- Email: [email protected]
Harm beliefs and coping expectancies in youth with specific phobias
Accepted Manuscript Harm beliefs and coping expectancies in youth with specific phobias Thomas H. Ollendick, Lars-Göran Öst, Sarah M. Ryan, Nicole N. Capriola, Lena Reuterskiöld PII:
S0005-7967(17)30017-7
DOI:
10.1016/j.brat.2017.01.007
Reference:
BRT 3084
To appear in:
Behaviour Research and Therapy
Received Date: 23 August 2016 Revised Date:
10 January 2017
Accepted Date: 17 January 2017
Please cite this article as: Ollendick, T.H., Öst, L.-G., Ryan, S.M., Capriola, N.N., Reuterskiöld, L., Harm beliefs and coping expectancies in youth with specific phobias, Behaviour Research and Therapy (2017), doi: 10.1016/j.brat.2017.01.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Harm Beliefs and Coping Expectancies in Youth with Specific Phobias Thomas H. Ollendick1, Lars-Göran Öst2, Sarah M. Ryan1, Nicole N. Capriola1, and Lena
1
RI PT
Reuterskiöld2
Virginia Tech
SC
Child Study Center Department of Psychology
M AN U
Blacksburg, VA 24061
Phone: (540) 231-6451; Fax: (540) 231-3652
2
Stockholm University
TE D
Department of Psychology
EP
SE-106 91 Stockholm, Sweden
Correspondence concerning this article should be addressed to:
AC C
Sarah Ryan
e-mail: [email protected] Tel (540) 231-3514
Fax (540) 231-3652
ACCEPTED MANUSCRIPT
Abstract Catastrophic beliefs and lowered coping expectancies are often present in individuals with specific phobias (SPs). The current study examined these beliefs and expectancies in 251
RI PT
youth who received One Session Treatment for one of the three most common types of SP in youth (animals, natural environment, and situational). We compared the children’s subjective beliefs to objective ratings of the likelihood of occurrence and the dangerousness of the feared
SC
events. Results revealed pre-treatment differences in the youths’ beliefs across phobia types and age. Specifically, children with animal phobias rated their beliefs as more likely to occur than did
M AN U
children with environmental and situational phobias. In addition, older children rated their beliefs as more dangerous than younger children. However, regardless of phobia type or child age, the beliefs improved following treatment. Changes in catastrophic beliefs and coping expectancies were related to changes in clinical severity following treatment but not 6-months following
TE D
treatment. Moreover, at pre-treatment, children viewed their beliefs as significantly more catastrophic and likely to occur than did independent coders of these beliefs; however, these differences were no longer evident following treatment. Clinical implications are discussed,
AC C
outcomes.
EP
highlighting how changes in beliefs and expectancies might be associated with treatment
Keywords: Phobic Beliefs, Specific Phobias, Children and Adolescents, Randomized Controlled Trial
2
ACCEPTED MANUSCRIPT
Harm Beliefs and Coping Expectancies in Youth with Specific Phobias Specific phobias (SPs) are characterized by excessive fear or anxiety about a specific object or situation (APA, 2000, 2013). The phobic object or situation is avoided or endured with
RI PT
intense fear or anxiety and, typically, the fear or anxiety persists for 6 months or longer and in many cases for decades. Moreover, the perceived fear or anxiety is typically unrealistic and out of proportion to the actual or “real” danger posed by the phobic stimulus (e.g., “the dog will
SC
jump on me and bite me, and I will bleed to death,” “the lightning will strike our home and kill me, my mom and my dad”). According to DSM-5 (APA, 2013), SPs are categorized into five
M AN U
subtypes: animal type (e.g., dogs, spiders, insects), natural environment type (e.g., heights, storms, water), situational type (e.g., flying, elevators, enclosed places), blood-injection-injury type (BII; e.g., seeing blood, needles, invasive procedures), and other type (e.g., choking, loud sounds, costumed characters). Clinically significant SPs are present in approximately 5% of
TE D
children in community samples and about 15-20% of children in anxiety disorder clinics (APA, 2000, 2013; Kessler, Berglund, Demler, Jin, & Walters, 2005). For many children, SPs result in considerable academic, social, and personal distress, as well as interference in day-to-day
EP
functioning (Essau, Conradt, & Petermann, 2000; Kendall, Safford, Flannery-Schroeder, & Webb, 2004; Ollendick, King, & Muris, 2004).
AC C
In a host of systematic reviews over the past 15 - 20 years, cognitive-behavioral therapy
(CBT) has been shown to be efficacious in treating youth with phobic and anxiety disorders, with about 65 – 80% of youth clinically improved and approximately 50 – 65% free of their diagnosis following treatment (e.g., In-Albon & Schneider, 2006; Higa-McMillan, Francis, Rith-Najarian, & Chorpita, 2016; Ollendick & King, 2012; Silverman, Pina, & Viswesvaran, 2008). For SPs in youth, One Session Treatment (OST), a variant of CBT developed by Öst (1989, 1997), has been
3
ACCEPTED MANUSCRIPT
found to be a rapid and effective treatment (see Ollendick & Davis, 2013, and Öst, 2012, for recent reviews). This 3-hour, one session treatment involves psychoeducation, in vivo exposure, social reinforcement, participant modeling, and behavioral experiments to challenge distorted
RI PT
and oftentimes catastrophic beliefs that characterize SPs. Although studies documenting the efficacy of OST have shown significant improvements in outcomes including fear levels,
avoidant behaviors, and, in some cases, reduced physiological arousal (e.g., Allen, Allen, Austin,
SC
Waldron, & Ollendick, 2015; Öst, Svensson, Hellström, & Lindwall, 2001), few studies have examined the distorted and catastrophic beliefs that characterize this disorder and are the central
M AN U
focus of long-standing information processing models of fear and anxiety (Beck, Emery, & Greenberg, 1985; Wright & Borden, 1991). In recent years, Öst (2012) and Ollendick and Muris (2015) have reaffirmed the role of these beliefs in the maintenance of phobias, suggesting that these catastrophic beliefs regarding the feared stimulus serve to maintain the phobic anxiety and
TE D
that the avoidance observed in SPs prevents disconfirmation of these beliefs. Although it is oftentimes assumed that these distorted beliefs change as a result of OST, this has not been investigated heretofore in clinical samples of youth with SPs.
EP
Based on the seminal work of Thorpe and Salkovskis (1995) and Anthony, Brown, and Barlow (1997) on phobic beliefs in the maintenance of SPs in adults and the pioneering work of
AC C
Bögels, Snieder, and Kindt (2003) on the specificity of dysfunctional thinking in youth with different anxiety disorders (e.g., children with separation anxiety disorder overestimate the probability of being abandoned and children with social anxiety disorder overestimate the danger of being rejected and negatively evaluated by others), we set about to examine the specificity of beliefs of youth with a SP by exploring the specific beliefs of children and adolescents with the three most prevalent subtypes of SPs (animal, natural environment, and situational subtypes; see
4
ACCEPTED MANUSCRIPT
Ollendick & Muris, 2015). In as much as previous work has shown that youth with situational/environmental phobias evince more impairment than youth with animal phobias (Ollendick, Raishevich, Davis, Sirby, & Öst, 2010), we specifically examined whether type of
RI PT
phobia was related to these beliefs. In Study 1, we examined “how likely” it would be that the anticipated belief would occur, “how bad” it would be if the content of the belief were to actually occur, and how well the child could cope with the anticipated outcome should it occur. These
SC
beliefs were obtained from youth who participated in two randomized clinical control trials examining the efficacy of OST with clinic-referred youth between 6 and 16 years of age
M AN U
(Ollendick et al., 2009, 2015). In addition, we rated the likelihood and probable outcomes of these beliefs based on independent, objective ratings provided by our research team. As a result, we were able to compare the subjective beliefs of the youth to more objective probabilities associated with these beliefs. In line with information processing models of fear and anxiety
TE D
(e.g., Beck et al., 1985), we anticipated that the subjective beliefs of the youth would be overestimated compared to the more objective evaluations determined by independent raters. Given the exploratory nature of Study 1, however, no directional hypotheses were offered.
EP
In Study 2, we sought to examine how changes in these beliefs were related to clinical outcomes, specifically changes in clinician-assigned severity ratings (CSR) for the child’s
AC C
specific phobia. Data for Study 2 were derived from the same two randomized clinical control trials used in Study 1. The beliefs were obtained from the youth themselves not only at pretreatment but also at post-treatment and 6-months following treatment in these studies. In as much as extinction learning is the theoretical basis of exposure therapy wherein repeated exposures to the phobic stimulus provide corrective information that challenges beliefs and danger expectancies (cf, Craske et al., 2008; Öst, 2012), we anticipated that these beliefs would
5
ACCEPTED MANUSCRIPT
change across time and be associated with changes in the clinical severity of the phobias following treatment. In addition to these dysfunctional beliefs, we also obtained ratings on coping
RI PT
expectancies of the youth. To assess this dimension, we asked the youth to rate how sure they would be that they could “handle or deal with” the feared event or situation should it actually occur. Early on, Beck and colleagues (1985) suggested that an underestimation of one’s abilities
SC
to cope with the danger or threat present in the phobic stimulus might result from both the
overestimation of the distorted belief and the continued avoidance of the feared stimulus. Here,
M AN U
we expected the coping statements of youth to be positively affected by our OST intervention and to show parallel changes to those observed with the dysfunctional beliefs. That is, the dysfunctional beliefs would be reduced and the youth’s coping estimates would be enhanced as a result of treatment. In as much as the dysfunctional beliefs and coping statements were obtained
TE D
at the same points in time, we did not make predictions about the order or sequencing of these two facets of the phobic response, only that they would both change in therapeutic directions. Method
EP
Participants
As noted, participants were drawn from two randomized clinical control trials (RCTs)
AC C
examining the effectiveness of OST for youth with a specific phobia (Ollendick et al., 2009, 2015). All youth were recruited through referrals from child psychiatric services, school health services, pediatricians, and print advertisements. For both RCTs, the following inclusion criteria were enlisted: (a) the participants had to be between 6 and 16 years of age, (b) the participants had to have a specific phobia diagnosis, according to DSM-IV (APA, 2004) criteria, (c) the phobia had to result in significant impairment/distress and be clinical in nature as established
6
ACCEPTED MANUSCRIPT
through a semi-structured diagnostic interview, (d) the duration of impairment needed to be at least 6 months, (e) the participants did not meet criteria for any specific disorders meeting exclusion criteria (i.e., primary major depressive disorder, pervasive developmental disorder,
RI PT
drug or alcohol abuse, and/or psychosis), and (f) the participants had to agree to discontinue other forms of psychotherapy and to be stable on any medications for the duration of the study. One hundred and sixty-five youth from Sweden and the United States of America
SC
between the ages of 7 and 15 with animal, situational and environment phobias were recruited between 2001 and 2006 for participation in the first RCT (Ollendick et al., 2009) comparing the
M AN U
effects of OST (n = 86) and Education Support Treatment (EST) (n = 79; Table 1; please note that youth 6 and 16 years of age were excluded for the current study due to the small number of youth at each of these ages). In addition, 86 youth between the ages of 7 and 15 from the United States of America with these three types of specific phobia were recruited between 2007 and
TE D
2013 for participation in a second RCT (Ollendick, et al., 2015) comparing the effects of standard, individual OST (n = 41) to a parent-augmented OST (i.e., A-OST; n = 45; please see below; Table 1). Study 1 combined data from the children who participated in both RCTs (n =
EP
165 + 86 = 251, mean age = 9.87 years, 89% Caucasian, mean family income = $75,165). In Study 2, only children and adolescents randomized to OST and A-OST (n = 172, mean age =
AC C
9.69 years, 89% Caucasian, mean family income = $73, 498) were included, excluding the 79 youth in the EST condition from Study 1. Since no differences were observed between the standard and augmented conditions in the second RCT, participants in the two treatment groups were combined for purposes in Study 2. Table 1 also provides both samples’ distributions across the specific phobia subtypes. Procedure
7
ACCEPTED MANUSCRIPT
Upon initial contact, the participants’ parents completed a brief telephone screen to determine eligibility (i.e., their child was free of any disorders which met exclusion criteria) and to confirm the presence of a probable specific phobia diagnosis. Following the phone screen,
RI PT
eligible families were invited to participate in a pre-treatment assessment session. All children and their parents provided informed written assent and consent, respectively, at the start of the study. During the assessment session, both children and parents participated in a clinical intake
SC
wherein a diagnostic interview (see below) was administered to the parent and the child
separately. Parents and children also completed a battery of questionnaires which are not
M AN U
analyzed for purposes of the present study. The entire assessment session lasted approximately 3 hours. Children provided their specific phobic beliefs during the interview with the assessment clinician.
