Model parameter estimates and predicted probabilities (%) for excess alcohol consumption and binge drinking for neighbourhood deprivation and SES
| Parameter estimate (SE) | Excess consumption, less than binge (%) | Binge (%) | |
|---|---|---|---|
| Model 1† | |||
| Neighbourhood deprivation quintiles | |||
| Least deprived | Reference | 22.2 | 9.7 |
| Less deprived | −0.2042* (0.0372) | 20.1 | 9.9 |
| Mid deprived | −0.4105* (0.0370) | 19.1 | 11.2 |
| More deprived | −0.6544* (0.0375) | 17.6 | 12.6 |
| Most deprived | −0.8526* (0.0391) | 17.2 | 12.6 |
| Interaction | |||
| Within to excess: less deprived | 0.2033* (0.0446) | ||
| Excess to binge: less deprived | 0.3254* (0.0565) | ||
| Within to excess: mid deprived | 0.5656* (0.0443) | ||
| Excess to binge: mid deprived | 0.7054* (0.0554) | ||
| Within to excess: more deprived | 0.9931* (0.0459) | ||
| Excess to binge: more deprived | 1.1510* (0.0563) | ||
| Within to excess: most deprived | 1.3587* (0.0489) | ||
| Excess to binge: most deprived | 1.3692* (0.0584) | ||
| Model 2‡ | |||
| Neighbourhood deprivation quintiles | |||
| Least deprived | Reference | 21.3 | 10.6 |
| Less deprived | −0.1973* (0.0387) | 19.5 | 11.1 |
| Mid deprived | −0.3879* (0.0386) | 18.8 | 13.0 |
| More deprived | −0.6073* (0.0395) | 17.5 | 15.3 |
| Most deprived | −0.7142* (0.0421) | 17.6 | 17.5 |
| Interaction | |||
| Within to excess: less deprived | 0.1954* (0.0470) | ||
| Excess to binge: less deprived | 0.3282* (0.0588) | ||
| Within to excess: mid deprived | 0.5720* (0.0467) | ||
| Excess to binge: mid deprived | 0.7296* (0.0577) | ||
| Within to excess: more deprived | 1.0157* (0.0483) | ||
| Excess to binge: more deprived | 1.2033* (0.0586) | ||
| Within to excess: most deprived | 1.3996* (0.0514) | ||
| Excess to binge: most deprived | 1.4615* (0.0608) | ||
| SES | |||
| Professional/managerial | Reference | 19.8 | 14.6 |
| Intermediate | −0.0973* (0.0265) | 19.0 | 13.0 |
| Routine occupations | −0.1519* (0.0226) | 18.6 | 12.2 |
| Never worked/long-term unemployed | −0.3339* (0.0614) | 17.1 | 9.7 |
*p<0.001.
†Model 1 included fixed effects terms for neighbourhood deprivation quintiles and the interaction with the binary transition explanatory variable for change in category of consumption, and random effects terms for household, LSOA and unitary authority.
‡Model 2 added social class, age group, gender, age group×gender, and adjusted for employment status, the highest educational qualification, ethnicity and housing tenure.
LSOA, lower super output areas; SES, socioeconomic status.