Why Are Men's Testosterone Levels Declining? What the Research Shows

Men today have lower average testosterone than men of the same age a few decades ago, and aging does not explain the drop. In a Boston study of men aged 45 to 79, levels at any given age fell about 1.2% per year between 1987 and 2004 (Travison et al., JCEM 2007). Rising body weight accounts for part of the decline. The rest has no confirmed cause.

A population average falling is not the same as your own level being low. Low testosterone is diagnosed one man at a time, with symptoms and repeat morning blood tests.

What the studies found

The evidence comes from observational studies that compared men of the same age measured in different years. No randomized trial can test this question, so these studies show a trend, not its cause.

• In the United States, the Massachusetts Male Aging Study measured 1,532 randomly selected Boston-area men aged 45 to 79 in three waves from 1987 to 2004. The age-matched decline was 1.2% per year, and changes in smoking, obesity, and other health factors did not explain it (Travison et al., JCEM 2007).
• In Denmark, four population surveys from 1982 to 2001 (5,350 samples) found lower total testosterone in more recently born men of the same age. After adjusting for the rise in body mass index (BMI) over the same years, the drop was no longer significant, and free testosterone (the unbound fraction) showed no age-independent change (Andersson et al., JCEM 2007).

• In Finland, surveys in 1972, 1977, and 2002 covered 3,271 men aged 25 to 74. Among men aged 60 to 69, average testosterone fell from 21.9 nmol/L in those born 1913 to 1922 to 13.8 nmol/L in those born 1942 to 1951, and the drop held after adjusting for BMI (Perheentupa et al., Eur J Endocrinol 2013). US labs usually report ng/dL rather than nmol/L.
• In Israel, tests from 102,334 men in one health system between 2006 and 2019 showed an age-independent decline in most age groups that rising obesity was unlikely to explain (Chodick et al., Reprod Biol Endocrinol 2020). These were men whose doctors ordered a testosterone test, not a random sample.
• A second US study used national survey data on 4,045 adolescent and young adult men and found lower average testosterone in 2011 to 2016 than in 1999 to 2000. Higher BMI went with lower levels, but the trend held in men with normal BMI too (Lokeshwar et al., Eur Urol Focus 2021). The authors noted that lab methods changed over those years, which limits the comparison.

The most recent data in these studies end in 2019, and the Boston figure of 1.2% per year describes men aged 45 to 79 from 1987 to 2004, not all men up to today.

How is this different from normal aging?

Testosterone also falls as each man gets older. In the same Boston cohort, total testosterone was 0.8% lower for each year of age when men of different ages were compared at one time. Following the same men for 7 to 10 years, their levels fell 1.6% per year (Feldman et al., JCEM 2002).

The population decline is a separate effect layered on top of that. In the studies above, a 60-year-old measured in a later year had lower testosterone than a 60-year-old measured earlier, and each man still declined as he aged. For how levels change across the lifespan, see testosterone levels by age.

What explains the decline?

No single cause has been proven. The factors below have evidence linking them to lower testosterone in individual men; how much each contributes to the population trend is not known.

Body weight

Weight is the best-supported explanation for part of the drop. In the Danish data, the decline was no longer significant once BMI was taken into account, and in the US young-men data higher BMI went with lower testosterone (Lokeshwar et al., 2021). In the Boston, Finnish, and Israeli studies, the decline remained after accounting for weight.

Losing weight raises testosterone. A meta-analysis of 24 studies found that both low-calorie diets and bariatric surgery raised total testosterone, with surgery raising it more, and the men who lost the most weight gained the most. Weight loss also lowered estradiol and raised the pituitary hormones that signal the testes (Corona et al., Eur J Endocrinol 2013).

General health

Men in the Boston cohort with no chronic illness, no prescription medication, no obesity, and no heavy drinking had levels of several androgens 10% to 15% higher than other men (Feldman et al., 2002). That study did not separate out which condition mattered most.

Opioids

A meta-analysis of 17 studies found lower testosterone in men who used opioids regularly than in controls, whatever the opioid. The authors rated the overall evidence quality as low (Bawor et al., Drug Alcohol Depend 2015). If you take an opioid and have symptoms of low testosterone, raise it with the prescriber rather than stopping on your own.

