Journal of Anxiety Disorders 32 (2015) 56–65

Contents lists available at ScienceDirect

Journal of Anxiety Disorders

The effect of maternal psychopathology on parent–child agreement of child anxiety symptoms: A hierarchical linear modeling approach Nicholas W. Affrunti ∗ , Janet Woodruff-Borden University of Louisville, United States

a r t i c l e

i n f o

Article history: Received 30 October 2014 Received in revised form 14 February 2015 Accepted 23 March 2015 Available online 2 April 2015 Keywords: Child anxiety Development Parent–child agreement Parent-report Psychopathology

a b s t r a c t The current study examined the effects of maternal anxiety, worry, depression, child age and gender on mother and child reports of child anxiety using hierarchical linear modeling. Participants were 73 mother–child dyads with children between the ages of 7 and 10 years. Reports of child anxiety symptoms, including symptoms of specific disorders (e.g., social phobia) were obtained using concordant versions of the Screen for Anxiety and Related Emotional Disorders (SCARED). Children reported significantly higher levels of anxiety symptoms relative to their mothers. Maternal worry and depression predicted for significantly lower levels of maternal-reported child anxiety and increasing discrepant reports. Maternal anxiety predicted for higher levels of maternal-reported child anxiety and decreasing discrepant reports. Maternal depression was associated with increased child-reported child anxiety symptoms. No significant effect of child age or gender was observed. Findings may inform inconsistencies in previous studies on reporter discrepancies. Implications and future directions are discussed. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Anxiety disorders are among the most prevalent psychiatric disorders experienced by children (Pérez-Edgar & Fox, 2005; Pine, 2007; Rapee, Schniering, & Hudson, 2009), with prevalence rates between 5.7% and 41.2% (Cartwright-Hatton, McNicol, & Doubleday, 2006; Costello et al., 1996). They are also associated with considerable impairment in school, social, and family domains (Essau, Conradt, & Petermann, 2000; Ginsburg, La Greca, & Silverman, 1998; Warren & Sroufe, 2004) and have long-term impacts on psychopathology, showing associations with anxiety, depression, externalizing disorders, and substance use problems later in life (Bittner et al., 2007; Kendall, Safford, FlannerySchroeder, & Webb, 2004; Pine, Cohen, Gurley, Brook, & Ma, 1998). Given the high prevalence and negative impact of these disorders in children, appropriate strategies for assessment are of paramount importance. A multiple informant (e.g., parent, child, and teacher) assessment is generally considered to be the preferred method for identifying and assessing childhood emotional disorders (Kendall & Flannery-Schroeder, 1998; Weems & Stickle, 2005). Yet, one

∗ Corresponding author at: University of Louisville, Department of Psychological and Brain Sciences, Life Sciences Building Room 317, Louisville, KY 40292, United States. Tel.: +1 502 852 6352. E-mail address: [email protected] (N.W. Affrunti). http://dx.doi.org/10.1016/j.janxdis.2015.03.010 0887-6185/© 2015 Elsevier Ltd. All rights reserved.

liability with this approach is a lack of agreement among informants. Indeed, disagreement between parent and child ratings of emotional problems is thought to be common (Achenbach, McConaughy, & Howell, 1987; De Los Reyes & Kazdin, 2004; Edelbrock, Costello, Dulcan, Conover, & Kala, 1986). This is true for children presenting to specialty clinics (Yeh & Weisz, 2001) or in community samples (Cosi, Canals, Hernández-Martinez, & Vigil-Colet, 2010). Studies specifically investigating parent–child concordance in the report of childhood anxiety have also found low agreement (Comer & Kendall, 2004; Manassis, Tannock, & Monga, 2009; Niditch & Varela, 2011; Safford, Kendall, FlannerySchroeder, Webb, & Sommer, 2005; Weems, Feaster, Horigian, & Robbins, 2011). Typically parents are found to report lower levels of intensity, frequency, and severity than their children, with a few notable exceptions (e.g., Krain & Kendall, 2000). Further, discrepancies remain across methods. That is, previous research on parent–child concordance in childhood anxiety, using either structured interviews or questionnaire methods, have found similarly discrepant findings (Choudhury, Pimentel, & Kendall, 2003; Cosi et al., 2010; Grills & Ollendick, 2002; Pereira et al., 2014). Indeed, the chosen method to examine these discrepancies may influence the discrepancies observed (De Los Reyes & Kazdin, 2004). In childhood anxiety research, the most common methods are interview and questionnaire based. Many studies have used the Anxiety Disorders Interview Schedule for Parents and Children (ADIS-P/C) to examine parent–child agreement (e.g., Comer & Kendall, 2004; Manassis et al., 2009; Niditch & Varela, 2011;

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

Reuterskiöld, Öst, & Ollendick, 2008). Others have used common questionnaires such as the Screen for Anxiety and Related Emotional Disorders parent and child versions (SCARED; Cosi et al., 2010; Pereira et al., 2014) to investigate discrepant reports. Though these are both frequently used within child anxiety research, comparing agreement across them may contribute to inconsistencies observed. Further, various statistical procedures employed in these studies may also affect results and discrepancies detected. Previous studies have not used the same analyses to determine agreement. Some studies have used correlations with raw scores, others standardized difference scores, and/or clinical agreement, the differences between which can further obfuscate comparisons across studies (De Los Reyes & Kazdin, 2004). Beyond methodological reasons, research has begun to investigate possible parent and child factors that may explain such disagreement. A significant factor may be the nature of anxiety in children. That is, while anxiety does have outward behavioral expressions (e.g., avoidance behaviors, reassurance seeking), anxiety symptoms that are salient and distressing may be internal to the child. Research has shown that higher rates of agreement are found among anxiety symptom dimensions with observable behavioral components (e.g., specific phobia) and lower rates are found for dimensions less observable (e.g., generalized anxiety disorder; Comer & Kendall, 2004; Pereira et al., 2014). Indeed, parents may be largely unaware of internal distress. Anxious children may also worry about how others perceive them, prompting socially desirous answers when questioned about their symptoms (Comer & Kendall, 2004; Dadds, Perrin, & Yule, 1998). Furthermore, anxiety can compromise cognitive processing, affecting memory retrieval and accuracy (Vasey & MacLeod, 2001). Lastly, children may lack the ability to accurately understand and convey their symptoms until they are older. This has led to a large number of studies examining child age as a possible correlate of parent–child discrepancies. Several studies have found that child age has an effect on reporting discrepancies in childhood anxiety. However, these effects have not always been in the same direction. That is, some studies have found that there are lower rates of disagreement (i.e., higher rates of agreement) between parents and older children (Grills & Ollendick, 2003; Rapee, Barrett, Dadds, & Evans, 1994; Wren, Bridge, & Birmaher, 2004). Other studies have found that parents demonstrated lower rates of disagreement (i.e., higher rates of agreement) with younger children (Krain & Kendall, 2000; Safford et al., 2005). Still other studies found no consistent effect of child age on parent–child discrepancies (Choudhury et al., 2003; Engel, Rodrigue, & Geffken, 1994; Reuterskiöld et al., 2008). Niditch and Varela (2011) provide some evidence that parental psychopathology may interact with child age in changing the effect of child age on parent–child reporting disagreement. That is, at lower levels of maternal anxiety discrepancies are relatively stable across child age. At higher levels of maternal anxiety child age influences discrepancies, with older children having mothers report higher levels of child anxiety and for younger children, child self-report higher levels of anxiety. Yet, there is no consensus on the effect of child age on discordant reports. There is similarly no agreement on where the effects of child age are occurring. That is, do children differ in selfreports on anxiety depending on age, do parents report different levels of child anxiety based on child age, or some combination of the two? The effect of child gender has also been investigated in parent–child discrepant reports. Similar to research on child age, research on child gender does not yield a consensus. Some studies have found no effect of child gender on disagreements between parent and child reports (Choudhury et al., 2003; Engel et al., 1994). Other studies have found that parent–daughter agreement was stronger than parent–son agreement (Reuterskiöld et al.,

