RT Journal Article
SR Electronic
T1 Causal relationships between substance use and insomnia
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2020.04.06.027003
DO 10.1101/2020.04.06.027003
A1 Joëlle A. Pasman
A1 Dirk J.A. Smit
A1 Lilian Kingma
A1 Jacqueline M. Vink
A1 Jorien L. Treur
A1 Karin J.H. Verweij
YR 2020
UL http://biorxiv.org/content/early/2020/04/06/2020.04.06.027003.abstract
AB Background Poor sleep quality and insomnia have been associated with the use of tobacco, alcohol, and cannabis, but it is unclear if there is a causal link. In this Mendelian Randomization (MR) study we examine if insomnia causes substance use and/or if substance use causes insomnia.Methods MR uses summary effect estimates from a genome-wide association study (GWAS) to create a genetic instrumental variable for a proposed ‘exposure’ variable and then identifies that same genetic instrument in an ‘outcome’ GWAS. With data of GWAS of insomnia, smoking (initiation, heaviness, cessation), alcohol use (drinks per week, dependence), and cannabis initiation, bi-directional causal effects were tested. Multiple sensitivity analyses were applied to assess the robustness of the findings.Results There was strong evidence for positive causal effects of insomnia on all substance use phenotypes (smoking traits, alcohol dependence, cannabis initiation), except alcohol per week. The effects on alcohol dependence and cannabis initiation were attenuated after filtering out pleiotropic SNPs. In the other direction, there was strong evidence that smoking initiation increased chances of insomnia (smoking heaviness and cessation could not be tested as exposures). We found no evidence that alcohol use per week, alcohol dependence, or cannabis initiation causally affect insomnia.Conclusions There were unidirectional effects of insomnia on alcohol dependence and cannabis initiation, and bidirectional effects between insomnia and smoking measures. Bidirectional effects between smoking and insomnia might give rise to a vicious circle. Future research should investigate if interventions aimed at insomnia are beneficial for substance use treatment.insominsomniasmok initsmoking initiationsmok heavsmoking heavinesssmok cessmoking cessationalc weekalcohol use per weekalc depalcohol dependencecan initcannabis initiationIVWinverse variance weighted meta-analysisORodds ratioGSMRgeneralised summary-data-based Mendelian randomizationNSNPsnumber of SNPs that was retained in the analyses after filtering for high LD, palindromic, and ambiguous SNPs, with additional HEIDI filtering in the GSMR and filtering for pleiotropic SNPs in the Steiger analyses.