Table 3.

Incidence Rate Ratios for Incident Diabetes by Diagnostic Criteria Among Adults Free of Diabetes at Baseline (1987–1989), Atherosclerosis Risk in Communities Study^a

Categorical Risk Factors	Incident Diabetes Case Definitionb			P Value
	ARIC Study (n = 1,441)	Self-Report (n = 293)	Multiple Evidence (n = 186)
Absolute cumulative incidence at 9 yearsc	0.123	0.027	0.017
Gender
Female	1.0	1.0	1.0
Male	1.3	1.1	1.3	0.19
Race
White	1.0	1.0	1.0
Black	2.0	1.9	2.4	0.06
High school graduate
No	1.0	1.0	1.0
Yes	0.72	0.63	0.62	0.54
Waist/hip ratio
Tertile 1 (<0.89)	1.0	1.0	1.0
Tertile 2 (0.89–0.95)	2.6	2.1	2.9	0.79
Tertile 3 (≥0.96)	5.7	4.9	6.9	0.37
Metabolic syndrome
No	1.0	1.0	1.0
Yes	4.4	3.2	4.5	<0.001
Baseline fasting glucose, mg/dL
<100	1.0	1.0	1.0
100–125	6.1	3.4	4.3	<0.001
Parental history of diabetes
No	1.0	1.0	1.0
Yes	1.8	2.2	2.2	0.15
Continuous risk factors (per standard deviation)
Age (6 years)	1.1	1.1	1.0	0.79
Systolic blood pressure (19 mm Hg)	1.37	1.18	1.30	<0.001
Diastolic blood pressure (11 mm Hg)	1.24	1.10	1.23	<0.001
Body mass index (5 kg/m2)	1.64	1.57	1.69	0.002
HDL cholesterol (7 mg/dL)	0.53	0.57	0.48	0.10
Log of insulin (0.77 pmol/L)	2.47	2.32	2.96	<0.001
Log of triglycerides (0.52 mg/dL)	1.64	1.60	1.88	0.007

Abbreviations: ARIC, Atherosclerosis Risk in Communities; HDL, high density lipoprotein.

^aGender, age, and race models are adjusted for each other; other phenotypes are all adjusted for age, race, and gender.

^bARIC Study: self-reported a physician’s diagnosis of diabetes, reported diabetes medication use, or had a fasting glucose measurement of ≥126 mg/dL or a nonfasting glucose measurement of ≥200 mg/dL; self-reported: self-reported a physician’s diagnosis of diabetes; multiple evidence: a minimum of 2 of the ARIC Study criteria listed above.

^cCumulative incidence at 9 years was computed by using the Kaplan-Meier estimator.