Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: J Clin Child Adolesc Psychol. 2020 Jun 8;51(2):219–229. doi: 10.1080/15374416.2020.1759075

Accommodation of anxiety in youth with autism spectrum disorder: Results from the TAASD study

Hannah E Frank a,*, Elana R Kagan a, Eric A Storch b, Jeffrey J Wood c, Connor Kerns d, Adam B Lewin e, Brent J Small e, Philip C Kendall a
PMCID: PMC7722072  NIHMSID: NIHMS1586424  PMID: 32511015

Abstract

Objective

Accommodation, or the ways in which families modify their routines and expectations in response to a child’s anxiety, is common and interferes with anxiety treatment outcomes. However, little research has examined family accommodation among youth with autism spectrum disorder and anxiety. The current study aimed to (a) identify pre-treatment correlates of accommodation, (b) examine changes in accommodation after treatment, and (c) assess relationships between accommodation and post-treatment anxiety severity.

Method

The sample consisted of 167 youth (mean age=9.90 years; 79.6% male; 18% Latinx) with clinically significant anxiety and a diagnosis of autism spectrum disorder who were enrolled in a randomized clinical trial comparing two cognitive behavioral therapy interventions for anxiety and treatment as usual. Participants were evaluated for symptom severity and family accommodation at pre- and post-treatment.

Results

Results indicated that clinician-rated anxiety severity and parent-rated externalizing behaviors and autism spectrum disorder severity significantly predicted pre-treatment accommodation. Accommodation significantly decreased from pre- to post-treatment and non-responders showed significantly higher accommodation at post-treatment compared to responders. Finally, youth with higher pre-treatment accommodation had higher post-treatment anxiety.

Conclusions

Findings indicate that accommodation for anxiety is common among youth with autism spectrum disorder and anxiety. Furthermore, accommodation is implicated in treatment outcomes and should be targeted in treatment for youth with autism spectrum disorder and anxiety.

Keywords: Anxiety, Autism spectrum disorder, Family accommodation, Cognitive-behavioral therapy

Anxiety disorders are common among youth, including youth with autism spectrum disorder (ASD). Approximately 20–40% of youth with ASD also meet criteria for one or more anxiety diagnoses (Lai et al., 2019; van Steensel, Bögels, & Perrin, 2011), and the percentage may be higher when considering distinct presentations of anxiety specific to ASD (Kerns et al., 2020). Among youth with ASD, anxiety presents in both traditional and distinct ways (Kerns et al., 2014) and is associated with a number of physical, mental, and functional impairments, including self-injury, depression, and family stress (Kerns et al., 2015). As a result, anxiety has been identified as an important focus of intervention among youth with ASD (Kerns et al., 2015; Selles, Rowa, McCabe, Purdon, & Storch, 2013; White, Oswald, Ollendick, & Scahill, 2009). Cognitive Behavioral Therapy (CBT) is well-established as an efficacious treatment of anxiety in children and adolescents, and has demonstrated benefits for anxious youth with ASD (e.g., Storch et al., 2013; see meta-analysis by Sukhodolsky, Bloch, Panza, & Reichow, 2013). Given the prevalence and impact of anxiety among youth with ASD, there is great public health utility in exploring factors that contribute to efficacious treatment outcomes in this population.

One factor associated with anxiety treatment outcomes in typically developing youth is family accommodation. Family accommodation refers to the ways in which family members change their behaviors or expectations in order to reduce a child’s distress in response to anxious stimuli. Family accommodation is ubiquitous among parents of youth with anxiety and obsessive-compulsive disorder (OCD), with 95–99% of parents endorsing some level of accommodation (Benito et al., 2015; Flessner et al., 2011). Common forms of accommodation include providing reassurance about worries, modifying of family routines, and participating in a child’s rituals (Lebowitz, Scharfstein, & Jones, 2014). Although these behaviors may be well-intentioned, they conflict with the CBT model for treating anxiety, which focuses on promoting approach rather than avoidance behaviors. Thus, it is unsurprising that accommodation is associated with higher symptom severity, greater functional impairment, and more family burden, as well as increased likelihood of a comorbid diagnosis (see Kagan, Frank, & Kendall, 2017 for review).

