Advertisement

Journal of Autism and Developmental Disorders

, Volume 47, Issue 12, pp 3949–3958 | Cite as

Intolerance of Uncertainty Predicts Anxiety Outcomes Following CBT in Youth with ASD

  • Amy Keefer
  • Nicole L. Kreiser
  • Vini Singh
  • Audrey Blakeley-Smith
  • Amie Duncan
  • Catherine Johnson
  • Laura Klinger
  • Allison Meyer
  • Judy Reaven
  • Roma A. Vasa
S.I. : Anxiety in Autism Spectrum Disorders

Abstract

Modified cognitive–behavioral therapy (MCBT) has been demonstrated to reduce anxiety in youth with autism spectrum disorder (ASD). However, non-response rates are fairly high. Few studies have investigated factors associated with response. Intolerance of uncertainty (IU) is a treatment target for anxiety and worry in neurotypical populations and has been linked to anxiety and ASD. We sought to examine whether IU affects outcomes following MCBT in 43 children, ages 8–14 years, with ASD without intellectual disability. Consistent with prior data, there was a significant reduction in parent reported anxiety following MCBT. Higher levels of pre-intervention IU predicted higher anxiety and worry pre- and post-intervention. These findings suggest that targeting IU may improve outcomes following MCBT in youth with ASD and anxiety.

Keywords

Autism spectrum disorders Anxiety Cognitive–behavioral therapy Intolerance of uncertainty 

Introduction

Evidence suggests that modified cognitive–behavioral therapy (MCBT) may be an efficacious treatment for a majority of children with Autism Spectrum Disorder (ASD) without intellectual disability who have co-occurring anxiety disorders (Sukhodolsky et al. 2013; Ung et al. 2015). Non-response rates, however, are fairly high with 20–50 % of participants demonstrating a limited treatment response (White et al. 2015). This high non-response rate highlights a critical need for research examining factors contributing to differential treatment response (Ung et al. 2015). Identifying prognostic factors can help characterize non-responders and modify CBT protocols to improve treatment outcomes.

Thus far, only one study has identified significant predictors of treatment response to MCBT in youth with ASD. White et al. (2013) found that greater severity of ASD symptoms and higher verbal intellectual ability predicts better response to MCBT. Ung et al. (2015) examined the role of anxiety informant (i.e., if the perception of CBT efficacy varies between child, parent, and clinician reported outcomes) and treatment modality (i.e., individual versus group interventions), but found that they did not moderate response to MCBT. Additionally, no studies have identified treatment predictors that are based on core psychological processes underlying anxiety in this population.

Recently, there has been interest in examining the construct of intolerance of uncertainty (IU) as it relates to anxiety in ASD. IU is conceptualized as a dimensional construct (Carleton 2012) that refers to “a tendency to react negatively on an emotional, behavioral, and cognitive level to uncertain situations and events” (Buhr and Dugas 2009). Individuals with higher IU prefer to “keep things the same” and therefore struggle with life situations that are unpredictable (Berenbaum et al. 2008). In typically developing (TD) youth and adults, IU is postulated to develop from atypical information processing, which leads to the interpretation of neutral stimuli as ambiguous (Dugas et al. 1998), and from negative beliefs about ambiguity which can lead to increased distress (Carleton et al. 2012).

IU is a well-established predictor of greater anxiety and worry in TD youth and adults (Buhr and Dugas 2006, 2009; Gentes and Ruscio 2011; Laugesen et al. 2003; Read et al. 2013). Although causal mechanisms for these relationships are yet to be identified, research has supported a directional relationship in which IU serves as a broad vulnerability factor for anxiety in TD youth and adults (Carleton 2012). Given the tight link between IU and anxiety, IU has been considered an efficacious target for CBT in TD youth and adults (Carleton 2012). Indeed, interventions specifically designed to increase uncertainty tolerance through the restructuring of maintaining cognitions (e.g., “uncertainty leads to negative outcomes”) and exposure to uncertain situations have been demonstrated to reduce anxiety severity in TD adults and children (Buhr and Dugas 2009; Dugas et al. 2012; Ladouceur et al. 2000). This has led some researchers to recommend that the assessment and potential treatment of IU should be a component of all anxiety interventions (Carleton 2012).

Several studies show that IU may be heightened in youth with ASD and linked to both anxiety and to core features of ASD. Boulter et al. (2014) found that IU and anxiety were positively correlated and that IU mediated the relationship between ASD and anxiety. These results suggest a causal model in which ASD is associated with higher levels of IU, which produces higher levels of anxiety (Boulter et al. 2014). In ASD, Chamberlain et al. (2013) found that IU was associated with greater physiological reactivity in response to uncertainty compared to individuals without ASD. Other studies report links between IU and the presence of more specific ASD features including repetitive motor behaviors, insistence on sameness, and sensory overreactivity (Neil et al. 2016; Wigham et al. 2015). Given preliminary evidence linking IU to both anxiety and ASD, the current study extends research on IU by examining whether pre-treatment IU is associated with treatment outcomes following MCBT for anxiety in youth with ASD. We first compared levels of IU, anxiety and worry in youth with ASD to ascertain change before and after MCBT. Following previous studies (Sukhodolsky et al. 2013; Ung et al. 2015), we hypothesized that youth will exhibit a significant reduction in anxiety following treatment. However, given the lack of data on the nature of IU and worry in individuals with ASD, we had no a priori hypothesis regarding a potential change in these constructs following the intervention. The second aim was to examine whether high and low levels of pre-intervention IU predicted response to MCBT. Since IU is positively correlated with anxiety and worry, we first examined whether these same relationships are present in youth with ASD. We then investigated the relationship between pre-intervention IU and treatment outcomes. As there are no prior data on the role of IU in the response to MCBT in individuals with ASD, our investigation of this relationship was exploratory.

