Table 6.

Association between PA variables and the risk of incident elevated blood- glucose

No. with incident elevated BG (N = 47, 7.3%)
	No. at risk (N = 641)	No. incident elevated BG (% of No. at risk)	Model 1 adjusted Hazard Ratio (95% CI)	p-value
Habitual PA level at baseline [min/week]
Low (< 75)	33	2 (0.3)	ref
Medium (75–149)	143	11 (1.7)	2.12 (0.46–9.85)	0.338
High (≥ 150)	465	34 (5.3)	1.39 (0.33–5.82)	0.656
Sports-related PA level at baseline [min/week]
Low (< 75)	345	31 (4.8)	ref
Medium (75–149)	142	10 (1.6)	0.97 (0.46–2.06)	0.942
High (≥ 150)	154	6 (0.9)	0.54 (0.22–1.32)	0.176
Sports-related intensity level at baseline [METh/week] {1}
Not intensively active (< 8.3)	349	31 (4.8)	ref
Moderately intensively active (8.3–16.5)	137	11 (1.7)	1.13 (0.55–2.28)	0.745
Highly intensively active (≥ 16.6)	154	5 (0.7)	0.50 (0.19–1.30)	0.154
Change in sports-related PA over time [min/week] {26}
Stable inactive (< 150 – < 150)	333	34 (5.3)	ref
Quits activity (≥ 150 – < 150)	52	5 (0.7)	0.97 (0.38–2.50)	0.950
Starts activity (< 150 – ≥ 150)	144	6 (0.9)	0.43 (0.17–1.04)	0.062
Stable active (≥ 150 – ≥ 150)	86	1 (0.1)	0.13 (0.02–0.92)	0.042
Change in sports-related intensity level over time [METh/week] {6}
Stable low (< 16.6 – < 16.6)	348	32 (5.0)	ref
Decreasing (≥ 16.6 – < 16.6)	54	4 (0.6)	1.03 (0.36–2.95)	0.962
Increasing (< 16.6 – ≥ 16.6)	135	10 (1.6)	0.76 (0.36–1.58)	0.457
Stable high (≥ 16.6 – ≥ 16.6)	98	1 (0.1)	0.13 (0.02–0.99)	0.049

BG blood-glucose, CI confidence interval, METh metabolic equivalent hours, min minutes, N number of participants, No. Number, PA physical activity, ref reference group, SES socio-economic status. Model 1: adjusted for age, sex und SES. Model 2 not reported due to small sample size of N 47 which does not allow for adjustment of more than three confounding variables. ^{} indicates number of N missing. Significant values (p < 0.05) are in [bold]; significant values after Bonferroni correction for multiple testing are indicated by [*]