Birth cohort effects on the quantity and heritability of alcohol consumption in adulthood: a Finnish longitudinal twin study

Aims: To estimate birth cohort effects on alcohol consumption and abstinence in Finland and to test differences between birth cohorts in genetic and environmental sources of variation in Finnish adult alcohol use. Design: The Older Finnish Twin Cohort longitudinal survey study 1975–2011. Setting: Finland. Participants: A total of 26 121 same-sex twins aged 18–95 years (full twin pairs at baseline n = 11 608). Measurements: Outcome variables were the quantity of alcohol consumption (g/month) and abstinence (drinking zero g/month). Predictor variables were 10-year birth cohort categories and socio-demographic covariates. In quantitative genetic models, two larger cohorts (born 1901–20 and 1945–57) were compared. Findings: Multi-level models in both sexes indicated higher levels of alcohol consumption in more recent birth cohorts and lower levels in earlier cohorts, compared with twins born 1921–30 (all P < 0.003). Similarly, compared with twins born 1921–30, abstaining was more common in earlier and less common in more recent cohorts (all P < 0.05), with the exception of men born 1911–20. Birth cohort differences in the genetic and environmental variance components in alcohol consumption were found: heritability was 21% [95% confidence interval (CI) = 0–56%] in the earlier-born cohort of women [mean age 62.8, standard deviation (SD) = 5.3] and 51% (95% CI = 36–56%) in a more recent cohort (mean age 60.2, SD = 3.7) at the age of 54–74. For men, heritability was 39% (95% CI = 27–45%) in both cohorts. In alcohol abstinence, environmental influences shared between co-twins explained a large proportion of variation in the earlier-born cohort (43%, 95% CI = 23–63%), whereas non-shared environmental (54%, 95% CI = 39–72%) and additive genetic influences (40%, 95% CI = 13–61%) were more important among more recent cohorts of men and women. Conclusion: The contribution of genetic and environmental variability to variability in alcohol consumption in the Finnish population appears to vary by birth cohort.