Methods

Participants and Procedures

This study was embedded in a larger investigation examining the impact of clinician training for the Facing Your Fears (FYF) anxiety treatment protocol (Reaven 2011; Reaven et al. 2012) on treatment outcomes in youth with ASD without intellectual disability. The FYF protocol is a 15-session MCBT intervention that utilizes a group therapy model for parents and youth. The intervention provides psychoeducation, behavioral coping strategies (e.g., relaxation), modified cognitive restructuring, and hierarchical exposures (e.g., facing fears a little at a time) to target specific fears identified by the child and parent. Intolerance of uncertainty is not systematically targeted by the intervention and, consistent with other MCBT interventions for youth with ASD, cognitive restructuring components were simplified compared to traditional CBT protocols. Clinician fidelity to the intervention was high, ranging from 74 to 99 % across sites and conditions. This study reports on data gathered during the final 2 years of the 3 year study.

Participants included 43 children and adolescents with ASD, ages 8–14 years, who were recruited through clinic or self-referral at three university based clinics (University of North Carolina TEACCH Autism Program, Cincinnati Children’s Hospital Medical Center, and Kennedy Krieger Institute). Families completed eligibility and pre-intervention measures 6 weeks prior to the treatment start date. Post-intervention measures were completed within 6 weeks of the final group session. All participants attended a minimum of 11 of 14 treatment sessions and, if deemed necessary, met individually with treatment leaders to review concepts missed due to absence. Participants had a well-characterized ASD diagnosis utilizing the Autism Diagnostic Observation Schedule (ADOS-G or ADOS-2; Lord et al. 2000, 2012) and a verbal IQ of 80 or above based on the Wechsler Abbreviated Scales of Intelligence (Weschler 2002), or an equivalent measure of cognitive ability. Participants also met criteria for at least one of the following DSM-IV anxiety disorders: generalized anxiety disorder, social anxiety disorder, separation anxiety disorder as measured by the Anxiety Disorders Interview Schedule-Parent (Silverman and Albano 1996) (Table 1). These three disorders frequently co-aggregate and have been shown to respond to similar treatments (Walkup 2008). Youth with specific phobia were included if they also met criteria for one of the primary anxiety disorder listed above. Ratings on the Screen for Childhood Anxiety Related Disorders: Child and Parent Versions (Birmaher et al. 1999) did not determine eligibility.
Table 1

Participant demographic characteristics

 

M (SD)

Age (in years)

11.18 (2.02)

Full scale IQ

102.6 (14.7)

 

n (% total)

Male

35 (81)

Maternal education

 

 High school

4 (10)

 Partial college

11 (27)

 College graduate

11 (27)

 Post graduate training

15 (36)

Types of anxiety disorders

 

 GADa

39 (91)

 Social anxiety disorder

38 (88)

 Separation anxiety disorder

16 (37)

 Specific phobiab

36 (84)

Number of anxiety disorders

 

 1

1 (2.3)

 2

9 (20.9)

 3+

33 (76.7)

a GAD generalized anxiety disorder

bCould not be primary anxiety disorder

Participant characteristics were similar across sites. Of the 43 parent–children pairs recruited for the study, 38 children and 36 parents completed all pre-intervention study measures. Within this group, 10 (26.3 %) children and 8 (22.2 %) parents did not complete the corresponding post-intervention measures either due to study drop-out or non-response to requests for these measures. Therefore, 28 parent-child pairs provided pre- and post-intervention data. To account for missing data, the analyses were run using listwise and pairwise deletion. Since the two analyses did not yield significantly different results the whole data was used for analysis (pairwise deletion) to maximize sample size. There were no significant differences in demographic characteristics or pre-intervention levels of outcome measures (i.e. intolerance of uncertainty, anxiety, or worry) between youth who completed and did not complete the study.

Measures

Modified Intolerance of Uncertainty Scale: Child and Parent Versions (IUS-C; IUS-P; Comer et al. 2009)

The IUS-C has 27 questions that assess children’s tendency to experience negative emotions, behaviors, and cognitions to uncertain situations and events (Buhr and Dugas 2002). It is based on a two-factor model consisting of items related to: 1) prospective intolerance (i.e., a desire for predictability) and 2) inhibitory intolerance (i.e., “uncertainty paralysis” or the inability to act when facing uncertainty; McEvoy and Mahoney 2011). The IUS-C has demonstrated strong internal consistency and convergent validity in TD youth (Comer et al. 2009). For the present study, the IUS-C was modified to support child comprehension and to accommodate common language difficulties in children with ASD. As such, specific modifications were made to reduce ambiguity as well as complexity of syntax and grammar. Some of the modifications included re-wording items from passive to active voice (e.g., changed “Surprise events upset me greatly.” to “I don’t like to be surprised by new plans or activities.”), and consistently using first person pronouns to reduce abstraction and increase personalization (e.g., changed “One should always think ahead to avoid surprises.” to “I always try to think ahead in order to avoid surprises.”). Additionally, the original 5-point Likert scale was reduced to 4 points to minimize complexity (1 = never true, 2 = sometimes true, 3 = almost always true, 4 = always true), resulting in total scores ranging from 27 to 108. A parallel parent-report version of the Intolerance to Uncertainty Scale (IUS-P) was administered for parents to report on their children. Modifications were approved by the scale developer. Internal consistency for the modified scales ranged from good to excellent in the present study (IUS-P α = 0.96; IUS-C α = 0.89). Due to modifications in the scale range, scores did not align with established clinical cut-off scores. Thus, pre-intervention IUS-C and IUS-P scores were divided using a median split (IUS-P median = 58; IUS-C median = 50) to compare treatment outcomes of those with higher IU (IUS-P M = 74.95; SD = 10.79; IUS-C M = 61.45; SD = 8.77) to those with lower IU in our sample (IUS-P M = 48.76; SD = 8.03; IUS-C M = 39.43; SD = 8.02).

Modified Penn State Worry Questionnaire-Children (PSWQ-C; Chorpita et al. 1997)

The PSWQ-C is a 14-item self-report worry scale in children and adolescents. The PSWQ-C has consistently demonstrated strong convergent and discriminant validity and reliability (Chorpita et al. 1997; Muris et al. 2001; Pestle et al. 2008). Modifications were made to the syntax and grammar including rewording items to increase precision of syntax. Items were rated using the original 4-point Likert scale (0 = never true, 1 = sometimes true, 2 = almost always true, 3 = always true) resulting in total scores ranging from 0 to 42. Directions were enhanced to ensure that the child understood the definition of worry and the Likert scale. Modifications were approved by the scale developers. The modified PSWQ-C demonstrated excellent internal consistency in this sample (α = .90).