Although family accommodation has primarily been explored in typically developing youth, research has begun to explore accommodation in families of youth with ASD and anxiety. Research has found that accommodation of anxiety is prevalent in families of youth with comorbid anxiety and ASD, with 97.5% of parents endorsing that they engage in one or more types of accommodation at least once a week (Storch et al., 2015). Accommodation of ASD core symptoms also appears to be relatively frequent, with research suggesting that families of youth with ASD frequently accommodate restrictive and repetitive behaviors as well as anxiety (Feldman et al., 2019). Additional research suggests that accommodation may be more frequent among youth with ASD than their typically developing peers. In a study of youth with OCD with and without ASD, youth with ASD had higher rates of family accommodation of anxiety (Griffiths, Farrell, Waters, & White, 2017b). Moreover, family accommodation has been found to mediate the relationship between ASD symptom severity, as measured by the social responsiveness scale (SRS), and functional impairment among a sample of youth with OCD (Griffiths, Farrell, Waters, & White, 2017a). This finding suggests that accommodation of anxiety symptoms may exacerbate ASD-related functional impairment. In addition, this study suggests that youth with more ASD traits experience higher rates of family accommodation than typically developing anxious peers.

Given the prevalence and impact of accommodation, attention has been given to how to best target and reduce accommodation. Reduced accommodation has been associated with improved treatment outcomes for youth with OCD (Merlo, Lehmkuhl, Geffken, & Storch, 2009) and youth with anxiety (Kagan, Peterman, Carper, & Kendall, 2016). Additionally, youth with lower pre-treatment accommodation showed greater post-treatment improvement (Garcia et al., 2010), with youth with lower pre-treatment accommodation more likely to be rated as a treatment responder controlling for pre-treatment anxiety (Kagan et al., 2016). Yet the majority of this research has been done in typically developing youth, with only one study examining the impact of CBT on accommodation among youth with ASD (N=24). In this preliminary study by Storch and colleagues (2015), family accommodation was significantly reduced following treatment of anxiety, with treatment responders reporting lower post-treatment accommodation. However, additional work with a larger sample is needed to better characterize pre- and post-treatment accommodation among youth with autism and anxiety.

The current study expanded upon previous work by examining family accommodation in youth with ASD who also met a cutoff score for interfering anxiety. Accommodation and symptom severity were evaluated at baseline and after 16 weeks of treatment within the Treatment of Anxiety in Autism Spectrum Disorder (TAASD) Study (Wood et al., 2019). Youth were randomly assigned to treatment-as-usual (TAU) or to receive one of two forms of CBT for anxiety. With regard to accommodation, it was hypothesized that (a) pre-treatment parental accommodation would be associated with ASD and anxiety symptom severity, (b) parental accommodation would be reduced following treatment, (c) reduction of parental accommodation would be greater in the CBT condition that entailed increased parental involvement, and (d) treatment response would be predicted by pre-treatment accommodation.

Methods

Participants

Participants were 167 children and adolescents (age 7–13 years) and their parents enrolled in the multi-site TAASD study (see Kerns et al., 2016 for detailed methodology and rationale). The three sites included Temple University, University of California, Los Angeles, and 145 University of South Florida. Families were self-referred, or referred from physicians, educators, or other mental health providers in the community. Participants were predominantly male (n=133; 79.6%) with a mean age of 9.90 years (SD=1.78). The majority of participants identified as White (n=128, 76.6%) and non-Latinx (n=135; 80.8%). Youth were included in the TAASD study if (a) they met diagnostic criteria for ASD, as described below; (b) they demonstrated clinically significant anxiety (i.e., severity score of 14 or greater on the Pediatric Anxiety Rating Scale (PARS; Research Units on Pediatric Psychopharmacology Anxiety Study, 2003); and (C) had an intelligence quotient (IQ) of 70 or more points. Participants in the CBT conditions completed a post-treatment assessment (n=127). All families underwent informed consent procedures, and youth provided assent. Additional details are available in the primary outcome paper for this trial (Wood et al., 2019).

