CitationLannoy, Severine; Stephenson, Mallory; & Edwards, Alexis (2023). The Roles of Aggregate Genetic Liability and Early Adversity In Risk for Suicidal Thought and Behavior. European Neuropsychopharmacology. vol. 75
AbstractBackground: Suicidal thoughts and behaviors are public health priorities, with a need to better understand suicide-related risk factors. Recent well-powered genetic studies (Genome-Wide Association Studies [GWAS]) have provided evidence of loci associated with suicidal behavior in different ancestry groups. This allows further explorations of genetic liability by the use of polygenic scores [PGS], such as looking at potential interactions between genetic and behavioral/environmental risk factors. The aim of this study is to evaluate how genetic risk and early adversity are implicated in the prediction of suicidal thoughts and behaviors.
Methods: We used data from the National Longitudinal Study of Adolescent to Adult Health, with information on suicide ideation and attempts collected between the end of adolescence to the middle of adulthood. We relied on the largest available GWAS on suicide attempts to compute PGS for participants from European and African ancestries (EA and AA) using PRS-CSx. Early adversity was documented by two groups of variables: (i) information about the home environment: availability of tobacco, alcohol, drugs, and guns; (ii) neglect and abuse: neglect of basic needs, physical abuse (hit by parents), sexual abuse (touch by parents), and intervention of social services. Logistic regressions were conducted separately for EA (N=5720) and AA (N=1950) participants. PGS were adjusted for the 10 first principal components, and analyses controlled for age and biological sex. We evaluated interactions between PGS and environment for variables reaching nominal significance (p < .05) in the multivariable model. Results: In the AA subgroup, none of the risk factors reached statistical significance. In the EA subgroup, multivariable analyses indicated that suicide attempt was related to PGS (OR=1.22, 95% CI = 1.03;1.44), availability of tobacco in the home environment (OR=1.54, 95% CI=1.06;2.23) as well physical abuse (OR=1.71, 95% CI=1.11;2.59), sexual abuse (OR=2.13, 95% CI=1.02;4.16), and intervention of social services (OR=2.58, 95% CI=1.39;4.65). Analyses with suicide ideation as an outcome in the EA subgroup supported and extended those results, showing the role of PGS for suicide attempts, availability of tobacco, alcohol, and drugs in the home environment, as well as physical abuse, sexual abuse, and intervention of social services. However, we found no evidence of gene by environment interaction.
Discussion: Results in EA participants demonstrate that both aggregate genetic liability and specific environmental factors contribute to risk for suicide ideation and attempt. We did not observe significant associations in AA participants, likely due to the low number of individuals endorsing ideation or attempts. Our findings suggest that beyond the experience of abuse, adversity in the home environment, such as substance availability, is associated with increased risk of suicidal thoughts and behaviors. These results also support the importance of considering genetic liability in suicide research, though we did not find evidence that participants’ genetic liability to suicide attempt exacerbated the adverse effects of environmental exposures. It is worth noting that PGS were computed based on a GWAS of mostly adult participants, and that genetic liability may vary across the life course. One step forward would be to evaluate the interaction between genetic risk specific to children/adolescents and early adversity.
Reference TypeJournal Article
Journal TitleEuropean Neuropsychopharmacology