Predicting the General P-factor of Psychopathology using Lower Levels of the Personality Hierarchy
Background: Mental health disorders share a substantial amount of variance, reflecting a generalised vulnerability to any and all mental health problems. Studies on personality-psychopathology associations have previously been mainly focused at the domain-level of the personality hierarchy even though research has indicated that lower level personality traits (facets and nuances) capture valid unique variance beyond domains. The current study investigated the associations between the general ‘p-factor’ of psychopathology and multiple levels of the personality hierarchy in order to gain finer-grained insights into their relations. Methods: First, the structure of psychopathology was modelled using an exploratory bi-factor model of 23 items measuring symptoms of mental health problems using the DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure and the ASSIST questionnaire in a sample of 1,853 Estonian adults. Factor scores for the p-factor and orthogonal specific factors were estimated and elastic net regression models trained to examine the predictive ability of the different levels of the personality hierarchy for these factor scores.Results: A bi-factor model including a general factor and three specific factors representing internalising problems, thought disorders and substance use best represented the structure of psychopathology. Elastic net regression analyses indicated that personality traits related to the vulnerability, depression and immoderation facets were most strongly positively associated with the p-factor while traits related to the friendliness facet and the achievement-striving facet showed the strongest negative associations. Nuance-level analyses had the highest predictive accuracy for all psychopathology factors, particularly for thought disorders and substance use. Conclusion: Lower levels of the personality hierarchy contain additional information about psychopathology. Utilising this information opens up avenues for clinical applications that may help identify individuals most at risk for developing mental health disorders.