Subsequent assessment sessions, each lasting about two hours, were conducted at one-
TE D
week post-treatment and six months following treatment. At each follow-up assessment session, the diagnostic modules endorsed at pre-treatment were re-administered, the phobic beliefs and self-efficacy expectancies were rated by the child on a 0 to 8 scale using an expectancy
EP
thermometer, and a battery of questionnaires were administered. Further details on the procedures of the original studies can be found in their respective publications. The children’s
AC C
phobic beliefs (how likely, how bad) were also independently coded by our research team (see below as well as Table 2). Treatments
The child focused OST treatment condition was based on principles developed by Öst
(1989, 1997) for adults and subsequently manualized for the treatment of SPs in youth (Ollendick et al., 2009; Öst et al., 2001). OST consisted of a three hour treatment session
8
ACCEPTED MANUSCRIPT
including gradual in-vivo exposure and challenging of the child’s distorted and catastrophic cognitions. Parents were not involved in the treatment session. The A-OST treatment condition (Ollendick et al., 2015) was developed to incorporate the
RI PT
primary caregiver into the treatment session. Similar to the OST condition, the clinician assigned to work with the child engaged the child in gradual in-vivo exposures and challenged the child’s catastrophic cognitions. The parent observed the treatment session with a second clinician and
SC
the clinician coached the parent on how to conduct exposures, how to reinforce approach
behaviors, and how to reduce reinforcement of avoidance behaviors. As noted above, differences
M AN U
between the standard OST and the augmented OST were not observed; hence, data were combined across the two treatment conditions. Measures
The Anxiety Disorders Interview Schedule for DSM-IV-Child and Parent Versions
TE D
(ADIS-IV-C/P; Silverman & Albano, 1996). The ADIS-IV-C (child version) and ADIS-IV-P (parent version) are semi-structured clinical interviews designed to facilitate diagnosis of anxiety and mood disorders as well as other psychiatric disorders in children and adolescents. Clinicians
EP
assign a severity rating (CSR) on a 9-point scale (0-8, with any rating ≥ 4 being indicative of probable diagnosis and a clinical level of interference). For the present study, separate clinicians
AC C
administered the ADIS-C and ADIS-P to the child and parent, respectively, with all clinicians trained to diagnostic criterion. Moreover, the parent and child assessors independently assigned their CSRs and were blind to the randomized treatment conditions, both prior to and following treatment. Consensus of CSR and diagnosis were determined at weekly meetings with each of the respective project directors (both licensed clinical psychologists). Based on Cohen’s kappa in the original studies, agreements on diagnoses ranged from .93 - .95 and .86 - .88 for primary and
9
ACCEPTED MANUSCRIPT
secondary diagnoses. ADIS-IV interviews were administered at pre-treatment as well as at both assessments following treatment. Results for the present analyses focused on longitudinal changes in the consensus CSRs following treatment as well as how these changes in CSR were
RI PT
related to the child’s specific phobia beliefs and self-efficacy for dealing with the phobic stimulus should it occur.
Phobic Beliefs Scale (PBS). At the pre-treatment assessment, the child-assigned clinician
SC
interviewed the child to identify specific beliefs associated with the child’s specific phobia (e.g., “lightning will catch my house on fire, and I won't get out in time,” “The spider will crawl on my
M AN U
body and bite and hurt me,” “When in the dark by myself I will see something that will scare me.”). Up to three beliefs were initially solicited from the children at pre-treatment and those same beliefs were queried at post-treatment and six-month follow-up assessments (range in number of beliefs = 1 – 3). Given that children ranged in their number of beliefs identified, the
TE D
first, or primary, belief was the focus of the present analyses. Following the identification of the specific phobic beliefs, the child was asked to indicate how likely it was that the feared belief would occur (i.e., probability estimate), how “bad” it would be if the feared content of the belief
EP
actually occurred (i.e., perceived danger or threat), and the degree to which the child felt he or she would be able to cope with the event if the feared belief did in fact occur (i.e., self-efficacy).
AC C
Each facet of the belief was rated by the child on a 0 to 8 scale using an expectancy thermometer. In order to examine how the children’s subjective, and often catastrophic, beliefs
compared to more “objective” ratings based on probability estimates, we developed two rating scales which examined the “how likely” and “how bad” facets of the Phobic Beliefs Scale (Tables 2 and 3). To examine the how likely facet of the Phobic Beliefs Scale, each child’s first belief was coded on a 0 to 8 scale by objective raters. Two graduate students in a clinical
10
ACCEPTED MANUSCRIPT
psychology science program (third and fourth authors) and one licensed clinical psychologist (first author) independently categorized each of the scales’ anchors based on a sub-sample of the specific beliefs that the children provided. During the development of the scale, beliefs
RI PT
associated with the three specific phobia types (i.e., animal, environmental, and situational) were equally represented. Anchors for the scale were as follows: 0 - Not at all likely to occur; 2 - Most unlikely to occur; 4 - Minimally likely to occur; 6 - Moderately likely to occur; and 8 - Highly
SC
likely to occur. When discrepancies were present, the research team discussed the discrepancies on how to objectively code the beliefs to match the scales’ anchors. All beliefs for both samples
sample with good reliability (ICC = .83).
M AN U
were then objectively coded by the two doctoral students. The students dual-coded 25% of the
In order to assess the “how bad” facet of the Phobic Beliefs Scale, each child’s primary belief was coded on a 0 to 8 scale developed by us for this study. Anchors included: 0 – Not bad
TE D
at all; 2 – A little bit bad but not too bad; 4 – Minor harm to self or others; 6 – Clear physical harm to self or others but death not likely; 8 – Death highly likely. This coding scale was also developed to accommodate differences in specific phobia types (i.e., animal, environment, and
EP
situational). Similar to the development of the “how likely” facet of the Phobic Beliefs Scale, the same two doctoral students in clinical psychology and the same licensed clinical psychologist
AC C
independently coded each of the scales’ anchors based on a sub-sample of the specific beliefs identified by the phobic children. During the development of the scale, the three specific phobia types (i.e., animal, environmental, and situational) were equally represented. As above, the research team discussed any discrepancies on how to objectively code the beliefs to match the scales’ anchors. Subsequently, all beliefs for the samples were objectively coded by the two
11
ACCEPTED MANUSCRIPT
doctoral students. The students dual-coded 25% of the sample with excellent reliability (ICC = .96). Self-efficacy estimates, the child’s rating of his/her personal ability to cope if the phobic
RI PT
belief were to actually occur, were not objectively coded by us. While the first two facets of the PBS are based on the actual event that the child believes would occur, coping estimates are more subjective and less amenable to an objective rating. For example, being bitten by a dog is
SC
objectively as likely to occur and objectively to be as "bad" for all children; however, each
child's individual ability to cope with that event depends on many internal and environmental
M AN U
factors that are specific to the individual child. Data Analyses
Data were analyzed with IBM SPSS Statistics Version 23. Missing values at both the post-treatment and 6-month follow-up assessments were imputed using the multiple imputation
TE D
procedure in SPSS (IBM Corp.). In order to control for sociodemographic factors, we calculated bivariate correlations between child age and gender with PBS items and treated phobia CSRs (See Table 3). Child age was significantly positively correlated with the “how bad” item of the
EP
PBS and was subsequently included in all hierarchical regression analyses and as a covariate in the Analyses of Variance (ANOVAs). Child gender did not significantly correlate with the PBS
AC C
items or with CSR ratings and was thus not included in any subsequent analyses. In Study 1, ANOVAs examining the effect of phobia type (animal, environmental,
situational) on each of the three components of the Phobic Beliefs Scale were computed. Additionally, repeated measures ANOVAs were used to compare the child ratings on the PBS to the objective coder ratings determined by our independent team. In Study 2, repeated measures ANOVAs were used to examine change in each of the PBS items across the three assessment
12
ACCEPTED MANUSCRIPT
time-points: pre-treatment, post-treatment, and 6-month follow-up. The change in each of the PBS items from the pre-treatment to post-treatment and pre-treatment to follow-up assessments was calculated and included in hierarchical regression analyses to determine the relations of
Results Study 1
RI PT
these changes to post-treatment and six-month follow-up CSR scores.
SC
Using one-way ANOVAs to examine phobia type differences across the facets of the PBS, while controlling for child age, a main effect of phobia type was found for “how likely” the
M AN U
children rated their beliefs to occur, F (2, 247) = 9.57, p < .001, η2 = .07, but not “how bad” they rated their beliefs, F (2, 247) = 0.07, p = .932, η2 = .00, or their reported ability to cope, F (2, 247) = 1.04, p = .355, η2 = .00. Post-hoc tests indicated that the phobic beliefs of animal phobic children were rated as more likely to occur than the phobic beliefs of the environmental and
TE D
situational phobic children, which did not differ from one another. Additionally, there was a significant effect of the covariate, child age, for “how bad” the children rated their phobic beliefs, F (1, 247) = 21.66, p < .001, η2 = .08; as noted above, child age significantly and
EP
positively correlated with the “how bad” item of the PBS. The effect of the covariate was not significant for “how likely” the children believed their beliefs were to occur, F (1, 247) = .14, p
AC C
= .706, η2 = .00, or for their reported ability to cope with the event should it occur, F (1, 247) = 0.50, p = .479, η2 = .00. Means and standard deviations for these variables can be found in Table 4.
At pre-treatment, when controlling for child age, participant ratings for how likely the
feared event would be to occur were significant higher than our independently coded scores, F
13
ACCEPTED MANUSCRIPT
(1, 247) = 80.98, p < .001, η2 = .25. The rater (i.e., participant or objective coder) by specific phobia type interaction, F (2, 247) = 1.03, p = .359, η2 = .01 was not significant. At pre-treatment, participant ratings for how bad it would be if the feared event were to
RI PT
occur were also significantly higher than our coded scores, F (1, 247) = 233.97, p < .001, η2 = .49. The rater by treated phobia type interaction was not significant, F (1, 247) = 0.17, p = .843, η2 = .00. The rater by age interaction was significant, F (1, 247) = 29.42, p < .001, η2 = .11. To
SC
further explore this interaction, we used a median-split to categorize our sample into two groups by age (7-9 year-olds and 10-15 year-olds). When age was included as a between-subject factor
M AN U
in the ANOVA the rater (i.e., participant or objective coder) by age interaction was significant, F (8, 224) = 2.94, p = .004, η2 = .10. Specifically, there was a greater difference between raters for older children than younger children. Study 2
TE D
Using repeated measures ANOVAs, significant time effects were found for all three items of the PBS (how likely, F (2, 167) = 120.99, p < .001, η2 = .42, how bad, F (2, 167) = 69.51, p < .001, η2 = .29, and ability to cope, F (2, 167) = 54.85, p < .001, η2 = .25),
EP
demonstrating that each improved following treatment. There were no differences between treated phobia types in the degree of PBS item improvement and the time by treated phobia type
AC C
interactions were not significant for any of the three items (how likely, F (4, 336) = 1.45, p = .219, η2 = .02, how bad, F (4, 336) = 0.58, p = .680, η2 = .01, and ability to cope, F (4, 336) = 0.41, p = .803, η2 = .01). Additionally, child age did not significantly influence degree of PBS item improvement and the time by age interactions were not significant for any of the three beliefs (how likely, F (2, 167) = 0.57, p = .564, η2 = .00, how bad, F (2, 167) = 0.58, p = .176, η2 = .01, and ability to cope, F (2, 167) = 0.41, p = .069, η2 = .01).
