Mood and anxiety profiles differentially associate with physical conditions in US adolescents.

Abstract Background Mood and anxiety are widely associated with physical conditions, but research and treatment are complicated by their overlap, clinical heterogeneity, and manifestation on a spectrum rather than as discrete disorders. In contrast to previous work relying on threshold-level disorders, we examined the association between empirically-derived profiles of mood and anxiety syndromes with physical conditions in a nationally-representative sample of US adolescents. Methods Participants were 2,911 adolescents (aged 13-18) from the National Comorbidity Survey-Adolescent Supplement who provided information on physical conditions and reported at least one lifetime mood-anxiety ‘syndrome’ based on direct interviews with the Composite International Diagnostic Interview Version 3.0. Mood-anxiety syndromes reflected 3-level ratings from subthreshold to severe distress/impairment, and subtyped mood episodes. Stepwise latent profile analysis identified mood-anxiety profiles and tested associations with physical conditions. Results Three mood-anxiety profiles were identified: “Mood-GAD” (25.6%)—non-atypical depression, mania, generalized anxiety; “Atypical-Panic” (11.3%)—atypical depression, panic; and “Reference” (63.1%)—lower mood and anxiety except specific phobia. Headaches were more prevalent in Mood-GAD and Atypical-Panic than Reference (47.9%, 50.1%, and 37.7%, respectively; p=0.011). Heart problems were more common in Mood-GAD than Atypical-Panic (7.4% v 2.2%, p=0.004) and Reference, with back/neck pain more prevalent in Mood-GAD than Reference (22.5% v 15.3%, p=0.016). Limitations Broad categories of physical conditions without information on specific diagnoses; replication regarding specificity is recommended. Conclusions Heart problems and pain-related conditions were differentially associated with specific mood-anxiety profiles. Subtyping depression and anxiety—inclusive of subthreshold syndromes—and their patterns of clustering may facilitate etiologic and intervention work in multimorbidity.