Procedure

Families completed a pre-treatment evaluation that included assessment of ASD and anxiety symptomology by an Independent Evaluator (IE). Parents completed questionnaires about parental accommodation and youth symptoms of ASD and anxiety. Following the assessment, participants were randomly assigned to receive 16 weeks of CBT for anxiety (n=148) or to the TAU condition (n=19) after which they selected and received their choice of CBT treatment. CBT was either Coping Cat (Kendall & Hedtke, 2006; n=72) or Behavioral Intervention for Anxiety in Children with Autism (BIACA; Wood et al., 2009; n=76). Advanced graduate students in clinical psychology provided CBT treatments under the supervision of licensed psychologists. After 16 weeks of CBT treatment or TAU, families completed questionnaires regarding accommodation and symptom severity and a post-treatment evaluation conducted by an IE to whom the treatment condition was unknown. On the basis of the parents’ report, the IE also assigned a score on the Clinical Global Impressions-Improvement Scale (CGI-I) to indicate the extent to which the youth had improved since beginning treatment.

CBT Conditions

BIACA

BIACA is a 16-week modular treatment designed to treat anxiety in youth with ASD. Each session is 90 minutes and includes 45 minutes with the child and 45 minutes with parents/caregivers. The manual includes modules focused on anxiety coping skills, hierarchy building, and in-vivo exposures, as well as on building friendships and social skills. Skills are designed to be practiced at school, in the community, and on playdates. Parents are also taught behavior management skills to address their child’s difficulties such as oppositional behavior. The use of rewards and completion of homework is emphasized throughout. Although family accommodation is not directly addressed in any module, it is consistent with the principles taught in BIACA (i.e., avoidance maintains anxiety) and could be discussed explicitly with parents as appropriate. BIACA employs a modular format guided by a treatment algorithm designed to address each child’s unique clinical needs. Decisions regarding module use were discussed on weekly supervision calls with study therapists and supervisors.

Coping Cat

Coping Cat is a manualized treatment that includes 16 weekly 60-minute sessions focused on treating anxiety. The first eight sessions focus on teaching youth skills for coping with anxiety, including recognizing anxious thoughts, feelings, and behaviors and creating a coping plan to face feared situations. The second eight sessions provide the child with the opportunity to practice newly learned skills through imaginal and in-vivo exposures in session and at home. Homework is assigned weekly and rewards are emphasized at home and in session. There are two dedicated parent sessions during which parents receive an overview of the coping skills taught to the child, receive psychoeducation about anxiety (e.g., the role of avoidance as a maintaining factor), provide input about anxiety symptoms, and ask questions. A discussion of family accommodation may occur more explicitly during the psychoeducation portion of these parent sessions if relevant to the child’s presentation and the parents’ concerns, but it is not required. All other sessions include a brief (10–15 minute) check in with parents during the session and/or parental inclusion when needed for conducting exposures.

Measures

Autism Diagnostic Observation Schedule, Second Edition

(ADOS-2; Lord et al., 2012). The ADOS-2 is a semi-structured observational assessment conducted to assess specific symptoms of ASD, including social interaction, language use, and restrictive/repetitive behaviors. The third module of the ADOS-2 was the primary screening tool used in this study and has demonstrated appropriate sensitivity (0.91) and specificity (0.84; Gotham, Risi, Pickles, & Lord, 2006). It demonstrated good internal consistency in the current sample (Cronbach’s α=0.77). IEs masked to treatment condition and trained to reliability completed the ADOS with youth at the pre-treatment assessment.

Child Behavior Checklist

(CBCL; Achenbach & Rescorla, 2001). Youth emotional/behavioral symptoms were assessed using the CBCL, a 118-item parent-measure. The CBCL has two empirically-derived broadband scales and eight syndrome scales subscales. The externalizing broadband scale, which measures behavioral problems, rule-breaking behaviors, and aggressive behaviors, was used in this study. The CBCL has demonstrated excellent psychometric properties in typically developing children (Cronbach’s α=0.62–0.92; Achenbach & Rescorla, 2001) and the externalizing subscale has demonstrated good criterion validity (α=0.92) in youth with ASD (Pandolfi, Magyar, & Dill, 2012). Reliability for the externalizing subscale in the current sample was also strong (Cronbach’s α=0.91).