14
ACCEPTED MANUSCRIPT
At both post-treatment and six-month follow-up assessments there was a significant difference between participant ratings and our objectively coded scores of how likely the identified phobic beliefs were to occur, F (1, 168) = 27.52, p < .001, η2 = .14 and F (1, 168) =
RI PT
63.86, p < .001, η2 = .28, respectively. Of critical importance, the participants rated their beliefs as less likely to occur than the objective coders rated the beliefs at these time points. At the posttreatment assessment, neither the rater by specific phobia type interaction, F (2, 168) = 0.34, p =
SC
.712, η2 = .01, nor rater by age interaction, F (1, 168) = 0.18, p = .676, η2 = .00, was significant. At the six-month follow-up assessment, the rater by specific phobia type interaction was
M AN U
significant, F (2, 168) = 3.68, p = .027, η2 = .04, however rater by age interaction was not, F (1, 168) = 0.04, p = .850, η2 = .00. Specifically, there was a greater difference between participant and coder ratings for the animal and environmental phobic children than the situational phobic children.
TE D
At both the post-treatment and six-month follow-up assessments there was not a significant difference between participant ratings and our objectively coded scores of how bad it would be if the phobic beliefs were to occur, F (1, 168) = 0.01, p = .944, η2 = .00 and F (1, 168)
EP
= 2.07, p = .152, η2 = .00, respectively. Additionally, at the post-treatment assessment, neither the rater by specific phobia type interaction, F (2, 168) = 0.75, p = .476, η2 = .01, nor rater by age
AC C
interactions, F (1, 168) = 1.41, p = .237, η2 = .01, were significant. Similar results were found for the rater by specific phobia type interaction, F (2, 168) = 1.92, p = .150, η2 = .02, and the rater by age interactions, F (1, 168) = 2.60, p = .11, η2 = .05, at the six-month follow-up assessment. CSR scores at the post-treatment and six-month follow-up assessments were regressed on
child age, treated phobia type dummy coded, and pre-treatment CSR scores (Step 1), and the change in each of the PBS items from the pre-treatment to post-treatment or follow-up
15
ACCEPTED MANUSCRIPT
assessments (Step 2). Each PBS item was included in a separate model. See Tables 5 and 6 for model outcomes. When the “how likely” item was examined in Step 2, the overall model was significant at
RI PT
post-treatment, (F(5, 166) = 4.89, p < .001), but not at six-month follow-up (F(5, 166) = 1.20, p = .310). However, change in “how likely” the children rated their beliefs did not predict CSR scores at either the post-treatment or 6-month follow-up assessments.
SC
When the “how bad” item was examined in Step 2, the overall model was again
significant at post-treatment, (F(5, 166) = 7.08, p < .001), but not at six-month follow-up (F(5,
M AN U
166) = 0.72, p = .606). At post-treatment, main effects showed that greater change in “how bad” the children rated their beliefs predicted lower CSR scores; however, at the 6-month follow-up assessment this effect was not significant.
When the children’s reported ability to cope was examined in Step 2, the overall model
TE D
was significant at post-treatment, (F(5, 166) = 5.52, p < .001), but not at six-month follow-up (F(5, 166) = 1.11, p = .359). Main effects showed that greater increases in ability to cope predicted lower CSR scores at the post-treatment assessment; however, at the 6-month follow-up
EP
assessment this effect was not significant.
Additionally, in the post-treatment analyses for all three PBS items the main effect of the
AC C
pre-treatment CSR of the treated phobia was significant, indicating that greater changes in CSRs were related to higher CSR scores at pre-treatment. This effect was not found in the six-month follow-up analyses, however. Discussion
Our findings are relatively straightforward in Study 1 and indicate important differences associated with phobia type as to “how likely” a phobic event would be to occur and important
16
ACCEPTED MANUSCRIPT
differences in age as to “how bad” it would be if the feared outcome were to occur. Youth with a specific phobia of animals endorsed that their feared belief would be more likely to occur than youth with environmental or situational phobias, independent of age. Older youth, however,
RI PT
reported a greater severity of “how bad” it would be if the feared event were to occur, not only for those with animal phobias but also those with environmental and situational phobias. This was true even though the belief ratings for all youth across the phobia types were generally in the
SC
elevated range (see Table 4), indicating that significant harm to the self or the child’s family would occur (e.g., the spider will bite me, and I will never be able to walk again; lightning will
M AN U
catch our house on fire and my mom will have to go to the hospital for five years; when I am in the dark my family will leave home and never come back). This effect may reflect that younger children are less likely to fully understand the extent of the feared outcome were it to occur whereas older children may have greater cognitive capacity to understand the magnitude of loss
TE D
should the feared outcome actually occur. Similar findings associated with cognitive capacity have been shown to be evident in studies examining meta-cognitions associated with other anxiety disorders in children differing in developmental levels (Cartwright-Hatton & Wells,
EP
1997; Wells & Cartwright-Hatton, 2004).
The dysfunctional quality and the specificity of these beliefs was readily evident when
AC C
we compared the beliefs of the youth to ratings by our coders who rated the beliefs on the basis of more objective probability estimates. In line with our hypotheses, our coders rated the beliefs as less likely to occur and less “bad” if they actually did occur than the youth themselves. This was true for all phobia types. It is also important to note that the beliefs of the children were clearly elevated and dysfunctional (see Table 4). These findings are comparable with past research which has suggested that youth with depression perceive themselves as qualitatively
17
ACCEPTED MANUSCRIPT
more negative when compared to their peers (Stark, Reynolds, & Kaslow, 1987). As such, these findings offer support for the dysfunctional nature and biased perceptions of youth with emotional difficulties. Of additional importance, youth with the three phobia types reported
RI PT
lower levels of self-efficacy associated with their ability to cope with the feared event should it occur. On average, they noted they were only a “little bit” confident that they would be able to cope with the phobic events at pre-treatment. Here, then, we show that an overestimation of the
SC
distorted beliefs and an underestimation of ability to cope with the danger of belief should it occur characterize specific phobias in youth and lend support to the information processing
M AN U
model of fear and anxiety as suggested early on by Beck and colleagues (1985) and more recently by Muris & Field (2008) when child fear and anxiety is specifically considered. In Study 2, we demonstrated that these distorted beliefs and coping estimates were changed following participation in OST. Examination of Table 4, for example, shows that beliefs
TE D
of how bad it would be if the feared event were to occur were more moderate and reflected a general belief of possible minor, non-catastrophic harm. This was observed for children of all ages and across phobia types. Notably, following treatment, the participants rated their beliefs as
EP
less likely to occur than the objective coders rated the beliefs. This indicates that the children held a less catastrophic view of their phobic beliefs following treatment.
AC C
Moreover, we showed that changes in beliefs about how bad it would be if the feared
event occurred and how the child could cope with the event if it did occur were both associated with changes in clinical severity ratings of the phobias at post-treatment, albeit not at 6 month follow-up. Although such changes might serve as mediators of change, we could not demonstrate such in this study since both the candidate mediators and the treatment outcomes were obtained at the same points in time, precluding the temporal precedence needed to establish mediation
18
ACCEPTED MANUSCRIPT
(Kraemer, Wilson, Fairburn, & Agras, 2002; Prins, Ollendick, Maric, & MacKinnon, 2015). Moreover, these changes were not associated with changes in clinical severity ratings at 6-month follow-up. Elsewhere we have commented on the many factors that might be associated with
RI PT
long term outcomes including intrapersonal and contextual factors associated with maintenance of treatment gains once they are demonstrated (Ollendick et al., 2015).
Our study is of course not without limitations. First, the youth in our clinical trials were
SC
referred to our specialty clinic and their phobias might be qualitatively different from those seen in either general clinics or in the community. Still, it is evident that our sample was one that was
M AN U
seriously impaired and that the phobias were long-standing. It should also be noted that youth with blood-injection-injury (BII) phobias were specifically excluded from these clinical trials. Recently, Oar, Farrell, Waters, and Ollendick (2016) have shown that youth with such phobias may be qualitatively different from youth with animal phobias. Youth with BII phobias were
TE D
shown to have more comorbid diagnoses, to report fears focused more on physical and somatic symptoms (e.g., faintness, nausea), and to report more exaggerated and dysfunctional danger expectancies than youth with animal phobias. However, whether youth with BII phobias differ
EP
from youth with situational or environmental phobias remains in terms of danger expectancies and coping expectancies remains to be determined. Second, our sample was largely Caucasian
AC C
and medium in family income. As a result, we are unsure whether the differences we observed here would be seen in children and adolescents from other ethnicities or socioeconomic strata. Third, the changes observed in distorted beliefs and coping expectancies were obviously examined only following the OST. Whether similar changes might be observed with other more standard psychosocial or pharmacological interventions remains to be seen (Walkup et al., 2008). Finally, trained graduate clinicians coded the child’s phobic cognition and served as our
19
ACCEPTED MANUSCRIPT
objective raters. Future research might examine whether differences exist in the ratings between youth with an SP and youth without a SP. These limitations notwithstanding, our finding are of import as they are the first to show
RI PT
the extent and type of dysfunctional beliefs associated with different types of specific phobias in youth and that these beliefs change with treatment and are associated with meaningful clinical outcomes. In sum, our findings offer additional evidence that a CBT-based treatment can
SC
effectively modify children’s beliefs and that these changes continue to be sustained post-
treatment. Given that changes in catastrophic beliefs are shown to be predictive of treatment
M AN U
success, these findings offer clinical utility by informing clinicians on the importance of identifying and targeting catastrophic beliefs during treatment. It will be important for future studies to examine whether such changes mediate treatment outcomes and what variables might
TE D
serve to moderate these relations.
AC C
EP
Acknowledgements: This work was supported by the National Institute of Mental Health, Grants R01 MH59308 and R01 MH074777.
20
ACCEPTED MANUSCRIPT
References Allen, K. B., Allen, B., Austin, K. E., Waldron, J. C., & Ollendick, T. H. (2015). Synchrony– desynchrony in the tripartite model of fear: Predicting treatment outcome in clinically
RI PT
phobic children. Behaviour Research and Therapy, 71, 54-64.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
SC
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.
M AN U
Anderson, S. E., Dallal, G. E., & Must, A. (2003). Relative weight and race influence average age at menarche: results from two nationally representative surveys of US girls studied 25 years apart. Pediatrics, 111(4), 844-850.
Anthony, M. M., Brown, T. A., & Barlow, D. H. (1997). Heterogeneity among specific phobia
TE D
types in DSM-IV. Behaviour Research and Therapy, 35, 1089-1100. Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders and phobias. A cognitive perspective. New York: Basic Book.
EP
Bögels, S. M., Snieder, N., & Kindt, M. (2003). Specificity of dysfunctional thinking in children with symptoms of social anxiety, separation anxiety and generalised anxiety. Behaviour
AC C
Change, 20(03), 160-169.
Cartwright-Hatton, S., & Wells, A. (1997). Beliefs about worry and intrusions: The Metacognitions Questionnaire and its correlates. Journal of Anxiety Disorders, 11, 279296.
Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N., & Baker, A.
21
ACCEPTED MANUSCRIPT
(2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46(1), 5-27. Essau, C. A., Conradt, J., & Petermann, F. (2000). Frequency, comorbidity, and psychosocial
RI PT
impairment of specific phobia in adolescents. Journal of Clinical Child Psychology, 29(2), 221-231.
Higa-McMillan, C. K., Francis, S. E., Rith-Najarian, L., & Chorpita, B. F. (2016).
