterone is what tells a clinician where in the axis the problem lives, and whether or not prescribing testosterone is in the best interest of the patient. 

How can you increase testosterone levels naturally?

Testosterone levels are dynamic, fluctuating wildly hour-to-hour and day-to-day. Raising testosterone levels to a meaningful level is not a small ask. While plenty of supplements and strategies advertise a “boost” in testosterone, most fail to produce anything the person would notice. 

Four interventions you should already be doing have data to support them and a meaningful effect size in men whose testosterone is suppressed by modifiable drivers. Unfortunately for your favorite influencer’s bottom line, they are not sexy or proprietary.

Losing body fat raises testosterone

Reducing visceral adipose tissue is the single most effective non-pharmacological tool for raising testosterone in men whose suppression is metabolic. Interventional loss trials, including lifestyle programs, GLP-1 receptor agonists, and bariatric surgery, consistently show testosterone increases proportional to the loss of visceral fat.²⁰ On weight loss targets, more is generally better. A 10% reduction in body weight through diet and exercise typically yields an average increase of about 84 ng/dL, the difference between being deficient and returning to a healthy range for many men. Anti-obesity medications like GLP-1 receptor agonists push these results further, generally improving testosterone levels by 20% to 30% while addressing sexual function through improved metabolic and vascular health.²¹ The most dramatic shifts, however, are seen following bariatric surgery; with an average weight loss of 30%, patients often see their testosterone levels climb by approximately 250 ng/dL. ²⁰

Outside of losing weight, reducing the waist circumference is another important target. A waist-to-height ratio below 0.5 is a good target for men. When losing weight, a weight-to-waist reduction ratio of about 0.4 to 0.6 kg per centimeter, achieved over weeks to months in a program combining resistance training, adequate protein, and aerobic conditioning, is the pattern that produces meaningful testosterone recovery without excessive lean mass loss.

Improving sleep raises testosterone levels

Because testosterone is sleep dependent, we recommend aiming for seven to eight hours of uninterrupted sleep per night. To achieve this, having consistent timing in and out of bed and limiting both alcohol and sedating substances that fragment sleep architecture are good ideas. 

It is also important to screen for obstructive sleep apnea in any man with snoring, witnessed apneas, morning headaches, treatment-resistant hypertension, or a BMI above 30. Treating sleep apnea improves testosterone, reduces the cardiovascular and metabolic risk that travels with it, and frequently relieves the fatigue and mood symptoms that the man came in worried about in the first place. In a meaningful subset of men with low testosterone in this phenotype, the right intervention is a sleep study and CPAP rather than a hormone prescription.

Exercise raises testosterone levels indirectly

Exercise does not increase resting testosterone levels directly. To the extent exercise aids in reducing visceral fat however, exercise can indirectly raise testosterone levels. 

Much has been made of the post-workout hormonal environment: elevated testosterone, growth hormone, and more. However, these brief post-workout hormonal changes are part of a larger picture involving energy availability and a recent stressor. They are not anabolic or a worthwhile target to chase. Most are here and gone within half an hour. This is true of the post-exercise testosterone spike, which is small and does not correlate to (or drive) muscle growth or resting testosterone levels in any meaningful way. 

The long-term story runs on a different pathway. Resistance training improves insulin sensitivity, builds and maintains muscle mass, and supports body composition changes that reduce visceral adiposity. Similarly, aerobic exercise does an even better job at reducing visceral adipose tissue. By combining both types of exercise, an individual’s testosterone level is likely to go up in response. The lever is the body composition change. 

A reasonable starting point is two to four resistance training sessions per week of self-selected exercises that hit all the major muscle groups of the body, performed at moderate-to-high relative intensity, with progressive overload, supported by adequate protein (1.4 to 1.6 g/kg/day from predominantly lean, minimally processed sources), and three to four sessions of moderate aerobic activity. 