Clinical Global Impressions Scale - Improvement

(CGI-I; Guy, 1976). The CGI-I is a rating of treatment response anchored by one (“very much improved) and seven (“very much worse”) and was rated by the IE at the post-treatment assessment. In keeping with previous trials of CBT for anxious youth (e.g., Walkup et al., 2008), youth who received as a CGI-I rating of one or two (much/very improved) were considered treatment responders. The CGI has demonstrated strong psychometric properties, with ratings strongly associated with clinician-rated and self-report measures of anxiety symptom severity and impairment (Zaider, Heimberg, Fresco, Schneier, & Liebowitz, 2003). Furthermore, the CGI has been used in other trials of CBT for anxious youth with and without ASD as an indicator of treatment response (Storch et al., 2013; Walkup et al., 2008).

Multidimensional Anxiety Scale for Children, Parent Version

(MASC; March, Parker, Sullivan, Stallings, & Conners, 1997). Youth anxiety symptom levels were assessed using the MASC, a 39-item, four-point Likert parent-report scale. It has four factors: physical symptoms, social anxiety, harm avoidance and separation/panic and a total scale. Only the total scale was used in this study. The MASC has demonstrated internal consistency (α=0.87) as well as predictive and discriminate validity (Villabo, Gere, Torgersen, March, & Kendall, 2012). Although some studies have shown that the MASC may not adequately capture anxiety among youth with ASD (White et al., 2015), other studies have found that the MASC may be sufficient for detecting anxiety among youth with high-functioning ASD (Kerns et al., 2020). Parents completed the MASC at pre- and post-treatment in the present study (Cronbach’s α=0.89).

Pediatric Accommodation Scale

(PAS; Benito et al. 2015). The PAS is a 14-item clinician-administered measure of family accommodation. It has three subscales that measure the frequency of accommodation (Frequency), the impact of accommodation on the parent (Parent Impact), and the impact of accommodation on the child (Child Impact). The PAS does not yield a total score. Because we were interested in examining both frequency and impact of accommodation, all three subscales were included in analyses. Of note, a critical cut off for these subscales has not been established. Items ask about the impairment from and frequency with which family members (a) reassured the child about their fears; (b) helped their child avoid anxious situations; (c) changed their family’s routine to reduce the child’s anxiety; (d) felt upset or stressed because of providing accommodation; and (e) did not provide accommodation, causing the child to become upset. The PAS has demonstrated internal consistency and inter-rater reliability (αs>0.80) as well as convergent and divergent validity. The PAS has been shown to be sensitive to CBT in anxious youth with ASD (Storch et al., 2015) and it demonstrated strong reliability in the current sample (Cronbach’s α=0.87).

PARS

(Research Units on Pediatric Pharmacology Anxiety Study Group, 2002). The PARS is a clinician-rated interview used to assess anxiety symptoms, severity, and impairment in the past week. PARS scores range from 0 to 25, with 14 used as a minimum cut-off score for clinically significant anxiety. The PARS, which has been shown to be sensitive to CBT in anxious youth with ASD (Storch et al., 2013), was administered by the IE at the pre- and post-treatment. The PARS has demonstrated strong psychometric properties in youth with ASD, including good inter-rater reliability (0.86), high test-retest reliability (0.83), and convergent validity (Storch et al., 2012). Reliability was good in the current sample (Cronbach’s α=0.74).

Social Responsiveness Scale

(SRS; Constantino, Davis, & Todd, 2003). The SRS is a parent-report measure of ASD symptomology and social deficits, rated on a 4-point Likert scale. It has demonstrated good psychometrics, including internal consistency (0.72–0.93), inter-rater reliability (0.80), and retest reliability (0.83). In addition, it has demonstrated convergent validity (0.70) with the ASD Diagnostic Interview-Revised. Parents completed the SRS at pre- and post-treatment (Cronbach’s α=0.96).