SC
Evidence base update: 50 years of research on treatment for child and adolescent anxiety.
M AN U
Journal of Clinical Child & Adolescent Psychology, 45(2), 91-113.
IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.
In-Albon, T., & Schneider, S. (2006). Psychotherapy of childhood anxiety disorders: A metaanalysis. Psychotherapy and Psychosomatics, 76, 15–24.
TE D
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 followup. Journal of Consulting and Clinical Psychology, 72(2), 276.
EP
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005).
AC C
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593-602.
Kraemer, H. C., Wilson, G. T., Fairburn, C. G., & Agras, W. S. (2002). Mediators and moderators of treatment effects in randomized clinical trials. Archives of General
Psychiatry, 59, 877 – 883. Muris, P., & Field, A. P. (2008). Distorted cognition and pathological anxiety in children and adolescents. Cognition and Emotion, 22(3), 395-421. 22
ACCEPTED MANUSCRIPT
Oar, E. L., Farrell, L. J., Waters, A. M., & Ollendick, T. H. (2016). Blood-injection-injury phobia and dog phobia in youth: Psychological characteristics and associated features in a clinical sample. Behavior Therapy, 47, 312-324.
RI PT
Ollendick, T. H., & Davis III, T. E. (2013). One-session treatment for specific phobias: A review of Öst's single-session exposure with children and adolescents. Cognitive Behaviour Therapy, 42(4), 275-283.
SC
Ollendick, T. H., Halldorsdottir, T., Fraire, M. G., Austin, K. E., Noguchi, R. J., Lewis, K. M., ... & Whitmore, M. J. (2015). Specific phobias in youth: A randomized controlled trial
Therapy, 46(2), 141-155.
M AN U
comparing one-session treatment to a parent-augmented one-session treatment. Behavior
Ollendick, T. H., & King, N. J. (2012). Evidence-based treatments for children and adolescents: Issues and controversies. Child and Adolescent Therapy: Cognitive-Behavioral
TE D
Procedures, 499-519.
Ollendick, T. H., King, N. J., & Muris, P. (2004). Phobias in children and adolescents. In M. Maj, H. S. Akiskal, J. J. Lopez-Ibor, & A. Okkasha (Eds.), Phobias (pp. 245–279). New
EP
York: Wiley.
Ollendick, T. H., & Muris, P. (2015). The scientific legacy of Little Hans and Little Albert:
AC C
Future directions for research on specific phobias in youth. Journal of Clinical Child & Adolescent Psychology, 44(4), 689-706.
Ollendick, T. H., Öst, L. G., Reuterskiöld, L., Costa, N., Cederlund, R., Sirbu, C., ... & Jarrett, M. A. (2009). One-session treatment of specific phobias in youth: a randomized clinical trial in the United States and Sweden. Journal of consulting and clinical psychology, 77(3), 504.
23
ACCEPTED MANUSCRIPT
Ollendick, T. H., Raishevich, N., Davis, T. E. III., Sirbu, C., & Öst, L-G. (2010). Phenomenology and psychological characteristics of youth with specific phobias. Behavior Therapy, 41, 133-141.
RI PT
Öst, L.-G. (1989). One-session treatment for specific phobias. Behaviour Research and Therapy, 27, 1–7.
Öst, L.-G. (1997). Rapid treatment of specific phobias. In G. C. L. Davey (Ed.), Phobias: A
SC
handbook of theory, research and treatment (pp. 227–247). Oxford, England: Wiley. Öst, L. G. (2012). One-session treatment: Principles and procedures with adults. In Intensive
M AN U
one-session treatment of specific phobias (pp. 59-95). New York, NY: Springer. Öst, L.-G., Svensson, L., Hellström, K., & Lindwall, R. (2001). One session treatment of specific phobias in youths: A randomized clinical trial. Journal of Consulting and Clinical Psychology, 69, 814–824.
TE D
Prins, P.J.M., Ollendick, T.H., Maric, M., & MacKinnon, D.P. (2015). Moderators and mediators in treatment outcome studies of childhood disorders: The what, why and how. In M. Maric, P.J.M. Prins, and T.H. Ollendick, (Eds.), Moderators and mediators of youth
EP
treatment outcomes (pp 1-19). Oxford: Oxford University Press. Silverman, W. K., & Albano, A. M. (1996). Anxiety Disorders Interview Schedule for DSM-IV:
AC C
Child and Parent Versions. San Antonio, TX: Psychological Corporation.
Silverman, W. K., Kurtines, W. M., Ginsburg, G. S., Weems, C. F., Rabian, B., & Serafini, L. T. (1999). Contingency management, self-control, and education support in the treatment of childhood phobic disorders: A randomized clinical trial. Journal of Consulting and Clinical Psychology, 67, 675–687. Silverman, W. K., Pina, A. A., & Viswesvaran, C. (2008). Evidence-based psychosocial
24
ACCEPTED MANUSCRIPT
treatments for phobic and anxiety disorders in children and adolescents. Journal of Clinical Child & Adolescent Psychology, 37(1), 105-130. Stark, K. D., Reynolds, W. M., & Kaslow, N. J. (1987). A comparison of the
RI PT
relative efficacy of self-control therapy and a behavioral problem-solving therapy for depression in children. Journal of Abnormal Child Psychology, 15(1), 91-113.
Thorpe, S. J., & Salkovskis, P. M. (1995). Phobic beliefs: Do cognitive factors play a role in
SC
specific phobias? Behaviour Research and Therapy, 33, 805-816.
Tremblay, L., & Frigon, J-Y. (2005). Precocious puberty in adolescent girls: A biomarker of later
M AN U
psychosocial adjustment problems. Child Psychiatry and Human Development, 36(1), 71−92.
Walkup, J. T., Albano, A., Piacentini, J., Birmaher, B., Compton, S. N., Sherrill, J. T., … Kendall, P.C. (2008). Cognitive behavioral therapy, sertraline, or a combination in
TE D
childhood anxiety. The New England Journal of Medicine, 359, 2753-2766. Wells, A., & Cartwright-Hatton, S. (2004). A short form of the metacognitions questionnaire: Properties of the MCQ-30. Behaviour Research and Therapy, 42, 385-396.
EP
Wright, J. H., & Borden, J. (1991). Cognitive therapy of depression and anxiety. Psychiatric
AC C
Annals, 21, 424-428.
25
M AN U
80 (46.5) 37 (21.5) 55 (32.0)
AC C
EP
TE D
Specific Phobia Subtypes Animal 129 (51.4) Environmental 57 (22.7) Situational 65 (25.9)
SC
Table 1 Participant Descriptive Data Mean (SD) / n (% of sample) Study 1 (n = 251) Study 2 (n = 172) Age 9.87 (2.15) 9.69 (2.19) Male 144 (57.4) 85 (55.2)
RI PT
ACCEPTED MANUSCRIPT
26
ACCEPTED MANUSCRIPT
Table 2. Objective Coding Scales for Children’s Phobic Beliefs Objective Coding Scale of How “Bad” Facet of Children’s Phobic Cognitions
RI PT
0 – Not bad at all, no harm indicated
2 – Belief low-level harm or dangerousness (e.g., I will be scared if I see an animal; Something scary will appear in the dark; The noise from storm will startle me.)
SC
4 – Belief of minor physical harm to self and possible others (e.g., The dog will knock me down; If alone and can't get anyone on phone, I will have a panic attack; Lightning will hit me or others)
8 – Death to self and possible others
M AN U
6 – Explicit and specific belief of major harm to self (e.g., I will catch a disease after being bitten by snake and have to go to hospital; Lightning will catch house on fire and my legs will get burnt; I will see things (e.g., heads) that might come and get you and bite feet off)
Objective Coding Scale of How Likely Facet of Children’s Phobic Cognitions
TE D
0 – Not likely at all
2- Most unlikely to occur (e.g., Dog will bite my finger off; Lightning will strike me; I will see things (e.g., heads) that might come and get you and bite feet off)
EP
4- Minimally likely to occur (e.g., The dog will jump on me and lick my face; Thunder and lightning will break the satellite on the house and the lights might go out; If my head gets under water, the water will get in my ears and it will hurt )
AC C
6- Moderately likely to occur (e.g., I could trip on something in the dark; I would be fearful if a snake crawled towards me; Lightening will strike somewhere) 8- Highly likely to occur (e.g., Being stung by a bee or wasp would hurt; I will hear tree branches moving outside during a strong storm; If being chased by a monster, I will run very fast.)
27
SC
Table 3. Pearson Correlations Among Child Age and Gender, and PBS Ratings and Treated Phobia CSR Child Age Child Gender PBS – “How Likely” .04 -.04 PBS – “How Bad” .28** -.12 PBS – Ability to Cope .04 .06 Pre-Treatment Phobia CSR -.05 .08 Post-Treatment Phobia CSR -.03 .03 6-Month Follow-up Phobia CSR .03 -.06 Note. ** p < .01
RI PT
ACCEPTED MANUSCRIPT
2.71(2.12) 4.89(2.40) 3.83(2.41)
1.63(1.84) 4.88(2.71) 4.40(2.75)
1.58(1.62) 5.12(2.61) 4.25(2.61)
2.17(2.01) 4.95(2.52) 4.07(2.54)
1.42(1.43) 4.44(2.68) 5.02(2.35)
1.62(1.30) 4.65(2.15) 4.43(2.37)
1.86(1.51) 4.43(2.30) 4.45(2.29)
2.56(0.98) 4.39(2.14)
2.18(0.68) 4.34(2.48)
2.71(1.09) 4.38(1.83)
EP
Post-Treatment How Likely How Bad Ability to Cope
TE D
M AN U
Table 4 Means and Standard Deviations for PBS items across Treated Phobia Type and Assessment Time Point Mean(SD) Animal Environmental Situational Total (n = 129) (n = 57) (n = 65) (n = 251) Pre-Treatment How Likely 4.69(1.85) 3.81(2.55) 3.37(2.04) 4.15(2.15) How Bad 6.59(1.81) 6.49(2.04) 6.65(1.91) 6.58(1.88) Ability to 2.36(1.94) 2.81(2.57) 2.66(2.22) 2.54(2.17) Cope
AC C
6-Month Follow-up How Likely 2.18(1.57) How Bad 4.32(2.21) Ability to 4.20(2.21) Cope Objective Codes How Likely 3.05(1.17) How Bad 4.40(1.20)
28
ACCEPTED MANUSCRIPT
SC
RI PT
Table 5 Regression of Post-treatment CSRs on Change in Each PBS Item How Likely How “Bad” Ability to Cope 2 2 ∆R β ∆R β ∆R2 β .11*** .11*** .11*** Step 1 Age -.06 -.03 -.07 Treated Phobia Type -.11 -.12 .11 (Animal) Treated Phobia Type .01 -.00 .01 (Situational) Pre-treatment CSR .27*** .27*** .28*** .01 .06** .03* Step 2 Change in PBS Item -.12 -.25** -.17* F-value 4.89*** 7.08*** 5.52***
M AN U
Note. * p < .05, ** p < .01, *** p < .001.