Screen for Childhood Anxiety Related Disorders: Child and Parent Versions (SCARED-C and SCARED-P; Birmaher et al. 1999)

The SCARED is a 41-item screener for symptoms of anxiety with five anxiety subscales (panic, generalized anxiety, separation anxiety, social anxiety, and school anxiety). Items were rated using a 3-point Likert Scale (0 = not true or hardly ever true, 1 = somewhat true or sometimes true, 2 = very true of often true), resulting in total scores ranging from 0 to 82. Parent and child versions are identical in content, and prior investigations have identified optimal cut-off scores of 25 (child) and 17 (parent) to best identify clinically significant anxiety (Birmaher et al. 1999). The tool has good convergent and divergent validity when compared to formal psychiatric diagnoses (Birmaher et al. 1999). Internal consistency for both parent and child report has been found to be adequate in ASD samples (Stern et al. 2014). In this sample, internal consistency was acceptable (parent report α = .71, child report α = .77).

Analysis

Shapiro–Wilk tests showed that although anxiety, worry, and IU data were normally distributed, parent-report SCARED scores were not. Thus, non-parametric statistics were used to minimize Type-I error. Aim 1 was examined using Wilcoxon signed-rank tests to assess the difference between pre- and post-intervention scores for each measure.

Aim 2 was examined using Spearman’s rank correlation coefficients to examine relationships among the predictor (IU) and outcome variables (SCARED-P, SCARED-C, PSWQ-C) before and following the intervention for each measure. Next linear mixed models with random slopes and fixed and random effects were developed. Three models were constructed to examine the association between the main predictor variable (IU) and each of the outcome variables. This approach was selected as it accounts for the correlated nature of the anxiety measurements, which were taken on the same individual at two time-points, and the heterogeneity within and between individuals in predicting the change in anxiety (Verbeke et al. 2010). The independent variables, namely pre-IU group (i.e., high versus low as identified using a median split), age, full scale IQ, and total number of anxiety disorders were entered as fixed effects. The variable “time,” which refers to the change in anxiety scores for each subject across the intervention, was treated as a random effect. This allowed for a randomly distributed intercept and slope of the pre-intervention to post-intervention anxiety score for every individual over time while modelling the mean response by IU group membership. In each model, univariate analyses were first conducted to assess the role of each independent variable in predicting the outcome. Multivariate analyses were then performed to assess the contribution of the other independent variables (i.e., age, full scale IQ, number of anxiety disorders, time) in explaining the relationship of the main predictor (IU) and each outcome measure. An interaction term (time by IU) was also included in each model to assess if membership in the high IU group was associated with a different trajectory or rate of change in outcome. The most parsimonious or best fitting model was selected by maximizing the variance in outcome explained by every variable (adjusted R-squared) and significance of the association in conjunction with the Akaike Information Criteria (AIC) and Bayesian Information Criteria (BIC).

Results

Aim 1: Change in IU, Anxiety, and Worry Pre- and Post-intervention

Figure 1 presents the data on change in pre- and post-intervention scores for each measure. Parent report data indicated no significant difference in pre- to post-intervention IUS scores (mean diff = −2.89; 95 % CI −7.22, 1.44). The child report data, however, indicated a near significant change in IUS scores (mean diff = −5.7; 95 % CI −11.9, 0.42).
Fig. 1

Changes in variables of interest from pre- to post-intervention

A significant decrease in pre- to post-intervention SCARED scores was present according to parent report (mean diff = −3.44; 95 % CI −6.26, −0.62), but not present according to child report (mean diff = −2.13; 95 % CI −6.25, 1.99). There was no significant change in PSWQ scores (mean diff = −2.93; 95 % CI −6.54, 0.7) from pre- to post-intervention.

Aim 2: Relationship of Pre-intervention IU to Anxiety Outcomes Following MCBT

Table 2 presents the results of the Spearman’s correlations. Main findings show pre-intervention IUS scores were positively correlated with pre-intervention SCARED scores (parent report = 0.69, child report = 0.36) and post-intervention SCARED scores within informant (parent report = 0.62, child report = 0.65). Post-intervention IUS scores also were positively correlated with post-intervention SCARED scores within information (parent report = 0.45, child report = 0.36). Child reported pre-intervention IUS scores were significantly positively correlated with pre-intervention (0.51) and post-intervention PSWQ-C scores (0.61). Child-reported post-intervention IUS scores were significantly positively correlated with post-intervention PSWQ-C scores (0.48).
Table 2

Correlations pre- and post-intervention

Variable

IUS-P pre

IUS-C pre

PSWQ-C pre

SCARED-P pre

SCARED-C pre

IUS-P post

IUS-C post

IUS-P-pre

      

IUS-C-pre

0.02

     

PSWQ-C pre

−0.18

0.51*

    

SCARED-P pre

0.69*

0.08

−0.15

   

SCARED-C pre

−0.21

0.36*

0.55*

−0.06

  

SCARED-P post

0.62*

0.18

−0.10

0.78*

0.06

0.45*

−0.21

SCARED-C post

0.08

0.65*

0.31

0.28

0.54*

0.50

0.36*

PSWQ-C post

0.10

0.61*

0.50*

0.15

0.24

0.24

0.48*

p < 0.5

Table 3 presents the results of the three linear mixed models examining whether pre-intervention IU predicts post-intervention parent reported anxiety, child reported anxiety, and child reported worry. When investigating parent reported anxiety (Model 1), univariate analysis showed that parent reported pre-IU group (high pre-IU versus low pre-IU) was the only variable that was significantly associated with post-intervention parent-reported SCARED scores (β = 14.72, 95 % CI 8.82, 20.62). The multivariate model also revealed a significant effect of parent reported pre-IU group on post-intervention parent reported SCARED scores. Age, IQ, and total number of anxiety disorders were not significant in the model. The IU by time interaction also was not significant and therefore excluded from the model. Thus, the most parsimonious model was selected over the multivariate model. This model indicated that those in the high IU group had significantly elevated SCARED scores post-intervention (β = 14.74, 95 % CI 8.85, 20.63) compared to those in the low IU group. There was a significant effect of time (β = −2.8, 95 % CI −5.14, −0.48), indicating that parent reported SCARED scores decreased significantly in both the high and low parent reported IU groups. Figure 2 illustrates this main finding.
Table 3

