Table 4.

Associations of Exposure to Neighborhood-Level Racial Residential Segregation with Incident Coronary Artery Calcification^*

Segregation in Young Adulthood+
	Model 1 Rate Ratio (95% Confidence Interval)	Model 2 Rate Ratio (95% Confidence Interval)	Model 3 Rate Ratio (95% Confidence Interval)
High Segregation in Young Adulthood	Reference	Reference	Reference
Medium Segregation in Young Adulthood	1.01 (0.75, 1.36)	1.04 (0.75, 1.42)	0.96 (0.66, 1.39)
Low Segregation in Young Adulthood	0.50 (0.27, 0.92)	0.52 (0.28, 0.98)	0.56 (0.29, 1.09)
Segregation Patterns from Young Adulthood to Midlife
	Model 1 Rate Ratio (95% Confidence Interval)	Model 2 Rate Ratio (95% Confidence Interval)	Model 3 Rate Ratio (95% Confidence Interval)
Continuously High/Medium Segregation	Reference	Reference	Reference
Declined from High / Medium to Low Segregation	1.17 (0.89, 1.55)	1.17 (0.89, 1.55)	1.29 (0.94, 1.77)
Increased from Low to High / Medium Segregation	0.42 (0.19, 0.95)	0.43 (0.19, 0.99)	0.47 (0.20, 1.07)
Continuous Low Segregation	0.75 (0.31, 1.83)	0.77 (0.31, 1.92)	1.00 (0.36, 2.74)

^*Estimated using Poisson regression with generalized estimating equations to account for clustering by census tract in young adulthood (baseline; 1985-1986). This model included participants who did not have coronary artery calcification in midlife (year 15 follow-up examination; N=1,125). Those who developed coronary artery calcification by year 20 or 25 were considered to have incident coronary artery calcification.

⁺Models were progressively adjusted: Model 1: baseline age, sex, field center, and baseline education. Model 2: Model 1 + baseline neighborhood poverty. Model 3: Model 2 + year 15 potential individual-level CAC risk factors (body mass index, smoking, alcohol use, fasting glucose, systolic blood pressure, and total cholesterol).