Parent cognitive-behavioral intervention for the treatment of childhood anxiety disorders: A pilot study

Behaviour Research and Therapy 61 (2014) 156e161

Contents lists available at ScienceDirect

Behaviour Research and Therapy journal homepage: www.elsevier.com/locate/brat

Shorter communication

Parent cognitive-behavioral intervention for the treatment of childhood anxiety disorders: A pilot study Allison M. Smith*, Ellen C. Flannery-Schroeder, Kathleen S. Gorman, Nathan Cook Department of Psychology, The University of Rhode Island, 10 Chafee Road, Kingston, RI 02881, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 January 2014 Received in revised form 18 August 2014 Accepted 19 August 2014 Available online 27 August 2014

Strong evidence supports cognitive-behavioral therapy (CBT) for the treatment of childhood anxiety. Many studies suggest that parents play an etiological role in the development and maintenance of child anxiety. This pilot study examined the efficacy of a cognitive-behavioral intervention delivered to the parents of 31 anxious children (ages 7e13). Parents were randomly assigned to an individual parent-only CBT intervention (PCBT, n ¼ 18) or wait-list control (WL, n ¼ 13). PCBT demonstrated significant reductions in children's number of anxiety disorder diagnoses, parent-rated interference and clinicianrated severity of anxiety, and maternal protective behaviors at post-treatment, which were maintained at 3-months. WL did not demonstrate significant changes. There were no significant differences between conditions in child self-reported or parent-report of child anxiety symptoms. Findings were replicated in a combined sample of treated participants, as well as in an intent-to-treat sample. Parentonly CBT may be an effective treatment modality for child anxiety, though future research is warranted. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Child anxiety Cognitive-behavioral therapy Parental behaviors Parenting Anxiety disorders

Anxiety disorders are the most common psychological conditions affecting youth (Kessler et al., 2005). Anxiety engenders significant functional impairment and distress in youth and their families (Ezpeleta, Keeler, Erkanli, Costello, & Angold, 2001) and does not remit without treatment (Costello, Angold, & March, 1995). However, anxiety disorders are highly treatable (James, Soler, & Weatherall, 2005). Cognitive-behavioral therapy (CBT) is designated a “probably efficacious” (Chambless & Hollon, 1998) first-line treatment for childhood anxiety disorders (Kendall, 2011). CBT is skill-based and traditionally delivered directly to the anxious child (e.g., (Kendall & Hedtke, 2006). Parents play role in the development and maintenance of child anxiety (Ginsburg & Schlossberg, 2002). Beyond the transmission of biological risk, environments fostered by parents promote or discourage adaptive coping. For instance, parents experiencing anxiety demonstrate increased cognitive biases towards threat, sensitivity to child distress, and apprehension while observing their child engage in age-normative tasks (Hudson & Rapee, 2004; Turner, Beidel, Roberson-Nay, & Tervo, 2003). Thus, parentechild

* Corresponding author. Permanent address: Department of Anesthesiology, Division of Pain Medicine, Boston Children's Hospital, 333 Longwood Avenue, 5th floor, Boston, MA 02215, USA. Tel.: þ1 857 218 5061; fax: þ1 617 730 0199. E-mail addresses: [email protected], [email protected] (A.M. Smith). http://dx.doi.org/10.1016/j.brat.2014.08.010 0005-7967/© 2014 Elsevier Ltd. All rights reserved.

interactions may model anxious responses to specific stimuli or avoidant/passive coping in general (Hadwin, Garner, & PerezOlivas, 2006). Maladaptive parent responses, such as overprotection, may also contribute to child anxiety (Ginsburg & Schlossberg, 2002; McLeod, Wood, & Avny, 2011; Wood, McLeod, Sigman, Hwang, & Chu, 2003). Protective behaviors include permitting or encouraging avoidance of feared stimuli, providing excessive reassurance, and/or rescuing the child from age-appropriate situations. These reduce autonomy and interfere with the development of adaptive coping strategies (Murray, Creswell, & Cooper, 2009; Simpson, Suarez, & Connolly, 2012) and are common among anxious parents (Murray et al., 2009; Waters, Zimmer-Gembeck, & Farrell, 2012). Despite the finding that the addition of parents to child-focused treatment has not been associated with differential treatment effectiveness (Reynolds, Wilson, Austin, & Hooper, 2012), some (Cobham, 2012; Thirlwall et al., 2013) suggest that CBT delivered solely to parents may be a viable, cost-effective modality for child anxiety. Using a transfer-of-control model (Silverman & Kurtines, 1996), skills are transferred from therapist to parent to child. Trials exploring parent-only CBT group interventions have demonstrated significant reductions in child anxiety (Cartwright-Hatton, McNally, & White, 2005; Heyne et al., 2002; Mendlowitz et al., 1999; Thienemann, Moore, & Tompkins, 2006) but are not without methodological limitations. Additional evidence suggests that individual parent-only CBT for child anxiety delivered via low-