Data Analytic Plan

Bivariate correlations examined associations between the three subscales of the PAS and youth age and IQ. T-tests were used to examine sex differences in PAS scores, and an ANOVA was used to examine site differences in PAS scores. Linear regression analyses evaluated the associations of pre-treatment accommodation with pre-treatment ASD and anxiety severity for participants in all conditions. Two sets of analyses were conducted – one examining parent-report measures as predictors, and one using clinician-report measures as predictors. Mixed-factorial ANOVAs were used to test the main effects of treatment condition and time, as well as the interaction between treatment condition and time, on all three accommodation subscales. Linear regression analyses examined the relationship between pre-treatment accommodation and post-treatment anxiety severity controlling for pre-treatment anxiety.1 Finally, t-tests examined whether responder status related to post-treatment accommodation. Because analyses were conducted with all three subscales of the PAS (given our interest in examining both frequency and interference), we assessed for significance using Bonferroni-adjusted alpha levels of .017 (.05/3). All variables were normally distributed (i.e. skewness and kurtosis within acceptable range). Regarding missingness, 72.5% of cases had complete data, and approximately 5% of the sample had one or two missing data points. Missing data analyses indicated that data were likely missing at random. To manage missing values, we performed multiple imputation using the predictive mean matching algorithm (Little & Rubin, 2002; Rubin, 1996). Twenty imputed datasets were generated, and reported results are based on the imputed data. Analyses were conducted using SPSS version 24.

Results

Pre-treatment Correlates of Accommodation

Accommodation was highly prevalent at pre-treatment, with 99.4% of parents endorsing that they provided at least one type of accommodation in the past week. More specifically, 92.8% of parents endorsed that they reassured their child at least once a week, 83.7% indicated that they facilitated anxious avoidance at least once a week, and 80.7% indicated that they reduced expectations at least once a week. Youth age was significantly associated with all three subscales of the PAS, such that accommodation was more frequent and impactful among younger youth (Frequency: r= −0.29, p< .001; Parent Impact: r= −0.26, p=.001; Child Impact: r= −0.21, p=.006), so it was included as a covariate in subsequent analyses. IQ was associated with accommodation on the Child Impact subscale of the PAS (r = −0.17, p = .03), and was also included as a covariate for analyses predicting that subscale even though its p-value was above the Bonferroni-adjusted alpha of .017. There were significant site differences on all pre- and post-PAS subscales (all ps < .001), so site was also included as a covariate in regression analyses. T-tests revealed no difference between girls and boys on accommodation.

Linear regressions examined the association between pre-treatment accommodation and symptoms of anxiety, ASD, and externalizing behaviors based on IE- and parent-rated measures. Predictors for the first set of regressions (Table 1) included IE-rated measures of anxiety (PARS) and ASD severity (ADOS-2) and used age and site as control variables. As noted above, IQ was included as a control variable only for the Child Impact subscale of the PAS. ASD severity was not a significant predictor of any of the PAS subscales. However, IE-rated anxiety significantly predicted all three subscales of the PAS such that higher anxiety was related to higher frequency and impact of accommodation.

Table 1.

Clinician-rated predictors of PAS accommodation subscales at pre-treatment.

Model 1 Model 2
Variable β T-value P-value ΔR2 β T-value P-value ΔR2
Frequency
 Age −0.26 −3.67 <.001 −1.16 −3.09 .002
 Temple a −0.32 −4.06 <.001 −5.38 −3.10 .002
 USF a −0.31 −3.91 <.001 .19** −3.39 −2.10 .04 .16**
 ADOS-2 - - - 0.02 0.05 .96
 PARS - - - 9.34 6.22 <.001
Parent Impact
 Age −0.44 −3.15 .002 −0.34 −2.57 .011
 Temple a −3.09 −4.78 .00 −2.43 −3.93 <.001
 USF a −2.17 −0.61 .00 .19** −1.26 −2.19 .03 .12**
 ADOS-2 - - - 0.02 0.11 .92
 PARS - - - 2.80 5.23 <.001
Child Impact
 Age −0.48 −2.21 .029 −0.29 −1.43 .155
 IQ −0.05 −1.81 .073 −0.06 −2.42 .016
 Temple a −3.75 −3.72 <.001 −2.80 −2.96 .004
 USF a −3.61 −4.10 <.001 .18** −1.87 −2.11 .037 .14**
 ADOS-2 - - - −0.19 −0.86 .39
 PARS - - - 4.40 5.44 <.001
Open in a new tab
**

p<.001

Abbreviations. CBCL = Child Behavior Checklist; MASC = Multidimensional Anxiety Scale for Children; PAS = Pediatric Accommodation Scale; SRS = Social Responsiveness Scale

Note: statistical significance was assessed using a Bonferroni-adiusted alpha of .017

a

UCLA is reference group

The second set of regressions (Table 2) included parent-rated measures of externalizing behaviors (CBCL), anxiety (MASC), and ASD (SRS). Once again, age and site were included as control variables for all subscales, and IQ was included as a control variable for the Child Impact scale only. Higher levels of externalizing behaviors significantly predicted greater accommodation for all three PAS subscales (frequency and parent and child impact of accommodation). Higher parent-reported ASD severity was significantly associated with greater accommodation. Parent-report of anxiety symptoms was not significantly related to any PAS subscales. Finally, although they were included as control variables, it is worth noting that site, age, and IQ were significant predictors in all of the models in which they were included with the exception of the IE-rated model predicting Child Impact of accommodation, in which age was not a significant predictor.