AC C
Note. † p < .05
EP
TE D
Table 6 Regression of 6-Month Follow-up CSRs on Change in Each PBS Item How “Bad” Ability to Cope How Likely ∆R2 β ∆R2 β ∆R2 β -.00 .02 .02 Step 1 Age .03 .02 .03 Treated Phobia Type .09 .10 .09 (Animal) Treated Phobia Type .02 .05 .04 (Situational) Pre-treatment CSR .14† .15† .15† .01 .00 .01 Step 2 Change in PBS Item -.12 -.02 -.11 F-value 1.20 0.72 1.11
29
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Highlights • Youth with a specific phobia identified feared beliefs and coping expectancies • Differences exist in youth’s catastrophic beliefs across phobia types and age • Youth viewed their beliefs as more catastrophic and likely to occur than did coders • Regardless of phobia type or child age, the beliefs improved following OST • Changes in beliefs were related to changes in clinical severity following treatment
S0005-7967(17)30017-7
DOI:
10.1016/j.brat.2017.01.007
Reference:
BRT 3084
To appear in:
Behaviour Research and Therapy
Received Date: 23 August 2016 Revised Date:
10 January 2017
Accepted Date: 17 January 2017
Please cite this article as: Ollendick, T.H., Öst, L.-G., Ryan, S.M., Capriola, N.N., Reuterskiöld, L., Harm beliefs and coping expectancies in youth with specific phobias, Behaviour Research and Therapy (2017), doi: 10.1016/j.brat.2017.01.007. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Harm Beliefs and Coping Expectancies in Youth with Specific Phobias Thomas H. Ollendick1, Lars-Göran Öst2, Sarah M. Ryan1, Nicole N. Capriola1, and Lena
1
RI PT
Reuterskiöld2
Virginia Tech
SC
Child Study Center Department of Psychology
M AN U
Blacksburg, VA 24061
Phone: (540) 231-6451; Fax: (540) 231-3652
2
Stockholm University
TE D
Department of Psychology
EP
SE-106 91 Stockholm, Sweden
Correspondence concerning this article should be addressed to:
AC C
Sarah Ryan
e-mail: [email protected] Tel (540) 231-3514
Fax (540) 231-3652
ACCEPTED MANUSCRIPT
Abstract Catastrophic beliefs and lowered coping expectancies are often present in individuals with specific phobias (SPs). The current study examined these beliefs and expectancies in 251
RI PT
youth who received One Session Treatment for one of the three most common types of SP in youth (animals, natural environment, and situational). We compared the children’s subjective beliefs to objective ratings of the likelihood of occurrence and the dangerousness of the feared
SC
events. Results revealed pre-treatment differences in the youths’ beliefs across phobia types and age. Specifically, children with animal phobias rated their beliefs as more likely to occur than did
M AN U
children with environmental and situational phobias. In addition, older children rated their beliefs as more dangerous than younger children. However, regardless of phobia type or child age, the beliefs improved following treatment. Changes in catastrophic beliefs and coping expectancies were related to changes in clinical severity following treatment but not 6-months following
TE D
treatment. Moreover, at pre-treatment, children viewed their beliefs as significantly more catastrophic and likely to occur than did independent coders of these beliefs; however, these differences were no longer evident following treatment. Clinical implications are discussed,
AC C
outcomes.
EP
highlighting how changes in beliefs and expectancies might be associated with treatment
Keywords: Phobic Beliefs, Specific Phobias, Children and Adolescents, Randomized Controlled Trial
2
ACCEPTED MANUSCRIPT
Harm Beliefs and Coping Expectancies in Youth with Specific Phobias Specific phobias (SPs) are characterized by excessive fear or anxiety about a specific object or situation (APA, 2000, 2013). The phobic object or situation is avoided or endured with
RI PT
intense fear or anxiety and, typically, the fear or anxiety persists for 6 months or longer and in many cases for decades. Moreover, the perceived fear or anxiety is typically unrealistic and out of proportion to the actual or “real” danger posed by the phobic stimulus (e.g., “the dog will
SC
jump on me and bite me, and I will bleed to death,” “the lightning will strike our home and kill me, my mom and my dad”). According to DSM-5 (APA, 2013), SPs are categorized into five
M AN U
subtypes: animal type (e.g., dogs, spiders, insects), natural environment type (e.g., heights, storms, water), situational type (e.g., flying, elevators, enclosed places), blood-injection-injury type (BII; e.g., seeing blood, needles, invasive procedures), and other type (e.g., choking, loud sounds, costumed characters). Clinically significant SPs are present in approximately 5% of
TE D
children in community samples and about 15-20% of children in anxiety disorder clinics (APA, 2000, 2013; Kessler, Berglund, Demler, Jin, & Walters, 2005). For many children, SPs result in considerable academic, social, and personal distress, as well as interference in day-to-day
EP
functioning (Essau, Conradt, & Petermann, 2000; Kendall, Safford, Flannery-Schroeder, & Webb, 2004; Ollendick, King, & Muris, 2004).
AC C
In a host of systematic reviews over the past 15 - 20 years, cognitive-behavioral therapy
(CBT) has been shown to be efficacious in treating youth with phobic and anxiety disorders, with about 65 – 80% of youth clinically improved and approximately 50 – 65% free of their diagnosis following treatment (e.g., In-Albon & Schneider, 2006; Higa-McMillan, Francis, Rith-Najarian, & Chorpita, 2016; Ollendick & King, 2012; Silverman, Pina, & Viswesvaran, 2008). For SPs in youth, One Session Treatment (OST), a variant of CBT developed by Öst (1989, 1997), has been
3
ACCEPTED MANUSCRIPT
found to be a rapid and effective treatment (see Ollendick & Davis, 2013, and Öst, 2012, for recent reviews). This 3-hour, one session treatment involves psychoeducation, in vivo exposure, social reinforcement, participant modeling, and behavioral experiments to challenge distorted
RI PT
and oftentimes catastrophic beliefs that characterize SPs. Although studies documenting the efficacy of OST have shown significant improvements in outcomes including fear levels,
avoidant behaviors, and, in some cases, reduced physiological arousal (e.g., Allen, Allen, Austin,
SC
Waldron, & Ollendick, 2015; Öst, Svensson, Hellström, & Lindwall, 2001), few studies have examined the distorted and catastrophic beliefs that characterize this disorder and are the central
M AN U
focus of long-standing information processing models of fear and anxiety (Beck, Emery, & Greenberg, 1985; Wright & Borden, 1991). In recent years, Öst (2012) and Ollendick and Muris (2015) have reaffirmed the role of these beliefs in the maintenance of phobias, suggesting that these catastrophic beliefs regarding the feared stimulus serve to maintain the phobic anxiety and
TE D
that the avoidance observed in SPs prevents disconfirmation of these beliefs. Although it is oftentimes assumed that these distorted beliefs change as a result of OST, this has not been investigated heretofore in clinical samples of youth with SPs.
EP
Based on the seminal work of Thorpe and Salkovskis (1995) and Anthony, Brown, and Barlow (1997) on phobic beliefs in the maintenance of SPs in adults and the pioneering work of
AC C
Bögels, Snieder, and Kindt (2003) on the specificity of dysfunctional thinking in youth with different anxiety disorders (e.g., children with separation anxiety disorder overestimate the probability of being abandoned and children with social anxiety disorder overestimate the danger of being rejected and negatively evaluated by others), we set about to examine the specificity of beliefs of youth with a SP by exploring the specific beliefs of children and adolescents with the three most prevalent subtypes of SPs (animal, natural environment, and situational subtypes; see
4
ACCEPTED MANUSCRIPT
Ollendick & Muris, 2015). In as much as previous work has shown that youth with situational/environmental phobias evince more impairment than youth with animal phobias (Ollendick, Raishevich, Davis, Sirby, & Öst, 2010), we specifically examined whether type of
RI PT
phobia was related to these beliefs. In Study 1, we examined “how likely” it would be that the anticipated belief would occur, “how bad” it would be if the content of the belief were to actually occur, and how well the child could cope with the anticipated outcome should it occur. These
SC
beliefs were obtained from youth who participated in two randomized clinical control trials examining the efficacy of OST with clinic-referred youth between 6 and 16 years of age
M AN U
(Ollendick et al., 2009, 2015). In addition, we rated the likelihood and probable outcomes of these beliefs based on independent, objective ratings provided by our research team. As a result, we were able to compare the subjective beliefs of the youth to more objective probabilities associated with these beliefs. In line with information processing models of fear and anxiety
TE D
(e.g., Beck et al., 1985), we anticipated that the subjective beliefs of the youth would be overestimated compared to the more objective evaluations determined by independent raters. Given the exploratory nature of Study 1, however, no directional hypotheses were offered.
EP
In Study 2, we sought to examine how changes in these beliefs were related to clinical outcomes, specifically changes in clinician-assigned severity ratings (CSR) for the child’s
AC C
specific phobia. Data for Study 2 were derived from the same two randomized clinical control trials used in Study 1. The beliefs were obtained from the youth themselves not only at pretreatment but also at post-treatment and 6-months following treatment in these studies. In as much as extinction learning is the theoretical basis of exposure therapy wherein repeated exposures to the phobic stimulus provide corrective information that challenges beliefs and danger expectancies (cf, Craske et al., 2008; Öst, 2012), we anticipated that these beliefs would
5
ACCEPTED MANUSCRIPT
change across time and be associated with changes in the clinical severity of the phobias following treatment. In addition to these dysfunctional beliefs, we also obtained ratings on coping
RI PT
expectancies of the youth. To assess this dimension, we asked the youth to rate how sure they would be that they could “handle or deal with” the feared event or situation should it actually occur. Early on, Beck and colleagues (1985) suggested that an underestimation of one’s abilities
SC
to cope with the danger or threat present in the phobic stimulus might result from both the
overestimation of the distorted belief and the continued avoidance of the feared stimulus. Here,
M AN U
we expected the coping statements of youth to be positively affected by our OST intervention and to show parallel changes to those observed with the dysfunctional beliefs. That is, the dysfunctional beliefs would be reduced and the youth’s coping estimates would be enhanced as a result of treatment. In as much as the dysfunctional beliefs and coping statements were obtained
TE D
at the same points in time, we did not make predictions about the order or sequencing of these two facets of the phobic response, only that they would both change in therapeutic directions. Method
EP
Participants
As noted, participants were drawn from two randomized clinical control trials (RCTs)
AC C
examining the effectiveness of OST for youth with a specific phobia (Ollendick et al., 2009, 2015). All youth were recruited through referrals from child psychiatric services, school health services, pediatricians, and print advertisements. For both RCTs, the following inclusion criteria were enlisted: (a) the participants had to be between 6 and 16 years of age, (b) the participants had to have a specific phobia diagnosis, according to DSM-IV (APA, 2004) criteria, (c) the phobia had to result in significant impairment/distress and be clinical in nature as established
6
ACCEPTED MANUSCRIPT
through a semi-structured diagnostic interview, (d) the duration of impairment needed to be at least 6 months, (e) the participants did not meet criteria for any specific disorders meeting exclusion criteria (i.e., primary major depressive disorder, pervasive developmental disorder,
RI PT
drug or alcohol abuse, and/or psychosis), and (f) the participants had to agree to discontinue other forms of psychotherapy and to be stable on any medications for the duration of the study. One hundred and sixty-five youth from Sweden and the United States of America
SC
between the ages of 7 and 15 with animal, situational and environment phobias were recruited between 2001 and 2006 for participation in the first RCT (Ollendick et al., 2009) comparing the
M AN U
effects of OST (n = 86) and Education Support Treatment (EST) (n = 79; Table 1; please note that youth 6 and 16 years of age were excluded for the current study due to the small number of youth at each of these ages). In addition, 86 youth between the ages of 7 and 15 from the United States of America with these three types of specific phobia were recruited between 2007 and
TE D
2013 for participation in a second RCT (Ollendick, et al., 2015) comparing the effects of standard, individual OST (n = 41) to a parent-augmented OST (i.e., A-OST; n = 45; please see below; Table 1). Study 1 combined data from the children who participated in both RCTs (n =
EP
165 + 86 = 251, mean age = 9.87 years, 89% Caucasian, mean family income = $75,165). In Study 2, only children and adolescents randomized to OST and A-OST (n = 172, mean age =
AC C
9.69 years, 89% Caucasian, mean family income = $73, 498) were included, excluding the 79 youth in the EST condition from Study 1. Since no differences were observed between the standard and augmented conditions in the second RCT, participants in the two treatment groups were combined for purposes in Study 2. Table 1 also provides both samples’ distributions across the specific phobia subtypes. Procedure
7
ACCEPTED MANUSCRIPT
Upon initial contact, the participants’ parents completed a brief telephone screen to determine eligibility (i.e., their child was free of any disorders which met exclusion criteria) and to confirm the presence of a probable specific phobia diagnosis. Following the phone screen,
RI PT
eligible families were invited to participate in a pre-treatment assessment session. All children and their parents provided informed written assent and consent, respectively, at the start of the study. During the assessment session, both children and parents participated in a clinical intake
SC
wherein a diagnostic interview (see below) was administered to the parent and the child
separately. Parents and children also completed a battery of questionnaires which are not
M AN U
analyzed for purposes of the present study. The entire assessment session lasted approximately 3 hours. Children provided their specific phobic beliefs during the interview with the assessment clinician.