A.M. Smith et al. / Behaviour Research and Therapy 61 (2014) 156e161

intensity modalities (e.g., bibliotherapy, telephone support) are promising (Lyneham & Rapee, 2006; Rapee, Abbott, & Lyneham, 2006; Thirlwall et al., 2013). The present study examined the efficacy of a ten-session parentonly CBT intervention delivered individually to parents of anxious children, versus a ten-week wait-list control, comparing child anxiety symptoms and diagnoses, as well as parents' self-reported anxiety and protective behaviors. We hypothesized: 1) child anxiety symptoms and diagnoses would be significantly reduced posttreatment, versus waitlist; 2) parents' self-reported anxiety and protective would be significantly reduced post-treatment, versus waitlist; and 3) post-treatment reductions would be maintained at three month follow-up and replicable in a sample of intervention completers. Method Participants Parents of anxious children were recruited via multiple community resources. Parents of 35 anxious children were eligible and consented. Of these, four withdrew prior to randomization, resulting in a final sample of 31 families (33 mothers, 23 fathers). Demographic and diagnostic characteristics are presented in Table 1. All families identified as White; one identified as Hispanic/ Latino. Mean age for mothers was 42.04 years (SD ¼ 6.03) and 45.07 years (SD ¼ 6.07) for fathers. Annual household incomes ranged from $55,000 to $300,000 (M ¼ $128,400, SD ¼ $77,730). Children (61.3% male ranged from 7 to 13 years (M ¼ 9.80, SD ¼ 1.78). They met DSM-IV diagnostic criteria for one or more of the following: Separation Anxiety Disorder, Social Anxiety Disorder, Specific Phobia, and/or Generalized Anxiety Disorder. Exclusionary criteria were comorbid pervasive developmental disorder, traumatic brain injury, organic brain damage, psychotic symptoms, or engagement in concurrent psychotherapy for anxiety. Children receiving concurrent pharmacological treatment for anxiety required stable dosages one month prior to and throughout participation. Table 1 Demographic & diagnostic characteristics of participants. Groups Variable Child's sex Male Female Family's ethnicity Hispanic/Latino Non-Hispanic/Latino Mean age of child in yrs (SD) Maternal mean age in yrs (SD) Paternal mean age in yrs (SD) Mean household income (SD) Parental status Married/Dom. Partnership Divorced, not remarried Comorbidity None þ1 Anxiety Dx þ2 or more Anxiety Dx þ Externalizing Dx

PCBT (n ¼ 18)

WL (n ¼ 13)

Difference

p

11 7

8 5

c2 ¼ .001

n.s

1 17 10.04 (1.80)

0 13 9.46 (1.76)

c2 ¼ .746 t(29) ¼ .893

n.s. n.s.

42.48 (7.49)

41.64 (2.50)

t(22) ¼ .334

n.s.

46.12 (7.21)

43.45 (3.39)

t(26) ¼ 1.141

n.s.

131,000 (82,417)

123,571 (74,202)

t(18) ¼ .199

n.s.

16

13

2

0

c2 ¼ 1.544

n.s.

