News · Nutrition & Diet
Coffee tracked higher total testosterone but lower free testosterone
A Finnish study of 2,264 people at age 46 found heavier coffee drinkers were leaner and had a distinct hormone pattern in men, with a catch buried in which testosterone rose.
Based on a peer-reviewed population cohort analysis in the European Journal of Nutrition
- Researchers analyzed 2,264 people (47% men) from the Northern Finland Birth Cohort 1966, all measured at age 46, publishing in the European Journal of Nutrition.
- Higher coffee intake groups had lower total and visceral fat and higher skeletal muscle mass, while BMI was comparable across groups.
- Coffee intake correlated inversely with circulating branched-chain amino acids in both sexes.
- In men, higher intake went with a more favorable glucose and insulin profile.
- In men, total testosterone rose with intake (beta = +0.29 nmol/L per cup/day), as did bioavailable testosterone and sex hormone-binding globulin, but free testosterone and the free androgen index went the other way.
- In women, only sex hormone-binding globulin rose, with free testosterone and the free androgen index falling.
- A single snapshot at one age. No one was followed over time, so the direction of any effect is unknown.
- Models adjusted for BMI, education, smoking, physical activity and alcohol, but leaner people may simply drink more coffee.
A study in the European Journal of Nutrition looked at what separates heavy coffee drinkers from light ones across body composition, blood sugar and sex hormones. The hormone result is the one that will get quoted, and it is the one most likely to be quoted wrongly.
The researchers investigated these associations in a large population-based cohort, with particular attention to sex-specific hormonal profiles and metabolomic signatures.
One birth year, one age
The sample is unusual and useful. The data came from 2,264 participants (47% men) of the Northern Finland Birth Cohort 1966 (NFBC1966) at age 46.
Everyone was born in the same region in the same year and measured at the same age. That removes age as a confounder entirely, which matters here because testosterone and its carrier protein both drift with age.
It also means this is a single snapshot. Nobody was followed forward, and nobody was asked to change anything.
The body composition finding
The first result is the cleanest, and it is not about hormones.
Higher coffee intake groups showed lower total and visceral fat and higher skeletal muscle mass, while BMI was comparable across groups.
That last clause is the interesting part. Body mass index did not differ between groups, but what the bodies were made of did. Same weight, less fat, more muscle. BMI is a blunt instrument and this is a neat demonstration of why.
The blood chemistry
Two metabolic signals pointed the same way.
Inverse correlations were observed between coffee intake and circulating branched-chain amino acids (BCAAs) in both sexes. Higher levels of those amino acids have been repeatedly tied to insulin resistance, so lower is generally the direction associated with better metabolic health.
And in men, higher coffee intake correlated with a more favorable glucose-insulin profile. Not in women, which is a pattern that recurs through this paper without an obvious explanation.
The testosterone result, read properly
Here is where care is required.
In men, the adjusted models revealed a distinct hormonal pattern: positive associations with total testosterone (beta = +0.29 nmol/L per cup/day), bioavailable testosterone, and sex hormone-binding globulin (SHBG), alongside inverse associations with free testosterone and the free androgen index (FAI).
Unpack that. Total testosterone went up. So did the protein that binds testosterone and keeps it out of action. And free testosterone, the fraction actually available to tissues, went down.
This is not a contradiction. It is arithmetic. If the carrier protein rises faster than the hormone does, the total climbs while the available share falls. Coffee did not straightforwardly raise or lower testosterone in this sample. It went with a shifted balance between bound and free.
In women, the picture was thinner. Hormonal associations were more limited, with positive associations for SHBG and inverse associations for free testosterone and FAI. The carrier protein moved; total testosterone did not.
What this cannot establish
Everything was measured at one moment, so nothing here shows direction. Coffee could shift the carrier protein. Or people who are leaner and more insulin-sensitive, which is itself associated with higher levels of that protein, could drink more coffee. A snapshot cannot separate the two.
The adjustments were reasonable, covering body mass index, educational attainment, smoking, physical activity and alcohol intake, and adjustment is never complete. Coffee drinking is bound up with sleep, work patterns, income and diet in ways no model fully captures.
The cohort is also one Finnish birth year. Finland has among the highest coffee consumption in the world, which is good for having enough heavy drinkers to compare, and poor for knowing whether the findings hold where a large cup is unusual.
The authors’ own framing
They do not overreach. Higher habitual coffee intake groups showed a leaner body composition, correlated inversely with circulating BCAAs in both sexes, and were independently associated with a sex-specific hormonal profile that may be metabolically relevant, particularly in men.
May be relevant. And then the request that most nutrition papers end with and most coverage ignores: future longitudinal and intervention studies are needed to establish causality and identify the coffee-derived bioactive compounds involved.
Which is to say nobody yet knows whether it is the caffeine, the chlorogenic acids, the diterpenes, or the kind of person who drinks four cups a day.
People also ask
Did coffee raise testosterone or not?
Both, depending on which testosterone you mean, and this is the whole story. Total testosterone counts every molecule in the blood, most of it bound to a carrier protein and biologically inactive. Free testosterone counts only what is unbound and available to tissues. In this study total went up and free went down, because the carrier protein went up faster. Anyone reporting this as a testosterone boost has read one number and stopped.
What is sex hormone-binding globulin?
A protein made in the liver that binds testosterone and estrogen in the bloodstream and holds them out of circulation. More of it means more hormone in transit but less hormone available to act. It rises with age, with liver health, with thyroid hormone and with lower insulin, which is one plausible route by which coffee could be involved here.
Were coffee drinkers just healthier people?
That is the obvious concern, and the models adjusted for BMI, education, smoking, physical activity and alcohol to try to address it. Adjustment helps but never fully settles the question. Coffee drinking clusters with a great many other habits, and a snapshot of 46-year-olds cannot tell you whether coffee produced the leaner body composition or the leaner people happened to drink more coffee.
What are branched-chain amino acids and why do they matter here?
Three amino acids, leucine, isoleucine and valine, that come from protein in food and are heavily marketed as supplements. Higher circulating levels have been repeatedly linked to insulin resistance and type 2 diabetes risk, which is why an inverse correlation with coffee is notable. It fits alongside the better glucose and insulin profile seen in men, though fitting together is not the same as explaining anything.
Should anyone change their coffee habit based on this?
This is general information rather than medical advice, and the answer is no. It is a snapshot of one Finnish birth year at one age, it cannot establish cause, and the hormone findings point in two directions at once. The authors themselves say longitudinal and intervention studies are needed. Anyone with a specific concern about testosterone or blood sugar should raise it with a clinician rather than adjusting their coffee intake.
References
- Verroest L, Jokelainen J, Choudhary S, et al. Associations of habitual coffee intake with testosterone and cardiometabolic markers: the Northern Finland birth cohort 1966 study. European Journal of Nutrition (2026).
- MedlinePlus. Testosterone Levels Test.
- National Institute of Diabetes and Digestive and Kidney Diseases. Insulin Resistance and Prediabetes.