Table 2.

Hazard ratios (95% CI) for risk of incident type-2 diabetes according to usual levels of lipoprotein(a) concentration

	No. incident type 2 diabetes cases	Median (range), Lp(a), mg/dL	HR (95% CI) Model 1	P value	HR (95% CI) Model 2	P value	HR (95% CI) Model 3	P value
Quintile
1	20	2.3 (0.8–3.6)	1.36 (0.74, 2.48)	0.326	1.35 (0.73, 2.49)	0.336	1.37 (0.74, 2.53)	0.311
2	26	5.1 (3.7–6.6)	1.90 (1.04, 3.45)	0.036	2.15 (1.18, 3.93)	0.013	2.24 (1.22, 4.10)	0.009
3	17	8.8 (6.7–12.3)	1.42 (0.76, 2.66)	0.269	1.34 (0.71, 2.51)	0.365	1.43 (0.75, 2.71)	0.276
4	18	17.5 (12.5–26.9)	1.05 (0.53, 2.08)	0.896	1.02 (0.51, 2.03)	0.961	1.01 (0.51, 2.01)	0.981
5	13	51.9 (27.1–316.2)	[Reference]		[Reference]		[Reference]
Per SD lower log Lp(a)			1.10 (0.93, 1.29)	0.253	1.12 (0.95, 1.32)	0.176	1.12 (0.95, 1.32)	0.171

Model 1 was adjusted for age and sex. Model 2 was additionally adjusted for alcohol consumption, BMI, smoking, SES and physical activity. Model 3 was adjusted for the same factors as Models 1 and 2 plus systolic blood pressure, HDL-C, log hsCRP and waist–hip ratio

Usual Lp(a) concentration are predicted long-term average levels of Lp(a) estimated by regressing the log-transformed Lp(a) values measured at the 5-year follow-up on the log-transformed Lp(a) baseline values

CI confidence intervals, BMI body mass index, HDL-C high density lipoprotein cholesterol, HR hazard ratio, hsCRP high sensitivity c-reactive protein, Lp(a) lipoprotein(a), SD standard deviation, SES socioeconomic status