Maternal phobic anxiety and child anxiety

Abstract

The present study examined the relation between maternal anxiety symptoms and child anxiety symptoms and evaluated whether a reporting bias is associated with maternal anxiety. Fifty-seven mother–child pairs participated. All children had features or diagnoses of separation anxiety disorder (SAD), generalized anxiety disorder, and/or social phobia. Measures of maternal symptomatology and child anxiety were administered. Higher levels of maternal phobic anxiety on the Brief Symptom Inventory were significantly associated with higher levels of separation anxiety in children. After controlling for clinician rating of SAD severity, maternal phobic anxiety emerged as a significant predictor of maternal ratings of child separation anxiety, accounting for 19% of the variance. Phobic mothers endorsed levels of separation anxiety in their children that exceeded levels endorsed by clinicians, suggesting maternal overreporting.

Introduction

It is widely accepted that anxiety disorders are among the most prevalent psychiatric disorders in children. Up to 15–20% of the general youth population has an anxiety disorder (Costello, Egger, & Angold, 2004). These disorders are strongly associated with risk for later developing mood disorders, other anxiety disorders, academic failure, and substance abuse problems (Ialongo, Edelsohn, Werthamer-Larsson, Crockett, & Kellam, 1995; Pine, Cohen, Gurley, Brooks, & Ma, 1998; Woodward & Fergusson, 2001). Several well-designed, controlled studies have documented that children of parents with anxiety disorders are at increased risk of developing anxiety disorders compared to children of normal control parents or children of parents with other psychiatric disorders (Beidel & Turner, 1997; Biederman et al., 2001; McClure, Brennan, Hammen, & Le Brocque, 2001; Merikangas, Dierker, & Szatmari, 1998; Merikangas, Avenevoli, Dierker, & Grillon, 1999).

Biederman et al. (2001) reported that children of parents with panic disorder and/or major depression were at increased risk for separation anxiety disorder (SAD). McClure et al. (2001) reported that maternal lifetime history of anxiety disorder more than doubled children's risk of anxiety disorder; this risk tripled for children of mothers with anxiety and comorbid depression. Similarly, Merikangas et al. (1999) reported that children with one parent with an anxiety disorder were three times more likely to have overanxious disorder and an additional threefold risk when both parents had an anxiety disorder. Studies demonstrate that transmission of anxiety from parent to child tends to be fairly specific with anxious parents having anxious children (Beidel & Turner, 1997; Merikangas et al., 1998).

Recently, maternal phobic anxiety has emerged above other forms of maternal anxiety as significantly related to child anxiety (Layne, Bernstein, & Fadness, 2002; Manassis & Hood, 1998). Phobic anxiety, as measured by the Brief Symptom Inventory (BSI; Derogatis, 1993) (SCL-90; Derogatis, 1994), “is defined as a persistent fear response-to a specific person, place, object, or situation—that is irrational and disproportionate to the stimulus and leads to avoidance or escape behavior” (Derogatis, 1993, p. 9). The Phobic Anxiety dimension focuses on phobic behavior rather than the internal state of anxiety. Elevated scores on Phobic Anxiety are typically associated with marked avoidant behavior; items on the scale describe impairing manifestations of phobic behavior (Derogatis, 1993). The content of the items relate closely to DSM-IV criteria for agoraphobia. The items on the Phobic Anxiety dimension are as follows: “Feeling afraid in open spaces or on the streets, Feeling afraid to travel on buses, subways, or trains, Having to avoid certain things, places, or activities because they frighten you, Feeling uneasy in crowds, such as shopping or at a movie, and Feeling nervous when you are left alone.”

Elevation on the Phobic Anxiety dimension of the SCL-90-R was the only measure of maternal psychopathology that significantly predicted elevation in clinician-rated impairment based on Global Assessment of Functioning scores in children with anxiety disorders (Manassis & Hood, 1998). The participants in the study were divided into two subgroups, children with phobic disorders (n = 38) and those with generalized anxiety disorder (GAD) (n = 36). The authors indicated that children in the phobic disorders group were those “presenting mainly with avoidant behavior” as opposed to those “presenting mainly with worries” (Manassis & Hood, 1998, p. 430). SAD was the primary diagnosis in 63% of the participants in the phobic disorders group. In children with phobic disorders, maternal phobic anxiety, maternal reports of child conduct problems, child depression, and developmental difficulties were significantly correlated with greater functional impairment in the child. In the GAD group, only psychosocial adversity was significantly correlated with greater impairment. Thus, maternal phobic anxiety was only associated with impairment among children presenting mainly with avoidant behavior.

