Skip to main content
. 2016 Sep 14;2:683–691. doi: 10.1016/j.ssmph.2016.09.006

Table 3.

Regression models for change in recreational physical activity between baseline and follow-up, according to quartile of greenspace.

Model 1 Adjusted for baseline PA (n=10852, adjusted R2 19.3%)
Model 2 Adjusted for baseline PA, age, sex, BMI, social class and marital status (n=10649, adjusted R2 21.7%)
95% CI
95% CI
Coeff. Lower Upper P P trend Coeff. Lower Upper P P trend
Quartile of greenspace:  quartile 1 (least green, ref) 1.00 <0.001 1.00 <0.001
 quartile 2 0.92 −0.75 2.59 0.279 0.87 −0.79 2.52 0.306
 quartile 3 2.02 0.35 3.69 0.018 1.53 −0.13 3.19 0.071
 quartile 4 (most green) 4.89 3.22 6.56 <0.001 4.03 2.36 5.71 <0.001
Baseline PA (MET h/wk) −0.52 −0.54 −0.50 <0.001 −0.53 −0.55 −0.51 <0.001
Age at 2HC (years) −0.50 −0.57 −0.43 <0.001
Sex (ref=female) 6.16 4.94 7.39 <0.001
BMI (kg/m2) −0.28 −0.43 −0.13 <0.001
Social class (ref=non-manual) −0.74 −1.97 0.48 0.233
Marital status (ref=not married) 2.33 0.76 3.91 0.004
Constant 16.79 15.45 18.13 <0.001 51.10 45.19 57.00 <0.001