Models of anxiety and worry by IU group-parent and child report (a) Model 1: Anxiety models by IU group-parent report, (b) Model 2: Anxiety models by IU group-child report, (c) Model 3: Worry models by IU group-child report

 

Univariate

Multivariate

Parsimonious

β(SE)

p

β(SE)

p

β(SE)

p

(a)

      

Pre-IU groupa

14.72 (3.01)

<0.001

15.57 (3.13)

<0.001

14.74 (3.01)

<0.001

Age

−0.65 (0.85)

0.45

0.60 (0.63)

0.34

  

IQ

−0.16 (0.11)

0.15

−0.08 (0.07)

0.24

  

Total anxiety disordersb

4.04 (4.61)

0.38

1.23 (3.04)

0.68

  

Time

−3.43 (1.38)

0.01

−2.81 (1.18)

0.018

−2.81 (1.18)

0.018

(b)

      

Pre-IU Groupa

9.54 (3.14)

0.002

8.38 (3.88)

0.03

9.36 (3.15)

0.03

Age

−1.11 (0.67)

0.1

−0.72 (0.87)

0.41

  

IQ

−0.21 (0.14)

0.17

−0.21 (0.15)

0.14

  

Total anxiety disordersb

3.54 (3.28)

0.28

−0.26 (3.19)

0.93

  

Time

2.18 (1.97)

0.27

1.23 (1.8)

0.49

1.96 (1.94)

0.31

(c)

      

Pre-IU groupa

8.81 (2.07)

<0.001

7.67 (2.45)

0.003

8.68 (2.08)

<0.001

Age

−1.17 (0.46)

0.01

−0.62 (0.47)

0.19

  

IQ

−0.02 (0.09)

0.80

−0.03 (0.08)

0.7

  

Total anxiety disordersb

4.09 (2.15)

0.06

0.01 (2.85)

0.99

  

Time

2.86 (1.60)

0.07

2.26 (1.63)

0.17

2.5 (1.66)

0.13

aPre-intervention high and low IU groups

bTotal number of DSM-IV anxiety disorders

Fig. 2

Change in anxiety following intervention in low versus high IU groups

Similar results emerged when predicting child-reported anxiety (Model 2) and worry (Model 3). The most parsimonious models supported a significant group effect for child reported pre-intervention IU on child reported SCARED and PSWQ-C scores, indicating that membership in the high IU group was associated with higher post-intervention child reported SCARED scores (β = 9.36, 95 % CI 3.17, 15.54) and PSWQ-C scores (β = 8.68, 95 % CI 4.60, 12.76) compared to those in the low IU group. However unlike parent-reported anxiety, there was no significant effect of time in child reported worry or anxiety scores, indicating that these variables did not change significantly following the intervention.

Post-hoc analysis investigated the percentage of individuals exhibiting clinical levels of anxiety according to parent reported SCARED scores at pre- and post-intervention. In the low IU group per parent report, 11 of the 17 (65 %) participants exhibited clinical levels of anxiety pre-intervention and 9 (53 %) exhibited clinical anxiety post-intervention. In comparison, all participants (n = 11) in the high IU group exhibited clinically significant levels of anxiety pre- and post-intervention.

Discussion

This study was an initial exploration of the role of IU in the response to MCBT in youth with ASD and co-occurring anxiety disorders. As demonstrated in previous studies of MCBT in individuals with ASD (Kreslins et al. 2015; Sukhodolsky et al. 2013), the FYF intervention was effective in reducing anxiety levels according to parent but not child ratings, which is consistent with data from prior MCBT studies (Wood et al. 2015) and may reflect the limited emotional insight of youth with ASD (White et al. 2012). Unlike findings in TD youth, levels of IU and worry did not decrease following the FYF intervention (Bomyea et al. 2015; Boswell et al. 2013).

The unique finding of this study is that, similar to TD individuals, IU was correlated with anxiety and worry and associated with greater anxiety and worry both preceding and following intervention. We found that anxiety decreased in both the high and low pre-intervention IU groups, however, youth with high levels of IU at baseline exhibited significantly higher levels of anxiety and worry preceding and following MCBT compared to youth with low levels of IU. Clinical anxiety was more likely to be present in youth with high pre-intervention IU. In fact, according to parent report, 100 % of participants in the high IU group exhibited clinical anxiety pre- and post-treatment; whereas, 65 and 53 % of individuals in the low IU group demonstrated clinical anxiety pre- and post-intervention, respectively. Taken together these findings suggest that IU is related to anxiety and worry. However, IU may not respond to current MCBT strategies and appears to be related to poorer outcomes following intervention.

There are several potential explanations as to why IU levels may not have changed significantly and why individuals with higher IU had higher levels of anxiety and worry following MCBT. One possibility is that MCBT may not impact IU in the same manner as traditional CBT strategies. CBT typically includes abstract and insight-driven processes such as the restructuring of maladaptive cognitions that may directly maintain IU (Beck 1970). Due to neuropsychological challenges (i.e., difficulty with abstract reasoning, limited insight) experienced by individuals with ASD, intensive cognitive restructuring strategies are typically modified in MCBT programs, which may limit efficacy in reducing IU. It is also possible that IU did not decrease because behavioral treatment components of MCBT (i.e., graded exposure) did not systematically target IU but rather addressed more concrete and specific anxiety stimuli. As a result, IU levels may have remained high and continued to drive high levels of anxiety and worry. It is also possible that current MCBT protocols are too short in duration and that a greater number of sessions are needed to reduce IU levels. Furthermore, MCBT protocols may not address associated physiological and psychological issues. Recent research has reported greater physiological reactivity to uncertainty in individuals with ASD when engaging in a lab based paradigm (Chamberlain et al. 2013). Heightened physiological reactions have been found to drive higher levels of anxiety and worry, which can compromise fear extinction (Morriss et al. 2015). High levels of IU have also been linked to cognitive avoidance and negative problem orientation in TD individuals (Laugesen et al. 2003) and deficits in information processing (Dugas and Koerner 2005). The presence of these associated challenges may interfere with the treatment process and inhibit the implementation of recommended strategies.