57

2008; Safford et al., 2005). Yet, Grills and Ollendick (2003) found that parent–son agreement was significantly stronger, for the diagnoses of social phobia and separation anxiety disorder, than parent–daughter agreement. These results also do not provide any agreement on the differential effects child gender can have on child self-report and parent-report. Indeed both child gender, and child age effects, while studied often, have yet to yield a strong empirical consensus. Parental factors have also been investigated for their effects on parent–child reporting discrepancies. The majority of this research has examined the effect of parental psychopathology on discordant reporting. Since anxiety tends to run in families (Beidel & Turner, 1997), it is likely that parents of anxious children may be anxious themselves, which may skew their reports of their child’s anxiety (Comer & Kendall, 2004; Kristensen & Torgersen, 2006). While some studies have found a positive association between parental anxiety and reporting discrepancies (Briggs-Gowan, Carter, & Schwab-Stone, 1996; Frick, Silverthorn, & Evans, 1994; Manassis et al., 2009), others have found no effect (Krain & Kendall, 2000; Reuterskiöld et al., 2008). Further, parental depression has also been theorized as affecting parents’ reports of their children’s anxiety (Berg-Nielsen, Vika, & Dahl, 2003; Renouf & Kovacs, 1994). Indeed, mothers self-reporting higher levels of depression have been found to report higher severity of their children’s anxiety (Briggs-Gowan et al., 1996; Garber, Van Slyke, & Walker, 1998). Although both the effects of parental anxiety and depression on parent–child discrepancies have been investigated often, only a single study has examined the effect of parental worry on such discrepancies. Lagattuta, Sayfan, and Bamford (2012) found that parent worry was associated with greater discrepancies in parent versus child self-reported anxiety. However, in this study worry was looked at in isolation without accounting for depression or other anxiety symptoms. Parental worry has shown to be a predictor of child anxiety, above parental anxiety (Fisak, Holderfield, Douglas-Osborn, & Cartwright-Hatton, 2012), and may have a specific cognitive effect on parent’s ability to interpret and recall events (Dugas et al., 2005; Mogg, Bradley, Millar, & White, 1995; Wells, 1995). Most previous studies have examined informant concordance using rates of disagreement. That is, the outcome variable for these studies was the amount of disagreement between parent and child and the effects observed are of specific variables (e.g., parent anxiety, child age, child gender) on that disagreement. The few studies that have examined for which reporter certain variables exert their effect (e.g., Frick et al., 1994; Garber et al., 1998; Manassis et al., 2009; Wren et al., 2004) have typically done so without accounting for effects of the other reporter. For example, the effect of parental anxiety on parent-report child anxiety does not account for the effect of parental anxiety on child self-reported anxiety. While foundational and important, these studies may miss certain effects when dyads are analyzed concurrently. Novel statistical techniques, such as hierarchical linear modeling, allow examination on where these variables have their effects while accounting for the effects on the other reporter. The current study is designed to examine not only the discrepancies observed, but also what factors influence each respective reporter. This allows for conclusions to be made about the individual effects of factors (e.g., parent anxiety, child age) that influence both the discrepancy as a whole, as well as their influence on reports within each parent–child dyad. In this way, the current study can replicate previous findings, while extending them using a unique statistical technique. The purpose of this study was to investigate the effects of maternal psychopathology, child gender, and child age on parent and child reports of child anxiety using hierarchical linear modeling. Specifically, the effects of maternal anxiety, depression, and worry, child age and gender, on both mother and child report of separate

58

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

child anxiety symptom domains (e.g., social phobia, generalized anxiety, somatic symptoms) were examined. Hierarchical linear modeling examines the respective effects of variables on each individual within the dyad and it is a superior technique when examining dyadic data in small samples as dyadic data may break assumptions of independence, which can influence significance testing (Kashy & Kenny, 2000; Kenny, 2011). Further hierarchical linear modeling allows effects to be determined for each informant. Based on extant literature we expected that (1) parents and children would show significantly discrepant reports, with different domains of anxiety symptoms, particularly those with overt versus covert symptoms, showing different levels of discrepancies, (2) maternal anxiety, depression, and worry would be associated with greater discrepancy in reports of child anxiety levels, (3) maternal psychopathology would show differential effects on discrepancies across different child anxiety symptom domains, (4) that child age would have an effect on child- and mother-reported child anxiety, and (5) that female children, compared to male children, would show higher levels of agreement with mothers. 2. Method 2.1. Participants and procedures Participants were 73 mother–child dyads from two moderately sized urban areas and their respective surrounding areas. Mothers ranged in age from 27 to 58 years old (M = 40.51, SD = 6.87), had an annual family income of 60,000 US dollars or more (66.2%), were married (86.1%), and had a bachelor’s degree or higher level of education (65.3%). Approximately half of parents were employed full-time (58.3%; 3.4% unemployed). Children were between the ages of 7 and 10 years (M = 9.38, SD = 1.28), were primarily Caucasian (94.4%), and there was approximately an even number of males and females (54.8% male). Based on suggested cut-offs for the questionnaires used, detailed below, approximately 17% of mothers had moderate or higher anxiety symptoms, 15% had moderate or higher depression symptoms, and 22% had clinically elevated worry. Further, 32% of children had clinically elevated symptoms of anxiety, based on either a parent or child report of symptoms that met or exceeded the suggested clinical cutoff. The current sample was a community sample. All participants were recruited as part of a larger, longitudinal study examining factors implicated in the development of child anxiety. The data presented here were collected as the first wave of the study. Participants were recruited through elementary and middle schools, contacts in the local community, and flyers distributed throughout the community. Families were sent questionnaires and instructed to complete parent and child versions of measures separately. Children who had difficulty completing measures could receive help from study staff during lab visits. The study was approved by the Institutional Review Board at the University of Louisville. Both child assent and parent consent were required for participation in the study. 2.2. Measures 2.2.1. Child anxiety Child anxiety was measured using the Screen for Child Anxiety Related Emotional Disorder parent and child versions (SCAREDP, SCARED-C; Birmaher et al., 1997, 1999). The SCARED-P and SCARED-C consist of 41-items and are complementary parent- and self-report measures of pediatric anxiety shown to differentiate between clinically anxious and nonanxious psychiatrically ill youth (Birmaher et al., 1997). Parents and children, separately, answer questions using a 3-point Likert scale indicating to what degree a

statement about them (e.g., “I worry about going to school”), or their child, is true, from 0 (not true) to 2 (very or often true). Since the scales are the same for each the SCARED-P and the SCARED-C, comparisons between them can be drawn. Further, the SCARED-P/C has been used often in discrepancy research (e.g., Cosi et al., 2010; Pereira et al., 2014) due to its robust nature in both clinical and community populations (Muris et al., 1998, 1999), ability to examine symptom and total level discrepancies, and use across cultures (e.g., German, Chinese; see Hale III, Crocetti, Raaijmakers, & Meeus, 2011 for a review). For the purposes of this study, the SCAREDP/C represented a valid and reliable questionnaire to measure child anxiety symptoms through both parent and child reports. Both the SCARED-P and SCARED-C yield a total score of anxiety symptoms, obtained by summing all 41-items. Further, each measure consists of five subscales corresponding to the following anxiety domains: separation anxiety disorder (8 items), panic disorder (13 items), school phobia (4 items), generalized anxiety disorder (9 items), and social phobia (7 items). Higher scores indicate higher levels of child anxiety symptoms. Clinically elevated scores are determined to be as follows: total scores at or above 25, greater than 5 for separation anxiety disorder, greater than 7 for panic disorder, greater than 3 for school phobia, greater than 9 for generalized anxiety disorder, and greater than 8 for social phobia (Birmaher et al., 1999). Cronbach’s alphas for both scales fell in the acceptable range with internal consistencies on the SCAREDC scales ranging from ˛ = .93 to ˛ = .66 and on the SCARED-P from ˛ = .93 to ˛ = .71. Table 1 presents Cronbach’s alphas for each subscale, for each reporter. Although the school phobia subscale for child report had an ˛ = .66, this may be due to the subscale consisting of only 4 items. However, results pertaining to this subscale should be interpreted with caution. 2.2.2. Maternal anxiety The Beck Anxiety Inventory (BAI; Beck et al., 1988a) was used to assess levels of maternal anxiety. The BAI is an extensively used self-report questionnaire containing 21 items, measuring the cognitive and physical symptoms of anxiety. The BAI measures primarily arousal associated with anxiety, leaving out some symptoms of negative affect that have been captured in the current study by the BDI and the PSWQ. Participants are asked to indicate the degree to which he/she has been bothered by each symptom (e.g. unable to relax) in the past week. The measure uses a 4-point likert scale from 0 (not at all) to 3 (severely) and yields a total score. Scores range from 0 to 63, where higher scores indicate greater anxiety. The recommended clinical classification is as follows: 0–7 suggests minimal anxiety, 8–15 suggests mild anxiety, 16–25 suggests moderate anxiety, and 26–63 suggests severe anxiety (Beck & Steer, 1990). The BAI has shown good test–retest reliability in a separate sample over a one week period (r = 0.75; Beck, Epstein, et al., 1988; Beck, Steer, & Carbin, 1988) and high internal consistency in the current sample (˛ = .88). 2.2.3. Maternal depression The Beck Depression Inventory (BDI; Beck et al., 1990) was used to assess depressive symptoms in mothers. The BDI is a widely used self-report measure consisting of 21 items, measuring cognitive, affective, vegetative, and somatic symptoms of depression. Participants are asked to indicate which statement in a specific symptom area (e.g., loss of pleasure, pessimism) best describes them for the past two weeks. The scale uses a 4-point Likert scale, from 0 to 3, denoting the severity of depressive symptoms in that domain. Scores range from 0 to 63 with higher scores indicating more severe depressive symptoms. The recommended clinical classification is as follows: scores of 0–13 suggest minimal or no depression, scores of 14–19 suggest mild depression, scores of 20–28 suggest moderate depression, and scores of 29 or above suggest severe depression