Treating underlying medical conditions can raise testosterone

An individual’s testosterone levels are an extension of their health status, with healthy individuals generally having higher levels than those with chronic medical conditions. By extension, making sure that any underlying medical condition is adequately treated can help raise testosterone levels indirectly. These conditions include major depression, hemochromatosis (iron overload), type 2 diabetes, substance use disorder, and more. 

These conditions are common contributors to low testosterone values in adult men presenting to primary care, and they are missed when the starting point is a wellness clinic drawing a number.

Why testosterone booster supplements are a bad place to spend money

No supplement reliably raises testosterone in well-nourished, healthy men. The exceptions to that statement are corrections of true deficiency. Severe zinc deficiency suppresses testosterone, and correction restores it.²² Vitamin D supplementation in men with confirmed deficiency produces small effects in some trials and none in others.²³ Ashwagandha appears to produce modest testosterone increases in stressed or hormonally suboptimal men, consistent with a cortisol-mediated mechanism, though the clinical significance remains understudied.²⁴ 

Outside of those narrow correction-of-deficiency cases, the category does not deliver. The common booster ingredients (D-aspartic acid, fenugreek, tongkat ali, tribulus) either fail in placebo-controlled trials or produce inconsistent small effects driven by publication bias and product heterogeneity. The dollars are better spent on the things that move the number at clinical scale: visceral fat loss, sleep, exercise, and addressing chronic medical conditions.

Signal: the full diagnostic and treatment framework

This guide is the short version of the case. Signal: The Complete Guide to Testosterone is the full system, including the history of how the field got here, the differential diagnosis, the evaluation protocols, the therapeutic options, and the case-based decision frameworks for men deciding whether to start, stop, or stay on testosterone replacement. Preorder coming soon! 

When is testosterone replacement therapy the right answer?

Testosterone replacement therapy (TRT) has a legitimate place in the treatment of testosterone deficiency. It also has a side-effect profile and requires appropriate monitoring. Whether TRT is the right answer depends on whether a diagnosis has been made appropriately and the individual’s preferences.

When is TRT appropriate?

Testosterone replacement therapy (TRT) is clearly indicated in men with primary testicular failure (Klinefelter syndrome, prior chemotherapy, severe testicular trauma, congenital absence of functional testes), men with confirmed central testosterone deficiency from structural pituitary disease, and men with confirmed, persistent testosterone deficiency in the setting of a syndromic presentation that has not responded to addressing the reversible drivers.¹⁰ 

In these men, the benefits of treatment can be substantial: improvements in sexual function, body composition, mood, anemia of chronic disease, and bone mineral density.¹⁹

Is testosterone safe for the heart?

The cardiovascular safety of testosterone replacement therapy dominated the field for more than a decade after observational signals raised concern. More recent data has emerged to assuage that concern.

The TRAVERSE trial was a randomized, placebo-controlled trial in 5,246 men aged 45 to 80 with testosterone deficiency and either preexisting cardiovascular disease or high cardiovascular risk. Subjects were treated with topical testosterone for an average of about two years.¹⁸ Major adverse cardiovascular events (cardiovascular death, non-fatal myocardial infarction, non-fatal stroke) occurred in 7.0% of the testosterone group and 7.3% of the placebo group. While the trial did not show a benefit on cardiovascular outcomes, it did not show the harm signal that had haunted the field. Secondary findings included higher rates of atrial fibrillation, acute kidney injury, and pulmonary embolism in the testosterone arm. Those imbalances were unadjusted secondary findings with multiple comparisons and no clear dose-response, so they should be flagged but not treated as established harms in the way the primary endpoint should be.

Even with that result, the TRAVERSE trial has limitations. The follow-up was approximately two years on average. Several of the harms TRT has been alleged to produce are decades-long: accumulated polycythemia (too many red blood cells) consequences, prostate cancer trajectories in older men, fertility recovery in men who try to come off years later. A two-year RCT cannot rule out long-tail harm, only establish that the medium-term cardiovascular risk is not significant. The trial also studied a specific population (men 45 to 80 with testosterone deficiency and high cardiovascular risk) using a specific topical formulation. Generalizing to younger, healthier men on injectable testosterone administered through a wellness clinic is outside the trial’s evidence base. 

