Table 3.

Odds ratio based on hierarchical linear modelling regression of being a drinker versus abstainer (Model 1) and unstandardised b coefficients based on hierarchical linear modelling regression of annual number of drinks consumed by drinkers (Model 2) on living with children (level 1), country level gender equality (reverse coded Gender Inequality Index × 10) (level 2) and cross-level interaction of gender equality with living with children (controlling for age)

	Men (n = 28 326)	Women (n = 35 585)
Model 1 Drinking status (drinker vs. abstainer)	Odds ratio (P-value)	Odds ratio (P-value)
Model 1a
Lives with children (level 1)	1.18 (P = 0.003)	0.90 (P = 0.036)
Gender equality (level 2)	1.60 (P <0.001)	2.11 (P <0.001)
Model 1b
Lives with children (level 1)	1.18 (P = 0.002)	0.90 (P = 0.039)
Gender equality (level 2)	1.59 (P <0.001)	2.18 (P <0.001)
Gender equality × lives with children (cross level interaction)	0.98 (P = 0.469)	1.02 (P = 0.649)
	Men (n = 22 415)	Women (n = 22 608)
Model 2 # drinks consumed annually by drinkers	b coefficient (P value)	b coefficient (P value)
Model 2 a
Lives with children (level 1)	−46.61 (P = 0.079)	−54.47 (P <0.001)
Gender equality (level 2)	−6.93 (P = 0.816)	−5.14 (P = 0.735)
Model 2b
Lives with children (level 1)	−38.65 (P = 0.117)	−53.64 (P = 0.002)
Gender equality (level 2)	−19.44 (P = 0.524)	−3.77 (P = 0.837)
Gender equality × lives with children (cross level interaction)	−32.76 (P = 0.036)	−2.39 (P = 0.868)