Table 3.

Spatial error regression model adjusting for demographic covariates and continuous smoking prevalence.

	SEM
	Univariate		Stepwise
	Est (95% CI)	P-value	Est (95% CI)	P-value
Air Toxics Hazard Score
[0]	Ref	--	Ref	--
Low, [0–1]	12.09 (6.56, 17.62)	<0.001	7.49 (2.19, 12.79)	0.006
High, [1–10]	17.89 (8.65, 27.14)	<0.001	11.42 (2.36, 20.47)	0.013
Hispanic (%)			−0.42 (−0.76, −0.08)	0.016
High school or less (%)			0.28 (0.01, 0.54)	0.043
Population density (100 ppL per sq. mile)			0.06 (−0.00, 0.11)	0.050
Smoking prevalence (%)			1.50 (0.41, 2.59)	0.007
Distance to major highway (km)
≤5 km			−3.82 (−5.78, −1.86)	<0.001
>5 km			0.59 (−3.55, 4.72)	0.781
Lambda	0.62 (0.49, 0.75)	<0.001	0.32 (0.11, 0.47)	0.004
Model Statistics
Log likelihood	−932.72		−907.72
R-squared	0.350		0.487
AIC	1875.4		1835.4
Moran’s Index	−0.054	0.868	−0.017	0.613

Est, regression coefficient estimates; ppl, people. P-values < 0.05 were bolded.