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

59

Table 1 Number of scale items, Cronbach’s alphas, and correlations between mother and child report of child anxiety.

Total anxiety symptoms Panic disorder Generalized anxiety disorder Separation anxiety disorder Social phobia School phobia * **

# of items

Child-report ˛

Parent-report ˛

r

41 13 9 8 7 4

0.93 0.88 0.83 0.81 0.72 0.66

0.93 0.78 0.85 0.82 0.88 0.71

0.53** 0.68** 0.43** 0.47** 0.39* 0.62**

p < .05. p < .001.

(Beck et al., 1990). The BDI has shown to be reliable over time, have appropriate correlations with other measures of depression, and to differentiate between clinically depressed and nondepressed patients (Beck, Steer, et al., 1988; Beck et al., 1990). The internal consistency in this sample was ˛ = .93. 2.2.4. Maternal Worry The Penn State Worry Questionnaire (PSWQ; Meyer et al., 1990) was used to assess worry in mothers. The PSWQ is a widely used measure assessing worry. The PSWQ is separate from the BAI in symptoms as the PSWQ focuses specifically on worry, whereas the BAI is primarily concerned with arousal symptoms. Participants are asked to rate how typical certain statements (e.g., “My worries overwhelm me”) are of them. It consists of 16 items that are rated on a 5 point Likert scale from 1 (not at all typical of me) to 5 (very typical of me) and yields a total score. Total scores range from 16 to 80, with higher scores indicating greater worry. Scores greater than 65 are considered to be in the clinical range (Fresco et al., 2003). The PSWQ has shown to be a valid and reliable measure in both clinical and nonclinical samples (Brown et al., 1992; Meyer et al., 1990). The internal consistency in this sample was ˛ = .95. 2.3. Statistical analyses Data were analyzed using the Statistical Package for Social Sciences (SPSS) version 21 and Hierarchical Linear Modeling version 7 with restricted maximum likelihood (HLM; Raudenbush et al., 2004). Descriptive statistics, correlations, and internal consistencies for study variables were calculated prior to conducting the hierarchical modeling analysis. Further, mean differences were explored through t-tests. Hierarchical linear modeling was used to resolve the issue of dependence when analyzing data from individuals within a relationship (e.g., mother–child). As noted above, family data may fail to meet the assumption of independence and influence significance testing (Kashy & Kenny, 2000; Kenny, 2011). Given the dyadic nature of the data, a hierarchical linear modeling framework was employed to capture the interdependence between mother and child and control for it when examining the influence of additional variables. Hierarchical linear modeling allows for the dyad to be considered, rather than ignored, within the analysis (Kenny, 2011). Scores are treated as nested within the same dyad. As such, the effect of maternal level variables (e.g., maternal depression, anxiety, and worry) and child age and gender can be estimated within the dyad, resulting in differential effects on parent-reported or child self-reported anxiety. Separate models are then created. A null model, with no predictors of the outcome (e.g., total child anxiety scores) is created to obtain an intraclass correlation (ICC), which is the proportion of variance explained by the nested structure (Hox, 2002). That is, the ICC provides the proportion of variance explained by the dyadic relationship, or the between-dyad variability in child anxiety. A second model is created including group member (i.e., parent or child) to obtain the proportion of variance explained by either

group member. The final model created then adds the dyad level variables to the previous model. This model includes group member, maternal variables (i.e., maternal anxiety, depression, and worry), and child age and gender, which provides their effects on study outcomes. This was subsequently repeated for each domain of anxiety symptoms (e.g., total anxiety, panic disorder, separation anxiety disorder, social phobia, generalized anxiety disorder, and school phobia). It is important to note that in all models, with all outcome variables, maternal anxiety, depression, worry, and child age were mean centered. All interpretations are then based on standard deviation distances from the mean, rather than raw score increases, and  represents a measure of effect size.

3. Results 3.1. Descriptive statistics and correlations Prior to conducting the hierarchical linear modeling analysis, preliminary correlation and mean differences were examined. All study variables were approximately normally distributed, with appropriate skewness and kurtosis values (West et al., 1995). Descriptive statistics for and correlations between maternal anxiety, worry, and depression and child anxiety reports are presented in Table 2. As expected there was significant intercorrelation between the maternal variables. However, multicollinearity did not appear to be of concern given the moderate correlations and appropriate tolerance statistics. Descriptive statistics and correlations for mother and child reports of child anxiety symptoms are presented in Tables 1 and 3. As expected, mother–child correlations were found to be significant and generally in the moderate range. The strongest correlation was for panic symptoms (r = .68, p < .001), while the weakest was for social phobia (r = .39, p = .001). However, despite significant correlations discrepancies remain. Table 3 presents the findings of independent sample t-tests. As can be seen in the table, total anxiety symptoms (t = 3.23, p = .002, d = 0.54), panic disorder (t = 3.07, p = .003, d = 0.51), separation anxiety disorder (t = 2.91, p = .004, d = 0.78), social phobia (t = 3.17, p = .002, d = 0.52), and school phobia (t = 3.38, p = .001, d = 0.56) all showed significant differences between mother and child report. Generalized anxiety disorder

Table 2 Correlations and descriptive statistics between and descriptive statistics for maternal anxiety, worry, and depression. 1 1. Maternal anxiety 2. Maternal worry 3. Maternal depression

–

Mean SD Range

8.95 6.99 0–36

**

p < .001.

2

3

0.58** –

0.51** 0.57** –

50.75 15.43 22–79

10.26 9.37 0–42

60

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

Table 3 Mean, standard deviations, and mean differences for mother and child report on domains of child anxiety.

Total anxiety symptoms Panic disorder Generalized anxiety disorder Separation anxiety disorder Social phobia School phobia *

Mother mean (SD)

Child mean (SD)

t

Cohen’s d

15.12 (11.08) 2.12 (2.76) 4.78 (3.85) 2.42 (3.31) 3.92 (3.35) 0.88 (1.31)

21.56 (12.89) 4.04 (4.57) 5.18 (3.88) 5.07 (3.52) 5.58 (2.94) 1.70 (1.61)

3.24* 3.07* 0.62 2.91* 3.17* 3.78*

0.54 0.51 0.10 0.78 0.52 0.56

p < .05.

was not significantly different between reporters (t = 0.62, p = .54, d = 0.10). 3.2. Hierarchical linear modeling As noted above, three models are created for each domain of child anxiety during the modeling process. The first model, a null model, is run to determine the ICC. The second model includes the reporter (i.e., mother and child) to determine the amount of variance explained by either reporter. The final model includes the reporter and maternal anxiety, worry, and depression, child age, and gender variables. Below are the equations for the final model: Level-1: Child anxiety = ˇ0 + ˇ1j (reporter) + rij Level-2: ˇ0j = 00 + 01 (maternal worry) + 02 (maternal anxiety) + 03 (maternal depression) + 04 (child age) +05 (child gender) + u0j

