Alcohol use and cognitive functioning in young adults: improving causal inference

Abstract

Background and aims: There have been few longitudinal studies of association between alcohol use and cognitive functioning in young people. We aimed to examine whether alcohol use is a causal risk factor for deficient cognitive functioning in young adults.

Design: Linear regression was used to examine the relationship between longitudinal latent class patterns of binge drinking and subsequent cognitive functioning. Two-sample Mendelian randomization (MR) tested evidence for the causal relationship between alcohol use and cognitive functioning.

Setting: South West England.

Participants: The observational study included 3155 adolescents and their parents (fully adjusted models) from the Avon Longitudinal Study of Parents and Children (ALSPAC). Genetic instruments for alcohol use were based on almost 1 000 000 individuals from the genome-wide association studies (GWAS) and Sequencing Consortium of Alcohol and Nicotine use (GSCAN). Genome-wide association studies for cognitive outcomes were based on 2500 individuals from ALSPAC.

Measurements: Binge drinking was assessed at approximately 16, 17, 18, 21 and 23 years. Cognitive functioning comprised working memory, response inhibition and emotion recognition assessed at 24 years of age. Ninety-nine independent genome-wide significant single nucleotide polymorphisms (SNPs) associated with 'number of drinks per week' were used as the genetic instrument for alcohol consumption. Potential confounders were included in the observational analyses.

Findings: Four binge drinking classes were identified: 'low-risk' (41.3%), 'early-onset monthly' (19.1%), 'adult frequent' (22.5%) and 'early-onset frequent' (17.0%). The association between early-onset frequent binge drinking and cognitive functioning: working memory (b = -0.42, 95% confidence interval (CI) = -1.24 to 0.41), response inhibition (b = 31.9, 95% CI = -25.3 to 89.2), and emotion recognition (b = 0.02, 95% CI = -0.07 to 0.10) in comparison to low-risk drinkers were inconclusive as to whether a difference was present. Two-sample MR analyses similarly provided little evidence that alcohol use is associated with deficits in working memory using the inverse variance weight (b = 0.29, 95% CI = -0.42 to 0.99), response inhibition (b = -0.32, 95% CI = -1.04 to 0.39) and emotion recognition (b = 0.03, 95% CI = -0.55 to 0.61).

Conclusions: Binge drinking in adolescence and early adulthood may not be causally related to deficiencies in working memory, response inhibition or emotion recognition in youths.

Keywords: ALSPAC; Alcohol; Mendelian randomization; cognition; longitudinal latent class analysis; prospective.