Another potential explanation for the lack of change in IU levels and the higher levels of anxiety and worry demonstrated in individuals with high IU following intervention is that IU may be related to core features of ASD. Under this conceptualization, IU would be expected to remain more stable over time and thus drive higher levels of anxiety and worry. Preliminary evidence suggests that IU may be associated with ASD features. In a recent study of youth with and without ASD, ASD diagnostic status was found to predict level of IU when controlling for anxiety (Kreiser et al. 2016) and another study demonstrated an association between IU and sensory sensitivities after adjusting for anxiety (Neil et al. 2016). Additionally, the two types of IU, a desire for predictability (i.e., a dislike of uncertainty and a desire to keep things the same) and uncertainty paralysis (i.e., the experience of feeling cognitively or behaviorally “stuck” when facing uncertainty; Carleton 2012) have a strong conceptual overlap with the DSM-5 ASD criterion of insistence on sameness and inflexible adherence to routines, behaviors which serve to prevent change in an individual’s environment. Collectively, these empirical and clinical data suggest that IU may be uniquely linked to the ASD phenotype. Under this hypothesis, the nature and etiology of IU may differ from that observed in the TD population and may require alternative treatment strategies (e.g., mindfulness based strategies to increase tolerance of IU) to reduce its severity as well as associated anxiety symptomatology.

Findings from this study have implications for developing more effective CBT treatment approaches for youth with ASD and co-occurring anxiety. To our knowledge, current MBCT approaches do not typically include recommendations specifically targeting IU. Rather, treatments tend to target anxiety more frequently through stimuli-focused exposure hierarchies (e.g., exposure to thunderstorms) rather than hierarchies eliciting exposures to uncertainty (e.g., distress about the uncertainty of changing weather patterns), which may maintain anxiety and worry. Given the repeated finding of high levels of IU in the ASD population (Boulter et al. 2014; Chamberlain et al. 2013; Kreiser et al. 2016), current MCBT models may be enhanced by emphasizing the importance of evaluating and potentially targeting IU through graded exposure in ASD youth. Additionally, the use of “exposure optimizing strategies” which are based in inhibitory learning models of anxiety (Craske et al. 2014) may enhance treatment response. Strategies such as “deepened extinction” in which individuals are exposed to two anxiety-triggering stimuli in isolation and then in combination have been demonstrated to reduce relapse of anxiety (Rescorla 2006). The use of this strategy to simultaneously target IU and co-occurring anxiety may produce greater reduction in anxiety and worry in ASD youth. Additionally, behavioral support strategies such as introducing periods of uncertainty throughout a child’s day (e.g., including a “question mark” in a visual schedule during a certain time period) to prepare individuals for the possibility of uncertainty and opportunities to utilize behavioral coping strategies (e.g., physical exercise, pleasurable sensory experiences) when facing uncertainty may be helpful in managing distress related to IU (Hodgson et al. 2016).

Another clinical implication of our findings pertains to assessment of IU. Clinicians may not routinely assess IU when evaluating anxiety in clinical practice. When it is reported, it may be dismissed as a feature of the ASD phenotype (e.g., behavioral rigidity), or misattributed to other comorbidities (e.g., behavioral noncompliance) by parents and clinicians. Additionally, due to the neuropsychological weaknesses associated with ASD, youth with ASD may have difficulty identifying and expressing the role of IU in their distress (Happe 2003; Wood et al. 2009). The current findings suggest that the direct assessment of IU is important to guide treatment planning regarding specific interventions for IU in youth with ASD and anxiety.

Strengths, Limitations, and Future Directions

One limitation to this study is its small sample size, which limits statistical power. This study did not directly examine how change in IU affected post-intervention outcomes, but rather provides preliminary evidence that post-treatment outcomes may be affected by pre-intervention IU levels. Therefore a larger sample is necessary to fully explore the possibility of moderation or mediation effects between these variables.

Other limitations pertain to measurement of target constructs. Although the IUS has demonstrated acceptable to excellent internal consistency in previous studies of individuals with ASD (Boulter et al. 2014), neither the IUS nor the PSWQ have been validated in the ASD population. To enhance feasibility, study authors modified the IUS and PSWQ to simplify item syntax and assess the child’s understanding of targeted constructs and the rating scale. All measures demonstrated good internal consistency and convergent validity (Table 2). The IUS also does not differentiate the potential overlap between IU and ASD symptoms, and there are longstanding concerns regarding the validity of rating scale instruments in the ASD population (Kerns and Kendall 2012). Additionally, the measures had a high number of within-informant correlations (i.e., child report associated with child report; parent report associated with parent report) which raises concern about the influence of shared method variance. This weakness is largely inevitable given the limited strategies to assess anxiety and IU in the ASD population and highlights the importance of developing more objective behavioral and physiological paradigms in future studies to better characterize the phenotype of IU and anxiety in ASD. Furthermore future studies would be strengthened by the inclusion of measures that provide data regarding the clinical response to the intervention (e.g., change in diagnostic status).

Another potential concern is that participants represented a wide age range that crossed developmental stages. However, age was entered and not found to be a significant contributor to the analyses. Additionally, the age range targeted in the study is consistent with current clinical practice for MCBT and, whenever possible, treatment groups were formed based on participant age (i.e., 8–11 year olds; 11–14 year olds). Future studies should examine these relationships using distinct developmental groups (e.g., pre-pubertal versus post-pubertal youth) to examine potential differences. Consistent with the current treatment literature, the current sample focused on youth with ASD who did not have intellectual disability and therefore the findings are not generalizable to other subgroups of youth with ASD. Finally, this study did not include a sample of youth with ASD and anxiety that did not receive the intervention to provide clinical comparison regarding treatment response.