That is, children reported significantly higher child total anxiety scores than their mothers. This effect did not vary across dyads, 2 (72) = 85.42, p = .13. Results of the full model detail the effects of maternal anxiety, worry, depression and child age and gender on each reporter of child anxiety total scores. Maternal worry was found to be the only significant predictor of mother-reported child total anxiety,  01 = −0.18, SE = 0.08, t(67) = −2.07, p = .04. Maternal anxiety ( 02 = 0.34, SE = 0.25, p = .17), maternal depression ( 03 = 0.22, SE = 0.19, p = .25), child age ( 04 = 0.58, SE = 0.85, p = .50), and child gender ( 05 = −0.18, SE = 1.05, p = .91) did not significantly predict mother-reported child total anxiety. Maternal depression was found to be the only significant predictor of child-reported child total anxiety,  13 = 0.35, SE = 0.17, t(67) = 2.05, p = .04. For every 1 standard deviation increase in maternal depression scores, the effect of child report increased by .35 points. Maternal anxiety ( 12 = 0.22, SE = 0.23, p = .34), maternal worry ( 11 = 0.14, SE = 0.08, p = .18), child age ( 14 = −0.24, SE = 0.98, p = .81), and child gender ( 15 = −0.21, SE = 1.95, p = .92) did not significantly predict childreported child total anxiety. Results also indicated that 47.5% of the between-dyad variance in child total anxiety scores was explained by maternal anxiety, worry, depression and child age and gender. Further, 10% of the variance in the effect of reporter on child total anxiety scores was explained by maternal anxiety, worry, depression, and child age and gender.

ˇ1j = 10 + 11 (maternal worry) + 12 (maternal anxiety) + 13 (maternal depression) + 14 (child age) +15 (child gender) + u0j

Further, HLM separates the variance observed into between and within dyad variances. Between dyad variance is variability that exists between different sets of dyads. For the purposes of this study, differences that explain this variance differ across dyads (e.g., maternal worry, anxiety, depression, child age, and gender). Within dyad variance is the variability that exists within the each dyad’s report of child anxiety. For the purposes of this study, the difference that explains this variability is the reporter (mother or child). The separation of variances, and associated variables explaining them, allows conclusions to be drawn about the proportion of variance explained by reporter and those explained by other factors (e.g., maternal worry). 3.2.1. Total anxiety symptoms Results of the null model revealed that 42.9% of the variance in child total anxiety scores was due to between-dyad differences (ICC = .429). The remaining variance (i.e., 57.1%) is within-dyads. That is, the majority of variance in child total anxiety scores was within the dyad, rather than between dyads. Results of the second model, with reporter added to the model, revealed that 34.6% of the variance in child total anxiety scores was explained by the reporter. Further, there was a significant effect of reporter on child total anxiety scores,  10 = 6.44, SE = 1.37, t(72) = 4.71, p < .001.

3.2.2. Panic disorder Results of the null model revealed that 50.9% of the variance in child panic disorder scores were due to between-dyad differences (ICC = .509). The remaining variance (i.e., 49.1%) exists within-dyads. Results of the second model, with reporter added to the model, revealed that 60.9% of the variance in child panic disorder scores was explained by the reporter. Further, there was a significant effect of reporter on child panic disorder scores,  10 = 1.92, SE = 0.39, t(72) = 4.86, p < .001. That is, children reported significantly higher child panic disorder scores than their mothers. However, this effect varied across dyads, 2 (72) = 140.58, p < .001. In testing the full model, maternal anxiety,  02 = 0.13, SE = 0.05, t(67) = 2.71, p = .009, and maternal depression,  03 = −0.13, SE = 0.04, t(67) = −3.16, p < .001, were found to significantly predict mother-reported child panic disorder scores. Maternal worry ( 01 = 0.01, SE = 0.02, p = .85), child age ( 04 = 0.22, SE = 0.21, p = .30), and child gender ( 05 = −0.36, SE = 0.53, p = .57) did not significantly predict mother-reported child panic disorder scores. There were no significant predictors of child-reported child panic disorder scores. Maternal depression ( 13 = 0.08, SE = 0.05, p = .067), maternal anxiety ( 12 = 0.11, SE = 0.08, p = .18), maternal worry ( 11 = 0.06, SE = 0.04, p = .16), child age ( 14 = 0.47, SE = 0.33, p = .15), and child gender ( 15 = −0.30, SE = 0.55, p = .58) did not significantly predict child-reported child panic disorder. Results also indicated that 52.8% of the between-dyad variance in child panic disorder scores was explained by maternal anxiety, worry, depression and child age and gender. Further, 1.8% of the variance in the effect of reporter on child panic disorder scores was

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

explained by maternal anxiety, worry, depression, and child age and gender. 3.2.3. Generalized anxiety disorder Results of the null model revealed that 42.3% of the variance in child generalized anxiety disorder scores were due to betweendyad differences (ICC = .423) and 57.7% is within-dyads. That is, the majority of variance in child generalized anxiety disorder scores was within the dyad, rather than between dyads. Results of the second model, with reporter added to the model, revealed that only 10.9% of the variance in child generalized anxiety disorder scores was explained by the reporter and there was no significant effect of reporter,  10 = 0.40, SE = 0.48, t(72) = 0.82, p = .42. In the full model, maternal worry was found to be the only significant predictor of mother-reported child generalized anxiety disorder scores,  01 = −0.09, SE = 0.03, t(67) = −2.43, p = .02. Maternal anxiety ( 02 = −0.01, SE = 0.08, p = .88), maternal depression ( 03 = 0.08, SE = 0.06, p = .17), child age ( 04 = 0.18, SE = 0.33, p = .57), and child gender ( 05 = 0.80, SE = 0.70, p = .21) did not significantly predict mother-reported child generalized anxiety disorder. There were no significant predictors of child-reported child generalized anxiety disorder scores: maternal anxiety ( 12 = 0.06, SE = 0.09, p = .48), maternal worry ( 11 = 0.02, SE = 0.04, p = .71), maternal depression ( 13 = −0.11, SE = 0.07, p = .10), child age ( 14 = −0.45, SE = 0.39, p = .25), and child gender ( 15 = 0.87, SE = 0.70, p = .17). Results also indicated that 36% of the between-dyad variance in child generalized anxiety disorder scores was explained by maternal anxiety, worry, depression and child age and gender. Further, 3.6% of the variance in the effect of reporter on child generalized anxiety disorder scores was explained by maternal anxiety, worry, depression, and child age and gender. 3.2.4. Separation anxiety disorder The null model indicated that 38.9% of the variance in child separation anxiety disorder scores were due to between-dyad differences (ICC = .389) and 62.1% is within-dyads. That is, the majority of variance in child separation anxiety disorder scores was within the dyad, rather than between dyads. Results of the second model, with reporter added to the model, revealed that 26.7% of the variance in child separation anxiety scores was explained by the reporter. There was a significant effect of reporter on child separation anxiety scores,  10 = 1.64, SE = 0.41, t(72) = 3.97, p < .001. That is, children reported significantly higher child total anxiety scores than their mothers. This effect did not vary across dyads, 2 (72) = 81.72, p = .20. Results of the full model revealed maternal anxiety to be the only significant predictor of mother-reported child separation anxiety disorder,  02 = 0.15, SE = 0.06, t(67) = 2.37, p = .02. Maternal worry ( 01 = 0.03, SE = 0.03, p = .32), maternal depression ( 03 = 0.05, SE = 0.05, p = .32), child age ( 04 = −0.09, SE = 0.28, p = .76), and child gender ( 05 = −0.43, SE = 0.61, p = .34) were not significant predictors. There were no significant predictors of child-reported child separation anxiety scores. Maternal depression ( 13 = 0.10, SE = 0.05, p = .056), maternal anxiety ( 12 = −0.02, SE = 0.09, p = .84), maternal worry ( 11 = 0.04, SE = 0.03, p = .21), child age ( 14 = −0.17, SE = 0.33, p = .61), and child gender ( 15 = −0.58, SE = 0.60, p = .33) were not significant predictors. Maternal anxiety, worry, depression and child age and gender explained 42.2% of the between-dyad variance in child separation anxiety disorder scores. Further, they explained approximately 1% of the variance in the effect of reporter on child separation anxiety disorder scores. 3.2.5. Social phobia Results of the null model revealed that 30% of the variance in child social phobia scores were due to between-dyad differences (ICC = .3). The remaining variance (i.e., 70%) is within-dyads.

