A predictive model relating daily fluctuations in summer temperatures and mortality rates

BMC Public Health

Table 5 Over-dispersion by population subgroup and temperature group

			Full Model	5 groups	4 groups	3 groups	2 groups	1 group	Model M₀
Population subgroup	55–74 years	Group		G1 _t-2	GCum1 _t-2	G2	GMA	GCum1 _t
		Over-dispersion	1.80	1.80	1.80	1.82	1.90	1.97	2.49
	75 years and more	Group		GCum1 _t-4	G1 _t-2	GCum1 _t	GMA	G2
		Over-dispersion	2.09	2.11	2.13	2.21	2.61	2.90	5.26
	Men	Group		GCum1 _t-1	G1 _t-2	GCum1 _t	GMA	G2
		Over-dispersion	1.46	1.47	1.47	1.49	1.66	1.84	2.61
	Women	Group		G1 _t-2	GCum1 _t-2	GMA	GCum1 _t	G2
		Over-dispersion	1.71	1.71	1.72	1.79	2.11	2.37	4.22
Three medical causes of death ^a		Group		GCum1 _t-1	G1 _t-2	GMA	GCum1 _t	G2
	Over-dispersion	1.74	1.76	1.79	1.84	2.10	2.33	3.81
Calibration group odd/even years subgroup	14 even years	Group		G1 _t-2	GCum1 _t-2	G2	GMA	GCum2'
		Over-dispersion	2.06	2.06	2.07	2.11	2.61	2.98	5.66
	14 odd years	Group		GCum1 _t-1	G1 _t-2	GCum1 _t	GMA	G2
		Over-dispersion	2.16	2.19	2.23	2.29	2.70	3.15	5.60

^a: Three Medical Causes of Death: heat-related deaths (hyperthermia, heatstroke, dehydration), cardiovascular disease and respiratory disease
The 5 groups of temperature indicators were the last 5 groups excluded from the model using the backward method.

ISSN: 1471-2458