1 5 12 4

3 2 8 2

157

Measures Parents completed the Anxiety Disorders Interview Schedule for Children e Parent Version (ADIS-C-IV-P; (Silverman & Albano, 1996)) with trained, blinded diagnosticians (n ¼ 9). A reliability criterion of 85% agreement (kappa) was set and obtained by all diagnosticians prior to the study. The ADIS yields cliniciangenerated severity ratings (CSRs) and parent-generated interference ratings (PIRs) for each diagnosis, ranging from 0 (least severe) to 8 (most severe). Here, CSRs were summed to create a total clinician severity score and PIRs were summed to create a total parent interference score. Children completed the 39-item Multidimensional Anxiety Scale for Children (MASC-C; (March, 1997)), the most widely used self-report measure of child anxiety (Langley, Bergman, & Piacentini, 2002). The MASC has excellent internal consistency (a ¼ .87 here). Parents then completed the MASC parent version (MASC-P). Its factor structure and internal consistency (a ¼ .89 here) parallel the child version (Baldwin & Dadds, 2007). Parents individually completed the 36-item Adult Manifest Anxiety Scale (AMAS; (Reynolds, Richmond, & Lowe, 2003)) to assess their own anxiety. The AMAS has good internal consistency (a ¼ .92 here) and high test-retest reliability (Lowe & Reynolds, 2004). Parent also completed the 25-item Parent Protection Scale (PPS; (Thomasgard, Metz, Edelbrock, & Shonkoff, 1995)) as a measure of the frequency of protective behaviors enacted toward the anxious child. The PPS has moderate internal consistency (a ¼ .63 here) in this age group (Mullins et al., 2004). Procedure All study procedures were approved by the university's Institutional Review Board and took place in the specialized child anxiety clinic housed within this institution. Interested families met with study staff to discuss the study and complete separate informed consent and assent processes. Parents then completed the ADIS to confirm eligibility and completed questionnaires. These data served as Time 1 for eligible participants. Ineligible participants were referred elsewhere. Eligible participants were then randomly assigned to the parent cognitive-behavioral treatment (PCBT, n ¼ 18) or waitlist control (WL, n ¼ 13). PCBT participants began the intervention with a randomly assigned therapist immediately. After treatment, they completed questionnaires and ADIS (Time 2). Three months later, they completed questionnaires and ADIS (Time 3). Parents assigned to WL began the waiting period. After ten weeks, they completed post-waitlist questionnaires and ADIS (Time 2). They then began with intervention with a randomly assigned therapist. After treatment, they completed questionnaires and ADIS (Time 3). Recruitment and retention rates did not differ significantly between groups at any time-point. Data from participants who pursued concurrent child-focused treatment while enrolled was deemed confounded and excluded. All other available data was used in analyses. Missing data was largely lost to follow-up; that is, families were unreachable or did not complete interviews and/or return questionnaires. Parent intervention

c2 ¼ 2.340 c2 ¼ .226

n.s. n.s.

Note. PCBT: Parent Cognitive-Behavioral Therapy group; WL: Waitlist control group; Dx: Disorder Diagnosis.

PCBT is a ten-module, individualized intervention for parents, engaging them as “consultants, collaborators, and co-clients” in the treatment of their child's anxiety (Kendall, 2011). The intervention provided psychoeducation about the nature of anxiety, discussed strategies for responding adaptively to child anxiety, and demonstrates essential cognitive-behavioral techniques for parents to teach their children (Albano & Kendall, 2002). Table 2 overviews

158

A.M. Smith et al. / Behaviour Research and Therapy 61 (2014) 156e161

Table 2 Topics and goals of intervention sessions. Module(s)

Topics/Goals

1

Orientation to treatment program; psychoeducation about nature, development, and treatment of anxiety; gathering of additional clinical information for tailoring treatment; motivational interviewing techniques to assess and enhance parent motivation and self-efficacy. Key factors in development and maintenance of anxiety; antecedents, behaviors, and consequences of anxiety; adaptive vs. maladaptive parent responses to child anxiety; rationale for facing fears (vs. escape or avoidance) and role of parents in facilitation; strategies for modeling adaptive coping in daily behavior. Skill Building: strategies for increasing affective awareness; physiological responses to anxiety; somatic symptom management strategies (e.g., deep breathing, relaxation training). Skill Building: psychoeducation regarding cognitive aspects of anxiety; relationships between thoughts, feelings, and behaviors; common thinking errors for anxious individuals; cognitive-restructuring techniques. Skill Building: problem solving strategies for parent and child; behavioral principles for shaping behavior; application of positive reinforcement for facing fears; summary and unification of skills. Skill Building: rationale for and behavioral principles guiding graded exposure tasks; development of fear hierarchy; planning and role-playing low-level exposures to be completed with child between sessions. Review of at-home exposures with attention to parents' behavioral responses to child's anxiety; trouble-shooting problem areas; planning and role-playing exposures to be completed between sessions. Summary of treatment gains; relapse prevention and maintenance; planning for future.

2

3

4

5

6

7e9

10

concepts and skills addressed in each module. Each module constituted one weekly one-hour session with the same therapist. Parent-child tasks were assigned between sessions. Therapists (n ¼ 8) were advanced doctoral psychology students who had received intensive training in CBT for anxiety and in this intervention. They participated in weekly supervision with a licensed clinical psychologist, who regularly reviewed digital session recordings to ensure treatment fidelity. Data analyses All analyses were conducted with IBM SPSS Statistics, v.19. All scales and subscale scores approximated linear, normal, and homoscedastic distributions. Mean substitution replaced missing values (<1% of cases). To test our hypotheses regarding group differences in outcomes, we conducted mixed factorial 2 (condition)  2 (time-point) analyses of variance (ANOVAs) with partial eta squared (h2p ) as a measure of effect size. Parent anxiety and protective behaviors data were first analyzed as a single group, then separately by gender. We then conducted repeated-measures ANOVAs with data collected via interview in the PCBT condition alone in order to assess maintenance of treatment gains. Based on an a-priori power analysis, low return of questionnaire data at Time 3 rendered our analyses underpowered to accurately estimate treatment effects in these data (i.e., MASC-C, MASC-P, AMAS, & PPS) across all three time-points and are therefore not included here. Next, in order to increase power, all participants who received the intervention were grouped together to compare all pre- and postintervention scores, using paired samples t-tests. Finally, treatment effects were also assessed using the intent-to-treat (ITT) sample. The last-observation-carried-forward (LOCF) method was used to address missing data at subsequent assessment time-points

