Abstract:Abstract IMPORTANCE Clinical epidemiological studies have found high rates of comorbidity between suicide attempts (SA) and opioid use disorder (OUD). However, the patterns of correlation and causation between them are still not clear due to psychiatric confounding. OBJECTIVE To investigate the pairwise associations and interrogate the potential bidirectional relationship between OUD and SA using genetically based methods. DESIGN, SETTING, AND PARTICIPANTS We utilized raw phenotypes and genotypes from UK Biobank, and summary statistics from Million Veteran Program, Psychiatric Genomic Consortium, iPSYCH, and International Suicide Genetics Consortium. Statistical genetics tools were used to perform epidemiological association, genetic correlation, polygenic risk score prediction, and Mendelian randomizations (MR). Analyses were conducted to examine the OUD-SA relationship with and without controlling for psychiatric disease status (e.g., major depressive disorder [MDD]). MAIN OUTCOMES AND MEASURES OUD and SA with or without major psychiatric disorders (schizophrenia, bipolar disorder, major depressive disorder, and alcohol use disorder). RESULTS Strong correlations between OUD and SA were observed at both phenotypic level (overall samples [OR=2.94, P =1.59 ×10 −14 ]; non-psychiatric subgroup [OR=2.15, P =1.07 ×10 −3 ]) and genetic level (r 2 =0.4 and 0.5 with or without conditioning on MDD). The higher genetic susceptibility to SA can increase the polygenic risk of OUD (OR=1.08, false discovery rate [FDR] =1.71 ×10 −3 ), while the higher susceptibility to OUD can also increase the risk of SA (OR=1.09, FDR =1.73 ×10 −6 ). However, predictive abilities for both were much weakened after controlling for influence of psychiatric diseases. A combination of different MR analyses suggested a possible causal association from SA to OUD (2-sample univariable MR: OR=1.14, P = 0.001; multivariable MR: OR=1.08, P = 0.001). CONCLUSIONS AND RELEVANCE This study provided new genetic evidence underlying the strong OUD-SA comorbidity. While controlling for the influence of psychiatric diseases, there is still some clue on possible causal association between SA genetic liability and the risk for OUD. Future prevention strategy for each phenotype needs to take into consideration of screening for the other one. Key Points Question Does opioid use disorder (OUD) have a potentially causal role in the risk for suicide attempts (SA), or vice versa? Findings In this observational study with raw data from >150,000 UK Biobank samples and genome-wide association summary statistics from >600,000 individuals, positive phenotypic and genetic associations were observed between OUD and SA, whether or not controlling for major psychiatric disorders. Comprehensive Mendelian randomization analyses (one-sample and two-sample) suggested the genetic liability for SA was associated with increased risk for OUD. However, they were still underpowered to reveal the putative causal association from OUD to SA. Meaning This genomics-based study supports a strong genetic association underlying the OUD-SA comorbidity. Though both phenotypes are intertwined with other psychiatric disorders, there also exists an independent bidirectional relationship between OUD and SA.

Understanding the Causal Relationships Between Opioid Dependence and Risk of Mental Disorders: A Comprehensive Two-Sample Mendelian Randomization Study

Genetic Liability for Prescription Opioid Use and Risk of Cardiovascular Diseases: a Multivariable Mendelian Randomization Study

Exploring the Causal Relationship Between Migraine and Insomnia Through Bidirectional Two-Sample Mendelian Randomization: A Bidirectional Causal Relationship

Cross-phenotype relationship between opioid use disorder and suicide attempts: new evidence from polygenic association and Mendelian randomization analyses (vol 28, pg 2913, 2023)

Exploring the bidirectional relationship between pain and mental disorders: a comprehensive Mendelian randomization study

Substance abuse and the risk of severe COVID-19: Mendelian randomization confirms the causal role of opioids but hints a negative causal effect for cannabinoids

Causal effects and immune cell mediators between prescription analgesic use and risk of infectious diseases: a Mendelian randomization study

Assessing the causal relationship between psychiatric disorders and obstructive sleep apnea: a bidirectional Mendelian randomization

Mendelian randomization highlights sleep disturbances mediated the effect of depression on chronic pain

Bidirectional causal relationship between psychiatric disorders and osteoarthritis: A univariate and multivariate Mendelian randomization study

The relationship between major depression and migraine: A bidirectional two-sample Mendelian randomization study

Cross-phenotype relationship between opioid use disorder and suicide attempts: new evidence from polygenic association and Mendelian randomization analyses

Correction: Cross-phenotype relationship between opioid use disorder and suicide attempts: new evidence from polygenic association and Mendelian randomization analyses

Mediators of the association between depression and migraine: a mendelian randomization study

Causal association between mental disorders and cerebrovascular diseases: Evidence from Mendelian randomization study

Genetically supported causality between gut microbiota, immune cells and morphine tolerance: a two-sample Mendelian randomization study

Mendelian randomization analyses identify bidirectional causal relationships of obesity with psychiatric disorders

The causal relationship between depression and obstructive sleep apnea: A bidirectional Mendelian randomization study

Causal links between major depressive disorder and insomnia: A Mendelian randomisation study

Mendelian randomization analyses support causal relationships between brain imaging-derived phenotypes and risk of psychiatric disorders

Investigating the Mediating Role of Mental Disorders in the Relationship Between Early Sexual Intercourse and Intentional Self‐Harm: A Two‐Step and Multivariable Mendelian Randomization Study