Short sleep

In a small study, 10 healthy men with an average age of 24 slept 5 hours a night for one week. Their daytime testosterone was 10% to 15% lower than after full nights (Leproult and Van Cauter, JAMA 2011). The study was one week long and did not show whether longer sleep restriction has a larger or lasting effect.

Chemicals in plastics

Phthalates are chemicals used in plastics and personal care products. In US national survey data from 2011 to 2012, higher urine phthalate levels went with lower testosterone, but in adult men the link appeared only at ages 40 to 60 (Meeker and Ferguson, JCEM 2014). This was a single snapshot, so it cannot show that phthalates lower testosterone or that cutting exposure would raise it.

For the part of the decline that weight does not explain, the Finnish researchers wrote that their study "offers no explanation" (Perheentupa et al., 2013). None of the other studies above identifies a cause either.

Does a lower population average mean your level is low?

No. A diagnosis needs both symptoms and a confirmed low level. The American Urological Association uses total testosterone below 300 ng/dL as the cutoff and requires two early-morning tests on separate days (AUA guideline, statements 1 and 2). The Endocrine Society asks for symptoms or signs plus "unequivocally and consistently low" levels, confirmed with a repeat fasting morning test, and then a search for the cause (Endocrine Society 2018).

For reference, the harmonized normal range in healthy, nonobese men aged 19 to 39 is 264 to 916 ng/dL (Travison et al., JCEM 2017).

Common symptoms include low sex drive, low energy, and mood changes, but none of them is specific to testosterone, so a symptom checklist cannot settle the question.

What can you do about it?

Several of the factors above are within your control:

• Lose excess weight. In the Corona meta-analysis, more weight lost meant a larger rise in testosterone.
• Protect your sleep. One week of 5-hour nights lowered daytime testosterone in young men.
• Review your medications. Opioids are linked to lower testosterone, and any change belongs with the clinician who prescribed them.
• Get tested if you have symptoms, with two early-morning tests.

When does testosterone therapy make sense?

The Endocrine Society recommends testosterone therapy for men with symptoms and confirmed low levels, after a discussion of benefits and risks (Endocrine Society 2018). A falling population average is not by itself a reason to treat anyone.

Some men should not start testosterone:

• Men trying to conceive. "Exogenous testosterone therapy should not be prescribed to men who are currently trying to conceive" (AUA statement 23, Strong Recommendation, Grade A).
• Men with any of the conditions where the Endocrine Society advises against starting: breast or prostate cancer, elevated hematocrit (the share of blood made of red cells), untreated severe sleep apnea, uncontrolled heart failure, a heart attack or stroke in the last 6 months, or a clotting disorder (Endocrine Society 2018).

Low testosterone from age alone is a separate case: FDA labeling for testosterone products keeps a limitation of use for age-related low testosterone (FDA 2025).

On safety, the TRAVERSE trial gave testosterone gel or placebo to 5,246 men aged 45 to 80 with, or at high risk of, heart disease. Major cardiac events occurred in 7.0% on testosterone and 7.3% on placebo over about 3 years. That result means testosterone did not raise the risk; the trial did not show that it protects the heart. Atrial fibrillation, acute kidney injury, and pulmonary embolism were more common with testosterone (Lincoff et al., NEJM 2023). In 2025 the FDA added a warning about increased blood pressure to testosterone products (FDA 2025).

Common questions

Is the decline still happening?

The Israeli data run through 2019 and the US young-men data through 2016, and both still showed a decline (Chodick et al., 2020; Lokeshwar et al., 2021). None of the studies above covers the years since.

Has testosterone really dropped 1% a year since 1987?

The figure comes from the Boston study, which found 1.2% per year in men aged 45 to 79 between 1987 and 2004 (Travison et al., 2007). Other countries found declines over different periods, but the rate has not been measured for all men up to today.

Is everyone's testosterone lower now?

No. These are shifts in averages, and only testing shows where your own level sits.

Can lifestyle changes reverse it?

For the part caused by excess weight, losing weight raises testosterone (Corona et al., 2013). For the part no study has explained, there is no proven fix.

Bottom line

Men's average testosterone has fallen over recent decades in studies from the US, Denmark, Finland, and Israel, beyond what aging explains. Weight gain accounts for part of it, and the rest is unexplained. For any one man, what counts is his own level, and calling it low takes symptoms plus two early-morning tests below 300 ng/dL.

If testing confirms low testosterone, AlphaMD's testosterone replacement therapy page explains how evaluation and treatment work.