Confirmatory factor structure of anxiety and depression: evidence of item variance across childhood

Abstract

The distinctiveness of anxiety and depressive symptoms in children has previously been questioned based on their high degree of comorbidity, shared risk factors, and treatment response. Developing children may show an unstable presentation of anxiety and depressive symptoms that would complicate interpretation of studies of comorbidity. The present study examined the measurement stability of anxiety and depressive symptoms across time and sex using a large epidemiologic sample of children. A nationally representative cohort of 1329 children (624 girls and 705 boys) aged four to seven in 1994 were drawn from the National Longitudinal Survey of Youth (NLSY). Using eight years of prospective data we examined whether a one or two factor structure of anxiety (five items) and depressive (four items) symptoms would be invariant across time and sex. Despite item variability within each factor across time, confirmatory factor analysis revealed distinct factors for anxiety and depression that were stable across time and sex. Results provide support that covariation between anxiety and depression is not likely the result of measurement overlap. However, items indicating factors of anxiety and depression in the NLSY may not be sufficient to permit developmentally-sensitive measurement of these factors.