61

That is, the majority of variance in child social phobia scores was within the dyad, rather than between dyads. Results of the second model, with reporter added to the model, revealed that 22.7% of the variance in child social phobia scores was explained by the reporter. Further, there was a significant effect of reporter on child total anxiety scores,  10 = 1.66, SE = 0.41, t(72) = 4.07, p < .001. That is, children reported significantly higher child total anxiety scores than their mothers. This effect did not vary across dyads, 2 (72) = 79.87, p = .25. In testing the full model, there were no significant predictors of mother-reported child social phobia. Maternal worry ( 01 = −0.05, SE = 0.02, p = .08), maternal anxiety ( 02 = 0.05, SE = 0.06, p = .43), maternal depression ( 03 = −0.07, SE = 0.04, p = .14), child age ( 04 = 0.26, SE = 0.34, p = .45), and child gender ( 05 = 0.06, SE = 0.60, p = .92) were not significant predictors. There were no significant predictors of child-reported child social phobia scores: maternal anxiety ( 12 = 0.02, SE = 0.06, p = .74), maternal worry ( 11 = 0.02, SE = 0.03, p = .58), maternal depression ( 13 = −0.02, SE = 0.06, p = .77), child age ( 14 = −0.14, SE = 0.30, p = .64), and child gender ( 15 = −0.10, SE = 0.08, p = .17). Maternal anxiety, worry, depression and child age and gender explained 8.4% of the betweendyad variance in child social phobia scores. Further, they explained approximately 25% of the variance in the effect of reporter on child social phobia scores. 3.2.6. School phobia The null model indicated that 49.4% of the variance in child school phobia scores were due to between-dyad differences (ICC = .494) and 50.6% is within-dyads. Results of the second model, with reporter added to the model, revealed that 40.4% of the variance in child school phobia scores was explained by the reporter. There was a significant effect of reporter on child school phobia scores,  10 = 0.82, SE = 0.15, t(72) = 5.42, p < .001. That is, children reported significantly higher child school phobia scores than their mothers. This effect did not vary across dyads, 2 (72) = 89.65, p = .08. Results of the full model revealed no significant predictors of mother-reported child school phobia: maternal anxiety, ( 02 = 0.03, SE = 0.02, p = .06), maternal worry ( 01 = 0.01, SE = 0.11, p = .45), maternal depression ( 03 = 0.03, SE = 0.03, p = .29), child age ( 04 = −0.09, SE = 0.10, p = .93), and child gender ( 05 = −0.20, SE = 0.28, p = .47) were not significant predictors. Maternal depression was found to be the only significant predictor of child-reported child school phobia scores,  13 = 0.34, SE = 0.02, t(67) = 2.07, p = .04. For every 1 standard deviation increase in maternal depression scores, the effect of child report increased by .34 points. That is, those children with mothers who had greater depressive symptoms were significantly further from the mother-report of their symptoms. Maternal anxiety ( 12 = 0.05, SE = 0.04, p = .18), maternal worry ( 11 = 0.01, SE = 0.01, p = .28), child age ( 14 = 0.05, SE = 0.10, p = .63), and child gender ( 15 = −0.42, SE = 0.88, p = .74) did not significantly predict child-reported child school phobia. Results also indicated that 22.1% of the between-dyad variance in child school phobia scores was explained by maternal anxiety, worry, depression and child age and gender. Further, 17.6% of the variance in the effect of reporter on child total anxiety scores was explained by maternal anxiety, worry, depression, and child age and gender. 4. Discussion The present study sought to investigate the role of parental psychopathology on mother- and child-reported child anxiety using hierarchical linear modeling. Based on extant literature (e.g., Comer & Kendall, 2004; Grills & Ollendick, 2003; Niditch & Varela, 2011; Pereira et al., 2014), we hypothesized that (1) mothers

62

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

and children would differ on reports of child anxiety, with different domains of anxiety symptoms, particularly those with overt versus covert symptoms, showing different levels of discrepancies (2) maternal anxiety, depression, and worry would be associated with greater discrepancy in reported child anxiety, (3) maternal psychopathology would show differential effects on discrepancies across different child anxiety symptom domains, (4) that child age would have an effect on child- and mother-reported child anxiety, and (5) that female children, compared to male children, would show higher levels of agreement with mothers. In testing our first hypothesis, correlations and t-tests between reporters suggested both that there was a significant association between mother- and child-reported child anxiety, but that their scores significantly differed. This is consistent with previous research (Cosi et al., 2010; Kraemer et al., 2003; Muris et al., 2002). However, some correlations observed in our data were higher than in previous studies (e.g., panic disorder). This may be due to the community nature of our sample, as Pereira et al. (2014) found that correlations between mother- and child-report decreased in a high anxiety subgroup, compared to a community sample. Mean differences were also in the expected direction, with children reporting higher levels of anxiety symptoms, in all domains except generalized anxiety disorder, than their mothers. The discrepancies detected are a common finding across childhood psychiatric disorders (see De Los Reyes & Kazdin, 2005 for a review), especially internalizing disorders (Achenbach et al., 1987; Duhig et al., 2000). Generalized anxiety disorder represented an exception to this pattern, as no significant difference between mother- and childreport was observed. This exception is noteworthy, as previous research has typically found more covert dimensions of anxiety to be those with higher levels of discrepant reporting between parents and children (e.g., Comer & Kendall, 2004). As noted above, this may have been due to sample characteristics, specifically it being a community sample. That is, both parents and children reported relatively low levels of generalized anxiety disorder symptoms, which may have decreased our ability to detect discrepancies. This is corroborated by Pereira et al. (2014) who found a larger effect size for discrepant reports in a high anxiety sample, compared with their overall sample, using the same SCARED measures. Evidence for our second and third hypotheses was mixed. While maternal worry and maternal depression generally predicted for higher discrepancies, maternal anxiety did not. More specifically, maternal worry predicted for decreased mother-reported levels of child total anxiety symptoms and generalized anxiety disorder. Maternal depression predicted for lower mother-reported levels of panic disorder. Lower levels of child anxiety symptoms indicate increased discrepancy as mothers reported lower levels of symptoms prior to taking into account these effects. This may indicate that cognitive symptoms (e.g., rumination) of maternal psychopathology decrease accuracy in reporting, as our measure of anxiety, the BAI, primarily assesses for arousal and panic disorder symptoms (Cox et al., 1996). Indeed, our results suggest that physiological arousal associated with anxiety in mothers was related to lower rates of discrepancies for child panic and separation anxiety disorder. That is, maternal anxiety significantly predicted for increased mother-reported levels of panic disorder and separation anxiety disorder. This effect reduces the discrepancy seen in those disorders. Mothers with increased symptoms of panic, as measured with the BAI, may be more aware of such symptoms in their children because of their own experience with those symptoms. The array of effects shown across child anxiety domains may inform why previous research has yet to come to a conclusion on the effect of parental psychopathology on discrepant reporting (e.g., Briggs-Gowan et al., 1996; Krain & Kendall, 2000). That is, different maternal psychopathologies in different domains of child anxiety were related to different directions of effects.