Table 2.

Parent-rated predictors of PAS accommodation subscales at pre-treatment.

Model 1 Model 2
Variable β T-value P-value ΔR2 β T-value P-value ΔR2
Frequency
 Age −1.49 −3.67 <.001 −1.42 −3.97 <.001
 Temple a −7.60 −4.06 <.001 −7.43 −4.54 <.001
 USF a −6.46 −3.90 <.001 .19** −4.76 −3.31 .001 .22**
 CBCL Ext. - - - 11.08 4.94 <.001
 SRS - - - 6.74 3.24 .001
 MASC - - - 0.06 1.38 .17
Parent Impact
 Age −0.44 −3.15 .002 −0.44 −3.31 .001
 Temple a −3.09 −4.77 <.001 −3.04 −5.01 <.001
 USF a −2.17 −3.79 <.001 .19** −1.72 −3.21 .002 .13**
 CBCL Ext. - - - 2.79 3.35 .001
 SRS - - - 2.18 2.83 .005
 MASC - - - 0.01 0.48 .63
Child Impact
 Age −0.48 −2.21 .029 −0.41 −2.17 .032
 IQ −0.05 −1.81 .073 −0.05 −3.78 .043
 Temple a −3.75 −3.72 <.001 −3.54 −3.64 <.001
 USF a −3.61 −4.10 <.001 .18** −2.98 3.26 <.001 .14**
 CBCL Ext. - - - 3.32 2.62 .01
 SRS - - - 3.87 3.26 .001
 MASC - - - 0.03 0.93 .35
Open in a new tab
**

p<.001

Abbreviations. CBCL Ext. = Child Behavior Checklist, Externalizing Subscale; MASC = Multidimensional Anxiety Scale for Children; PAS = Pediatric Accommodation Scale; SRS = Social Responsiveness Scale

Note: statistical significance was assessed using a Bonferroni-adiusted alpha of .017

a

UCLA is reference group

Change in Accommodation following Treatment

Changes in accommodation were examined using mixed-factorial ANOVAs with a Time (pre-treatment, post-treatment) x Condition (TAU versus BIACA versus Coping Cat) design. For the Frequency subscale, there was a significant main effect of time, F(1, 164)= 109.55, p < .001, partial η2= 0.40, but not of condition (p= .36). However, there was a significant interaction between time and condition, F(2, 164)= 5.33, p = .006, partial η2 = 0.06. This indicates that there was a significantly greater change in accommodation frequency over time for BIACA and Coping Cat than for TAU (see Figure 1). For the Parent Impact scale, there was a significant main effect of time, F(1, 164)= 67.18, p < .001, partial η2= 0.29, but not a significant main effect of condition (p= .37). The time by condition interaction approached significance, F(2, 164)= 2.99, p= .053, and trended toward indicating a greater decrease in parent impact of accommodation for the CBT and BIACA conditions relative to TAU. For the Child Impact scale, there was a significant main effect of time, F(1, 164)= 77.09, p < .001, partial η2= 0.32. There was not a significant main effect of condition (p= .57) nor an interaction between time and condition (p= .18).

Figure 1.

Figure 1.

Open in a new tab

Change in accommodation over time on PAS Frequency subscale

Relation between Accommodation and Treatment Response

Linear regressions examined the relationship between pre-treatment accommodation and post-treatment anxiety (MASC, PARS), controlling for pre-treatment anxiety (MASC, PARS) among participants receiving CBT. Post-treatment anxiety on the MASC was not significantly predicted by pre-treatment accommodation on any subscale of the PAS. However, post-treatment anxiety on the PARS was significantly predicted by pre-treatment accommodation Frequency, β= 0.03, p< .001, Parent Impact, β= 0.06, p= .001, and Child Impact, β= 0.05, p< .001. In addition, t-tests indicated that treatment responders as rated by the IE on the CGI-I reported lower accommodation at post-treatment, Frequency, t(122)= 4.55, p< .001, Parent Impact, t(17.02)= 2.55, p= .02, and Child Impact, t(122)= 3.87, p< .001. Using the Bonferroni-adjusted alpha of .017, the Parent Impact scale only trended toward significance.

