(A–F) Association of 24-h urine steroid hormone metabolite excretion with lean mass. Multivariable regression models are visualized by showing the relationship between the steroid hormone metabolite and lean mass while holding the effect of all other covariables in the model constant. All models in (A)–(F) are adjusted for sex, age, BH, daily physical activity, current smoking, regular caffeine and alcohol consumption, hypertension, diabetes, serum albumin, estimated glomerular filtration rate, hemoglobin, and study center. The relationships are visualized separately for male and female sex and for belonging to the nearest class of age and BH with breaks at 30, 50, and 70 y for age and at 160, 170, and 180 cm for height. Solid lines represent regression lines, and shaped areas represent the corresponding 95% CI. Different slopes of regression lines in (A) indicate a significant interaction between androsterone and age in the model. Different slopes of regression lines in (E) indicate a significant interaction between estriol and sex. The statistical R package visreg was used to generate all figures (60).