Associations of cause-specific mortality with area level deprivation and travel time to health care in France from 1990 to 2007, a multilevel analysis

Table 2 Association between all-cause mortality and sex, individual education, contextual deprivation and time to health care

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Variables	HR	HR	HR	HR	HR	HR	HR	HR
Individual effects
Men	ref	ref	ref	ref	ref	ref	ref	ref
Women	0.49***	0.48***	0.49***	0.49***	0.48***	0.48***	0.49***	0.48***
Upper and post-secondary		ref			ref	ref		ref
Lower secondary		1.31***			1.31***	1.31***		1.31***
Completed elementary		1.42***			1.42***	1.42***		1.42***
Incompleted elementary		1.75***			1.74***	1.75***		1.75***
Contextual Effects
Area Deprivation
Q1 (least deprived)			ref		ref		ref	ref
Q2			1.05		1.01		1.08*	1.05
Q3			1.08**		1.02		1.14***	1.09**
Q4			1.13***		1.04		1.19***	1.13***
Q5			1.17***		1.06*		1.26***	1.17***
Trend			1.10***		1.05***		1.15***	1.10***
Travel time to health care
Q1 (shortest time)				ref		ref	ref	ref
Q2				0.99		0.97	0.96	0.95
Q3				1.04		0.98	0.96	0.93*
Q4				1.05		0.98	0.94	0.91**
Q5				0.98		0.91**	0.87***	0.84***
Trend				0.98		0.93***	0.89***	0.87***
Spatial Variance (ZE94)	0.013	0.011	0.010	0.012	0.011	0.010	0.008	0.008

The Trend is estimated from a separate model in which the concerned contextual effect has been introduced as a linear effect (log-linear). The displayed trend effect is the linear effect multiplied by the mean score difference between the fifth and the first quintile
HR Hazard Ratio. Q ordered population-weighted quintile
*p-value <0.050, **p-value <0.010, ***p-value <0.001

ISSN: 1471-2458