Genetic and shared couple environmental contributions to smoking and alcohol use in the UK population

Alcohol use and smoking are leading causes of death and disability worldwide. Both genetic and environmental factors have been shown to influence individual differences in the use of these substances. In the present study we tested whether genetic factors, modelled alongside common family environment, explained phenotypic variance in alcohol use and smoking behaviour in the Generation Scotland (GS) family sample of up to 19,377 individuals. SNP and pedigree-associated effects combined explained between 18% and 41% of the variance in substance use. Shared couple effects explained a significant amount of variance across all substance use traits, particularly alcohol intake, for which 38% of the phenotypic variance was explained. We tested whether the within-couple substance use correlations were due to assortative mating by testing the association between partner polygenic risk scores in 34,987 couple pairs from the UK Biobank (UKB). Only couples' smoking status PRSs were significantly associated (b=0.01, S.E=0.005, p=0.02). However, an individual's alcohol PRS was associated with their partner's phenotype (b=0.04, S.E=0.007, p < 2 x 10-7). In support of this, G carriers of a functional ADH1B polymorphism (rs1229984), known to be associated with greater alcohol intake, were found to consume less alcohol if they had a partner who carried an A allele at this SNP. Together these results show that the shared couple environment contributes significantly to patterns of substance use. It is unclear whether this is due to shared environmental factors, assortative mating, or indirect genetic effects. Future studies would benefit from longitudinal data and larger sample sizes to assess this further.

Download data

• Downloaded 317 times
• Download rankings, all-time:
  • Site-wide: 53,948 out of 94,912
  • In genetics: 3,014 out of 4,824
• Year to date:
  • Site-wide: 84,063 out of 94,912
• Since beginning of last month:
  • Site-wide: 51,612 out of 94,912

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide