Bi-directional effects between loneliness and substance use: Evidence from a Mendelian randomisation study

Aims Loneliness and social isolation are associated with cigarette smoking and problematic alcohol use. Observational evidence suggests these associations arise because loneliness increases substance use, however there is potential for reverse causation (problematic drinking causing damage to social networks, leading to loneliness). With conventional epidemiological methods, controlling for (residual) confounding and reverse causality is difficult. In this study, we apply Mendelian randomisation (MR) to assess bi-directional causal effects between loneliness on the one hand and smoking behaviour and alcohol (ab)use on the other. Design We applied bi-directional MR using summary-level data of the largest available genome-wide association studies of loneliness (n=511,280), smoking (initiation (n=249,171), cigarettes-per-day (n=249,171) and cessation (n=143,852)), alcoholic drinks-per-week (n=226,223) and alcohol dependence (n=46,568), using independent samples. For each relationship, we selected genetic variants predictive of the exposure variable as instruments and tested their association with the outcome variable. Effect estimates for individual variants were combined with inverse-variance weighted regression (gene-outcome/gene-exposure association) and the robustness of these findings was assessed with five different sensitivity methods. Findings There was weak evidence of increased loneliness leading to higher likelihood of initiating smoking and smoking more cigarettes, and a lower likelihood of quitting smoking. Additionally, there was evidence that initiating smoking increases loneliness. We found no evidence of a causal effect between loneliness and alcohol (ab)use. Conclusions We report tentative evidence for causal, bidirectional, increasing effects between loneliness and cigarette smoking. These findings improve our understanding of the interrelatedness of smoking and loneliness, however, replication with better powered genetic instruments is recommended.