Interestingly, maternal anxiety and maternal worry showed no significant associations with child-reported child anxiety. However, maternal depression did have an effect on child-reported total anxiety symptoms and school phobia. This may further help explain inconsistent findings across the effects of parental psychopathology on parent–child discrepant reporting (De Los Reyes & Kazdin, 2005). If, as seen here, parental psychopathology exerts effects not only on parent-report but also different effects on childreport, discrepancies may vary greatly across different patterns of parental psychopathology and domains of child anxiety. Further, these discrepancies appear to grow larger the greater the parental psychopathology. For example, we would predict a maternal report of child total anxiety symptoms for a mother 1 standard deviation above the mean on worry, depression, and anxiety would be 12.96, while a child-report would be 23.29. At 2 standard deviations above the mean on maternal worry, depression, and anxiety, we would predict a maternal-report of 10.51 and a child-report of 24.75. Discrepancies of approximately 14 and 11 points are considerably higher than the discrepancy seen at mean levels of maternal worry, depression, and anxiety, 6 points. The growing discrepancies observed suggest mothers with greater worry and depressive symptoms report less problems in their children. Mothers with higher arousal symptoms of anxiety report increased symptoms in their children. Importantly, these effects are not uniform across symptom areas. However, this pattern may broadly suggest that parents with cognitive symptoms that increase self-focus, observed in both depressed and worried individuals (Watkins & Moulds, 2005; Watkins & Teasdale, 2001; Watkins & Teasdale, 2004) may be less aware of child distress and report less symptoms. This is inconsistent with the depressiondistortion hypothesis (Chi & Hinshaw, 2002; Richters, 1992), which theorizes that depression promotes negative biases in parent’s perception of their children’s behavior and emotions. Current findings may clarify previously mixed support for the depression-distortion hypothesis (Gartstein et al., 2009; Richters & Pellegrini, 1989; Youngstrom et al., 2000). While both parental depression and worry may increase discrepancies in reporting, they may do so by reducing parents’ accuracy through a distancing of their awareness of child distress, instead focusing more on their internal processes (e.g., rumination, worry). However, the same effect was not seen for anxious mothers. That is, increased parental anxiety increased the amount of agreement with their children. It is possible that anxious mothers are more vigilant of, or more familiar with, anxious symptoms in their children. Contrary to both our fourth and fifth hypotheses, no effect of child age or child gender was observed for either mother- or child-reported child anxiety. Although some previous research has identified these variables as relevant factors in the discrepant reporting (Rapee et al., 1994; Reuterskiöld et al., 2008; Safford et al., 2005; Wren et al., 2004), there are others that have not found such effects (e.g., Choudhury et al., 2003; Pereira et al., 2014). It is important to note that our study used a limited age range of children, which may not have allowed age-related effects to be observed. Yet, both child gender and child age have been previously used as possible explanations for child discrepancies with girls and older children providing more accuracy (De Los Reyes & Kazdin, 2005). For example, it has been hypothesized that younger children may be unable to report reliably and validly the symptoms they experience (Grills & Ollendick, 2002). Contrary findings observed in our sample may suggest that child characteristics only play a limited role in influencing child self-reports, after accounting for parental effects. That is, maternal characteristics may also influence child self-report. Maternal depression may play a specific role in child self-reported anxiety through emotion socialization. While maternal depression is associated with increased child internalizing

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

symptoms (see Goodman et al., 2011), our findings suggest it may have an effect on the child’s own rating of his/her anxiety symptoms. Depressed mothers are less responsive, display less positive affect, and children of depressed mothers tend to exhibit higher levels of negative affect (Feng et al., 2007; Field, 1984; Field et al., 1988; Pickens & Field, 1993). There is also evidence that emotionregulation strategies have been shown to differ in children of depressed, versus non-depressed mothers (Eisenberg et al., 1998; Silk et al., 2006). Further, children tend to integrate parent characteristics into their own self-concepts (Eder & Mangelsdorf, 1997; Thompson, 1998) and when exposed to intense negative emotion may decrease self-reflection (Denham, 1998). As such, emotion socialization may be an important variable in why maternal depressive symptoms had an effect on child self-reported total anxiety symptoms and school phobia but only mother-reported panic symptoms. That is, parental depression may lead to specific emotion socialization deficits that influence children’s self-reported anxiety, increasing discrepancies between parent and child. Importantly, the effects observed were specifically related to total anxiety symptoms and school phobia, which suggests effects are not uniform across symptom areas. And despite evidence that parents may socialize children to emotions differently based on gender (Chaplin et al., 2005; Garner et al., 1997; Klimes-Dougan et al., 2007), it appears that both children and mothers may be able to report similar symptoms of anxiety regardless of gender. Yet, understanding the role that emotion socialization has on child self-reported anxiety remains an important direction for further research. Findings from the current study are subject to certain limitations. Although instructions indicated that questionnaires should be completed autonomously and internal consistencies were generally acceptable, we were unable to guarantee all respondents did so while understanding all items. Method variance may have affected outcomes as maternal anxiety, worry, depression, and mother-reported child anxiety were all questionnaires completed by mothers. Additionally, other measures are available with concordant parent and child reports of child-anxiety (e.g., Anxiety Disorders Interview Schedule, Multidimensional Anxiety Scale for Children). These measures may yield separate findings and determining the effect of the measurement tool on these discrepancies remains an important avenue for continued investigation. Further research should attempt to include more objective measures of parent symptoms (e.g., independent evaluation). The currently study also only included reports from mothers and children. Examining discrepancies across further informants (e.g., teachers) may hold important information for the assessment of anxiety in children. Sample characteristics may have also limited the generalizability of the results. Our sample included only mothers and was almost exclusively Caucasian. Replicating these effects in fathers (see Bögels & Phares, 2008 for a review) and a wider range of ethnicities remains an important future direction for research. Further, our sample had elevated rates of children with clinically relevant anxiety symptoms (32%) and mothers with clinically relevant worry symptoms (22%) for a community sample. Results may differ in samples with higher or lower rates of child anxiety symptoms. Although determining the effect of parental psychopathology on a continuous range of child anxiety symptoms, rather than only diagnostic agreement, is important, our study cannot conclude what effects maternal variables have on diagnostic agreement. Future studies should aim to look at the agreement between diagnostic groups across both parents and children. Lastly, the current study chose to investigate discrepancies as outcomes. Yet, there are also compelling reasons for future research to examine discrepant reports as predictors using statistical techniques such as polynomial regression (see Laird & De Los Reyes, 2013; Laird & Weems, 2011).

63

In sum, maternal anxiety, depression, and worry symptoms affected differently both child- and mother-reported anxiety levels. In analyzing the data using hierarchical linear modeling, specific patterns became apparent in the reporting of child anxiety. That is, different domains of maternal psychopathology showed different effects on parent- or child-report in different domains of anxiety. The directions of effects were generally consistent with previous research, with children reporting significantly higher anxiety in all symptom domains except generalized anxiety disorder. Maternal depression appeared to have a specific effect on child self-reported anxiety as well. This is the first study, to our knowledge, that has specifically investigated the effect of maternal worry on parentreported child anxiety. Based on our findings, maternal worry may serve to increase discrepancies between parent- and child-report child anxiety. Interestingly, maternal worry showed a distinct and opposite effect on reporting discrepancies from maternal anxiety. No effect was observed for child characteristics such as age and gender. The current study emphasizes the importance of a multiinformant approach in the evaluation of childhood anxiety. The low level of agreement between parents and children makes obtaining both reports, and preferably others (e.g., teacher, independent evaluator), vital to understanding the child’s level of anxiety. Both perspectives provide important data in a complete evaluation of child anxiety. Further, understanding parental variables may provide important context in developing a broader interpretation of child anxiety symptoms and may provide evaluators with necessary information to come to a more reliable and valid conclusion.

References Achenbach, T. M., McConaughy, S. H., & Howell, C. T. (1987). Child/adolescent behavioral and emotional problems: implications of cross-informant correlations for situational specificity. Psychological Bulletin, 101, 213–232. Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety: psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893–897. Beck, A. T., & Steer, R. A. (1990). Manual for the Beck Anxiety Inventory. San Antonio, TX: Psychological Corporation. Beck, A. T., Steer, R. A., & Brown, G. K. (1990). Beck Depression Inventory. San Antonio, TX: Psychological Corporation. Beck, A. T., Steer, R. A., & Carbin, M. G. (1988). Psychometric properties of the Beck Depression Inventory: twenty-five years of evaluation. Clinical Psychology Review, 8, 77–100. Beidel, D. C., & Turner, S. M. (1997). At risk for anxiety: I. Psychopathology in the offspring of anxious parents. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 918–924. Berg-Nielsen, T. S., Vika, A., & Dahl, A. A. (2003). When adolescents disagree with their mothers: CBCL-YSR discrepancies related to maternal depression and adolescent self-esteem. Child: Care Health and Development, 29, 207–213. Birmaher, B., Brent, D. A., Chiappetta, L., Bridge, J., Monga, S., & Baugher, M. (1999). Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): a replication study. Journal of the American Academy of Child & Adolescent Psychiatry, 38, 1230–1236. Birmaher, B., Khetarpal, S., Brent, D., Cully, M., Balach, L., Kaufman, J., et al. (1997). The screen for child anxiety related emotional disorders (SCARED): scale construction and psychometric characteristics. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 545–553. Bittner, A., Egger, H. L., Erkanli, A., Costello, E. J., Foley, D. L., & Angold, A. (2007). What do childhood anxiety disorders predict? Journal of Child Psychology and Psychiatry, 48, 1174–1183. Bögels, S., & Phares, V. (2008). Fathers’ role in the etiology, prevention and treatment of child anxiety: a review and new model. Clinical Psychology Review, 28, 539–558. Briggs-Gowan, M. J., Carter, A. S., & Schwab-Stone, M. (1996). Discrepancies among mother, child, and teacher reports: examining the contributions of maternal depression and anxiety. Journal of Abnormal Child Psychology, 24, 749–765. Brown, T. A., Antony, M. M., & Barlow, D. H. (1992). Psychometric properties of the Penn State Worry Questionnaire in a clinical anxiety disorders sample. Behaviour Research and Therapy, 30, 33–37. Cartwright-Hatton, S., McNicol, K., & Doubleday, E. (2006). Anxiety in a neglected population: prevalence of anxiety disorders in pre-adolescent children. Clinical Psychology Review, 26, 817–833. Chaplin, T. M., Cole, P. M., & Zahn-Waxler, C. (2005). Parental socialization of emotion expression: gender differences and relations to child adjustment. Emotion, 5, 80–88.

