. 2021 Dec 22;37(1):49–65. doi: 10.1007/s10654-021-00828-3

Table 2.

Levels of GDF-15, NT-proBNP, HbA1C, CRP, cystatin-C among different strata of the sub-cohort, EPIC-Heidelberg (n = 3792)

	GDF-15			NT-proBNP			HbA1C			CRP			Cystatin-C
	Mean ± SE	P*	R²	Mean ± SE	P*	R²	Mean ± SE	P*	R²	Mean ± SE	P*	R²	Mean ± SE	P*	R²
Age		< .001	0.13		< .001	0.04		< .001	0.07		< .001	0.02		< .001	0.13
< 40	518 ± 18			212 ± 21			33.3 ± 0.5			2.1 ± 0.3			337 ± 18
[40—45[	533 ± 12			205 ± 13			33.9 ± 0.3			2.6 ± 0.2			350 ± 11
[45—50[	586 ± 13			218 ± 14			34.5 ± 0.3			2.2 ± 0.2			356 ± 12
[50—55[	645 ± 9			180 ± 11			35.8 ± 0.3			3.1 ± 0.2			563 ± 9
[55 -60[	731 ± 9			224 ± 11			36.7 ± 0.3			3.7 ± 0.2			589 ± 9
≥ 60	803 ± 10			315 ± 11			38.3 ± 0.3			4.0 ± 0.2			597 ± 10
Sex		< .001	0.13		< .001	0.04		< .001	0.07		0.07	0.02		0.94	0.13
Men	661 ± 7			196 ± 8			35.9 ± 0.2			2.8 ± 0.1			465 ± 7
Women	611 ± 6			255 ± 7			34.9 ± 0.2			3.1 ± 0.1			465 ± 6
Level of education		< .001	0.14		< .001	0.04		< .001	0.08		< .001	0.03		0.80	0.13
None/primary school	676 ± 8			225 ± 10			36.5 ± 0.2			3.6 ± 0.1			470 ± 8
Technical/professional	631 ± 8			243 ± 9			35.8 ± 0.2			2.9 ± 0.1			459 ± 8
Secondary school	655 ± 18			222 ± 21			35.1 ± 0.5			2.9 ± 0.3			467 ± 17
Longer (incl. University)	601 ± 8			210 ± 9			34.2 ± 0.2			2.4 ± 0.1			465 ± 8
Physical activity level		0.60	0.13		0.64	0.04		0.05	0.07		0.45	0.02		0.53	0.13
Inactive	630 ± 10			218 ± 12			34.8 ± 0.5			2.7 ± 0.2			477 ± 9
Moderately inactive	640 ± 9			231 ± 10			35.5 ± 0.2			3.1 ± 0.2			464 ± 8
Moderately active	640 ± 7			230 ± 9			35 .7 ± 0.2			3.0 ± 0.1			461 ± 7
Active	620 ± 15			210 ± 18			35.5 ± 0.4			3.1 ± 0.3			454 ± 15
BMI		< .001	0.13		0.31	0.04		< .001	0.09		< .001	0.09		0.53	0.13
Normal	642 ± 7			225 ± 8			34.7 ± 0.2			2.0 ± 0.1			464 ± 7
Overweight	614 ± 8			219 ± 9			35.2 ± 0.2			3.3 ± 0.1			463 ± 7
Obese	674 ± 11			243 ± 14			38.1 ± 0.3			5.1 ± 0.2			477 ± 11
Smoking status		< .001	0.21		0.67	0.04		< .001	0.08		< .001	0.04		0.01	0.13
Never	579 ± 7			218 ± 8			34.8 ± 0.2			2.6 ± 0.1			459 ± 7
Long time quitter (> 10 y)	598 ± 9			229 ± 11			35.0 ± 0.2			2.7 ± 0.2			450 ± 9
Short time quitter (≤ 10 y)	643 ± 14			247 ± 17			35.4 ± 0.4			3.2 ± 0.3			469 ± 14
Current, light (≤ 10 cig /d)	679 ± 16			225 ± 19			36.2 ± 0.5			2.6 ± 0.3			470 ± 16
Current, heavy (> 10 cig/d)	824 ± 11			225 ± 14			37.3 ± 0.3			4.3 ± 0.2			501 ± 12
Alcohol consumption^£		0.008	0.14		0.32	0.04		< .001	0.07		0.42	0.02		0.72	0.13
Low consumers	624 ± 7			220 ± 7			35.9 ± 0.2			2.9 ± 0.1			467 ± 7
High consumers	647 ± 7			231 ± 7			35.0 ± 0.2			3.0 ± 0.1			464 ± 6
Baseline diabetes		< .001	0.15		0.99	0.04		< .001	0.35		< .001	0.02		0.27	0.13
No	629 ± 5			226 ± 5			34.6 ± 0.1			2.9 ± 0.1			466 ± 5
Yes	841 ± 23			226 ± 28			59.6 ± 0.5			4.7 ± 0.4			441 ± 23

GDF-15: n = 3701; NT-proBNP: n = 3430; HbA1C: n = 3717; CRP: n = 3635; Cystatin-C: n = 3665

HbA1C in % can be obtained from mmol/mol as follows: HbA1C (%) = (0.0915 × (HbA1C mmol/mol)) + 2.15%

Values are means (adjusted for age and sex) ± SE

^*p-value for lifestyle factor, obtained using a generalized linear model adjusted for age and sex

R²: adjusted R² obtained using age and sex-adjusted linear regressions, interpreted as the variance in the biomarker jointly explained by age, sex and the specific lifestyle factor

^£Lifetime alcohol consumption according to sex-specific median: median in men = 20.6 g/day and in women = 4.1 g/day

R² for a model including all of the above factors: 0.23 for GDF-15, 0.04 for NT-proBNP, 0.37 for HbA1C, 0.12 for CRP and 0.14 for Cystatin-C