Summary

This is one of the first studies to stratify outcomes of MCBT according to pre-intervention IU levels. Results indicate that higher levels of pre-intervention IU are associated with higher levels of anxiety and worry post-intervention. Current MCBT protocols may need to be modified to target IU and to enhance response to anxiety intervention for ASD youth.

Notes

Acknowledgments

Drs. Reaven, Keefer, Duncan, and Klinger and Ms. Johnson were supported by NIH Grant 4R33MH089291-03 which was awarded to Dr. Reaven. Recruitment at TEACCH Autism Program was supported by NICHD Grant U54HD079124. Dr. Vasa’s effort was supported by Autism Speaks Grant 8790.

Funding

This study was funded in part by NIH Grant #4R33MH089291-03, Autism Speaks Grant #8790, and NICHD Grant #U54HD079124.

Author Contributions

AK conceived of the study, participated in its design and coordination, and drafted the manuscript; NK participated in the design and interpretation of the data and drafting the manuscript; VS participated in the design and performed the statistical analysis; ABS participated in drafting the manuscript; AD participated in the measurement and drafting the manuscript; CJ participated in the measurement and design; LK participated in drafting the manuscript; AM participated in the measurement and drafting the manuscript; JR contributed to the interpretation of the data and drafting the manuscript; RV conceived of the study, participated in its design and drafting the manuscript. All authors read and approved the final manuscript.

Compliance with Ethical Standards

Conflict of interest

Drs. Keefer, Kreiser, Duncan, Klinger, Meyer, and Vasa and Ms. Johnson and Ms. Singh have no conflicts of interest. Drs. Reaven and Blakeley-Smith receive royalties from Paul H. Brookes, publisher of the Facing Your Fears treatment manual.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.

