Table 3. Multilevel linear regression analysis examining the association between neighborhood characteristics and cardiovascular risk among older people living in Chita peninsula, Aichi Prefecture, Japan.

	Male participants (n = 1,805)				Female participants (n = 2,005)
	Model 1		Model 2		Model 1		Model 2
	coef.	95% CI	coef.	95% CI	coef.	95% CI	coef.	95% CI
1. Proximity of shops selling fresh fruits and vegetables
1a. Perceived	−0.02	−0.48, 0.45	−0.06	−0.63, 0.50	−0.15	−0.67, 0.37	−0.15	−0.66, 0.36
1b. GIS-based	0.12	−0.69, 0.92	−0.04	−0.95, 0.87	0.14	−0.63, 0.92	0.07	−0.70, 0.83
2. Proximity of exercise facilities
2a. Perceived	−0.77	−1.48, −0.07*	−1.00	−1.78, −0.21*	−0.16	−0.96, 0.64	−0.22	−1.01, 0.56
2b. GIS-based	−0.14	−0.59, 0.30	−0.09	−0.61, 0.43	0.15	−0.34, 0.64	0.10	−0.38, 0.59
3. Walking environment in the neighborhood
3a. Perceived	−0.58	−1.25, 0.08†	−0.62	−1.39, 0.14	−0.30	−0.98, 0.38	−0.31	−0.98, 0.35
3b. GIS-based	−0.21	−0.76, 0.34	−0.16	−0.83, 0.51	0.08	−0.55, 0.71	0.09	−0.54, 0.71
4. Social Capital
4a. Cognitive	−0.16	−0.57, 0.25	−0.26	−0.71, 0.19	−0.14	−0.55, 0.27	−0.17	−0.57, 0.23
4b. Structural	−0.27	−0.94, 0.41	−0.02	−0.77, 0.74	0.07	−0.61, 0.74	0.08	−0.59, 0.75
5. Personal safety and risk of traffic accidents
5a. Perceived	0.81	0.11, 1.52*	1.08	0.30, 1.86**	0.40	−0.37, 1.17	0.44	−0.31, 1.19

**: p < 0.01

*: p < 0.05

†: p < 0.10

Model 1 adjusted for basic individual socio-demographic variables (i.e., age in years, age-squared, sex, body mass index, marital status, educational status, employment status, equivalent household income and depression) and neighborhood SES. When investigating the effect of neighborhood cognitive and structural social capital (4a and 4b), we also included individual-level cognitive and structural social capital measures as covariates, respectively.

Model 2 further adjusted for individual behavioral variables (i.e., time spent walking, fruit and vegetable consumption, alcohol consumption, smoking and frequency of meeting friends).