Overall, cardiovascular safety at two years in the studied population looks reasonable, but long-term safety remains an open question that no current trial answers.

What are the side effects of TRT?

Exogenous testosterone suppresses the HPG axis, meaning that in nearly all men, it will reduce LH and FSH, suppress endogenous testosterone production, and suppress spermatogenesis. Fertility is reversibly suppressed in most cases, but recovery can take six to twelve months or longer after discontinuation, and a meaningful minority of men do not recover baseline sperm production. Any man considering TRT who desires children should treat fertility preservation as a primary part of the pre-treatment conversation, including sperm banking and the use of fertility-preserving alternatives discussed below.

Other expected effects include erythrocytosis (elevated red blood cell count), acne, fluid retention, possible worsening of obstructive sleep apnea, and gynecomastia in some men. Monitoring includes periodic measurements of total testosterone, hematocrit, prostate-specific antigen, and lipids at a minimum, as well as a clinical assessment of symptoms and side effects. The monitoring burden is real and indefinite for as long as therapy continues.

What about SERMs and HCG?

Selective estrogen receptor modulators (SERMs) and human chorionic gonadotropin (hCG) are sometimes used to treat testosterone deficiency.

SERMs (clomiphene, enclomiphene) work upstream of the testes by blocking the negative feedback of estradiol on the hypothalamus and pituitary. The brain raises its LH output, and the testes produce more testosterone endogenously. HCG acts directly on the LH receptor on Leydig cells, stimulating endogenous production from below the brain.

These options will not work in primary testicular failure, where the testes themselves cannot respond regardless of the signal. They also produce a more variable response and the long-term safety data is not as good when compared to TRT. These options remain off-label in the U.S. for testosterone deficiency. Still, both preserve fertility and are reasonable options to discuss in men with secondary hypogonadism who want to avoid the suppression that exogenous testosterone produces. 

When is TRT the wrong choice?

TRT in a man with borderline testosterone, non-specific symptoms, and unaddressed metabolic disease may make him feel somewhat better in the short run. Exogenous testosterone in supraphysiologic doses produces a euphoric and energizing effect that often masks the underlying problem. 

But TRT does not treat the visceral fat that suppressed his testosterone in the first place and it will commit him to lifelong therapy that suppresses his fertility and requires long-term monitoring. Starting TRT without an appropriate diagnosis and weighing the risks and benefits is the wrong decision most of the time, which is unfortunately common in the current age of TRT clinics and online pill mills.

Frequently asked questions

Do testosterone booster supplements work?

Not at the population level, and rarely at the individual level. The 2020 review of 50 popular products found that 62% had no published data supporting the claims on the label, and the products with positive data showed modest and inconsistent effects.³ 12% of muscle-building supplements associated with FDA warnings contained undisclosed synthetic steroids.⁴ If your goal is to raise testosterone, your dollars are better spent on the things that actually move the number: losing visceral fat, fixing sleep, exercise, and treating chronic medical conditions.

What is a normal testosterone level?

The Endocrine Society defines the lower limit of normal as 264 ng/dL measured by LC-MS/MS on a morning fasting sample.¹⁰ The previous threshold of 300 ng/dL was based on older immunoassay data that systematically overestimated true testosterone, especially at low concentrations. Reference ranges for free testosterone and SHBG depend on the laboratory and the assay. A number in the reference range does not rule out symptomatic testosterone deficiency, and a number below the reference range does not establish a diagnosis without symptoms and a confirmatory second draw.

Can you raise testosterone naturally?

In most men whose testosterone is low in the context of metabolic disease, sleep disorders, or other modifiable drivers, yes. Meaningful weight loss, improved sleep, treatment of obstructive sleep apnea, consistent exercise, and addressing reversible secondary causes can produce increases in total testosterone in the range of 50 to 150 ng/dL or more.²⁰ Men whose low testosterone is driven by primary testicular failure or structural pituitary disease will not respond to lifestyle changes, which is why a diagnostic workup matters before starting either lifestyle or pharmacologic treatment.

