Skip to main content
. 2016 Aug 11;71(2):137–145. doi: 10.1136/jech-2016-207753

Table 3.

Growth models. 2009–2014 alcohol-related recorded sex crime rate (per 1000 people)

Parameter Unadjusted (SE) Adjusted 1 (SE) Adjusted 2 (SE)
Linear trend Quadratic trend Linear trend Quadratic trend Linear trend Quadratic trend
Baseline (year 2009) −2.136 (0.051) −2.476 (0.074) −1.881 (0.136)
Trend (2009–2014) −0.144 (0.013) 0.020 (0.001) −0.144 (0.013) 0.020 (0.001) −0.388 (0.037) 0.043 (0.003)
p<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001
Effect on intercept
 No policy Ref Ref Ref
 Low policy −0.185 (0.100) −0.218 (0.093) −0.167 (0.092)
 Medium 0.116 (0.079) 0.043 (0.074) 0.149 (0.075)
 High policy 0.481 (0.095) 0.245 (0.091) 0.488 (0.099)
p<0.001 p<0.001 p<0.001
Deprivation at baseline 2.361 (0.240) 3.074 (0.535)
p<0.001 p<0.001
Population at baseline −0.013 (0.014) −0.204 (0.032)
p=0.373 p<0.001
Outlet density at baseline 0.003 (0.001) 0.002 (0.001)
p<0.001 p=0.104
Alcohol-related hospital admissions at baseline −0.001 (0.000) −0.004 (0.001)
p=0.101 p<0.001
Effects on slope
 Low policy 0.095 (0.025) −0.008 (0.025) 0.095 (0.023) −0.008 (0.002) 0.078 (0.025) −0.007 (0.002)
 Medium 0.034 (0.020) −0.004 (0.002) 0.034 (0.020) −0.004 (0.002) −0.003 (0.020) −0.001 (0.002)
 High policy −0.048 (0.024) 0.004 (0.002) −0.048 (0.024) 0.004 (0.002) −0.136 (0.027) 0.012 (0.002)
p<0.001 p<0.001 p<0.001 p<0.001 p<0.001 p<0.001
Deprivation at baseline −0.319 (0.146) 0.032 (0.014)
p=0.029 p=0.019
Population at baseline 0.062 (0.087) −0.005 (0.001)
p<0.001 p<0.001
Outlet density at baseline 0.001 (0.000) −0.000 (0.000)
p=0.060 p=0.011
Alcohol-related hospital admissions at baseline 0.001 (0.000) −0.000 (0.000)
p<0.001 p<0.001
Variance Estimates Explained† Estimates Explained† Estimates Explained†
 Intercept 0.218 9% 0.178 25% 0.172 28%
 Slope 0.003 0.3% 0.003 0.3% 0.003 5%
 Residual 0.028 0.4% 0.028 0.4% 0.028 1.5%
 R2 84.1% 84.1% 84.3%

Bold indicates p<0.05.

All models also adjusted for quarter (ie, 3-month period).

†Explained variance relative to model without explanatory variables (ie, only time trends as fixed effects).