Subsequent assessment sessions, each lasting about two hours, were conducted at one-
TE D
week post-treatment and six months following treatment. At each follow-up assessment session, the diagnostic modules endorsed at pre-treatment were re-administered, the phobic beliefs and self-efficacy expectancies were rated by the child on a 0 to 8 scale using an expectancy
EP
thermometer, and a battery of questionnaires were administered. Further details on the procedures of the original studies can be found in their respective publications. The children’s
AC C
phobic beliefs (how likely, how bad) were also independently coded by our research team (see below as well as Table 2). Treatments
The child focused OST treatment condition was based on principles developed by Öst
(1989, 1997) for adults and subsequently manualized for the treatment of SPs in youth (Ollendick et al., 2009; Öst et al., 2001). OST consisted of a three hour treatment session
8
ACCEPTED MANUSCRIPT
including gradual in-vivo exposure and challenging of the child’s distorted and catastrophic cognitions. Parents were not involved in the treatment session. The A-OST treatment condition (Ollendick et al., 2015) was developed to incorporate the
RI PT
primary caregiver into the treatment session. Similar to the OST condition, the clinician assigned to work with the child engaged the child in gradual in-vivo exposures and challenged the child’s catastrophic cognitions. The parent observed the treatment session with a second clinician and
SC
the clinician coached the parent on how to conduct exposures, how to reinforce approach
behaviors, and how to reduce reinforcement of avoidance behaviors. As noted above, differences
M AN U
between the standard OST and the augmented OST were not observed; hence, data were combined across the two treatment conditions. Measures
The Anxiety Disorders Interview Schedule for DSM-IV-Child and Parent Versions
TE D
(ADIS-IV-C/P; Silverman & Albano, 1996). The ADIS-IV-C (child version) and ADIS-IV-P (parent version) are semi-structured clinical interviews designed to facilitate diagnosis of anxiety and mood disorders as well as other psychiatric disorders in children and adolescents. Clinicians
EP
assign a severity rating (CSR) on a 9-point scale (0-8, with any rating ≥ 4 being indicative of probable diagnosis and a clinical level of interference). For the present study, separate clinicians
AC C
administered the ADIS-C and ADIS-P to the child and parent, respectively, with all clinicians trained to diagnostic criterion. Moreover, the parent and child assessors independently assigned their CSRs and were blind to the randomized treatment conditions, both prior to and following treatment. Consensus of CSR and diagnosis were determined at weekly meetings with each of the respective project directors (both licensed clinical psychologists). Based on Cohen’s kappa in the original studies, agreements on diagnoses ranged from .93 - .95 and .86 - .88 for primary and
9
ACCEPTED MANUSCRIPT
secondary diagnoses. ADIS-IV interviews were administered at pre-treatment as well as at both assessments following treatment. Results for the present analyses focused on longitudinal changes in the consensus CSRs following treatment as well as how these changes in CSR were
RI PT
related to the child’s specific phobia beliefs and self-efficacy for dealing with the phobic stimulus should it occur.
Phobic Beliefs Scale (PBS). At the pre-treatment assessment, the child-assigned clinician
SC
interviewed the child to identify specific beliefs associated with the child’s specific phobia (e.g., “lightning will catch my house on fire, and I won't get out in time,” “The spider will crawl on my
M AN U
body and bite and hurt me,” “When in the dark by myself I will see something that will scare me.”). Up to three beliefs were initially solicited from the children at pre-treatment and those same beliefs were queried at post-treatment and six-month follow-up assessments (range in number of beliefs = 1 – 3). Given that children ranged in their number of beliefs identified, the
TE D
first, or primary, belief was the focus of the present analyses. Following the identification of the specific phobic beliefs, the child was asked to indicate how likely it was that the feared belief would occur (i.e., probability estimate), how “bad” it would be if the feared content of the belief
EP
actually occurred (i.e., perceived danger or threat), and the degree to which the child felt he or she would be able to cope with the event if the feared belief did in fact occur (i.e., self-efficacy).
AC C
Each facet of the belief was rated by the child on a 0 to 8 scale using an expectancy thermometer. In order to examine how the children’s subjective, and often catastrophic, beliefs
compared to more “objective” ratings based on probability estimates, we developed two rating scales which examined the “how likely” and “how bad” facets of the Phobic Beliefs Scale (Tables 2 and 3). To examine the how likely facet of the Phobic Beliefs Scale, each child’s first belief was coded on a 0 to 8 scale by objective raters. Two graduate students in a clinical
10
ACCEPTED MANUSCRIPT
psychology science program (third and fourth authors) and one licensed clinical psychologist (first author) independently categorized each of the scales’ anchors based on a sub-sample of the specific beliefs that the children provided. During the development of the scale, beliefs
RI PT
associated with the three specific phobia types (i.e., animal, environmental, and situational) were equally represented. Anchors for the scale were as follows: 0 - Not at all likely to occur; 2 - Most unlikely to occur; 4 - Minimally likely to occur; 6 - Moderately likely to occur; and 8 - Highly
SC
likely to occur. When discrepancies were present, the research team discussed the discrepancies on how to objectively code the beliefs to match the scales’ anchors. All beliefs for both samples
sample with good reliability (ICC = .83).
M AN U
were then objectively coded by the two doctoral students. The students dual-coded 25% of the
In order to assess the “how bad” facet of the Phobic Beliefs Scale, each child’s primary belief was coded on a 0 to 8 scale developed by us for this study. Anchors included: 0 – Not bad
TE D
at all; 2 – A little bit bad but not too bad; 4 – Minor harm to self or others; 6 – Clear physical harm to self or others but death not likely; 8 – Death highly likely. This coding scale was also developed to accommodate differences in specific phobia types (i.e., animal, environment, and
EP
situational). Similar to the development of the “how likely” facet of the Phobic Beliefs Scale, the same two doctoral students in clinical psychology and the same licensed clinical psychologist
AC C
independently coded each of the scales’ anchors based on a sub-sample of the specific beliefs identified by the phobic children. During the development of the scale, the three specific phobia types (i.e., animal, environmental, and situational) were equally represented. As above, the research team discussed any discrepancies on how to objectively code the beliefs to match the scales’ anchors. Subsequently, all beliefs for the samples were objectively coded by the two
11
ACCEPTED MANUSCRIPT
doctoral students. The students dual-coded 25% of the sample with excellent reliability (ICC = .96). Self-efficacy estimates, the child’s rating of his/her personal ability to cope if the phobic
RI PT
belief were to actually occur, were not objectively coded by us. While the first two facets of the PBS are based on the actual event that the child believes would occur, coping estimates are more subjective and less amenable to an objective rating. For example, being bitten by a dog is
SC
objectively as likely to occur and objectively to be as "bad" for all children; however, each
child's individual ability to cope with that event depends on many internal and environmental
M AN U
factors that are specific to the individual child. Data Analyses
Data were analyzed with IBM SPSS Statistics Version 23. Missing values at both the post-treatment and 6-month follow-up assessments were imputed using the multiple imputation
TE D
procedure in SPSS (IBM Corp.). In order to control for sociodemographic factors, we calculated bivariate correlations between child age and gender with PBS items and treated phobia CSRs (See Table 3). Child age was significantly positively correlated with the “how bad” item of the
EP
PBS and was subsequently included in all hierarchical regression analyses and as a covariate in the Analyses of Variance (ANOVAs). Child gender did not significantly correlate with the PBS
AC C
items or with CSR ratings and was thus not included in any subsequent analyses. In Study 1, ANOVAs examining the effect of phobia type (animal, environmental,
situational) on each of the three components of the Phobic Beliefs Scale were computed. Additionally, repeated measures ANOVAs were used to compare the child ratings on the PBS to the objective coder ratings determined by our independent team. In Study 2, repeated measures ANOVAs were used to examine change in each of the PBS items across the three assessment
12
ACCEPTED MANUSCRIPT
time-points: pre-treatment, post-treatment, and 6-month follow-up. The change in each of the PBS items from the pre-treatment to post-treatment and pre-treatment to follow-up assessments was calculated and included in hierarchical regression analyses to determine the relations of
Results Study 1
RI PT
these changes to post-treatment and six-month follow-up CSR scores.
SC
Using one-way ANOVAs to examine phobia type differences across the facets of the PBS, while controlling for child age, a main effect of phobia type was found for “how likely” the
M AN U
children rated their beliefs to occur, F (2, 247) = 9.57, p < .001, η2 = .07, but not “how bad” they rated their beliefs, F (2, 247) = 0.07, p = .932, η2 = .00, or their reported ability to cope, F (2, 247) = 1.04, p = .355, η2 = .00. Post-hoc tests indicated that the phobic beliefs of animal phobic children were rated as more likely to occur than the phobic beliefs of the environmental and
TE D
situational phobic children, which did not differ from one another. Additionally, there was a significant effect of the covariate, child age, for “how bad” the children rated their phobic beliefs, F (1, 247) = 21.66, p < .001, η2 = .08; as noted above, child age significantly and
EP
positively correlated with the “how bad” item of the PBS. The effect of the covariate was not significant for “how likely” the children believed their beliefs were to occur, F (1, 247) = .14, p
AC C
= .706, η2 = .00, or for their reported ability to cope with the event should it occur, F (1, 247) = 0.50, p = .479, η2 = .00. Means and standard deviations for these variables can be found in Table 4.
At pre-treatment, when controlling for child age, participant ratings for how likely the
feared event would be to occur were significant higher than our independently coded scores, F
13
ACCEPTED MANUSCRIPT
(1, 247) = 80.98, p < .001, η2 = .25. The rater (i.e., participant or objective coder) by specific phobia type interaction, F (2, 247) = 1.03, p = .359, η2 = .01 was not significant. At pre-treatment, participant ratings for how bad it would be if the feared event were to
RI PT
occur were also significantly higher than our coded scores, F (1, 247) = 233.97, p < .001, η2 = .49. The rater by treated phobia type interaction was not significant, F (1, 247) = 0.17, p = .843, η2 = .00. The rater by age interaction was significant, F (1, 247) = 29.42, p < .001, η2 = .11. To
SC
further explore this interaction, we used a median-split to categorize our sample into two groups by age (7-9 year-olds and 10-15 year-olds). When age was included as a between-subject factor
M AN U
in the ANOVA the rater (i.e., participant or objective coder) by age interaction was significant, F (8, 224) = 2.94, p = .004, η2 = .10. Specifically, there was a greater difference between raters for older children than younger children. Study 2
TE D
Using repeated measures ANOVAs, significant time effects were found for all three items of the PBS (how likely, F (2, 167) = 120.99, p < .001, η2 = .42, how bad, F (2, 167) = 69.51, p < .001, η2 = .29, and ability to cope, F (2, 167) = 54.85, p < .001, η2 = .25),
EP
demonstrating that each improved following treatment. There were no differences between treated phobia types in the degree of PBS item improvement and the time by treated phobia type
AC C
interactions were not significant for any of the three items (how likely, F (4, 336) = 1.45, p = .219, η2 = .02, how bad, F (4, 336) = 0.58, p = .680, η2 = .01, and ability to cope, F (4, 336) = 0.41, p = .803, η2 = .01). Additionally, child age did not significantly influence degree of PBS item improvement and the time by age interactions were not significant for any of the three beliefs (how likely, F (2, 167) = 0.57, p = .564, η2 = .00, how bad, F (2, 167) = 0.58, p = .176, η2 = .01, and ability to cope, F (2, 167) = 0.41, p = .069, η2 = .01).