for non-completers (i.e., those who completed the Time 1 assessment and were randomized to condition. Results Means and standard deviations of all dependent variables are presented by condition in Table 3. Chi-square analyses and independent samples t-tests indicated no pre-existing significant differences between conditions. Treatment outcome (PCBT vs. WL) There were significant Time  Group interactions for total number of child anxiety disorder diagnoses, F (1, 28) ¼ 15.40, p < .01, h2p ¼ .36, parent interference scores, F (1, 28) ¼ 15.92, p < .001, h2p ¼ .36, and clinician severity scores, F (1, 28) ¼ 12.01, p < .01, h2p ¼ .30. Mean scores for each of these variables decreased significantly in PCBT, but did not change significantly in WL. There were no significant interactions or main effects for MASC-C and MASC-P scores, or for AMAS and PPS scores, when analyzing parents as one group. There was, however, a significant Time  Group interaction for maternal (but not paternal) PPS scores, F (1, 21) ¼ 8.05, p < .05, h2p ¼ .28, which decreased significantly in PCBT, but did not change significantly in WL. AMAS scores were not significantly different between conditions when analyzed separately by gender. Maintenance of PCBT treatment gains: 3-month follow-up Examination of the PCBT condition at all three time-points indicated significant changes in total number of child anxiety disorder diagnoses over time, F (2, 10) ¼ 16.92, p < .01, h2p ¼ .77, with a significant reduction in diagnoses from pre-to post-treatment and a further yet non-significant (p ¼ .08) reduction in scores from posttreatment to follow-up. There were also significant changes in parent interference scores over time, F (2, 10) ¼ 10.74, p < .01, h2p ¼ .68, with a significant reduction in scores from pre-to posttreatment and no significant change in scores from post-treatment to follow-up (p ¼ .32). Finally, clinician severity scores also differed significantly over time, F (2,10) ¼ 13.43, p < .01, h2p ¼ .73, with a significant reduction in scores from pre-to post-treatment and another significant reduction from post-treatment to follow-up (p < .05). Combined sample: pre-post intervention analyses When examining the combined sample, we similarly found significant pre-to post-treatment decreases in total number of child anxiety disorder diagnoses, t (25) ¼ 7.28, p < .001, parent interference scores, t (25) ¼ 5.42, p < .001, and clinician severity scores, t (25) ¼ 5.80, p < .001. We also found no significant pre-to posttreatment changes in MASC-P or MASC-C scores in the combined sample. However, contrary to earlier findings, in the combined sample, we found significant pre-to post-treatment decreases in parent PPS scores, t (35) ¼ 3.75, p < .01, and AMAS scores, t (35) ¼ 2.93, p < .01. Intent-to-treat analyses Similar to the original analyses, in the ITT sample, there were significant Time  Group interactions for total number of child anxiety disorder diagnoses, F (1, 29) ¼ 14.13, p < .01, h2p ¼ .33, parent interference scores, F (1, 29) ¼ 14.81, p < .01, h2p ¼ .34, and clinician severity scores, F (1, 29) ¼ 11.16, p < .01, h2p ¼ .28. Mean scores for each of these variables decreased significantly in PCBT, but did not

A.M. Smith et al. / Behaviour Research and Therapy 61 (2014) 156e161

159

Table 3 Means & standard deviations for diagnostic interview & parent/child report measures. PCBT

Waitlist

Pre-Tx

Post-Tx

3.67 (1.68) 21.28 (10.05) 20.06 (8.75)

1.88 (1.62) 12.71 (8.59) 12.71 (7.27)

57.71 13.53 50.02 58.44

54.32 11.73 47.19 52.21

3-mo f/u

Pre-WL

Post-WL/Pre-Tx

Post-Tx

2.92 (1.50) 15.15(6.03) 16.77 (6.73)

3.23 (1.59) 17.46 (7.88) 18.15 (7.37)

1.33 (1.73) 10.11 (9.91) 9.78 (9.36)

51.29 15.89 51.42 53.50

52.98 17.61 51.00 51.11

52.86 11.58 46.67 55.00

a

Diagnostic interview Total number of diagnoses Parent interference score Clinician severity score Parent & child report Child anxiety e parent reportb Total parent anxietyc Protective behaviorsc Child anxiety e child reportd Notes. a N for b N for c N for d N for

(14.84) (7.65) (5.82) (16.14)