How long does it take to see changes to testosterone levels?

Sleep changes can affect testosterone within a week. Weight loss and training adaptations take longer. A reasonable interval to reassess after a coordinated lifestyle intervention is three to six months, with a repeat morning draw under the same conditions as the baseline. A response in the range of 50 to 150 ng/dL is typical when the intervention is meaningful. Men who do not respond have either not made a meaningful change in the upstream driver or have a different diagnosis than the one being addressed.

Is TRT dangerous?

TRT has a defined side-effect profile, e.g. erythrocytosis, fertility suppression, possible worsening of sleep apnea, acne, and gynecomastia in some men. A recent large trial, TRAVERSE, did not show an increased risk of major adverse cardiovascular events in men with testosterone deficiency and high cardiovascular risk over an average of about two years.¹⁸ TRT is relatively safe in a man with properly diagnosed testosterone deficiency and appropriate monitoring. Conversely, TRT in a man with an unconfirmed diagnosis and untreated metabolic disease is riskier. 

What is the best time of day to test testosterone?

Testosterone levels should be checked in the morning, before 10 a.m., fasted, on a day when you have not been acutely ill or sleep deprived. Testosterone follows a diurnal rhythm and can fall by 20 to 50% from morning to afternoon. A second confirmatory draw under the same conditions, along with active symptoms, is required for diagnosis of testosterone deficiency.

Does lifting weights raise testosterone?

Testosterone levels generally go up in the ~30 to 60 minutes following a workout, yes, and the increase does not matter. Multiple studies have shown that the magnitude of that acute rise is not predictive of muscle hypertrophy or strength gain. The supplement industry built a category around the post-workout spike. The category is selling something the body does anyway, and the spike is not doing what the marketing implies. The long-term effect of training on testosterone is real and runs almost entirely through body composition: leaner men with more skeletal muscle and better insulin sensitivity carry higher resting testosterone over years. This is mediated primarily by the reduction in visceral fat, as discussed earlier in this article.²⁷

Should I worry about the population-level decline in testosterone?

The population-level decline in testosterone, after correcting for measurement errors related to testing and BMI, is likely negligible. The largest studies on the topic using the highest-quality measurement method (LC-MS/MS) showed no statistically significant decline.^{7, 8} To be clear, the population-level decline in testosterone is almost entirely attributable to measurement error, not something in the water. Your trajectory is set by your metabolic health, your sleep, your training, and your underlying conditions. 

Where to go from here

A note on incentives before the recommendation. Barbell Medicine sells coaching, training programs, and our upcoming book, Signal. We benefit financially when men preorder the book or engage our services. The structure of that conflict is the same as the structure of the wellness clinics critiqued above; the difference is what we are selling. We sell education and behavior-change support; they sell prescriptions for products that suppress endogenous production. Read our recommendations with that asymmetry in mind.

If 50% of low testosterone levels normalize on their own, will mine? 

Return to the figure that opens this article. Half of men with an initial low testosterone value come back normal on a repeat draw without any treatment. By this point in the article, the reasons are plain to see rather than a mystery. Many of those men were measured at the wrong time of day, or after a stressful week, or during an illness they did not know they had. Many had a single immunoassay value that overstated true deficiency at the low end of the curve. Many had a reversible upstream driver (a few weeks of poor sleep, a recent weight gain, a course of opioids, a worsening apnea pattern) that resolved on its own. The shared feature is that the lab number was a snapshot of a system, and the system was telling a story that one draw could not capture.

The principle that ties the rest of the article together is the one that finding makes concrete. Testosterone is a downstream readout of your body’s health. It reflects sleep, body composition, chronic illness burden, medications, and the integrity of the HPG axis. Treat the upstream system and the readout improves on its own in most adult men. Replace the hormone without addressing the upstream system and the readout improves while the system underneath may or may not change. An appropriate diagnosis is what separates the man for whom replacement is the right answer from the man for whom it is the wrong one. The current system is not doing a good job, as it simultaneously overtreats individuals who haven’t been properly worked up, and undertreats those who never received the workup in the first place.