References

  1. Beck, A. T. (1970). Cognitive therapy: Nature and relation to behavior therapy. Behavior Therapy, 1(2), 184–200. doi: 10.1016/S0005-7894(70)80030-2.CrossRefGoogle Scholar
  2. Berenbaum, H., Bredemeier, K., & Thompson, R. J. (2008). Intolerance of uncertainty: Exploring its dimensionality and associations with need for cognitive closure, psychopathology, and personality. Journal of Anxiety Disorders, 22(1), 117–125. doi: 10.1016/j.janxdis.2007.01.004.CrossRefPubMedGoogle Scholar
  3. Birmaher, B., Brent, D., Chiapetta, L., Bridge, J., Monga, S., & Baugher, M. (1999). Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED). Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1230–1236.CrossRefPubMedGoogle Scholar
  4. Bomyea, J., Ramsawh, H., Ball, T. M., Taylor, C. T., Paulus, M. P., Lang, A. J., & Stein, M. B. (2015). Intolerance of uncertainty as a mediator of reductions in worry in a cognitive behavioral treatment program for generalized anxiety disorder. Journal of Anxiety Disorders, 33, 90–94. doi: 10.1016/j.janxdis.2015.05.004.CrossRefPubMedPubMedCentralGoogle Scholar
  5. Boswell, J. F., Thompson-Hollands, J., Farchione, T. J., & Barlow, D. H. (2013). Intolerance of uncertainty: A common factor in the treatment of emotional disorders. Journal of Clinical Psychology, 69(6), 630–645. doi: 10.1002/jclp.21965.CrossRefPubMedGoogle Scholar
  6. Boulter, C., Freeston, M., South, M., & Rodgers, J. (2014). Intolerance of uncertainty as a framework for understanding anxiety in children and adolescents with autism spectrum disorders. Journal of Autism and Developmental Disorders, 44(6), 1391–1402. doi: 10.1007/s10803-013-2001-x.CrossRefPubMedGoogle Scholar
  7. Buhr, K., & Dugas, M. J. (2002). The intolerance of uncertainty scale: Psychometric properties of the English version. Behaviour Research and Therapy, 40(8), 931–945.CrossRefPubMedGoogle Scholar
  8. Buhr, K., & Dugas, M. J. (2006). Investigating the construct validity of intolerance of uncertainty and its unique relationship with worry. Journal of Anxiety Disorders, 20(2), 222–236. doi: 10.1016/j.janxdis.2004.12.004.CrossRefPubMedGoogle Scholar
  9. Buhr, K., & Dugas, M. J. (2009). The role of fear of anxiety and intolerance of uncertainty in worry: An experimental manipulation. Behaviour Research and Therapy, 47(3), 215–223. doi: 10.1016/j.brat.2008.12.004.CrossRefPubMedGoogle Scholar
  10. Carleton, R. N. (2012). The intolerance of uncertainty construct in the context of anxiety disorders: Theoretical and practical perspectives. Expert Review of Neurotherapeutics, 12(8), 937–947. doi: 10.1586/ern.12.82.CrossRefPubMedGoogle Scholar
  11. Carleton, R. N., Mulvogue, M. K., Thibodeau, M. A., McCabe, R. E., Antony, M. M., & Asmundson, G. J. (2012). Increasingly certain about uncertainty: Intolerance of uncertainty across anxiety and depression. Journal of Anxiety Disorders, 26(3), 468–479. doi: 10.1016/j.janxdis.2012.01.011.CrossRefPubMedGoogle Scholar
  12. Chamberlain, P. D., Rodgers, J., Crowley, M. J., White, S. E., Freeston, M. H., & South, M. (2013). A potentiated startle study of uncertainty and contextual anxiety in adolescents diagnosed with autism spectrum disorder. Molecular Autism, 4(1), 31. doi: 10.1186/2040-2392-4-31.CrossRefPubMedPubMedCentralGoogle Scholar
  13. Chorpita, B. F., Tracey, S. A., Brown, T. A., Collica, T. J., & Barlow, D. H. (1997). Assessment of worry in childre and adolescents: An adaptation of the Penn State Worry Questionnaire. Behavior Research and Therapy, 35, 569–581.CrossRefGoogle Scholar
  14. Comer, J. S., Roy, A. K., Furr, J. M., Gotimer, K., Beidas, R. S., Dugas, M. J., & Kendall, P. C. (2009). The intolerance of uncertainty scale for children: A psychometric evaluation. Psychological Assessment, 21(3), 402–411. doi: 10.1037/a0016719.CrossRefPubMedPubMedCentralGoogle Scholar
  15. Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Maximizing exposure therapy: An inhibitory learning approach. Behaviour Research and Therapy, 58, 10–23. doi: 10.1016/j.brat.2014.04.006.CrossRefPubMedPubMedCentralGoogle Scholar
  16. Dugas, M. J., Gagnon, F., Ladouceur, R., & Freeston, M. H. (1998). Generalized anxiety disorder: A preliminary test of a conceptual model. Behaviour Research and Therapy, 36(2), 215–226.CrossRefPubMedGoogle Scholar
  17. Dugas, M. J., & Koerner, N. (2005). Cognitive behavioral treatment for generalized anxiety disorder: Current status and future directions. Journal of Cognitive Psychotherapy: An International Quarterly, 19, 61–81.CrossRefGoogle Scholar
  18. Dugas, M. J., Laugesen, N., & Bukowski, W. M. (2012). Intolerance of uncertainty, fear of anxiety, and adolescent worry. Journal of Abnormal Child Psychology, 40(6), 863–870. doi: 10.1007/s10802-012-9611-1.CrossRefPubMedGoogle Scholar
  19. Gentes, E. L., & Ruscio, A. M. (2011). A meta-analysis of the relation of intolerance of uncertainty to symptoms of generalized anxiety disorder, major depressive disorder, and obsessive–compulsive disorder. Clinical Psychology Review, 31(6), 923–933. doi: 10.1016/j.cpr.2011.05.001.CrossRefPubMedGoogle Scholar
  20. Happe, F. (2003). Cognition in autism: One deficit or many? Novartis Foundation Symposium, 251, 198–207 (discussion 207–112, 281–197).Google Scholar
  21. Hodgson, A. R., Freeston, M. H., Honey, E., & Rodgers, J. (2016). Facing the unknown: Intolerance of uncertainty in children with autism spectrum disorder. Journal of Applied Research in Intellectual Disabilities. doi: 10.1111/jar.12245.Google Scholar
  22. Kerns, C. M., & Kendall, P. C. (2012). The presentation and classification of anxiety in autism spectrum disorder. Clinical Psychology: Science and Practice, 19(4), 323–347. doi: 10.1111/cpsp.12009.Google Scholar
  23. Kreiser, N., Keefer, A., Singh, V., Mostofsky, S., & Vasa, R. (2016). The relationship between intolerance of uncertainty, anxiety, and ASD features. Manuscript submitted for publication.Google Scholar
  24. Kreslins, A., Robertson, A. E., & Melville, C. (2015). The effectiveness of psychosocial interventions for anxiety in children and adolescents with autism spectrum disorder: A systematic review and meta-analysis. Child and Adolescent Psychiatry and Mental Health, 9, 22. doi: 10.1186/s13034-015-0054-7.CrossRefPubMedPubMedCentralGoogle Scholar
  25. Ladouceur, R., Gosselin, P., & Dugas, M. J. (2000). Experimental manipulation of intolerance of uncertainty: A study of a theoretical model of worry. Behaviour Research and Therapy, 38(9), 933–941.CrossRefPubMedGoogle Scholar
  26. Laugesen, N., Dugas, M. J., & Bukowski, W. M. (2003). Understanding adolescent worry: The application of a cognitive model. Journal of Abnormal Child Psychology, 31(1), 55–64.CrossRefPubMedGoogle Scholar
  27. Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L., DiLavore, P. C., et al. (2000). The autism diagnostic observation schedule-generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism Developmental Disorders, 30, 205–223.CrossRefPubMedGoogle Scholar