Discussion

Consistent with findings from previous studies in both typically developing youth and youth with ASD, pre-treatment accommodation was common, impairing, and associated with IE-rated anxiety severity. Accommodation was also found to be associated with externalizing symptoms and parent-reported ASD severity. Although the reduction in accommodation did not differ between the two CBT conditions, accommodation frequency was significantly reduced for both CBT conditions at the post-treatment assessment. Also of interest, pre-treatment accommodation predicted treatment response. Taken together, these findings indicate that family accommodation of anxiety in youth with ASD is largely similar to that of anxiety in typically developing youth, and that CBT is efficacious in reducing both the frequency and detrimental impact of accommodation.

This study is the largest to examine accommodation in youth with ASD and anxiety. Consistent with previous research (Kagan et al. 2017; Storch et al. 2015), the accommodation Frequency scale significantly decreased following CBT for anxiety. There was a significant time by condition interaction, such that both CBT conditions had a greater decrease in accommodation frequency than the TAU condition. Of interest, this improvement was found for both CBT conditions (i.e., Coping Cat; BIACA) even though BIACA involved substantially more parent involvement. Specifically, with regard to parental involvement over the 16 weeks of treatment, Coping Cat included at least two 60-minute parent sessions and approximately ten minutes with parents during all other sessions, while BIACA included 45 minutes with parents at each of the 16 sessions. This finding suggests that relatively limited parent involvement during sessions is sufficient to lead to a significant decrease in the frequency of accommodation for the majority of families of youth with ASD and anxiety. It may be that parents simply have less need to accommodate once anxious symptoms are reduced in treatment; alternatively, findings may indicate that the emphasis on exposure and reduction of avoidance is sufficient to lead many parents to reduce accommodation without extended focus in treatment. Another potential interpretation is that a more direct and extended focus on accommodation may only be needed for some families, perhaps those with very high levels of accommodation. Further research is needed on how and to what extent therapists should specifically target accommodation in the treatment of anxiety, particularly among families of youth with ASD.

Findings indicate that higher levels of clinician-rated pre-treatment accommodation was associated with higher post-treatment anxiety and greater likelihood of being a non-responder. This suggests that starting treatment with high levels of frequency and impact of accommodation may be detrimental to the efficacy of treatment. This finding is consistent with other findings on accommodation in typically developing youth with anxiety (see Kagan, Frank, & Kendall, 2017 for further discussion of this issue). Although this relationship was clear when examining clinician report, there did not appear to be a significant relationship between parent-report of anxiety symptoms (using the MASC) and accommodation at pre-treatment or at post-treatment. This may be a result of measurement limitations, given evidence that the MASC may not capture the full and distinctive expression of anxiety in this population (White et al., 2015), especially among youth with intellectual impairment and ASD (Kerns et al., 2020). This suggests that adapted measures assessing distinct presentations of anxiety among youth with ASD might be better suited to measuring parent-reported anxiety in future work. In addition, future research is needed to examine what modifications may be made to treatment to assist families who start treatment with high levels of accommodation in maximizing treatment gains. One possibility is that in cases that involve extensive family accommodation, treatment outcomes would be improved by spending more time with parents during sessions and/or placing more emphasis on addressing accommodation specifically. However, it is important to note that the direction of the association between improvement in youth anxiety symptoms and reduction in parental accommodation has not been established among youth with anxiety disorders. There is evidence that reductions in parental accommodation temporally precede improvement of OCD symptoms (Piacentini et al., 2011), but questions of directionality remain an area of needed research and will inform how to optimize outcomes for those families with high initial rates of accommodation.