If you want the full story on testosterone, you are now able to preorder Signal: What Testosterone Levels Are Telling You About Your Health. If you want help addressing the modifiable factors, i.e. training, body composition, sleep, and the diagnostic conversation with your physician, the Barbell Medicine coaching team works with men through exactly this process.

References

1. Baillargeon J, Urban RJ, Ottenbacher KJ, Pierson KS, Goodwin JS. Trends in androgen prescribing in the United States, 2001-2011. JAMA Intern Med. 2013;173(15):1465-1466. https://doi.org/10.1001/jamainternmed.2013.6895

2. Rao PK, Boulet SL, Mehta A, et al. Trends in testosterone replacement therapy use from 2003 to 2013 among reproductive-age men in the United States. J Urol. 2017;197(4):1121-1126. https://doi.org/10.1016/j.juro.2016.10.063

3. Clemesha CG, Thaker H, Samplaski MK. ‘Testosterone boosting’ supplements composition and claims are not supported by the academic literature. World J Mens Health. 2020;38(1):115-122. https://doi.org/10.5534/wjmh.190043

4. Tucker J, Fischer T, Upjohn L, Mazzera D, Kumar M. Unapproved pharmaceutical ingredients included in dietary supplements associated with US Food and Drug Administration warnings. JAMA Netw Open. 2018;1(6):e183337. https://doi.org/10.1001/jamanetworkopen.2018.3337

5. Travison TG, Araujo AB, O’Donnell AB, Kupelian V, McKinlay JB. A population-level decline in serum testosterone levels in American men. J Clin Endocrinol Metab. 2007;92(1):196-202. https://doi.org/10.1210/jc.2006-1375

6. Travison TG, Shackelton R, Araujo AB, et al. The natural history of symptomatic androgen deficiency in men: onset, progression, and spontaneous remission. J Am Geriatr Soc. 2008;56(5):831-839. https://doi.org/10.1111/j.1532-5415.2008.01679.x

7. Santi D, Cignarelli A, Baldi M, et al. Secular trends in serum testosterone concentrations: a systematic review and meta-analysis of 1,256 studies published between 1971 and 2024. 2025. https://doi.org/10.1016/j.eclinm.2025.102871

8. Nyante SJ, Graubard BI, Li Y, et al. Trends in sex hormone concentrations in US males: 1988-1991 to 1999-2004. Int J Androl. 2012;35(3):456-466. https://doi.org/10.1111/j.1365-2605.2011.01230.x

9. Trost LW, Mulhall JP. Challenges in testosterone measurement, data interpretation, and methodological appraisal of interventional trials. J Sex Med. 2016;13(7):1029-1046. https://doi.org/10.1016/j.jsxm.2016.04.068

10. Bhasin S, Brito JP, Cunningham GR, et al. Testosterone therapy in men with hypogonadism: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. https://doi.org/10.1210/jc.2018-00229

11. Wu FCW, Tajar A, Beynon JM, et al. Identification of late-onset hypogonadism in middle-aged and elderly men. N Engl J Med. 2010;363(2):123-135. https://doi.org/10.1056/NEJMoa0911101

12. Mulligan T, Frick MF, Zuraw QC, Stemhagen A, McWhirter C. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract. 2006;60(7):762-769. https://doi.org/10.1111/j.1742-1241.2006.00992.x

13. Corona G, Goulis DG, Huhtaniemi I, et al. European Academy of Andrology and European Society of Endocrinology recommendations on the diagnosis and management of male hypogonadism. Andrology. 2020;8(5):970-987. https://doi.org/10.1111/andr.12770

14. Leproult R, Van Cauter E. Effect of 1 week of sleep restriction on testosterone levels in young healthy men. JAMA. 2011;305(21):2173-2174. https://doi.org/10.1001/jama.2011.710

