General psychopathology and substance use disorders: An indicator of functional impairment

IntroductionResearch represented in the Hierarchical Taxonomy of Psychopathology (HiTOP) suggests that broader, dimensional constructs of Internalizing (INT; Fear and Distress), Externalizing (EXT; Disinhibition and Antagonism), and Thought Disorder (TD) measures better characterize the structure of common mental health disorders. Shared variance across these dimensions has long been noted, suggesting an even broader general psychopathology factor (p-factor). Polysubstance use (PSU) is defined as the use of two or more addictive drugs simultaneously or concurrently and polysubstance use disorder (PSUD) is the presence of two or more substance use disorders (SUDs). PSU is associated with broad increases across problem behaviors and psychopathology. Objective. The goal of the present project was to identify the relationship between different underlying HiTOP factors and SUDs/PSUDs across a treatment sample and a nationally representative sample. MethodsThe first sample contained 2617 participants from a residential substance use treatment center in the DC area, serving primarily low-income, African American population. The second sample, National Comorbidity Survey-Replication (NCS-R), included 5,692 adults, aged 18 and older, selected using a multistage stratified cluster sampling design representative of the US household population (Kessler et al. 2004). To accomplish the primary objective, SUDs were separated from the rest of the DSM diagnoses, forming a separate SUDs/PSUDs latent factor. Then the HiTOP model was constructed from the remaining diagnoses leaving the subfactors TD, Fear, Distress, and Antagonism, in addition to the p-factor. Structural equation modeling, completed in MPlus software, was used to develop the latent HiTOP and SUDs/PSUDs factors and conduct mediation models. ResultsDirect effects indicated robust relationships between SUDs/PSUDs and the p-factor, as well as each of the sub-factors. Mediation models indicated that p-factor fully mediated the relationship between SUDs/PSUDs and TD and Distress, as well as the majority of the relationship between SUDs/PSUDs and Fear and Antagonism. Indirect effects indicated that the relationship between Antagonism and SUDs/PSUDs remained significantly positive relative to p-factor as a mediator, and Fear maintained some negative prediction in the treatment sample but was fully mediated in the NCS-R data. DiscussionConcisely, there is a similar and robust relationship between the latent HiTOP p-factor measure and SUDs/PSUDs across the samples. Beyond the broad p-factor direct effects accounting for most of SUDs/PSUDs, indirect effects indicate that there is some unique contributions to SUDs/PSUDs from Antagonism (positive) and Fear (negative). Results support the inference of a practically important and statistically robust relationship between mental health and substance use, as well as the presence of smaller differential contributions to SUDs/PSUDs from INT and EXT.