Preschool Anxiety Disorders: Comprehensive Assessment of Clinical, Demographic, Temperamental, Familial, and Life Stress Correlates

Abstract

This study examined correlates of preschoolers' anxiety disorders using a comprehensive, multimethod design. Participants included a community sample of 541 three-year-old children, of whom 106 (19.6%) met criteria for at least 1 anxiety disorder. Child and parental psychopathology and life stress were assessed with clinical interviews. Child temperament and parenting behavior were assessed with laboratory observations. Mothers and fathers reported on their parenting styles. Compared to preschoolers with no anxiety disorder, preschoolers with an anxiety disorder were more likely to meet criteria for comorbid depressive and oppositional defiant disorders and to exhibit greater temperamental behavioral inhibition and lower positive affectivity, and more sleep problems. Children with anxiety disorders also experienced more stressful life events in the previous 6 months, and their mothers had a higher rate of current anxiety disorders. Compared to children with other anxiety disorders, children with only specific phobia exhibited a somewhat different pattern of associations than children with other anxiety disorders. Overall, the findings suggest that many of the correlates observed in older youth with anxiety disorders are also observed in preschoolers.

Acknowledgments

Authors report no conflicts of interest. This work was supported by the following grants: NIMH grants RO1 MH069942 (DNK) and a GCRC Grant no. M01-RR10710 to Stony Brook University from the National Center for Research Resources.

Notes

Note: Of parents who participated in the observational assessment of parenting behaviors, 93.5% were mothers; temperament variable means and standard deviations are reported as z scores. PSDQ = Parenting Styles and Dimensions Questionnaire.

^a n = 435.

^b n = 106.

^c Fisher's exact test.

^d Remained significant after Benjamini and Hochberg's (1995) procedure to correct for multiple comparisons within domains was employed. Benjamini and Hochberg-adjusted p values for statistical significance were as follows: demographic variables p < .006; child psychopathology p < .038; child temperament p < .033; parental psychopathology p < .025; observed parenting behavior p < .017; maternal PSDQ p < .033; paternal PSDQ p < .017; and stressful life events p < .05 (no adjustment necessary).

Note: OR = odds ratio; CI = confidence interval; PSDQ = Parenting Styles and Dimensions Questionnaire.

†p < .10. *p < .05. **p < .01.

¹We did not include paternal reports of permissiveness in the model, as its inclusion reduced the sample from 466 to 376 children. When we included paternal reports of permissiveness in the model, the results were similar and paternal permissiveness was not significant (OR = 1.07, 95% CI = .98–1.18, p = .114).