14
ACCEPTED MANUSCRIPT
At both post-treatment and six-month follow-up assessments there was a significant difference between participant ratings and our objectively coded scores of how likely the identified phobic beliefs were to occur, F (1, 168) = 27.52, p < .001, η2 = .14 and F (1, 168) =
RI PT
63.86, p < .001, η2 = .28, respectively. Of critical importance, the participants rated their beliefs as less likely to occur than the objective coders rated the beliefs at these time points. At the posttreatment assessment, neither the rater by specific phobia type interaction, F (2, 168) = 0.34, p =
SC
.712, η2 = .01, nor rater by age interaction, F (1, 168) = 0.18, p = .676, η2 = .00, was significant. At the six-month follow-up assessment, the rater by specific phobia type interaction was
M AN U
significant, F (2, 168) = 3.68, p = .027, η2 = .04, however rater by age interaction was not, F (1, 168) = 0.04, p = .850, η2 = .00. Specifically, there was a greater difference between participant and coder ratings for the animal and environmental phobic children than the situational phobic children.
TE D
At both the post-treatment and six-month follow-up assessments there was not a significant difference between participant ratings and our objectively coded scores of how bad it would be if the phobic beliefs were to occur, F (1, 168) = 0.01, p = .944, η2 = .00 and F (1, 168)
EP
= 2.07, p = .152, η2 = .00, respectively. Additionally, at the post-treatment assessment, neither the rater by specific phobia type interaction, F (2, 168) = 0.75, p = .476, η2 = .01, nor rater by age
AC C
interactions, F (1, 168) = 1.41, p = .237, η2 = .01, were significant. Similar results were found for the rater by specific phobia type interaction, F (2, 168) = 1.92, p = .150, η2 = .02, and the rater by age interactions, F (1, 168) = 2.60, p = .11, η2 = .05, at the six-month follow-up assessment. CSR scores at the post-treatment and six-month follow-up assessments were regressed on
child age, treated phobia type dummy coded, and pre-treatment CSR scores (Step 1), and the change in each of the PBS items from the pre-treatment to post-treatment or follow-up
15
ACCEPTED MANUSCRIPT
assessments (Step 2). Each PBS item was included in a separate model. See Tables 5 and 6 for model outcomes. When the “how likely” item was examined in Step 2, the overall model was significant at
RI PT
post-treatment, (F(5, 166) = 4.89, p < .001), but not at six-month follow-up (F(5, 166) = 1.20, p = .310). However, change in “how likely” the children rated their beliefs did not predict CSR scores at either the post-treatment or 6-month follow-up assessments.
SC
When the “how bad” item was examined in Step 2, the overall model was again
significant at post-treatment, (F(5, 166) = 7.08, p < .001), but not at six-month follow-up (F(5,
M AN U
166) = 0.72, p = .606). At post-treatment, main effects showed that greater change in “how bad” the children rated their beliefs predicted lower CSR scores; however, at the 6-month follow-up assessment this effect was not significant.
When the children’s reported ability to cope was examined in Step 2, the overall model
TE D
was significant at post-treatment, (F(5, 166) = 5.52, p < .001), but not at six-month follow-up (F(5, 166) = 1.11, p = .359). Main effects showed that greater increases in ability to cope predicted lower CSR scores at the post-treatment assessment; however, at the 6-month follow-up
EP
assessment this effect was not significant.
Additionally, in the post-treatment analyses for all three PBS items the main effect of the
AC C
pre-treatment CSR of the treated phobia was significant, indicating that greater changes in CSRs were related to higher CSR scores at pre-treatment. This effect was not found in the six-month follow-up analyses, however. Discussion
Our findings are relatively straightforward in Study 1 and indicate important differences associated with phobia type as to “how likely” a phobic event would be to occur and important
16
ACCEPTED MANUSCRIPT
differences in age as to “how bad” it would be if the feared outcome were to occur. Youth with a specific phobia of animals endorsed that their feared belief would be more likely to occur than youth with environmental or situational phobias, independent of age. Older youth, however,
RI PT
reported a greater severity of “how bad” it would be if the feared event were to occur, not only for those with animal phobias but also those with environmental and situational phobias. This was true even though the belief ratings for all youth across the phobia types were generally in the
SC
elevated range (see Table 4), indicating that significant harm to the self or the child’s family would occur (e.g., the spider will bite me, and I will never be able to walk again; lightning will
M AN U
catch our house on fire and my mom will have to go to the hospital for five years; when I am in the dark my family will leave home and never come back). This effect may reflect that younger children are less likely to fully understand the extent of the feared outcome were it to occur whereas older children may have greater cognitive capacity to understand the magnitude of loss
TE D
should the feared outcome actually occur. Similar findings associated with cognitive capacity have been shown to be evident in studies examining meta-cognitions associated with other anxiety disorders in children differing in developmental levels (Cartwright-Hatton & Wells,
EP
1997; Wells & Cartwright-Hatton, 2004).
The dysfunctional quality and the specificity of these beliefs was readily evident when
AC C
we compared the beliefs of the youth to ratings by our coders who rated the beliefs on the basis of more objective probability estimates. In line with our hypotheses, our coders rated the beliefs as less likely to occur and less “bad” if they actually did occur than the youth themselves. This was true for all phobia types. It is also important to note that the beliefs of the children were clearly elevated and dysfunctional (see Table 4). These findings are comparable with past research which has suggested that youth with depression perceive themselves as qualitatively
17
ACCEPTED MANUSCRIPT
more negative when compared to their peers (Stark, Reynolds, & Kaslow, 1987). As such, these findings offer support for the dysfunctional nature and biased perceptions of youth with emotional difficulties. Of additional importance, youth with the three phobia types reported
RI PT
lower levels of self-efficacy associated with their ability to cope with the feared event should it occur. On average, they noted they were only a “little bit” confident that they would be able to cope with the phobic events at pre-treatment. Here, then, we show that an overestimation of the
SC
distorted beliefs and an underestimation of ability to cope with the danger of belief should it occur characterize specific phobias in youth and lend support to the information processing
M AN U
model of fear and anxiety as suggested early on by Beck and colleagues (1985) and more recently by Muris & Field (2008) when child fear and anxiety is specifically considered. In Study 2, we demonstrated that these distorted beliefs and coping estimates were changed following participation in OST. Examination of Table 4, for example, shows that beliefs
TE D
of how bad it would be if the feared event were to occur were more moderate and reflected a general belief of possible minor, non-catastrophic harm. This was observed for children of all ages and across phobia types. Notably, following treatment, the participants rated their beliefs as
EP
less likely to occur than the objective coders rated the beliefs. This indicates that the children held a less catastrophic view of their phobic beliefs following treatment.
AC C
Moreover, we showed that changes in beliefs about how bad it would be if the feared
event occurred and how the child could cope with the event if it did occur were both associated with changes in clinical severity ratings of the phobias at post-treatment, albeit not at 6 month follow-up. Although such changes might serve as mediators of change, we could not demonstrate such in this study since both the candidate mediators and the treatment outcomes were obtained at the same points in time, precluding the temporal precedence needed to establish mediation
18
ACCEPTED MANUSCRIPT
(Kraemer, Wilson, Fairburn, & Agras, 2002; Prins, Ollendick, Maric, & MacKinnon, 2015). Moreover, these changes were not associated with changes in clinical severity ratings at 6-month follow-up. Elsewhere we have commented on the many factors that might be associated with
RI PT
long term outcomes including intrapersonal and contextual factors associated with maintenance of treatment gains once they are demonstrated (Ollendick et al., 2015).
Our study is of course not without limitations. First, the youth in our clinical trials were
SC
referred to our specialty clinic and their phobias might be qualitatively different from those seen in either general clinics or in the community. Still, it is evident that our sample was one that was
M AN U
seriously impaired and that the phobias were long-standing. It should also be noted that youth with blood-injection-injury (BII) phobias were specifically excluded from these clinical trials. Recently, Oar, Farrell, Waters, and Ollendick (2016) have shown that youth with such phobias may be qualitatively different from youth with animal phobias. Youth with BII phobias were
TE D
shown to have more comorbid diagnoses, to report fears focused more on physical and somatic symptoms (e.g., faintness, nausea), and to report more exaggerated and dysfunctional danger expectancies than youth with animal phobias. However, whether youth with BII phobias differ
EP
from youth with situational or environmental phobias remains in terms of danger expectancies and coping expectancies remains to be determined. Second, our sample was largely Caucasian
AC C
and medium in family income. As a result, we are unsure whether the differences we observed here would be seen in children and adolescents from other ethnicities or socioeconomic strata. Third, the changes observed in distorted beliefs and coping expectancies were obviously examined only following the OST. Whether similar changes might be observed with other more standard psychosocial or pharmacological interventions remains to be seen (Walkup et al., 2008). Finally, trained graduate clinicians coded the child’s phobic cognition and served as our
19
ACCEPTED MANUSCRIPT
objective raters. Future research might examine whether differences exist in the ratings between youth with an SP and youth without a SP. These limitations notwithstanding, our finding are of import as they are the first to show
RI PT
the extent and type of dysfunctional beliefs associated with different types of specific phobias in youth and that these beliefs change with treatment and are associated with meaningful clinical outcomes. In sum, our findings offer additional evidence that a CBT-based treatment can
SC
effectively modify children’s beliefs and that these changes continue to be sustained post-
treatment. Given that changes in catastrophic beliefs are shown to be predictive of treatment
M AN U
success, these findings offer clinical utility by informing clinicians on the importance of identifying and targeting catastrophic beliefs during treatment. It will be important for future studies to examine whether such changes mediate treatment outcomes and what variables might
TE D
serve to moderate these relations.
AC C
EP
Acknowledgements: This work was supported by the National Institute of Mental Health, Grants R01 MH59308 and R01 MH074777.
20
ACCEPTED MANUSCRIPT
References Allen, K. B., Allen, B., Austin, K. E., Waldron, J. C., & Ollendick, T. H. (2015). Synchrony– desynchrony in the tripartite model of fear: Predicting treatment outcome in clinically
RI PT
phobic children. Behaviour Research and Therapy, 71, 54-64.
American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.
SC
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.
M AN U
Anderson, S. E., Dallal, G. E., & Must, A. (2003). Relative weight and race influence average age at menarche: results from two nationally representative surveys of US girls studied 25 years apart. Pediatrics, 111(4), 844-850.
Anthony, M. M., Brown, T. A., & Barlow, D. H. (1997). Heterogeneity among specific phobia
TE D
types in DSM-IV. Behaviour Research and Therapy, 35, 1089-1100. Beck, A. T., Emery, G., & Greenberg, R. L. (1985). Anxiety disorders and phobias. A cognitive perspective. New York: Basic Book.
EP
Bögels, S. M., Snieder, N., & Kindt, M. (2003). Specificity of dysfunctional thinking in children with symptoms of social anxiety, separation anxiety and generalised anxiety. Behaviour
AC C
Change, 20(03), 160-169.
Cartwright-Hatton, S., & Wells, A. (1997). Beliefs about worry and intrusions: The Metacognitions Questionnaire and its correlates. Journal of Anxiety Disorders, 11, 279296.
Craske, M. G., Kircanski, K., Zelikowsky, M., Mystkowski, J., Chowdhury, N., & Baker, A.
21
ACCEPTED MANUSCRIPT
(2008). Optimizing inhibitory learning during exposure therapy. Behaviour Research and Therapy, 46(1), 5-27. Essau, C. A., Conradt, J., & Petermann, F. (2000). Frequency, comorbidity, and psychosocial
RI PT
impairment of specific phobia in adolescents. Journal of Clinical Child Psychology, 29(2), 221-231.
Higa-McMillan, C. K., Francis, S. E., Rith-Najarian, L., & Chorpita, B. F. (2016).
