Same BMI, different bodies: a Finnish cohort finds coffee tracks with fat and muscle, not weight

Researchers at the University of Oulu report that among 2,264 members of the Northern Finland Birth Cohort 1966, all aged 46, people with higher habitual coffee intake had lower total body fat and lower visceral fat plus greater skeletal muscle mass than lighter drinkers ^[1][2]. The groups had similar body mass index ^[2]. That is the part worth keeping: a single number on a chart said the two groups were the same, and the tissue measurements said they were not.

The metabolite result points the same way. Higher coffee intake was associated with lower circulating branched-chain amino acids in both men and women ^[3]. When chronically elevated, those markers have previously been linked to insulin resistance and greater risk of developing type 2 diabetes ^[4]. So the composition signal and the biochemical signal agree, which is more than most single-cohort nutrition findings manage.

The hormone findings are messier and sex-specific. In men, greater coffee consumption was associated with a more favorable glucose-insulin profile, higher total and bioavailable testosterone, and higher sex hormone-binding globulin, while free testosterone and the free androgen index were modestly lower ^[5]. In women, the associations were mainly higher SHBG and lower measures of free androgens ^[6]. Higher SHBG with lower free androgen fractions is the pattern you would expect if binding capacity rose, so the "higher testosterone" headline and the "lower free testosterone" footnote are probably the same phenomenon seen from two ends.

Lead author Luca Verroest, a doctoral researcher at Oulu, says the hormonal signature persisted after accounting for BMI and lifestyle factors, and that several associations differed between men and women ^[7]. The release does not list which lifestyle factors, does not report effect sizes, does not give the coffee intake categories in cups, and does not report confidence intervals ^[11]. Without magnitudes, none of this can be judged against measurement noise in a cross-sectional cohort.

The design limit is stated plainly by the researchers: the study is observational and cannot establish that coffee caused any of the differences ^[8]. The arrow could run the other way, since leaner and more muscular 46-year-olds may simply drink more coffee, or a third habit may drive both. Earlier research had already associated coffee drinking with reduced risk of type 2 diabetes and cardiovascular disease without explaining the biology ^[9], and this study adds candidate pathways rather than closing them.

Setting matters for generalisation. Finland ranks among the world's highest coffee-consuming countries, at roughly 11.8 kilograms, or 26 pounds, per person per year ^[10], which works out to about 32 grams a day ^[12]. A cohort where near-universal heavy consumption is the norm is a good place to find a dose gradient and a poor place to learn what happens to light drinkers elsewhere.

What to watch: the paper itself, published in the European Journal of Nutrition as "Associations of habitual coffee intake with testosterone and cardiometabolic markers: the Northern Finland Birth Cohort 1966 study" ^[13], for the effect sizes the release omits. The Oulu group says it is now pursuing these questions in animal models, with the longer-term aim of human intervention studies ^[14], and that more research is needed before the findings can shape dietary recommendations ^[15]. Until an intervention exists, the transferable finding is methodological rather than dietary: if your screening stops at BMI, you are blind to the fat and muscle differences these authors found underneath it ^[2].