Nitrogen dioxide levels estimated from land use regression models several years apart and association with mortality in a large cohort study

Table 2 Association between logarithm of nitrogen dioxide (NO ₂ ) concentrations and land-use variables, from univariate analysis and multiple linear regression model. Rome 1995/96 and 2007

			1995/96			2007
Land-use variables	Slope^a	p value	Slope^b	p value	Slope^a	p value	Slope^b	p value
Intercept			(24.5)	0.023			(21.8)	0.199
Proximity to a green urban area
No	-
Yes	−0.0067371	0.895			−0.0482182	0.517
Altitude (m)	−0.0009000	0.239	−0.0013084	0.002	−0.0017940	0.138	−0.0025286	0.001
GIS Coordinate x (m)	0.0000031	0.233	0.0000031	0.054	0.0000060	0.148	0.0000069	0.009
GIS Coordinate y (m)	0.0000090	0.002	−0.0000046	0.049	0.0000164	0.001	−0.0000042	0.252
Distance from the city centre (m)	−0.0000218	<0.001	−0.0000176	<0.001	−0.0000355	<0.001	−0.0000301	<0.001
Number of residents in the census block	−0.0000500	0.295			−0.0001610	0.004
Size of the census block (m²)	−0.0000004	<0.001			−0.0000004	<0.001	−0.0000002	0.001
Inverse population density (m2/person)	−0.0002000	<0.001	−0.0001024	0.004	−0.0000072	0.248
Meters of high traffic road in a 150 m buffer	0.0003840	0.001			0.0005537	<0.001	0.0003620	<0.001
Traffic density in a 150 m buffer (cars/m)	0.0008000	0.001	0.0003290	0.032	0.0013315	<0.001
Distance from the closest high traffic road (m)	−0.0005000	<0.001	−0.0002021	0.003	−0.0006183	<0.001
Adjusted R-square			0.737				0.704

^aThe estimate derived from a simple linear regression model with only one independent variable and logarithm of NO₂ concentration (μg/m³) as dependent variable.
^bThe estimate derived from a multiple linear regression model with logarithm of NO₂ concentration (μg/m³) as dependent variable.

ISSN: 1476-069X