Skip to main content

Table 7 Tests for linear trend in log odds of exceeding heavy episodic drinking thresholds: adjusted excess odds ratios of moving down one SES level, 2008–2013 (95 % CI; p-value)

From: Unravelling the alcohol harm paradox: a population-based study of social gradients across very heavy drinking thresholds

  48 g+/64 g+ 96 g+/128 g+ 144 g+/192 g+ 192 g+/256 g+
Income 0.91 (0.89–0.93;<0.001) 0.98 (0.95–1.01; 0.187) 1.03 (0.98–1.09; 0.242) 1.03 (0.94–1.13; 0.560)
Education 0.93 (0.91–0.95;<0.001) 1.01 (0.98–1.04; 0.425) 1.09 (1.04–1.15;<0.001) 1.16 (1.05–1.27; 0.002)
Occupation 0.97 (0.95–0.98;<0.001) 1.03 (1.01–1.05; 0.004) 1.07 (1.03–1.10;<0.001) 1.10 (1.05–1.16;<0.001)
Deprivation 0.99 (0.97–1.01; 0.192) 1.05 (1.02–1.08;<0.001) 1.14 (1.09–1.20;<0.001) 1.19 (1.10–1.30;<0.001)
  1. Adjusted for age, sex, ethnicity and year of survey
  2. ‘Other’ categories were excluded from education and occupation
  3. Note that SES indicators have different numbers of levels. For example, ‘deprivation’ is based on quintiles, so an excess odds ratio of 1.19 represents an estimated odds ratio of 2.01 comparing bottom and top quintiles. Occupation has eight levels, so an excess odds ratio of 1.10 represents an estimated odds ratio of 1.95 between ‘unemployed’ and ‘higher managerial’
  4. Thresholds are grams of pure alcohol in one day for women/men