Skip to main content

Table 2 Total and direct effects of neighbourhood environmental characteristics on cognitive function

From: From urban neighbourhood environments to cognitive health: a cross-sectional analysis of the role of physical activity and sedentary behaviours

Environmental characteristic (units)

Effect

Memory

(CVLT raw score)

Processing speed

(SDMT raw score)

 

b (95% CI)

b (95% CI)

Population density (10 persons/ha)

Total

0.121 (0.045, 0.197)

0.151 (-0.146, 0.447)

Direct

0.069 (-0.045, 0.184)

-0.064 (-0.505, 0.378)

Street intersection density (10 intersections/km2)

Total

-0.011 (-0.288, 0.286)

-0.026 (-0.137, 0.086)

Direct

0.007 (-0.022, 0.037)

-0.001 (-0.114, 0.113)

Percentage of commercial land use (10%)

Total

-0.004 (-0.124, 0.116)

0.109 (-0.365, 0.583)

Direct

-0.054 (-0.176, 0.069)

-0.073 (-0.553, 0.407)

Non-commercial land use mix (0.10 score)

Total

-0.002 (-0.064, -0.061)

0.118 (-0.114, 0.349)

Direct

-0.014 (-0.078, 0.050)

0.111 (-0.121, 0.342)

Percentage of parkland (10%)

Total

0.078 (0.014, 0.142)

0.243 (0.006, 0.480)

Direct

0.078 (0.014, 0.141)

0.211 (0.021, 0.401)

Percentage of blue space (10%)

Total

-0.150 (-0.490, 0.189)

0.225 (-1.154, 1.604)

Direct

-0.097 (-0.436, 0.243)

0.492 (-0.886, 1.870)

NO2 (ppb)

Total

0.011 (-0.040, 0.061)

0.159 (-0.029, 0.347)

Direct

0.007 (-0.043, 0.057)

0.218 (0.032, 0.404)

PM2.5 (μg/m3)

Total

0.096 (0.046, 0.146)

0.013 (-0.171, 0.197)

Direct

0.085 (0.035, 0.135)

-0.059 (-0.238, 0.121)

  1. CVLT California Verbal Learning Test; SDMT Symbol-Digit Modalities Test; b regression coefficient; CI confidence intervals. Total-effect models are not adjusted for physical activity and sedentary behaviour mediators, while direct-effect models are. Statistically significant effects in bold (p<.05). Details on regression models, including confounders, are in the Additional file 1 (Table S1)