(17.14) (6.71) (6.36) (20.48)

1.08 (0.90) 7.83 (5.36) 7.83 (5.13) 54.25 14.78 47.56 47.00

(13.37) (10.74) (6.88) (16.78)

(15.00) (8.32) (5.42) (17.68)

(14.08) (7.63) (5.92) (23.63)

(14.58) (6.96) (3.96) (24.85)

diagnostic interview variables ¼ 31 at Time 1, 30 at Time 2, 21 at Time 3. child anxiety e parent report ¼ 29 at Time 1, 21 at Time 2, 13 at Time 3. parent anxiety & protective behaviors ¼ 50 at Time 1, 39 at Time 2, 21 at Time 3. child anxiety e child report ¼ 26 at Time 1, 21 at Time 2, 13 at Time 3.

change significantly in WL. There were no significant Time  Group interactions for MASC-C and MASC-P scores, or for AMAS and PPS scores, when analyzing parents as a single group. There was, however, a significant Time  Group interaction for maternal (but not paternal) PPS scores, F (1, 29) ¼ 8.99, p < .01, h2p ¼ .24, which decreased significantly in PCBT, but did not change significantly in WL. AMAS scores were not significantly different between conditions when analyzed separately by gender. Discussion This pilot study examined the efficacy of a ten-session cognitivebehavioral intervention delivered individually to the parents of anxious children versus a ten week wait-list control condition, in an effort to replicate and extend previous parent-only intervention trials. After participating in the intervention, the PCBT condition reported significant decreases in the total number of child anxiety disorder diagnoses, parent's ratings of their child's functional impairment, and clinician's ratings of child anxiety disorder severity, after receiving the intervention, while the WL group did not report significant changes after the waiting period. These gains were maintained at 3-month follow-up, with clinician severity ratings decreasing significantly again at follow-up. In the combined sample of treated participants from both conditions, we again found that these three variables decreased significantly from pretreatment to post-treatment, as well as significant decreases in parent anxiety and parent protective behaviors. Given the considerable number of non-completers of the treatment (primarily due to study design factors that could not be controlled), we confirmed the findings described above by repeating the analyses in our intent-to-treat sample, in order to better reflect any failures to respond to treatment. Naturally, as this is the more conservative analysis, the effect size estimates are smaller but likely represent the most accurate estimate of the likely effect size. These findings, though preliminary, suggest that it may be possible to reduce child anxiety without directly involving the child in therapy. Like previous parent-only interventions (Thienemann et al., 2006; Waters, Ford, Wharton, & Cobham, 2009), this intervention required parents to engage in home activities with their child to facilitate treatment goals, while also modifying behaviors known to elicit and maintain anxiety. Thus, parents participating in such interventions may take greater responsibility for modifying their own behaviors (Waters et al., 2009) and working with their child between sessions than they might in child-only treatment modality. Because parents naturally interact with their child more

frequently than would a therapist, parents can routinely process challenges and reinforce the child's adaptive coping in real-time. While we were unable to assess mechanisms of change, it is theoretically plausible that decreases in parent protective behaviors contributed to decreases observed in child anxiety diagnoses, functional impairment, and severity. It is viewed as natural for parents to want to protect or remove their child from distressing situations. However, when such situations are developmentally appropriate, such behaviors may inadvertently preclude children from learning adaptive skills and convey that the child is incapable of coping independently (Simpson et al., 2012). This intervention aim to decrease protective behaviors (and relatedly, the distress associated with allowing their child to face fears) through education. For some, this may have provided the child with increased opportunities to develop and practice coping skills. Contrary to our hypotheses and findings reported via structured interview, there were no significant differences between conditions in parent-report and child self-report of child anxiety symptoms. Particularly curious is the fact that parent reports varied by measurement tool (significant change via ADIS vs. no change via MASCP). Such differences highlight the limitation of relying exclusively on one form of reporting to detect change. While the ADIS appeared sensitive to treatment change, the MASC did not detect treatment effects in the smaller subset of the sample that completed questionnaires at Times 2 and 3. Further, the ADIS and MASC measure slightly different, though related, outcomes. Some (e.g., (Manassis, 2000)) suggest that anxiety-related impairment (e.g., ADIS parent interference ratings) is a more meaningful measure of anxiety severity than symptom counts. Further, with exposure-based treatment, increases in function (i.e., reductions in impairment) often occur prior to meaningful decreases in anxiety symptomatology, as anxiety may linger until fully habituating. Therapeutic change may reflect “sowing and reaping” (Nauta, Scholing, Emmelkamp, & Minderaa, 2001), whereby skills are transferred (“sown”) from therapists to parents to children during treatment (i.e., sowing seeds), but meaningful decreases in child anxiety (“reaping”) are not observed until sufficiently practicing new skills and managing anxiety independently. We were unable to assess these effects here. Contrary to our hypotheses, there were no significant differences between conditions in parents' own anxiety symptoms, which were examined as a secondary outcome of this child-focused intervention, rather than a treatment target. However, parent anxiety, particularly anxiety in relation to their child, remains relevant, given that the transfer-of-control model requires parents to manage distress in order to effectively communicate and model

