Skip to main content

Table 4 Associations with fruit and vegetable consumption (outcome) and additional exposures of interest

From: Investigating the importance of the local food environment for fruit and vegetable intake in older men and women in 20 UK towns: a cross-sectional analysis of two national cohorts using novel methods

 

N (%)

N (%) low fruit and veg

N (%) medium fruit and veg

N (%) high fruit and veg

Unadjusted

Confounder adjusted

Mutually adjusted

OR

95% CI

p-value (trend)

OR

95% CI

p-value (trend)

OR

95% CI

p-value (trend)

Area-level exposures of interest

 Unhealthy food marketing environmenta

      

0.40

  

0.42

  

0.27

  0 (no adverts present)

1180 (58.8)

382 (32.4)

414 (35.1)

384 (32.5)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1 (fewer adverts)

481 (24.0)

157 (32.6)

154 (32.0)

170 (35.3)

0.92

0.75, 1.13

 

0.91

0.74, 1.12

 

0.84

0.66, 1.06

 

  2 (more adverts)

346 (17.2)

129 (37.3)

97 (28.0)

120 (34.7)

0.92

0.73, 1.16

 

0.93

0.73, 1.17

 

0.89

0.68, 1.17

 

 Road quality scoreb

      

0.76

  

0.91

  

0.61

  0 (worst walking environment)

586 (29.2)

188 (32.1)

183 (31.2)

215 (36.7)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1

716 (35.7)

262 (36.6)

227 (31.7)

227 (31.7)

0.87

0.71, 1.07

 

0.89

0.72, 1.10

 

0.90

0.72, 1.12

 

  2 (best walking environment)

705 (35.1)

218 (30.9)

255 (36.2)

232 (32.9)

0.96

0.78, 1.20

 

0.98

0.79, 1.23

 

1.06

0.84, 1.33

 

 Transport

      

0.10

  

0.26

  

0.69

  0 (fewest bus stops)

620 (30.9)

183 (29.5)

213 (34.4)

224 (36.1)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1

763 (38.0)

266 (34.9)

234 (30.7)

263 (34.5)

0.87

0.72, 1.07

 

0.91

0.74, 1.11

 

0.92

0.72, 1.16

 

  2 (most bus stops)

624 (31.1)

219 (35.1)

218 (34.9)

187 (30.0)

0.84

0.68, 1.03

 

0.88

0.71, 1.10

 

0.94

0.72, 1.24

 

 Area-level incomec

      

<0.001

  

0.009

  

0.003

  0 (least deprived)

990 (49.3)

289 (29.2)

316 (31.9)

385 (38.9)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1

598 (29.8)

206 (34.5)

210 (35.1)

182 (30.4)

0.82

0.67, 0.99

 

0.86

0.71, 1.06

 

0.82

0.66, 1.01

 

  2 (most deprived)

419 (20.9)

173 (41.3)

139 (33.2)

107 (25.5)

0.67

0.53, 0.84

 

0.73

0.58, 0.93

 

0.69

0.54, 0.89

 

 Walkabilityd

      

0.27

  

0.41

  

0.52

  0 (lowest walkability)

528 (26.3)

158 (29.9)

176 (33.3)

194 (36.7)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1

772 (38.5)

256 (33.2)

252 (32.6)

264 (34.2)

0.93

0.75, 1.15

 

0.94

0.76, 1.17

 

1.09

0.82, 1.46

 

  2 (highest walkability)

707 (35.2)

254 (35.9)

237 (33.5)

216 (30.6)

0.88

0.71, 1.10

 

0.91

0.73, 1.14

 

1.13

0.79, 1.60

 

 Population densitye

      

0.02

  

0.07

  

0.20

  0 Lowest population density

604 (30.1)

176 (29.1)

197 (32.6)

231 (38.3)

1.00

(ref)

 

1.00

(ref)

 

1.00

(ref)

 

  1

755 (37.6)

256 (33.9)

243 (32.2)

256 (33.9)

0.88

0.72, 1.08

 

0.89

0.72, 1.09

 

0.87

0.65, 1.15

 

  2 Highest population density

648 (32.3)

236 (36.4)

225 (34.7)

187 (28.9)

0.77

0.62, 0.95

 

0.81

0.66, 1.01

 

0.79

0.56, 1.13

 
  1. Multilevel ordinal logistic regression models with random intercepts at the town and LSOA/data zone levels. Restricted to study members non-missing for all variables in the Table (N = 2007 (1124 men, 883 women) study members across 740 LSOAs/data zones with median 2 (range 1-20) study members per LSOA/data zone and with median 105 (range 31-158) study members per town)
  2. aUnhealthy marketing environment defined from a count of unhealthy food and drink adverts within an area including those promoting sugary drinks, unhealthy snacks/junk food and alcohol
  3. bRoad quality score calculated from latent class analysis including 10 variables: ‘quality of pavement’; ‘lowered curbs’; ‘barriers on pavement’; ‘pavement width’; ‘pedestrian traffic’; ‘road use’; ‘road connectivity’; ‘traffic calming measures’; ‘lamp posts’ and ‘road crossings’ (full details in Additional file 1)
  4. cIncome deprivation score and crime score generated from the 2010 Index of Multiple Deprivation LSOA rank (IMD: https://www.gov.uk/government/statistics/english-indices-of-deprivation-2010) or the 2009 Scottish Index of Multiple Deprivation datazone rank (SIMD: http://www.gov.scot/Topics/Statistics/SIMD) to define relative deprivation of an area for England and Scotland respectively
  5. dArea walkability generated from street connectivity defined as the number of road nodes/interconnections per km2 within an LSOA/datazone obtained from 2015 Ordinance Survey (Digimap Meridian 2 National)
  6. ePopulation densitiy obtained from mid-year population estimates from 2010 from the Office of National Statistics (www.ons.gov.uk) and the Scottish Neighbourhood Statistics (www.sns.gov.uk). Estimates used to generate population density per km2 at the area level smoothed using a 5 km radius buffer