Self-reported life satisfaction and alcohol use: a 15-year follow-up of healthy adult twins.

AIMS To study the bidirectional relationships between life satisfaction (LS) and alcohol use. METHODS Health questionnaires were administered in 1975, 1981 and 1990 to a population-based sample of healthy Finnish twins aged 18-45 at baseline (n = 14,083). These included a LS scale and three indicators for adverse alcohol use: binge drinking, passing out and high consumption (women/men ≥400/800 g/month). In longitudinal analyses, logistic regression, pair-wise case-control analyses and growth models were applied. RESULTS All alcohol indicators increased the age-adjusted risk of becoming dissatisfied regardless of study period [binge drinking odds ratio (OR)(1975-1990 )= 1.29; 95% confidence interval (CI) 1.12-1.50; high consumption OR(1975-1990 )= 1.60; 1.29-1.99 and passing out OR(1981-1990 )= 2.01; 1.57-2.57]. Also, the dissatisfied had an increased subsequent risk for adverse alcohol use. The risk for passing out due to drinking (OR(1975-1990 )= 1.50; 1.22-1.86) was increased regardless of study period, while high consumption (OR(1975-1981 )= 1.97; 1.40-2.77; OR(1981-1990 )= 2.48; 1.50-4.12) and binge drinking (OR(1975-1981 )= 1.37; 1.12-1.67) showed some variation by the study period. Predictions remained after multiple adjustments. Longitudinally, high consumption predicted dissatisfaction somewhat more strongly than vice versa. The change/levels within the whole range of LS and alcohol consumption were only slightly associated in the entire study population. CONCLUSION Life dissatisfaction and adverse alcohol use reciprocally predict each other prospectively. The heavier the alcohol use the stronger the relationship.