SC
Evidence base update: 50 years of research on treatment for child and adolescent anxiety.
M AN U
Journal of Clinical Child & Adolescent Psychology, 45(2), 91-113.
IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.
In-Albon, T., & Schneider, S. (2006). Psychotherapy of childhood anxiety disorders: A metaanalysis. Psychotherapy and Psychosomatics, 76, 15–24.
TE D
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 followup. Journal of Consulting and Clinical Psychology, 72(2), 276.
EP
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005).
AC C
Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593-602.
Kraemer, H. C., Wilson, G. T., Fairburn, C. G., & Agras, W. S. (2002). Mediators and moderators of treatment effects in randomized clinical trials. Archives of General
Psychiatry, 59, 877 – 883. Muris, P., & Field, A. P. (2008). Distorted cognition and pathological anxiety in children and adolescents. Cognition and Emotion, 22(3), 395-421. 22
ACCEPTED MANUSCRIPT
Oar, E. L., Farrell, L. J., Waters, A. M., & Ollendick, T. H. (2016). Blood-injection-injury phobia and dog phobia in youth: Psychological characteristics and associated features in a clinical sample. Behavior Therapy, 47, 312-324.
RI PT
Ollendick, T. H., & Davis III, T. E. (2013). One-session treatment for specific phobias: A review of Öst's single-session exposure with children and adolescents. Cognitive Behaviour Therapy, 42(4), 275-283.
SC
Ollendick, T. H., Halldorsdottir, T., Fraire, M. G., Austin, K. E., Noguchi, R. J., Lewis, K. M., ... & Whitmore, M. J. (2015). Specific phobias in youth: A randomized controlled trial
Therapy, 46(2), 141-155.
M AN U
comparing one-session treatment to a parent-augmented one-session treatment. Behavior
Ollendick, T. H., & King, N. J. (2012). Evidence-based treatments for children and adolescents: Issues and controversies. Child and Adolescent Therapy: Cognitive-Behavioral
TE D
Procedures, 499-519.
Ollendick, T. H., King, N. J., & Muris, P. (2004). Phobias in children and adolescents. In M. Maj, H. S. Akiskal, J. J. Lopez-Ibor, & A. Okkasha (Eds.), Phobias (pp. 245–279). New
EP
York: Wiley.
Ollendick, T. H., & Muris, P. (2015). The scientific legacy of Little Hans and Little Albert:
AC C
Future directions for research on specific phobias in youth. Journal of Clinical Child & Adolescent Psychology, 44(4), 689-706.
Ollendick, T. H., Öst, L. G., Reuterskiöld, L., Costa, N., Cederlund, R., Sirbu, C., ... & Jarrett, M. A. (2009). One-session treatment of specific phobias in youth: a randomized clinical trial in the United States and Sweden. Journal of consulting and clinical psychology, 77(3), 504.
23
ACCEPTED MANUSCRIPT
Ollendick, T. H., Raishevich, N., Davis, T. E. III., Sirbu, C., & Öst, L-G. (2010). Phenomenology and psychological characteristics of youth with specific phobias. Behavior Therapy, 41, 133-141.
RI PT
Öst, L.-G. (1989). One-session treatment for specific phobias. Behaviour Research and Therapy, 27, 1–7.
Öst, L.-G. (1997). Rapid treatment of specific phobias. In G. C. L. Davey (Ed.), Phobias: A
SC
handbook of theory, research and treatment (pp. 227–247). Oxford, England: Wiley. Öst, L. G. (2012). One-session treatment: Principles and procedures with adults. In Intensive
M AN U
one-session treatment of specific phobias (pp. 59-95). New York, NY: Springer. Öst, L.-G., Svensson, L., Hellström, K., & Lindwall, R. (2001). One session treatment of specific phobias in youths: A randomized clinical trial. Journal of Consulting and Clinical Psychology, 69, 814–824.
TE D
Prins, P.J.M., Ollendick, T.H., Maric, M., & MacKinnon, D.P. (2015). Moderators and mediators in treatment outcome studies of childhood disorders: The what, why and how. In M. Maric, P.J.M. Prins, and T.H. Ollendick, (Eds.), Moderators and mediators of youth
EP
treatment outcomes (pp 1-19). Oxford: Oxford University Press. Silverman, W. K., & Albano, A. M. (1996). Anxiety Disorders Interview Schedule for DSM-IV:
AC C
Child and Parent Versions. San Antonio, TX: Psychological Corporation.
Silverman, W. K., Kurtines, W. M., Ginsburg, G. S., Weems, C. F., Rabian, B., & Serafini, L. T. (1999). Contingency management, self-control, and education support in the treatment of childhood phobic disorders: A randomized clinical trial. Journal of Consulting and Clinical Psychology, 67, 675–687. Silverman, W. K., Pina, A. A., & Viswesvaran, C. (2008). Evidence-based psychosocial
24
ACCEPTED MANUSCRIPT
treatments for phobic and anxiety disorders in children and adolescents. Journal of Clinical Child & Adolescent Psychology, 37(1), 105-130. Stark, K. D., Reynolds, W. M., & Kaslow, N. J. (1987). A comparison of the
RI PT
relative efficacy of self-control therapy and a behavioral problem-solving therapy for depression in children. Journal of Abnormal Child Psychology, 15(1), 91-113.
Thorpe, S. J., & Salkovskis, P. M. (1995). Phobic beliefs: Do cognitive factors play a role in
SC
specific phobias? Behaviour Research and Therapy, 33, 805-816.
Tremblay, L., & Frigon, J-Y. (2005). Precocious puberty in adolescent girls: A biomarker of later
M AN U
psychosocial adjustment problems. Child Psychiatry and Human Development, 36(1), 71−92.
Walkup, J. T., Albano, A., Piacentini, J., Birmaher, B., Compton, S. N., Sherrill, J. T., … Kendall, P.C. (2008). Cognitive behavioral therapy, sertraline, or a combination in
TE D
childhood anxiety. The New England Journal of Medicine, 359, 2753-2766. Wells, A., & Cartwright-Hatton, S. (2004). A short form of the metacognitions questionnaire: Properties of the MCQ-30. Behaviour Research and Therapy, 42, 385-396.
EP
Wright, J. H., & Borden, J. (1991). Cognitive therapy of depression and anxiety. Psychiatric
AC C
Annals, 21, 424-428.
25
M AN U
80 (46.5) 37 (21.5) 55 (32.0)
AC C
EP
TE D
Specific Phobia Subtypes Animal 129 (51.4) Environmental 57 (22.7) Situational 65 (25.9)
SC
Table 1 Participant Descriptive Data Mean (SD) / n (% of sample) Study 1 (n = 251) Study 2 (n = 172) Age 9.87 (2.15) 9.69 (2.19) Male 144 (57.4) 85 (55.2)
RI PT
ACCEPTED MANUSCRIPT
26
ACCEPTED MANUSCRIPT
Table 2. Objective Coding Scales for Children’s Phobic Beliefs Objective Coding Scale of How “Bad” Facet of Children’s Phobic Cognitions
RI PT
0 – Not bad at all, no harm indicated
2 – Belief low-level harm or dangerousness (e.g., I will be scared if I see an animal; Something scary will appear in the dark; The noise from storm will startle me.)
SC
4 – Belief of minor physical harm to self and possible others (e.g., The dog will knock me down; If alone and can't get anyone on phone, I will have a panic attack; Lightning will hit me or others)
8 – Death to self and possible others
M AN U
6 – Explicit and specific belief of major harm to self (e.g., I will catch a disease after being bitten by snake and have to go to hospital; Lightning will catch house on fire and my legs will get burnt; I will see things (e.g., heads) that might come and get you and bite feet off)
Objective Coding Scale of How Likely Facet of Children’s Phobic Cognitions
TE D
0 – Not likely at all
2- Most unlikely to occur (e.g., Dog will bite my finger off; Lightning will strike me; I will see things (e.g., heads) that might come and get you and bite feet off)
EP
4- Minimally likely to occur (e.g., The dog will jump on me and lick my face; Thunder and lightning will break the satellite on the house and the lights might go out; If my head gets under water, the water will get in my ears and it will hurt )
AC C
6- Moderately likely to occur (e.g., I could trip on something in the dark; I would be fearful if a snake crawled towards me; Lightening will strike somewhere) 8- Highly likely to occur (e.g., Being stung by a bee or wasp would hurt; I will hear tree branches moving outside during a strong storm; If being chased by a monster, I will run very fast.)
27
SC
Table 3. Pearson Correlations Among Child Age and Gender, and PBS Ratings and Treated Phobia CSR Child Age Child Gender PBS – “How Likely” .04 -.04 PBS – “How Bad” .28** -.12 PBS – Ability to Cope .04 .06 Pre-Treatment Phobia CSR -.05 .08 Post-Treatment Phobia CSR -.03 .03 6-Month Follow-up Phobia CSR .03 -.06 Note. ** p < .01
RI PT
ACCEPTED MANUSCRIPT
2.71(2.12) 4.89(2.40) 3.83(2.41)
1.63(1.84) 4.88(2.71) 4.40(2.75)
1.58(1.62) 5.12(2.61) 4.25(2.61)
2.17(2.01) 4.95(2.52) 4.07(2.54)
1.42(1.43) 4.44(2.68) 5.02(2.35)
1.62(1.30) 4.65(2.15) 4.43(2.37)
1.86(1.51) 4.43(2.30) 4.45(2.29)
2.56(0.98) 4.39(2.14)
2.18(0.68) 4.34(2.48)
2.71(1.09) 4.38(1.83)
EP
Post-Treatment How Likely How Bad Ability to Cope
TE D
M AN U
Table 4 Means and Standard Deviations for PBS items across Treated Phobia Type and Assessment Time Point Mean(SD) Animal Environmental Situational Total (n = 129) (n = 57) (n = 65) (n = 251) Pre-Treatment How Likely 4.69(1.85) 3.81(2.55) 3.37(2.04) 4.15(2.15) How Bad 6.59(1.81) 6.49(2.04) 6.65(1.91) 6.58(1.88) Ability to 2.36(1.94) 2.81(2.57) 2.66(2.22) 2.54(2.17) Cope
AC C
6-Month Follow-up How Likely 2.18(1.57) How Bad 4.32(2.21) Ability to 4.20(2.21) Cope Objective Codes How Likely 3.05(1.17) How Bad 4.40(1.20)
28
ACCEPTED MANUSCRIPT
SC
RI PT
Table 5 Regression of Post-treatment CSRs on Change in Each PBS Item How Likely How “Bad” Ability to Cope 2 2 ∆R β ∆R β ∆R2 β .11*** .11*** .11*** Step 1 Age -.06 -.03 -.07 Treated Phobia Type -.11 -.12 .11 (Animal) Treated Phobia Type .01 -.00 .01 (Situational) Pre-treatment CSR .27*** .27*** .28*** .01 .06** .03* Step 2 Change in PBS Item -.12 -.25** -.17* F-value 4.89*** 7.08*** 5.52***
M AN U
Note. * p < .05, ** p < .01, *** p < .001.
AC C
Note. † p < .05
EP
TE D
Table 6 Regression of 6-Month Follow-up CSRs on Change in Each PBS Item How “Bad” Ability to Cope How Likely ∆R2 β ∆R2 β ∆R2 β -.00 .02 .02 Step 1 Age .03 .02 .03 Treated Phobia Type .09 .10 .09 (Animal) Treated Phobia Type .02 .05 .04 (Situational) Pre-treatment CSR .14† .15† .15† .01 .00 .01 Step 2 Change in PBS Item -.12 -.02 -.11 F-value 1.20 0.72 1.11
29
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Highlights • Youth with a specific phobia identified feared beliefs and coping expectancies • Differences exist in youth’s catastrophic beliefs across phobia types and age • Youth viewed their beliefs as more catastrophic and likely to occur than did coders • Regardless of phobia type or child age, the beliefs improved following OST • Changes in beliefs were related to changes in clinical severity following treatment