Despite multiple variables linked to psychosis through hypothesis-driven research, the one-exposure-to-one-outcome paradigm ignores the diversity of exposures. For a study, researchers sought to find real signals and systematic procedures required, comparable to agnostic genome-wide analysis. Through data-driven agnostic studies and genetically informed ways to analyze relationships, it is possible to examine nongenetic correlates of psychotic symptoms systematically.
From January 1 to June 1, 2021, cohort research examined data from the UK Biobank Mental Health Survey. In 2 equally sized split discovery and replication data sets, an exposome-wide association investigation was conducted. The exposome-wide analysis variables that were linked to psychotic events were put to the test in a multivariable model. The single-nucleotide variant-based heritability and genetic overlap with psychotic experiences using linkage disequilibrium score regression were estimated for the variables associated with psychotic experiences in the final multivariable model, and mendelian randomization (MR) approaches were used to test potential causality. Multiple sensitivity tests, including collider-correction MR, 2-sample MR, and multivariable MR analyses, were used to further examine the substantial relationships found in 1-sample MR analyses.
There were 1,55,247 participants in the research (87 896 [57%] female; mean [SD] age, 55.94 [7.74] years). About 162 variables (66%) in the discovery data set were linked to psychotic events. Around 148 (91%) of them were repeated. In about 36 factors were found in the multivariable analysis to be linked to psychotic events. About 28 of them exhibited a substantial genetic connection to psychotic events. Forward connections with three variables and reverse relationships with three were found via one-sample MR analysis. Sensitivity tests supported the forward associations with having ever experienced sexual assault, pleiotropy of risk-taking behavior, reverse associations without pleiotropy of having experienced a physically violent crime, cannabis use, and the reverse association with pleiotropy of worrying too long after embarrassment. As a result, connections between several correlated factors that have been extensively investigated and others that have not were discovered. In the final multivariable and MR analyses, the direction of the connection for a number of factors was reversed.
The study’s results highlighted the necessity of systematic techniques and the triangulation of evidence to create a knowledge foundation from continuously expanding observational data to inform population-level psychosis prevention initiatives.