  28. Lord, C., Rutter, M., DiLavore, P. C., Risi, S., Gotham, K., & Bishop, S. (2012). Autism Diagnostic Observation Schedule (2nd ed.). Torrance, CA: Western Psychological Services.Google Scholar
  29. McEvoy, P. M., & Mahoney, A. E. (2011). Achieving certainty about the structure of intolerance of uncertainty in a treatment-seeking sample with anxiety and depression. Journal of Anxiety Disorders, 25, 112–122.CrossRefPubMedGoogle Scholar
  30. Morriss, J., Christakou, A., & van Reekum, C. M. (2015). Intolerance of uncertainty predicts fear extinction in amygdala-ventromedial prefrontal cortical circuitry. Biology of Mood and Anxiety Disorders, 5, 4. doi: 10.1186/s13587-015-0019-8.CrossRefPubMedPubMedCentralGoogle Scholar
  31. Muris, P., Meesters, C., & Gobel, M. (2001). Reliability, validity, and normative data of the Penn State Worry Questionnaire in 8–12-yr-old children. Journal of Behavior Therapy and Experimental Psychiatry, 32, 63–72.CrossRefPubMedGoogle Scholar
  32. Neil, L., Olsson, N. C., & Pellicano, E. (2016). The relationship between intolerance of uncertainty, sensory sensitivities, and anxiety in autistic and typically developing children. Journal of Autism and Developmental Disorders. doi: 10.1007/s10803-016-2721-9.PubMedPubMedCentralGoogle Scholar
  33. Pestle, S. L., Chorpita, B. F., & Schiffman, J. (2008). Psychometric properties of the Penn State Worry Questionnaire for Children in a Large Clinical Sample. Journal of Clinical Child & Adolescent Psychology, 37, 465–471.CrossRefGoogle Scholar
  34. Read, K. L., Comer, J. S., & Kendall, P. C. (2013). The intolerance of uncertainty scale for children (IUSC): Discriminating principal anxiety diagnoses and severity. Psychological Assessment, 25(3), 722–729. doi: 10.1037/a0032392.CrossRefPubMedGoogle Scholar
  35. Reaven, J. (2011). The treatment of anxiety symptoms in youth with high-functioning autism spectrum disorders: Developmental considerations for parents. Brain Research, 1380, 255–263. doi: 10.1016/j.brainres.2010.09.075.CrossRefPubMedGoogle Scholar
  36. Reaven, J., Blakeley-Smith, A., Culhane-Shelburne, K., & Hepburn, S. (2012). Group cognitive behavior therapy for children with high-functioning autism spectrum disorders and anxiety: A randomized trial. Journal of Child Psychology and Psychiatry, 53(4), 410–419.CrossRefPubMedPubMedCentralGoogle Scholar
  37. Rescorla, R. A. (2006). Deepened extinction from compound stimulus presentation. Journal of Experimental Psychology: Animal Behavior Processes, 32(2), 135–144. doi: 10.1037/0097-7403.32.2.135.PubMedGoogle Scholar
  38. Silverman, W., & Albano, A. (1996). Anxiety disorders interview schedule for children for DSM-IV: (child and parent versions). San Antonio, TX: Psychological Corporation.Google Scholar
  39. Stern, J., Gadgil, M., Blakeley-Smith, A., Reaven, J., & Hepburn, S. (2014). Psychometric properties of the SCARED in youth with autism spectrum disorder. Research in Autism Spectrum Disorders, 8, 1225–1234.CrossRefPubMedPubMedCentralGoogle Scholar
  40. Sukhodolsky, D. G., Bloch, M. H., Panza, K. E., & Reichow, B. (2013). Cognitive–behavioral therapy for anxiety in children with high-functioning autism: A meta-analysis. Pediatrics, 132(5), e1341–e1350. doi: 10.1542/peds.2013-1193.CrossRefPubMedPubMedCentralGoogle Scholar
  41. Ung, D., Selles, R., Small, B. J., & Storch, E. A. (2015). A systematic review and meta-analysis of cognitive–behavioral therapy for anxiety in youth with high-functioning autism spectrum disorders. Child Psychiatry and Human Development, 46(4), 533–547. doi: 10.1007/s10578-014-0494-y.CrossRefPubMedGoogle Scholar
  42. Verbeke, G., Molenberghs, G., & Rizopoulos, D. (2010). Random effects models for longitudinal data. In K. von Montfort, J. H. L. Oud, & A. Satorra (Eds.), Longitudinal research with latent variables (pp. 37–96). New York: Springer.CrossRefGoogle Scholar
  43. Walkup, J. T. (2008). Cognitive behavioral therapy, sertraline, or a combination in childhood anxiety. New England Journal of Medicine, 359(26), 2753–2766.CrossRefPubMedPubMedCentralGoogle Scholar
  44. Weschler, D. (2002). Wechsler Abbreviated Scales of Intelligence (WASI). San Antonio, TX: Psychological Corporation.Google Scholar
  45. White, S. W., Ollendick, T., Albano, A. M., Oswald, D., Johnson, C., Southam-Gerow, M. A., et al. (2013). Randomized controlled trial: Multimodal anxiety and social skill intervention for adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 43(2), 382–394. doi: 10.1007/s10803-012-1577-x.CrossRefPubMedPubMedCentralGoogle Scholar
  46. White, S. W., Schry, A. R., & Maddox, B. B. (2012). Brief report: The assessment of anxiety in high-functioning adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 42(6), 1138–1145. doi: 10.1007/s10803-011-1353-3.CrossRefPubMedGoogle Scholar
  47. White, S. W., Schry, A. R., Miyazaki, Y., Ollendick, T. H., & Scahill, L. (2015). Effects of verbal ability and severity of autism on anxiety in adolescents with ASD: One-year follow-up after cognitive behavioral therapy. Journal of Clinical Child and Adolescent Psychology, 44(5), 839–845. doi: 10.1080/15374416.2014.893515.CrossRefPubMedGoogle Scholar
  48. Wigham, S., Rodgers, J., South, M., McConachie, H., & Freeston, M. (2015). The interplay between sensory processing abnormalities, intolerance of uncertainty, anxiety and restricted and repetitive behaviours in autism spectrum disorder. Journal of Autism and Developmental Disorders, 45(4), 943–952. doi: 10.1007/s10803-014-2248-x.CrossRefPubMedGoogle Scholar
  49. Wood, J. J., Drahota, A., Sze, K., Har, K., Chiu, A., & Langer, D. A. (2009). Cognitive behavioral therapy for anxiety in children with autism spectrum disorders: A randomized, controlled trial. Journal of Child Psychology and Psychiatry, 50(3), 224–234. doi: 10.1111/j.1469-7610.2008.01948.x.CrossRefPubMedPubMedCentralGoogle Scholar
  50. Wood, J. J., Ehrenreich-May, J., Alessandri, M., Fujii, C., Renno, P., Laugeson, E., et al. (2015). Cognitive behavioral therapy for early adolescents with autism spectrum disorders and clinical anxiety: A randomized, controlled trial. Behavior Therapy, 46, 7–19.CrossRefPubMedGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Amy Keefer
    • 1
    • 2
  • Nicole L. Kreiser
    • 1
    • 2
  • Vini Singh
    • 1
  • Audrey Blakeley-Smith
    • 3
  • Amie Duncan
    • 4
  • Catherine Johnson
    • 1
  • Laura Klinger
    • 5
  • Allison Meyer
    • 6
  • Judy Reaven
    • 3
  • Roma A. Vasa
    • 1
    • 2
  1. 1.Center for Autism and Related DisorderKennedy Krieger InstituteBaltimoreUSA
  2. 2.Department of Psychiatry and Behavioral SciencesJohns Hopkins University School of MedicineBaltimoreUSA
  3. 3.University of Colorado School of MedicineAuroraUSA
  4. 4.Cincinnati Children’s Hospital Medical CenterCincinnatiUSA
  5. 5.TEACCH Autism Program, Department of PsychiatryUniversity of North Carolina at Chapel HillChapel HillUSA
  6. 6.TEACCH Autism Program, Department of PsychologyUniversity of North Carolina at Chapel HillChapel HillUSA

Personalised recommendations