> Elevated BMI was associated with reduced total testosterone levels (P < .0001), with the mean BMI increasing from 25.83 in 1999-2000, to 27.96 in 2015-2016 (P = 0.0006).
They do say: > Lokeshwar noted that even men with a normal BMI (18.5-24.9) had declining total testosterone levels (P < .05) during the same time frames.
However the range of 18.5-24.9 is wider than the mean increase that they think is the most noteworthy independent variable. It's fully possible that within the BMI band of 18.5-24.9, average BMI nonetheless increased and is the primary driver. Note that the effect size over time is muuuuch more significant when not controlling for binned BMI.
That doesn't mean whatever environmental shit you're not fond of is safe of course. Heck, in some roundabout way it may even be contributing to obesity, possibly even through other hormonal changes. Wouldn't be my first guess though.