Background Non-drinkers have been repeatedly shown to have worse health than moderate drinkers in later life particularly as regards conditions such as coronary heart disease, leading some researchers to suggest that moderate alcohol consumption has beneficial effects on health. However the causal direction between non-drinking and worse health has been contested. Whether poor health is associated with non-drinking among young adults in relation to social and health factors will be investigated. Such a finding would imply that poor health may precede non-drinking even at early stages of the life course.

Methods We performed logistic regression analysis of cross-sectional national survey data on 18 to 34 year olds, collected from The Health Survey for England 2006 and 2008. Data was collected through face to face interviews and is self-reported. Logistic regression analysis of longitudinal data collected from the National Child Development Study 1958 were also conducted. This is a follow up study to assess whether poor health at age 16 is associated with non-drinking at age 23, and whether these people are more likely to abstain at subsequent age sweeps.

Results HSE 06 & 08: Having a limiting longstanding illness during early adulthood increased the odds of being a non-drinker 1.74 times for men (p<0.01), and 1.45 times for women (p<0.01). In both men and women belonging to the lowest income quintile or having no qualifications was associated with increased odds of being a non-drinker (p<0.001) indicating that the social gradient in non-drinking begins at an early age. Men and women aged 18 to 34 years with the lowest activity levels were also more likely to be non-drinkers (p<0.01).

NCDS: Preliminary analysis reveals that non-drinkers at age 23 have higher rates of medical conditions at age 16 as assessed by a medical officer such as having a mental condition (p<0.001) a physical condition and heart and haematological condition (p<0.01).

Conclusion Young adults who have a limiting longstanding illness are more likely not to drink alcohol even after adjusting for a range of social and demographic measures. Studies on the putative health benefits of moderate alcohol consumption later in life need to take account of early life history. Further analysis using longitudinal data will explore whether poor health precedes non-drinking right at the start of drinking history and whether people with poor health continue to abstain from alcohol at older ages.