160

A.M. Smith et al. / Behaviour Research and Therapy 61 (2014) 156e161

appropriate coping for their child. Notably, in the combined sample, there was a significant pre-to post-treatment decrease in parent anxiety, suggesting that with a larger sample, an effect was in fact detectable. This is consistent with studies (Cobham, Dadds, & Spence, 1998) reporting beneficial effects for parents involved in child treatment. Overall, this study bridges gaps in the literature by evaluating an individualized parent-only child anxiety intervention that explicitly teaches cognitive-behavioral strategies to be transferred from therapist to parent to child. It answers the call for studies that link maladaptive parenting behaviors with the development and maintenance of child anxiety (McLeod et al., 2011; Wood et al., 2003) and studies that measure individual treatment elements (Breinholst, Esbjørn, Reinholdt-Dunne, & Stallard, 2012). Specifically, this intervention attends to parents' protective behaviors as a target of both intervention and assessment and emphasizes parent promotion of adaptive coping (Simpson et al., 2012). The individual treatment modality of this intervention allowed for targeted treatment (versus generalized group treatment or bibliotherapy). The study employed a multimodal assessment from multiple perspectives. Further, because the intervention was conducted with parents, there was minimal disruption to typical childhood activities. Finally, given that CBT is standard practice for many clinicians, the intervention can be widely implemented. Several study limitations call for a cautious interpretation of the findings described here. While the study was sufficiently powered to detect large treatment effects, the small sample size reduced power to detect small to moderate between-group differences. As such, mediation/moderation analyses were considered impractical. Missing follow-up data precluded the assessment of change in data collected via questionnaire and increased potential attrition bias. There is also risk of demand characteristics on parent-reported variables. Simply by participating in the intervention, parents could not be blinded to their condition and time-point. Further, after investing in an intervention aimed at reducing anxiety, parents may have felt motivated to report improvements (rather than deterioration or no change). While including the child ADIS may have mitigated this potential for bias on the ADIS, we prioritized reducing burden and avoiding inadvertent therapeutic intervention (e.g., spontaneous psychoeducation). Similarly, with regard to parent protective behaviors, since this intervention targeted responses to child anxiety across sessions, parents were likely aware of what behaviors “should” change over time. The sample's lack of racial/ethnic and socioeconomic diversity also limits the generalizability of these findings to White, middle to upper class families in suburban settings. Cultural differences in parenting and methods of coping, as well as perceived importance of and access to behavioral health care, may influence intervention engagement and implementation. Still, race and ethnicity have not been found to be strong correlates of child anxiety, with presentation appearing consistent across gender, family size, parents' marital status, education level, or race/ethnicity (Canino et al., 2004; Kendall et al., 2010; Pina, Zerr, Villalta, & Gonzales, 2012). Additionally, while the wait-list control accounts for the passage of time and spontaneous remission, without an active comparison group, we cannot conclude how PCBT would compare with other treatment approaches for child anxiety. On primarily ethical grounds, we utilized a wait-list control to first establish this new intervention and unique modality as effective in comparison to no treatment. Finally, the method of randomization (i.e., coin toss) may have increased risk of bias, though research staff was not privy to participant clinical information and made no attempts to balance groups. Taken together, these limitations necessitate that results from this pilot study be considered preliminary until a large randomized controlled trial with an active comparison group is undertaken.

