§ Claim under review · Health
"People who drank more coffee had less body fat, less visceral fat, and more lean mass than less-frequent or non-drinkers. The biggest benefit was seen at 5 or more cups per day. And for men, every extra cup per day was associated with a ~10 ng/dL higher total testosterone. Higher coffee intake was also associated with better glucose and insulin regulation in men."
Verdict
Partially accurate but misleading
Confidence
HighSummary
This post is based on a real and recent peer-reviewed study of 2,264 Finnish adults, all aged 46, published in the European Journal of Nutrition. The study did find that people reporting higher coffee intake had less total and visceral fat and more skeletal muscle mass despite similar BMI, and that men with higher intake had better fasting glucose and insulin measures. However, the post leaves out half of the hormone result. While total and bioavailable testosterone were higher per cup, free testosterone and the free androgen index were actually lower, which complicates the simple "more coffee, more testosterone" reading. The stated figure of about 10 ng/dL per cup is also somewhat inflated; the published coefficient converts to roughly 8 ng/dL. The study authors explicitly warned that these are population-level patterns, not evidence that drinking more coffee will change an individual's hormones, and that reverse causation and confounding cannot be ruled out. The post's added claim that experimental evidence supports coffee's hormonal benefits is weak: the main randomized trial on coffee and sex hormones found no significant differences after eight weeks.
The readings
key figures from the evidencetotal testosterone increase per cup/day (published coefficient)
converted testosterone increase per cup/day, vs claimed ~10
total cohort sample size analyzed
Why this verdict
Evidence
The study exists and the core body composition finding is reported accurately. Higher coffee intake groups showed lower total and visceral fat and higher skeletal muscle mass, while BMI was comparable across groups, and inverse correlations were observed between coffee intake and circulating branched-chain amino acids in both sexes . In men, higher coffee intake correlated with a more favourable glucose-insulin profile.
On testosterone, the picture is more mixed than the post presents. Fully adjusted regression models showed that each additional cup of coffee per day was associated with higher total testosterone (β = +0.29 nmol/L), bioavailable testosterone (β = +0.08 nmol/L), and SHBG (β = +0.57 nmol/L), but lower FAI (β = −0.36) and free testosterone (β = −0.01 nmol/L), after full adjustment. The university's own release states the same two-sided result: higher coffee consumption was linked to a more favourable glucose-insulin profile, higher total and bioavailable testosterone and increased SHBG, while at the same time free testosterone and the free androgen index were modestly lower.
Converting the reported coefficient: 0.29 nmol/L equals approximately 8.4 ng/dL, not ~10 ng/dL.
Findings
✓ What's accurate 5
- The study is real, peer-reviewed, and recent, and the body composition direction is correctly reported: less total fat, less visceral fat, more skeletal muscle mass at higher intake, with BMI similar across groups.
- ≥5 cups per day was genuinely the highest-intake category and showed the most favorable body composition profile in group comparisons.
- Total testosterone in men was positively associated with intake per cup per day, in fully adjusted models.
- The glucose/insulin claim for men is supported: the paper reports a more favorable glucose-insulin profile in men with higher intake.
- The poster correctly labels it observational.
≈ What's misleading 6
- Omitted qualifier (most significant): the post reports only the testosterone measures that went up. Free testosterone and the free androgen index went DOWN per additional cup. The paper's hormonal signal is a mixed pattern driven substantially by higher SHBG, not a clean "coffee raises testosterone" result. A clinician-facing appraisal notes that the simultaneous inverse associations with free androgen measures make a simple testosterone-mediated explanation for greater muscle mass difficult to support .
- Exaggeration (minor, quantitative): the paper's coefficient of +0.29 nmol/L per cup/day converts to roughly 8.4 ng/dL, not "~10 ng/dL." Rounding upward by about 20%.
- Framing as "benefit": the post calls the 5+ cups result "the biggest benefit." The study measured differences between groups, not benefits produced by coffee. The paper itself frames these as population-level associations.
- Unsupported causal inference by implication: the post explicitly says it is willing to believe "there's a real link" rather than accept confounding, but offers no evidence for choosing causation over the reverse-causation and residual-confounding explanations the authors say cannot be excluded.
- Overstated support from experimental evidence: the post asserts that "mechanistic and experimental evidence actually supports coffee's benefits on metabolic and hormonal health." For hormonal health this is weak. The only randomized trial directly on this question found that after an 8-week parallel-arm trial in 42 overweight adults, no significant differences were found between treatment groups for any of the studied outcomes at week 8 , and the authors concluded their data do not indicate a consistent effect of caffeinated coffee consumption on SHBG in men or women, with results to be interpreted cautiously given the small sample . Separately, NHANES analyses found no linear relationship between caffeine intake and testosterone levels in the total population, though a non-linear association was present . Trial evidence on metabolic outcomes is also partial: a 24-week Harvard trial in 126 adults found coffee did not affect participants' sensitivity to insulin, though drinking coffee was associated with a modest loss of body fat .
- Omitted comparator weakness: the "non-drinker" reference group was only 70 people out of 2,264, making the non-drinker contrast the least statistically robust part of the comparison.
? What's uncertain 4
- Whether the body composition group differences remained statistically significant in fully adjusted models, or whether they are primarily descriptive group comparisons. The abstract describes them at group level; I could not retrieve the adjusted body composition regression table.
- The exact magnitude of the body fat difference. One secondary outlet reports people drinking five or more cups per day had an average body-fat percentage of about 27%, compared with roughly 30% among those drinking one to four cups , but I could not confirm these figures against the paper's tables directly.
- Which coffee component, if any, is responsible. The study was not designed to answer this, and cup size, brew method, and additions were not recorded.
- Whether reverse causation explains part of the pattern, for example leaner or healthier people tolerating and drinking more coffee, or people with metabolic symptoms cutting back.
Sources
6 of 6 linked to recordsVerroest 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;65(5):215. DOI: 10.1007/s00394-026-04038-z
University of Oulu press release, 16 July 2026
News-Medical detailed study write-up
2 Minute Medicine study appraisal
Wedick NM, et al. RCT on coffee and sex hormones, Nutrition Journal 2012
NHANES caffeine-testosterone analyses (Tandfonline 2018)