Table 3.

Results from the multivariable regression models with body shape as outcome−baseline. Model 1 uses the CTS score as predictor for body shape (WtHR) only. To compare the strength of the effect, model 2 includes known sociodemographic variables predicting an increased WHtR. Follow-up: Concerning the CTS sum score as predictor of the WHtR in men, the strength of the effect was similar to the baseline. However, in women the effect was reversed. Age (increased) and educational years (decreased) are the strongest predictors for changes in body shape.

Pooled data model	Variables	Women		Men
		Beta		Beta
Model 1	CTS sum	0.074**		0.133***
Model 2	CTS sum	0.028		0.063*
	Age	0.333***		0.327***
	Education years	−0.157***		−0.137***
	Occuation	−0.092***		0.003
	Current depression	− 0.002		0.051
Follow-up
Model 1a	WHtR ~ CTS sum	0.056*		0.116***
Model 2a	CTS sum	0.007		0.043*
	Age	0.321***		0.261***
	Education years	−0.162***		−0.143***
	Occupation	− 0.050		− 0.071*
	Current depression	0.012		0.064*

⁺p <  = 0.1, *p <  = 0.05, **p <  = 0.01, ***p <  = 0.001. Statistical significant results are highlighted in bold.