The causal interplay between depression and alcohol use from adolescence to young adulthood: a Mendelian randomization study

Abstract Background Depression is often comorbid with alcohol use problems, and sex differences may further complicate this interplay. Methods We conducted a longitudinal study using a large European adolescent cohort assessed at ages 14 (baseline, BL), 16 (follow-up 1, FU1), 19 (follow-up 2, FU2), and 23 (follow-up 3, FU3). Depression and alcohol use were measured using standardized behavioral scales. Cross-lagged analysis, improved Mendelian randomization (MR) analysis, and mediation analysis were conducted to infer the causal interplay. Results 2110 adolescents were included at baseline (49% male). Depression and alcohol consumption demonstrated a significant positive correlation ( r BL = 0.094, p BL = 1.58E-05, 95% CI = [0.052, 0.137]), which gradually diminished over time and eventually became significantly negative. Depression and alcohol use problems remained strongly correlated across three timepoints ( r > 0.074, p < 6.76E-03). Cross-lagged analysis suggested that depression predicted future alcohol use problems: β BL-FU1 = 0.058, p = 0.021, 95% CI = [0.009, 0.108]; β FU2-FU3 = 0.142, p = 8.34E-07, 95% CI = [0.113, 0.263]. MR analyses confirmed this causal interplay ( r mean = 0.043, longitudinal p permuation < 0.001). Interestingly, MR analyses also indicated that alcohol consumption might alleviate depression ( r mean = −0.022, longitudinal p permutation = 0.043), particularly in females at FU3, of which the anxiety status and the personality trait neuroticism largely mediated the effect. These findings were validated in an independent matched sample (N = 562) from Human Connectome Project. Conclusions Depression may predict future alcohol use problems, whereas moderate alcohol consumption might alleviate depressive symptoms, especially in females.