Skip to main content

Table 3 Two-level logistic mixed effects models with LSOA-level random effects for psychological distress measured by GHQ-30

From: Urban built environment configuration and psychological distress in older men: Results from the Caerphilly study

Model predictors

Model 1†

Model 2†

Model 3†

O.R. (95% C.I.) p-value

O.R. (95% C.I.) p-value

O.R. (95% C.I.) p-value

Built environment morphometrics

   

Dwelling level variables

   

Dwelling centred density

1.02 (−0.02, 0.06) p = 0.20

 

1.01 (−0.03, 0.05) p = 0.32

Plot exposure (none vs. one bldg face)

0.92 (−0.78, 0.57) p = 0.40

 

0.94 (−0.78, 0.61) p = 0.43

Plot exposure (more than one faces vs. one bldg face)

0.78 (−0.79, 0.27) p = 0.18

 

0.79 (−0.81, 0.30) p = 0.20

Dwelling type (semi-detached vs. detached)

0.72 (−0.89, 0.23) p = 0.12

 

0.76 (−0.86, 0.30) p = 0.18

Dwelling type (terraced vs. detached)

0.55 (−1.35, 0.15) p = 0.06*

 

0.48 (−1.51, 0.02) p = 0.03**

Dwelling type (flat vs. detached)

0.72 (−1.34, 0.64) p = 0.25

 

0.82 (−1.19, 0.79) p = 0.35

Land use configuration

   

Land use mix (z-score)

   

T2 vs. T1

0.72 (−0.93, 0.27) p = 0.14

 

0.63 (−1.10, 0.18) p = 0.08*

T3 vs. T1

0.51 (−1.51, 0.19) p = 0.06*

 

0.42 (−1.77, 0.04) p = 0.03**

Density of bus stops

1.04 (−0.01, 0.10) p = 0.07*

 

1.04 (−0.02, 0.10) p = 0.07*

Density of retail

0.99 (−0.04, 0.02) p = 0.31

 

1.00 (−0.04, 0.03) p = 0.45

Density of community services

1.01 (−0.04, 0.06) p = 0.42

 

1.00 (−0.06, 0.06) p = 0.47

Density of recreation & leisure facilities

0.98 (−0.08, 0.04) p = 0.24

 

0.98 (−0.08, 0.05) p = 0.33

Density of business & offices

1.02 (0.00, 0.03) p = 0.06*

 

1.02 (−0.01, 0.04) p = 0.08*

Topological accessibility of streets (z-score)

   

Street movement potential R1200 m

0.56 (−1.13, -0.01) p = 0.02**

 

0.54 (−1.28, 0.03) p = 0.03**

Street movement potential R3000 m

0.95 (−0.61, 0.47) p = 0.43

 

1.14 (−0.69, 0.94) p = 0.38

Street movement potential RN m

1.53 (0.04, 0.81) p = 0.02**

 

1.24 (−0.39, 0.83) p = 0.25

Connectivity

1.10 (−0.20, 0.39) p = 0.25

 

1.18 (−0.17, 0.49) p = 0.16

Natural environment

   

Topography (Standard deviation in slope)

 

1.24 (−0.03, 0.47)p = 0.04**

1.38 (−0.07, 0.71) p = 0.05**

Greenness (Mean NDVI within 500 m)

 

0.82 (−0.51, 0.10) p = 0.10*

0.79 (−0.66, 0.21) p = 0.14

Neighbourhood deprivation

   

WIMD domains

   

Income deprivation

 

1.03 (0.00, 0.07) p = 0.04**

1.03 (−0.02, 0.07) p = 0.11

Employment deprivation

 

0.97 (−0.07, 0.00) p = 0.03**

0.96 (−0.08, 0.00) p = 0.02**

Health deprivation

 

0.99 (−0.03, 0.01) p = 0.13

0.99 (−0.03, 0.02) p = 0.31

Education deprivation

 

1.00 (−0.04, 0.03) p = 0.42

1.02 (−0.03, 0.06) p = 0.21

Housing deprivation

 

1.00 (−0.04, 0.04) p = 0.47

1.00 (−0.05, 0.04) p = 0.46

Physical environment

 

1.02 (0.00, 0.04) p = 0.01**

1.02 (0.00, 0.04) p = 0.04**

Random effects

   

Between LSOA variance (S.E.)

0.054 (0.083)

0.025 (0.039)

0.042 (0.079)

Model fit

   

Bayesian DIC

690.02

675.53

695.05

  1. Results are expressed as odds ratio, 95% confidence interval and p-value for the logistic regression. All models have been adjusted for individual level variables of age, alcohol consumption, social class, education and prevalence of chronic vascular morbidities.
  2. T Tertile (T1, T2, T3 represents the lower, middle and upper tertiles respectively).
  3. *p < 0.10; **p < 0.05.
  4. †Model 1 comprises of built environmental morphometrics; Model 2 included neighbourhood deprivation captured by six domains of Welsh index of multiple deprivation and natural environment captured by standard deviation in slope and mean greenness index NDVI; Model 3 indicates the fully adjusted model.
Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us