64

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65

Chi, T. C., & Hinshaw, S. P. (2002). Mother–child relationships of children with ADHD: the role of maternal depressive symptoms and depression-related distortions. Journal of Abnormal Child Psychology, 30, 387–400. Choudhury, M. S., Pimentel, S. S., & Kendall, P. C. (2003). Childhood anxiety disorders: parent–child (dis)agreement using a structured interview for the DSM-IV. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 957–964. Comer, J. S., & Kendall, P. C. (2004). A symptom-level examination of parent–child agreement in the diagnosis of anxious youths. Journal of the American Academy of Child & Adolescent Psychiatry, 43, 878–886. Cosi, S., Canals, J., Hernández-Martinez, C., & Vigil-Colet, A. (2010). Parent–child agreement in SCARED and its relationship to anxiety symptoms. Journal of Anxiety Disorders, 24, 129–133. Costello, E. J., Angold, A., Burns, B. J., Stangl, D. K., Tweed, D. L., Erkanli, A., et al. (1996). The Great Smoky Mountains Study of Youth: goals, design, methods, and the prevalence of DSM-III-R disorders. Archives of General Psychiatry, 53, 1129–1136. Cox, B. J., Cohen, E., Direnfeld, D. M., & Swinson, R. P. (1996). Does the Beck Anxiety Inventory measure anything beyond panic attack symptoms? Behaviour Research and Therapy, 34, 949–954. Dadds, M. R., Perrin, S., & Yule, W. (1998). Social desirability and self-reported anxiety in children: an analysis of the RCMAS Lie Scale. Journal of Abnormal Child Psychology, 26, 311–317. De Los Reyes, A., & Kazdin, A. E. (2004). Measuring informant discrepancies in clinical child research. Psychological Assessment, 16, 330–334. De Los Reyes, A., & Kazdin, A. E. (2005). Informant discrepancies in the assessment of childhood psychopathology: a critical review, theoretical framework, and recommendations for further study. Psychological Bulletin, 131, 483–509. Denham, S. A. (1998). Emotional development in young children. New York, NY: Guilford Press. Dugas, M. J., Hedayati, M., Karavidas, A., Buhr, K., Francis, K., & Phillips, N. A. (2005). Intolerance of uncertainty and information processing: evidence of biased recall and interpretations. Cognitive Therapy and Research, 29, 57–70. Duhig, A. M., Renk, K., Epstein, M. K., & Phares, V. (2000). Interparental agreement on internalizing externalizing, and total behavior problems: a meta-analysis. Clinical Psychology: Science and Practice, 7, 435–453. Edelbrock, C., Costello, A. J., Dulcan, M. K., Conover, N. C., & Kala, R. (1986). Parent–child agreement on child psychiatric symptoms assessed via structured interview. Journal of Child Psychology and Psychiatry, 27, 181–190. Eder, R. A., & Mangelsdorf, S. C. (1997). The emotional basis of early personality development: implications for the emergent self-concept. In: R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 209–240). San Diego, CA: Academic Press. Eisenberg, N., Cumberland, A., & Spinrad, T. L. (1998). Parental socialization of emotion. Psychological Inquiry, 9, 241–273. Engel, N. A., Rodrigue, J. R., & Geffken, G. R. (1994). Parent–child agreement on ratings of anxiety in children. Psychological Reports, 75, 1251–1260. Essau, C. A., Conradt, J., & Petermann, F. (2000). Frequency, comorbidity, and psychosocial impairment of anxiety disorders in German adolescents. Journal of Anxiety Disorders, 14, 263–279. Feng, X., Shaw, D. S., Skuban, E. M., & Lane, T. (2007). Emotional exchange in mother–child dyads: stability, mutual influence, and associations with maternal depression and child problem behavior. Journal of Family Psychology, 21, 714–725. Field, T. M. (1984). Early interactions between infants and their postpartum depressed mothers. Infant Behavior and Development, 7, 517–522. Field, T., Healy, B., Goldstein, S., Perry, S., Bendell, D., Schanberg, S., et al. (1988). Infants of depressed mothers show depressed behavior even with nondepressed adults. Child Development, 59, 1569–1579. Fisak, B., Holderfield, K. G., Douglas-Osborn, E., & Cartwright-Hatton, S. (2012). What do parents worry about? Examination of the construct of parent worry and the relation to parent and child anxiety. Behavioural and Cognitive Psychotherapy, 40, 542–557. Fresco, D. M., Mennin, D. S., Heimberg, R. G., & Turk, C. L. (2003). Using the Penn State Worry Questionnaire to identify individuals with generalized anxiety disorder: a receiver operating characteristic analysis. Journal of Behavior Therapy and Experimental Psychiatry, 34, 283–291. Frick, P. J., Silverthorn, P., & Evans, C. (1994). Assessment of childhood anxiety using structured interviews: patterns of agreement among informants and association with maternal anxiety. Psychological Assessment, 6, 372–379. Garber, J., Van Slyke, D. A., & Walker, L. S. (1998). Concordance between mothers’ and children’s reports of somatic and emotional symptoms in patients with recurrent abdominal pain or emotional disorders. Journal of Abnormal Child Psychology, 26, 381–391. Garner, P. W., Robertson, S., & Smith, G. (1997). Preschool children’s emotional expressions with peers: the roles of gender and emotion socialization. Sex Roles, 36, 675–691. Gartstein, M. A., Bridgett, D. J., Dishion, T. J., & Kaufman, N. K. (2009). Depressed mood and maternal report of child behavior problems: another look at the depression–distortion hypothesis. Journal of Applied Developmental Psychology, 30, 149–160. Ginsburg, G. S., La Greca, A. M., & Silverman, W. K. (1998). Social anxiety in children with anxiety disorders: relation with social and emotional functioning. Journal of Abnormal Child Psychology, 26, 175–185. Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D. (2011). Maternal depression and child psychopathology: a meta-analytic review. Clinical Child and Family Psychology Review, 14, 1–27.