15. Wittert G. The relationship between sleep disorders and testosterone in men. Asian J Androl. 2014;16(2):262-265. https://doi.org/10.4103/1008-682X.122586

16. Chung F, Abdullah HR, Liao P. STOP-Bang questionnaire: a practical approach to screen for obstructive sleep apnea. Chest. 2016;149(3):631-638. https://doi.org/10.1378/chest.15-0903

17. Reddy RG, Aung T, Karavitaki N, Wass JAH. Opioid induced hypogonadism. BMJ. 2010;341:c4462. https://doi.org/10.1136/bmj.c4462

18. Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular safety of testosterone-replacement therapy. N Engl J Med. 2023;389(2):107-117. https://doi.org/10.1056/NEJMoa2215025

19. Snyder PJ, Bhasin S, Cunningham GR, et al. Effects of testosterone treatment in older men. N Engl J Med. 2016;374(7):611-624. https://doi.org/10.1056/NEJMoa1506119

20. Corona G, Rastrelli G, Monami M, et al. Body weight loss reverts obesity-associated hypogonadotropic hypogonadism: a systematic review and meta-analysis. Eur J Endocrinol. 2013;168(6):829-843. https://doi.org/10.1530/EJE-12-0955

21.  Endocrine Society. (2025, July 14). Anti-obesity medications can normalize testosterone levels in men. https://www.endocrine.org/news-and-advocacy/news-room/endo-annual-meeting/endo-2025-press-releases/portillo-canales-press-release 

22. Prasad AS, Mantzoros CS, Beck FW, Hess JW, Brewer GJ. Zinc status and serum testosterone levels of healthy adults. Nutrition. 1996;12(5):344-348. https://doi.org/10.1016/S0899-9007(96)80058-X

23. Pilz S, Frisch S, Koertke H, et al. Effect of vitamin D supplementation on testosterone levels in men. Horm Metab Res. 2011;43(3):223-225. https://doi.org/10.1055/s-0030-1269854

24. Lopresti AL, Drummond PD, Smith SJ. A randomized, double-blind, placebo-controlled, crossover study examining the hormonal and vitality effects of ashwagandha (Withania somnifera) in aging, overweight males. Am J Mens Health. 2019;13(2):1557988319835985. https://doi.org/10.1177/1557988319835985

25. Rastrelli G, Carter EL, Ahern T, et al. Development of and recovery from secondary hypogonadism in aging men: prospective results from the EMAS. J Clin Endocrinol Metab. 2015;100(8):3172-3182. https://doi.org/10.1210/jc.2015-1571

26. Morgentaler A, Traish AM. Shifting the paradigm of testosterone and prostate cancer: the saturation model and the limits of androgen-dependent growth. Eur Urol. 2009;55(2):310-320. DOI: 10.1016/j.eururo.2008.09.024

27. Roberts BM, Nuckols G, Krieger JW. Sex Differences in Resistance Training: A Systematic Review and Meta-Analysis. J Strength Cond Res. 2020;34(5):1448-1460. DOI: 10.1519/JSC.0000000000003521

28. Zitzmann M. The role of the CAG repeat androgen receptor polymorphism in andrology. Front Horm Res. 2009;37:52-61. DOI: 10.1159/000175843 

29. Romero-Corral A, Somers VK, Sierra-Johnson J, et al. Accuracy of body mass index in diagnosing obesity in the adult general population. Int J Obes. 2008;32(6):959-966. DOI: 10.1038/ijo.2008.11

30. Sartorius G, Spasevska S, Idan A, et al. Serum testosterone, dihydrotestosterone and estradiol concentrations in older men self-reporting very good health: the healthy man study. Clin Endocrinol (Oxf). 2012;77(5):755-763. DOI: 10.1111/j.1365-2265.2012.04432.x

31. Harman SM, Metter EJ, Tobin JD, Pearson J, Blackman MR. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab. 2001;86(2):724-731. DOI: 10.1210/jcem.86.2.7219