Background: Anxiety disorder (AD) and substance use disorder (SUD) highly co-occur with bipolar disorder (BD). AD and/or SUD co-occurrence is associated with poorer clinical outcomes in BD. However, respective associations between AD and/or SUD diagnoses and BD outcomes require clarification. Baseline data from the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) were therefore utilized to investigate independent and simultaneous contributions of ADs and SUDs on clinical variables in BD. Methods: Two latent factors, "pathological anxiety" and "substance use problems," were derived from presence/absence of lifetime AD and SUD diagnoses. Latent dimensions' associations with clinical variables, obtained from the Affective Disorders Evaluation, Mini-International Neuropsychiatric Interview and Range of Impaired Functioning, were estimated via structural equation modeling (SEM). Results: Modeled independently, pathological anxiety and substance use problems were significantly associated with several variables. Yet when modeled simultaneously, pathological anxiety's associations with functional impairment, past-year rapid cycling, and past-year %time spent anxious and depressed remained while most variables' associations with substance use problems became non-significant. The only significant latent-factor interaction evidenced was for age of BD onset.

This cross-sectional study examined behavioral and physiological indicators of the parent-child relationship as moderators of the link between maternal depressive and child psychopathology symptoms. Ninety-seven mothers (M age ϭ 35.38 years) and their 9-to 12-year-old children (M age ϭ 10.32 years, 53.7% girls, 78.1% African American) from economically impoverished backgrounds participated. Mothers reported on their own depressive symptoms using the Beck Depression Inventory-2 (BDI-2) and their child's emotional and behavioral problems using the Child Behavior Checklist (CBCL). Physiological activity (i.e., respiratory sinus arrhythmia) of both mother and child was collected during a dyadic conflict discussion task, and interactions were coded independently for mother and child positivity, negativity, and engagement. Physiological synchrony was computed using multilevel modeling, and dyad-level behavioral indicators (i.e., dyadic positivity, negativity, and engagement) were created by averaging individual mother and child behavioral codes. Moderation analyses indicated that the link between mother and child symptoms was exacerbated for dyads who displayed low levels of positivity and low levels of engagement, as well as in the context of positive physiological synchrony. In contrast, high levels of positivity and engagement and negative physiological synchrony buffered the link between symptoms. Findings have implications for conceptual models of intergenerational risk for psychopathology and for clinical prevention and intervention efforts that target the parent-child relationship.