Structural Models of Comorbid Anxiety and Depression in a Primary-Care Older Adult Sample: Effect of Medical Illness Severity, Threat, Chronicity, and Progressiveness on Model Fit

by Palmer, William Michael

Recent research suggests that anxiety disorders may be more common in later life than previously thought. Among other factors, the presence of comorbid mood disorders and medical illness confounds accurate assessment and diagnosis of these conditions in the elderly. There have been few studies, however, examining the structural relationships between anxiety and depression with older-adult samples, and even fewer have considered the effect of medical illness on these relationships. This study examined three established structural models of anxiety and depression, using a clinical sample of older adults seeking treatment in a primary-care setting (N = 2,163). It was hypothesized that the presence of comorbid medical illness would act as a moderating variable in evaluating the fitness of these models. Results indicated that a hierarchical model represented the most parsimonious fit to the full sample. Tests of factorial invariance revealed variance in model fit as a function of illness severity and threat, and as a function of illness chronicity and progressiveness. Specifically, the relationship of somatic symptoms to anxiety varied by combined severity/threat, as well as by chronicity/progressiveness. These findings support previous conceptualizations of the relationship between anxiety and depression. Implications of these results for taxonomy, assessment, and intervention are discussed. Palmer, William Michael, 2007, UMSL p. 3