There is some controversy whether T(4) treatment is indicated in obese humans with hyperthyrotropinemia. The objective of this study was to examine whether hyperthyrotropinemia is a cause or a consequence of obesity. The study was designed as a cross-sectional comparison between obese and lean children and includes a 1-yr follow-up study. The study was set in a primary care facility. The patients were 246 obese and 71 lean children. The 1-yr intervention program was based on exercise, behavior therapy, and nutrition education. The main outcome measures were TSH, free T(3) (fT3), free T(4) (fT4), high-density lipoprotein, low-density lipoprotein, and total cholesterol at baseline and 1 yr later. TSH (P = 0.009) and fT3 (P = 0.003) concentrations were significantly higher in obese children than in normal weight children, whereas there was no difference in fT4 levels (P = 0.804). Lipids did not correlate significantly to thyroid hormones in cross-sectional and longitudinal analyses. fT3, fT4, and lipids did not differ significantly in the 43 (17%) children with TSH levels above the normal range from the children with TSH levels within the normal range. Substantial weight loss in 49 obese children led to a significant reduction of TSH (P = 0.035) and fT3 (P = 0.036). The 197 obese children without substantial weight loss demonstrated no significant changes of thyroid hormones. Because fT3 and TSH were moderately increased in obese children and weight loss led to a reduction, the elevation of these hormones seems to be rather a consequence of obesity than a cause of obesity. Because fT3 and TSH were both increased in obesity and thyroid hormones were not associated to lipids, we put forward the hypothesis that there is no necessity for thyroxine treatment.