Table 2.

Cohort Effect: Results From Cox Proportional Hazards Models Adjusting for Sex and Age

Model number	Variables	Dementia		Alzheimer’s disease
		Hazard ratio (95% CI)	p-Value	Hazard ratio (95% CI)	p-Value
1	2001 cohort vs 1992 cohort	0.62 (0.48–0.81)	.0004	0.57 (0.43–0.77)	.0002
2a	2001 cohort vs 1992 cohort	0.64 (0.49–0.84)	.0009	0.60 (0.44–0.80)	.0006
	Rural residence vs urban residence	1.41 (1.11–1.80)	.0054	1.55 (1.19–2.02)	.0011
2b	2001 cohort vs 1992 cohort	0.74 (0.57–0.98)	.0356	0.72 (0.53–0.98)	.0346
	Years of education	0.92 (0.88–0.95)	<.0001	0.90 (0.86–0.94)	<.0001
2c	2001 cohort vs 1992 cohort	0.66 (0.51–0.87)	.0026	0.62 (0.46–0.83)	.0017
	White collar vs. blue collar	0.70 (0.54–0.91)	.0080	0.66 (0.49–0.89)	.0064
3	2001 cohort vs 1992 cohort	0.75 (0.57–0.99)	.0436	0.72 (0.53–0.99)	.0407
	Years of education	0.93 (0.89–0.97)	.0011	0.91 (0.87–0.96)	.0002
	Rural vs urban residence	1.26 (0.98–1.62)	.0746	1.36 (1.03–1.79)	.0297
	White collar vs blue collar	0.90 (0.67–1.20)	.4651	0.91 (0.66–1.27)	.5858
4	2001 cohort vs 1992 cohort	0.75 (0.57–0.99)	.0393	0.72 (0.53–0.98)	.0363
	Rural vs urban residence	2.34 (1.10–4.98)	.0264	2.97 (1.30–6.75)	.0095
	White collar job vs nonwhite collar	0.86 (0.64–1.16)	.3284	0.86 (0.62–1.21)	.3965
	Years of education for urban residence	0.96 (0.91–1.02)	.1656	0.95 (0.89–1.02)	.1371
	Years of education for rural residence	0.90 (0.85–0.95)	.0852*	0.87 (0.82–0.93)	.0477*

*p-Values for testing the differences in hazard ratios for years of education between participants with urban and rural residence.