Aims
We estimated the degree to which the relationship between socio‐economic position (SEP) and alcohol‐related disorders is attenuated after adjustment for levels and patterns of drinking, behavioural, material and social factors.
Design
A longitudinal cohort study with baseline in 2002, with linkage to register data on patient care and deaths in 2002–11 to yield the outcome measures.
Setting
Stockholm County, Sweden.
Participants
Respondents to baseline survey aged 25–64 (n = 17 440) with information on all studied covariates.
Measurements
Occupational class was the studied SEP indicator and a combined measure of volume of weekly alcohol consumption and frequency of heavy episodic drinking, smoking, employment status, income, social support, marital status and education, all at baseline, were the studied covariates. Alcohol‐related disorders (n = 388) were indicated by first register entries on alcohol‐related medical care or death during the follow‐up.
Findings
Unskilled workers had an approximately four times greater risk of alcohol‐related disorders than higher non‐manual employees, hazard ratio (HR) = 4.08 (2.78, 5.98). After adjustment for alcohol use, the SEP difference in risk for alcohol‐related harm fell by a fourth for the same group, HR = 2.91 (1.96, 4.33). The difference was reduced further when behavioural factors and material factors were taken into account, HR = 2.09 (1.34, 3.26), whereas adjusting for social factors and attained education resulted in smaller reductions.
Conclusions
Socio‐economic differences in alcohol use explain one fourth of the socio‐economic position differences in alcohol‐related disorders in Stockholm, Sweden. Hazardous alcohol use and other behavioural, material and social factors together explain nearly 60% of the socio‐economic position differences in alcohol‐related disorders.