Similarly, in a treatment study of 51 adolescents with anxiety-based school-refusal, phobic anxiety in the mothers, as measured with the SCL-90-R at baseline, was the only measure of maternal symptomatology significantly correlated with clinician ratings of adolescents’ anxiety posttreatment (Layne et al., 2002). In addition, a study demonstrated that mothers with elevations on the Phobic Anxiety, Obsessive-Compulsive, and Depression dimensions of the SCL-90-R were significantly more likely to have young children with clinical SAD than to have children with subclinical SAD or no SAD symptoms (Kearney, Sims, Pursell, & Tillotson, 2003). These studies show that elevated maternal phobic anxiety is associated with greater functional impairment in phobic youths (Manassis & Hood, 1998), greater number of SAD symptoms in children (Kearney et al., 2003), and greater anxiety posttreatment among anxious adolescents with school avoidance (Layne et al., 2002). It is notable that an association between maternal phobic anxiety and generalized anxiety in children has not been demonstrated. The results across the studies described above suggest a more specific relationship: phobic anxiety in mothers is associated with separation anxiety and phobic avoidance in children.

Regarding the relationship between maternal anxiety and child anxiety, the following question has been raised: do children of anxious parents genuinely have more anxiety than children of parents without anxiety or do anxious parents endorse more anxiety in their children than parents without anxiety, or are both true? Debate has focused on whether maternal symptomatology is associated with a reporting bias, specifically whether symptomatic mothers overreport child symptomatology (Chilcoat & Breslau, 1997; Krain & Kendall, 2000). Krain and Kendall (2000) examined the role of parental anxiety in parent ratings of child anxiety using the State-Trait Anxiety Inventory (STAI) and State-Trait Anxiety Inventory for Children. No significant relationships emerged between parental anxiety and parental ratings of child anxiety. However, maternal depression was a significant predictor of maternal ratings of child anxiety.

Chilcoat and Breslau (1997) investigated whether maternal psychiatric history was predictive of greater reporting of internalizing or externalizing problems in children. Participants included 801 mothers with 17% (n = 140) having a lifetime history of major depression, 36% (n = 291) with a history of anxiety disorder, and 13% (n = 105) with substance abuse or dependence history. The study utilized teacher ratings as a comparison standard to determine whether significant differences in child symptomatology were “real” or were due to maternal overreporting. Mothers with a history of an anxiety disorder and mothers with a history of major depression and comorbid anxiety disorder (depression/anxiety) endorsed internalizing symptoms in excess of symptoms endorsed by teachers. The authors interpreted this as mothers overstating their child's symptoms. However, because teacher ratings for children of mothers with depression/anxiety were elevated, the authors concluded that children of mothers with depression/anxiety genuinely had higher levels of internalizing symptoms but that these mothers also overstate the severity of these symptoms. Teacher ratings for children of anxious, nondepressed mothers did not differ from ratings of children with no maternal disorder. The paper was criticized for its conclusions about maternal reporting biases because teacher report was the comparative standard (Biederman, Mick, & Faraone, 1998). Biederman et al. (1998) indicated that teacher report is not an appropriate gold standard in child anxiety assessment.

The present study was designed to (a) examine the relation between maternal anxiety and child anxiety, (b) investigate a possible specific relation between maternal phobic anxiety and child separation anxiety, and (c) determine whether a reporting bias is associated with maternal anxiety or maternal depression, using a more appropriate gold standard (i.e., clinician ratings) than has been used previously. Clinician rating offers a better gold standard because a trained clinician integrates information from both parental and child perspectives to arrive at a composite score. It is hypothesized that higher levels of maternal anxiety will be associated with higher levels of child anxiety. In addition, it is hypothesized that a specific relationship will emerge between maternal phobic anxiety and measures of separation anxiety in children. It is not known whether maternal ratings of child anxiety will be in excess of that which is accounted for by clinician ratings (i.e., indicative of a maternal reporting bias), as previous research has not adequately addressed this issue.