The extant literature suggests that parents may indeed have a considerable part to play in the treatment of childhood anxiety. Our study piloted one potential approach involving parents with promising yet very preliminary findings. The limitations described above offer several future directions for replication studies utilizing larger samples: 1) Establish that treatment gains observed by parents and clinicians here are replicable in more diverse samples. 2) Demonstrate child-reported reductions in anxiety. 3) Compare treatment gains in parent-only interventions with those observed in evidence-based CBT modalities. Such studies should elucidate specific treatment elements that contribute to therapeutic gains, allowing researchers to examine mechanisms of therapeutic change (e.g., modification of parent behavior vs. traditional CBT facets). Selecting of outcomes measures should extend beyond symptom checklists to include measures of functional outcomes and quality of life, as well as objective behavior measures. Finally, complete follow-up data would allow researchers to fully examine potential benefits beyond post-treatment and three months. To conclude, this pilot study examined a new treatment modality for child anxiety in which parents were trained parents to “transfer” CBT strategies from the therapist to their child, utilizing externally valid situations to practice those skills with the child in real-time. Most parent participants reported observing decreased anxiety and functional impairment in their child upon parent completion of the intervention. With sufficient replication and extension of these findings, we may find that empowering parents as lay-therapists increases sustainability and transportability of CBT for child anxiety. Conflicts of interest None. References Albano, A. M., & Kendall, P. C. (2002). Cognitive behavioural therapy for children and adolescents with anxiety disorders: clinical research advances. International Review of Psychiatry, 14(2), 129e134. Baldwin, J. S., & Dadds, M. R. (2007). Reliability and validity of parent and child versions of the multidimensional anxiety scale for children in community samples. Journal of the American Academy of Child & Adolescent Psychiatry, 46(2), 252e260. Breinholst, S., Esbjørn, B. H., Reinholdt-Dunne, M. L., & Stallard, P. (2012). CBT for the treatment of child anxiety disorders: a review of why parental involvement has not enhanced outcomes. Journal of Anxiety Disorders, 26(3), 416e424. Canino, G., Shrout, P. E., Rubio-Stipec, M., Bird, H. R., Bravo, M., Ramirez, R., et al. (2004). The DSM-IV rates of child and adolescent disordersin Puerto Rico: prevalence, correlates, Service Use, and the effects of impairment. Archives of General Psychiatry, 61(1), 85e93. Cartwright-Hatton, S., McNally, D., & White, C. (2005). A new cognitive behavioural parenting intervention for families of young anxious children: a pilot study. Behavioural and Cognitive Psychotherapy, 33(02), 243e247. Chambless, D. L., & Hollon, S. D. (1998). Defining empirically supported therapies. Journal of Consulting and Clinical Psychology, 66(1), 7. Cobham, V. E. (2012). Do anxiety-disordered children need to come into the clinic for efficacious treatment? Journal of Consulting and Clinical Psychology, 80(3), 465. Cobham, V. E., Dadds, M. R., & Spence, S. H. (1998). The role of parental anxiety in the treatment of childhood anxiety. Journal of Consulting and Clinical Psychology, 66(6), 893. Costello, E., Angold, A., & March, J. (1995). Anxiety disorders in children and adolescents. Epidemiology (pp. 109e124). New York, London: Guilford. Ezpeleta, L., Keeler, G., Erkanli, A., Costello, E. J., & Angold, A. (2001). Epidemiology of psychiatric disability in childhood and adolescence. Journal of Child Psychology and Psychiatry, 42(7), 901e914. Ginsburg, G. S., & Schlossberg, M. C. (2002). Family-based treatment of childhood anxiety disorders. International Review of Psychiatry, 14(2), 143e154. Hadwin, J. A., Garner, M., & Perez-Olivas, G. (2006). The development of information processing biases in childhood anxiety: a review and exploration of its origins in parenting. Clinical Psychology Review, 26(7), 876e894. Heyne, D., KIng, N. J., Tonge, B. J., Rollings, S., Young, D., Pritchard, M., et al. (2002). Evaluation of child therapy and caregiver training in the treatment of school refusal. Journal of the American Academy of Child & Adolescent Psychiatry, 41(6), 687e695.

A.M. Smith et al. / Behaviour Research and Therapy 61 (2014) 156e161 Hudson, J. L., & Rapee, R. M. (2004). From anxious temperament to disorder: An etiological model. James, A., Soler, A., & Weatherall, R. (2005). Cognitive behavioural therapy for anxiety disorders in children and adolescents. Cochrane Database of Systematic Reviews, 4. Kendall, P. C. (2011). Child and adolescent therapy: Cognitive-behavioral procedures. Guilford Press. Kendall, P. C., Compton, S. N., Walkup, J. T., Birmaher, B., Albano, A. M., Sherrill, J., et al. (2010). Clinical characteristics of anxiety disordered youth. Journal of Anxiety Disorders, 24(3), 360e365. Kendall, P. C., & Hedtke, K. A. (2006). Cognitive-behavioral therapy for anxious children: Therapist manual. Workbook Pub. Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593e602. Langley, A. K., Bergman, R. L., & Piacentini, J. C. (2002). Assessment of childhood anxiety. International Review of Psychiatry, 14(2), 102e113. Lowe, P. A., & Reynolds, C. R. (2004). Psychometric analyses of the Adult Manifest Anxiety ScaleeAdult version among young and middle-aged adults. Educational and Psychological Measurement, 64(4), 661e681. Lyneham, H. J., & Rapee, R. M. (2006). Evaluation of therapist-supported parentimplemented CBT for anxiety disorders in rural children. Behaviour Research and Therapy, 44(9), 1287e1300. Manassis, K. (2000). Childhood anxiety disorders: lessons from the literature. Canadian Journal of Psychiatry, 45(8), 724e730. March, J. (1997). Manual for the Multidimensional Anxiety Scale for Children (MASC). Toronto: Multi-Health Systems. McLeod, B. D., Wood, J. J., & Avny, S. B. (2011). Parenting and child anxiety disorders. Handbook of child and adolescent anxiety disorders (pp. 213e228). Springer. Mendlowitz, S. L., Manassis, K., Bradley, S., Scapillato, D., Miezitis, S., & Shaw, B. E. (1999). Cognitive-behavioral group treatments in childhood anxiety disorders: the role of parental involvement. Journal of the American Academy of Child & Adolescent Psychiatry, 38(10), 1223e1229. Mullins, L. L., Fuemmeler, B. F., Hoff, A., Chaney, J. M., Van Pelt, J., & Ewing, C. A. (2004). The relationship of parental overprotection and perceived child vulnerability to depressive symptomotology in children with type 1 diabetes mellitus: the moderating influence of parenting stress. Children's Health Care, 33(1), 21e34. Murray, L., Creswell, C., & Cooper, P. (2009). The development of anxiety disorders in childhood: an integrative review. Psychological Medicine, 39(9), 1413. Nauta, M. H., Scholing, A., Emmelkamp, P. M., & Minderaa, R. B. (2001). Cognitivebehavioural therapy for anxiety disordered children in a clinical setting: does additional cognitive parent training enhance treatment effectiveness? Clinical Psychology & Psychotherapy, 8(5), 330e340.

