Introduction Life expectancy in Russia among men is extremely low for an industrialised country (62 years in 2008), which is partly explained by hazardous alcohol consumption. We investigated the relationship between socio-demographic variables and dimensions of alcohol use.

Methods In 2008–2009 a population sample of 996 men aged 25–59 years living in Izhevsk, Russia took part in a survey that collected information on socio-demographic circumstances and a self-completed questionnaire that included the WHO Alcohol Use Disorders Identification Test (AUDIT). Latent dimensions of the total AUDIT score were determined using confirmatory factor analysis (CFA) and structural equation modelling was used to estimate the strength of association of these dimensions with socio-demographic variables.

Results Confirmatory factor analysis supported a two factor structure for the AUDIT: alcohol consumption and alcohol-related problems. Consumption was highest among men who were unemployed seeking work (p=0.001) but was not related to education (p=0.91) or marital status (p=0.76). The dimension capturing alcohol-related problems was greatest among men who were unemployed seeking work (p<0.001) and those who were less educated (p=0.02) but was not associated with marital status (p=0.43).

Conclusions The main socio-demographic variables available in this study were not associated with the two dimensions of the AUDIT in the same way. This together with the failure to show an association of either AUDIT dimension with marital status is surprising and warrants further investigation. It may indicate limitations in the sensitivity of AUDIT to important aspects of Russian drinking patterns.