Table 3.

Linear regression analysis of statin use and log-transformed thyroid-stimulating hormone (TSH) levels at the end of follow-up.

	B	SE	95% CI of B	Beta	p value
Statin use in total population
Univariable model	−0.073	0.027	−0.125 to −0.021	−0.136	0.006
Multivariable model 1	−0.061	0.020	−0.100 to −0.022	−0.114	0.002
Multivariable model 2	−0.053	0.020	−0.093 to −0.013	−0.099	0.010
Statin use in euthyroid subjects
Univariable model	−0.068	0.026	−0.119 to −0.017	−0.139	0.009
Multivariable model 1	−0.052	0.021	−0.093 to −0.011	−0.106	0.014
Multivariable model 2	−0.045	0.021	−0.087 to −0.003	−0.092	0.035
Statin use in subjects with SCH
Univariable model	−0.108	0.053	−0.214 to −0.001	−0.276	0.047
Multivariable model 1	−0.131	0.066	−0.265 to 0.002	−0.337	0.053
Multivariable model 2	−0.116	0.071	−0.261 to 0.028	−0.298	0.111

TSH, thyroid stimulating hormone.

Data are unstandardized coefficients (B), corresponding standard error (SE), 95% confidence interval (CI) of B, standardized coefficients (Beta), and significance (p value).

Multivariable model 1 was adjusted for basal age, sex, free triiodothyronine, free thyroxine, thyroid stimulating hormone, thyroid autoimmunity, body mass index, systolic blood pressure, fasting plasma glucose, alanine aminotransferase, and estimated glomerular filtration rate;

Multivariable model 2 was further adjusted for the presence of cardiovascular disease and follow-up time.