161

Pina, A. A., Zerr, A. A., Villalta, I. K., & Gonzales, N. A. (2012). Indicated prevention and early intervention for childhood anxiety: a randomized trial with Caucasian and Hispanic/Latino youth. Journal of Consulting and Clinical Psychology, 80(5), 940. Rapee, R. M., Abbott, M. J., & Lyneham, H. J. (2006). Bibliotherapy for children with anxiety disorders using written materials for parents: a randomized controlled trial. Journal of Consulting and Clinical Psychology, 74(3), 436. Reynolds, C. R., Richmond, B. O., & Lowe, P. A. (2003). The Adult Manifest Anxiety Scale (AMAS): Manual. Western Psychological Services (WPS). Reynolds, S., Wilson, C., Austin, J., & Hooper, L. (2012). Effects of psychotherapy for anxiety in children and adolescents: a meta-analytic review. Clinical Psychology Review, 32(4), 251e262. Silverman, W., & Albano, A. (1996). The anxiety disorders interview schedule for DSMIV: Child and parent versions. San Antonio, TX: The Psychological Corporation. Silverman, W. K., & Kurtines, W. M. (1996). Anxiety and phobic disorders: A pragmatic approach. Springer. Simpson, D., Suarez, L., & Connolly, S. (2012). Treatment and outcomes for anxiety disorders among children and adolescents: a review of coping strategies and parental behaviors. Current Psychiatry Reports, 14(2), 87e95. Thienemann, M., Moore, P., & Tompkins, K. (2006). A parent-only group intervention for children with anxiety disorders: pilot study. Journal of the American Academy of Child & Adolescent Psychiatry, 45(1), 37e46. Thirlwall, K., Cooper, P. J., Karalus, J., Voysey, M., Willetts, L., & Creswell, C. (2013). Treatment of child anxiety disorders via guided parent-delivered cognitivebehavioural therapy: randomised controlled trial. The British Journal of Psychiatry, 203(6), 436e444. Thomasgard, M., Metz, W. P., Edelbrock, C., & Shonkoff, J. P. (1995). Parent-child relationship disorders. Part I. Parental overprotection and the development of the Parent Protection Scale. Journal of Developmental & Behavioral Pediatrics, 16(4), 244e250. Turner, S. M., Beidel, D. C., Roberson-Nay, R., & Tervo, K. (2003). Parenting behaviors in parents with anxiety disorders. Behaviour Research and Therapy, 41(5), 541e554. Waters, A. M., Ford, L. A., Wharton, T. A., & Cobham, V. E. (2009). Cognitivebehavioural therapy for young children with anxiety disorders: comparison of a child þ parent condition versus a parent only condition. Behaviour Research and Therapy, 47(8), 654e662. Waters, A. M., Zimmer-Gembeck, M. J., & Farrell, L. J. (2012). The relationships of child and parent factors with children's anxiety symptoms: parental anxious rearing as a mediator. Journal of Anxiety Disorders, 26(7), 737e745. Wood, J. J., McLeod, B. D., Sigman, M., Hwang, W. C., & Chu, B. C. (2003). Parenting and childhood anxiety: theory, empirical findings, and future directions. Journal of Child Psychology and Psychiatry, 44(1), 134e151.