Table 3.

Ordered logistic regression of self-rated poor health status on observer ratings of neighborhood disorder and inflammatory burden.

Dependent variable:	Model 1 (ND)		Model 2 (ND, IB)
Self-rated poor health status (0–4)	Coeff. (SE)	Odds Ratio	Coeff. (SE)	Odds Ratio
Independent variable
Observer ratings of neighborhood disorder (ND)	0.123 *	1.131	0.109 †	1.115
	(0.058)		(0.058)
Mediator
Inflammatory burden (IB)			0.094 **	1.099
			(0.037)
Control variables
Childhood adversity	0.074 †	1.077	0.073 †	1.076
	(0.041)		(0.041)
Family SES	−0.215 **	0.807	−0.228 **	0.796
	(0.072)		(0.076)
Age	0.023 *	1.023	0.021 †	1.021
	(0.011)		(0.011)
South	0.045	1.046	0.086	1.090
	(0.173)		(0.177)
Length of residence	−0.001	0.999	−0.001	0.999
	(0.009)		(0.009)
Neighborhood cohesion	0.000	1.000	−0.001	0.999
	(0.006)		(0.007)
Concentrated disadvantage	0.050	1.051	0.060	1.062
	(0.054)		(0.055)
Residential stability	−0.047	0.954	−0.063	0.939
	(0.058)		(0.063)
Thresholds
Threshold parameter 0–1	−1.434 **	0.238	−1.424 **	0.241
	(0.178)		(0.180)
Threshold parameter 1–2	−0.325 *	0.723	−0.306 †	0.736
	(0.158)		(0.158)
Threshold parameter 2–3	0.590 **	1.804	0.619 **	1.857
	(0.155)		(0.157)
Threshold parameter 3–4	1.976 **	7.214	2.015 **	7.501
	(0.178)		(0.177)
R-Square	0.072		0.088 *
∆R-Square			0.016 **

Notes: Unstandardized coefficients are shown with robust standard errors in parentheses; SES, age, neighborhood cohesion, concentrated disadvantage, and residential stability are centered by mean subtraction; observer ratings of neighborhood disorder are standardized by z-transformation (mean = 0 and SD =1); N = 325. † p ≤ 0.100; * p ≤ 0.050; ** p ≤ 0.010 (two-tailed tests).