Grills, A. E., & Ollendick, T. H. (2002). Issues in parent–child agreement: the case of structured diagnostic interviews. Clinical Child and Family Psychology Review, 5, 57–83. Grills, A. E., & Ollendick, T. H. (2003). Multiple informant agreement and the anxiety disorders interview schedule for parents and children. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 30–40. Hale, W. W., III, Crocetti, E., Raaijmakers, Q. A., & Meeus, W. H. (2011). A metaanalysis of the cross-cultural psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED). Journal of Child Psychology and Psychiatry, 52, 80–90. Hox, J. J. (2002). Multilevel analysis: techniques and applications. Mahwah, NJ: Lawrence Erlbaum Associates. Kashy, D. A., & Kenny, D. A. (2000). The analysis of data from dyads and groups. In: H. T. Reis, & C. M. Judd (Eds.), Handbook of research methods in social and personality psychology (pp. 451–477). New York, NY: Cambridge University Press. Kendall, P. C., & Flannery-Schroeder, E. C. (1998). Methodological issues in treatment research for anxiety disorders in youth. Journal of Abnormal Child Psychology, 26, 27–38. Kendall, P. C., Safford, S., Flannery-Schroeder, E., & Webb, A. (2004). Child anxiety treatment: outcomes in adolescence and impact on substance use and depression at 7.4-year follow-up. Journal of Consulting and Clinical Psychology, 72, 276–287. Kenny, D. A. (2011). Commentary: dyadic analyses of family data. Journal of Pediatric Psychology, 36, 630–633. Klimes-Dougan, B., Brand, A. E., Zahn-Waxler, C., Usher, B., Hastings, P. D., Kendziora, K., et al. (2007). Parental emotion socialization in adolescence: differences in sex, age and problem status. Social Development, 16, 326–342. Kraemer, H. C., Measelle, J. R., Ablow, J. C., Essex, M. J., Boyce, W. T., & Kupfer, D. J. (2003). A new approach to integrating data from multiple informants in psychiatric assessment and research: mixing and matching contexts and perspectives. American Journal of Psychiatry, 160, 1566–1577. Krain, A. L., & Kendall, P. C. (2000). The role of parental emotional distress in parent report of child anxiety. Journal of Clinical Child Psychology, 29, 328–335. Kristensen, H., & Torgersen, S. (2006). Social anxiety disorder in 11–12 year-old children. European Child & Adolescent Psychiatry, 15, 163–171. Lagattuta, K. H., Sayfan, L., & Bamford, C. (2012). Do you know how I feel? Parents underestimate worry and overestimate optimism compared to child self-report. Journal of Experimental Child Psychology, 113, 211–232. Laird, R. D., & De Los Reyes, A. (2013). Testing informant discrepancies as predictors of early adolescent psychopathology: why difference scores cannot tell you what you want to know and how polynomial regression may. Journal of Abnormal Child Psychology, 41, 1–14. Laird, R. D., & Weems, C. F. (2011). The equivalence of regression models using difference scores and models using separate scores for each informant: implications for the study of informant discrepancies. Psychological Assessment, 23, 388–397. Manassis, K., Tannock, R., & Monga, S. (2009). Anxious by maternal-versus selfreport: are they the same children? Journal of the Canadian Academy of Child and Adolescent Psychiatry, 18, 103–109. Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 28, 487–495. Mogg, K., Bradley, B. P., Millar, N., & White, J. (1995). A follow-up study of cognitive bias in generalized anxiety disorder. Behaviour Research and Therapy, 33, 927–935. Muris, P., Meesters, C., & Schouten, E. (2002). A brief questionnaire of DSM-IVdefined anxiety and depression symptoms among children. Clinical Psychology & Psychotherapy, 9, 430–442. Muris, P., Merckelbach, H., Schmidt, H., & Mayer, B. (1998). The revised version of the Screen for Child Anxiety Related Emotional Disorders (SCARED-R): factor structure in normal children. Personality and Individual Differences, 26, 99–112. Muris, P., Merckelbach, H., Van Brakel, A., & Mayer, A. B. (1999). The revised version of the screen for child anxiety related emotional disorders (SCARED-R): further evidence for its reliability and validity. Anxiety Stress & Coping, 12, 411–425. Niditch, L. A., & Varela, R. E. (2011). Mother–child disagreement in reports of child anxiety: effects of child age and maternal anxiety. Journal of Anxiety Disorders, 25, 450–455. Pereira, A. I., Muris, P., Barros, L., Goes, R., Marques, T., & Russo, V. (2014). Agreement and discrepancy between mother and child in the evaluation of children’s anxiety symptoms and anxiety life interference. European Child & Adolescent Psychiatry, http://dx.doi.org/10.1007/s00787-014-0583-2 Pérez-Edgar, K., & Fox, N. A. (2005). Temperament and anxiety disorders. Child and Adolescent Psychiatric Clinics of North America, 14, 681–706. Pickens, J., & Field, T. (1993). Facial expressivity in infants of depressed mothers. Developmental Psychology, 29, 986–988. Pine, D. S. (2007). Research review: a neuroscience framework for pediatric anxiety disorders. Journal of Child Psychology and Psychiatry, 48, 631–648. Pine, D. S., Cohen, P., Gurley, D., Brook, J., & Ma, Y. (1998). The risk for early-adulthood anxiety and depressive disorders in adolescents with anxiety and depressive disorders. Archives of General Psychiatry, 55, 56–64. Rapee, R. M., Barrett, P. M., Dadds, M. R., & Evans, L. (1994). Reliability of the DSMIII-R childhood anxiety disorders using structured interview: interrater and parent–child agreement. Journal of the American Academy of Child & Adolescent Psychiatry, 33, 984–992.

N.W. Affrunti, J. Woodruff-Borden / Journal of Anxiety Disorders 32 (2015) 56–65 Rapee, R. M., Schniering, C. A., & Hudson, J. L. (2009). Anxiety disorders during childhood and adolescence: origins and treatment. Annual Review of Clinical Psychology, 5, 311–341. Raudenbush, S. W., Bryk, A. S., & Congdon, R. (2004). HLM 7 for Windows [Computer software]. Lincolnwood, IL: Scientific Software International. Renouf, A. G., & Kovacs, M. (1994). Concordance between mothers’ reports and children’s self-reports of depressive symptoms: a longitudinal study. Journal of the American Academy of Child & Adolescent Psychiatry, 33, 208–216. Reuterskiöld, L., Öst, L. G., & Ollendick, T. (2008). Exploring child and parent factors in the diagnostic agreement on the Anxiety Disorders Interview Schedule. Journal of Psychopathology and Behavioral Assessment, 30, 279–290. Richters, J. E. (1992). Depressed mothers as informants about their children: a critical review of the evidence for distortion. Psychological Bulletin, 112, 485–499. Richters, J., & Pellegrini, D. (1989). Depressed mothers’ judgments about their children: an examination of the depression-distortion hypothesis. Child Development, 60, 1068–1075. Safford, S. M., Kendall, P. C., Flannery-Schroeder, E., Webb, A., & Sommer, H. (2005). A longitudinal look at parent–child diagnostic agreement in youth treated for anxiety disorders. Journal of Clinical Child and Adolescent Psychology, 34, 747– 757. Silk, J. S., Shaw, D. S., Skuban, E. M., Oland, A. A., & Kovacs, M. (2006). Emotion regulation strategies in offspring of childhood-onset depressed mothers. Journal of Child Psychology and Psychiatry, 47, 69–78. Thompson, R. A. (1998). Emotional competence and the development of self. Psychological Inquiry, 9, 308–309. Vasey, M. W., & MacLeod, C. (2001). Information-processing factors in childhood anxiety: a review and developmental perspective. In: M. W. Vasey, & M. R. Dadds (Eds.), The developmental psychopathology of anxiety (pp. 253–277). New York, NY: Oxford University Press. Warren, S. L., & Sroufe, L. A. (2004). Developmental issues. In: T. H. Ollendick, & J. S. March (Eds.), Phobic and anxiety disorders in children and adolescents: a clinician’s guided to effective psychosocial and pharmacological interventions (pp. 92–115). New York, NY: Oxford University Press.

65

Watkins, E., & Moulds, M. (2005). Distinct modes of ruminative self-focus: impact of abstract versus concrete rumination on problem solving in depression. Emotion, 5, 319–328. Watkins, E., & Teasdale, J. D. (2001). Rumination and overgeneral memory in depression: effects of self-focus and analytic thinking. Journal of Abnormal Psychology, 110, 353–357. Watkins, E., & Teasdale, J. D. (2004). Adaptive and maladaptive self-focus in depression. Journal of Affective Disorders, 82, 1–8. Weems, C. F., & Stickle, T. R. (2005). Anxiety disorders in childhood: casting a nomological net. Clinical Child and Family Psychology Review, 8, 107–134. Weems, C. F., Feaster, D. J., Horigian, V. E., & Robbins, M. S. (2011). Parent and child agreement on anxiety disorder symptoms using the DISC predictive scales. Assessment, 18, 213–216. Wells, A. (1995). Meta-cognition and worry: a cognitive model of generalized anxiety disorder. Behavioural and Cognitive Psychotherapy, 23, 301–320. West, S. G., Finch, J. F., & Curran, P. J. (1995). Structural equation models with nonnormal variables: problems and remedies. In: R. H. Hoyle (Ed.), Structural equation modeling: concepts issues, and applications (pp. 56–75). Thousand Oaks, CA: Sage Publications. Wren, F. J., Bridge, J. A., & Birmaher, B. (2004). Screening for childhood anxiety symptoms in primary care: integrating child and parent reports. Journal of the American Academy of Child & Adolescent Psychiatry, 43, 1364–1371. Yeh, M., & Weisz, J. R. (2001). Why are we here at the clinic? Parent–child (dis)agreement on referral problems at outpatient treatment entry. Journal of Consulting and Clinical Psychology, 69, 1018–1025. Youngstrom, E., Loeber, R., & Stouthamer-Loeber, M. (2000). Patterns and correlates of agreement between parent, teacher, and male adolescent ratings of externalizing and internalizing problems. Journal of Consulting and Clinical Psychology, 68, 1038–1050.