The effect of modifiable risk factors on geographic mortality differentials: a modelling study

BMC Public Health

Table 5 Results of Cox proportional hazards regression model, females aged 40 and over

Parameter	DF	Estimate	Standard Error	Chi-Square	Pr > ChiSq	Hazard Ratio	95% confidence limits
score	1	2.32	0.80	8.42	0.0037	10.20	2.13	48.95
Age 40-44	1	-3.23	0.75	18.41	<.0001	0.04	0.01	0.17
Age 45-49	1	-2.80	0.63	19.72	<.0001	0.06	0.02	0.21
Age 50-54	1	-2.22	0.51	18.81	<.0001	0.11	0.04	0.30
Age 55-59	1	-2.55	0.63	16.57	<.0001	0.08	0.02	0.27
Age 60-64	1	-2.09	0.55	14.27	0.0002	0.12	0.04	0.37
Age 65-69	1	-1.35	0.42	10.50	0.0012	0.26	0.12	0.59
Age 70-74	1	-0.27	0.28	0.90	0.3433	0.76	0.44	1.33
Adjusted Hosmer-Lemeshow goodness of fit
Chi-Square		DF		Pr > ChiSq
9.37		8		0.40

Notes
1. Results of a Cox proportional hazards regression model predicting risk of death prior to 1 June 2004 among female AusDiab participants aged 40 and over
2. Age group 75 and over taken as baseline group for the age parameter
C statistic: 84%

ISSN: 1471-2458