Another notable finding is that in the present study, parent-reported externalizing symptoms (CBCL-externalizing) were associated with greater frequency and impact of pre-treatment accommodation. It is possible that anxiety may frequently present as externalizing symptoms among youth with ASD, and that parents are particularly likely to accommodate anxiety in order to avoid a behavioral outburst. Some research has shown that parents of anxious children worry about disruptive behavior if they do not accommodate (Lebowitz et al., 2013). Although the relationship between externalizing disorders and accommodation has been inconsistent in typically developing children with anxiety disorders, this relationship is clear among children with OCD. There may be certain externalizing symptom profiles associated with more accommodation, which may also be more common among youth with ASD. It is plausible that externalizing symptoms, which may be perceived by parents as particularly disruptive or even dangerous, may elicit more accommodation than anxiety symptoms elicit. An alternative explanation is that parents may struggle to accurately differentiate symptoms of anxiety behavioral disorders. In our experience, this is particularly difficult in youth with ASD, who may be less likely to articulate cognitive aspects of anxiety (e.g. worry), leaving parents to infer anxiety in situations where behavioral symptoms such as conflict and defiance are present. Thus, the relationship between externalizing symptoms and accommodation of anxiety in the current study could simply be a reflection of parents’ mistaken conceptualization of externalizing symptoms as indicative of anxiety. Efforts to develop measures that aim to specifically assess accommodation of ASD, externalizing, and anxiety symptoms may facilitate future work in disentangling these relationships.

The significant reduction in accommodation for both CBT conditions is promising, but those in TAU also evidenced reductions in accommodation comparable to the CBT conditions when measured by the Parent Impact and Child Impact scales. Thus, findings on the reduction of accommodation during treatment may, in some cases, be linked to how accommodation is measured (i.e., frequency vs. impact). However, the magnitude of the reduction in accommodation impact for those in TAU was smaller than that seen in the two CBT conditions. In addition, decreases in anxiety symptoms from pre- to post-treatment for youth in the TAU condition were less than for the CBT conditions. It is important to note that TAU was an active treatment comparison condition, in which participants were allowed to seek treatment in the community over the 16-week TAU period; for some, this may have included intervention components that addressed anxiety and/or accommodation.

In partial support of hypotheses, we found an association between ASD severity and accommodation on the parent-rated SRS but not on the IE-rated ADOS-2. This may suggest that parents’ perception of ASD symptoms is a more potent predictor of increased accommodation than observer-rated assessments of ASD symptom severity. This is consistent with the findings of Griffiths et al (2017b), who found that parents of youth with ASD were more likely to provide more accommodation than parents of youth without ASD. As noted above, there were also discrepancies between parent- and IE-rated measures of anxiety, further highlighting the importance of examining these questions with multiple instruments. Given the confluence of factors that may lead to accommodation of anxiety among youth with ASD (i.e., anxiety severity, externalizing symptoms, ASD severity, comorbid disorders, parental distress), parent-report alone may not be sufficient to understand the complex clinical picture.

Future research should examine family accommodation as a potential mediator of the relationship between ASD severity and anxiety severity. Griffiths and colleagues (2017a) found that family accommodation mediated the relationship between ASD traits (as reported by parents on the SRS) and OCD symptom severity among youth who had OCD and symptoms of ASD; a similar relationship may be found in youth with comorbid anxiety and ASD. Future work should also examine the degree to which accommodation is provided in response to behaviors related specifically to ASD, anxiety, and comorbid disruptive behaviors. It is also important to determine what types of accommodations at school and at home are most appropriate for youth with ASD. Although research suggests that providing excessive accommodation to typically developing youth with anxiety disorders is counter-productive in the long-term (Kagan et al. 2017), there are not clear guidelines for the extent to which accommodations that may reduce anxiety are appropriate for youth with ASD. “Ideal” levels of accommodation of anxiety may differ for youth with ASD compared to typically developing youth. In addition, accommodation of anxiety and other behaviors may be appropriate when they function to promote approach behaviors; thus, the function of accommodation should always be considered.

Finally, results should be interpreted in the context of potential limitations. First, the sample was predominantly white and non-Hispanic/Latino. Greater racial and ethnic diversity is needed in future research to increase generalizability. In addition, this study included high-functioning youth with ASD; there is a need for further exploration of the relationships between accommodation, anxiety, and ASD severity within the full range of ASD severity. We did not directly compare youth with and without ASD, which limits our ability to form conclusions about differences in accommodation between youth with and without ASD.

Despite its limitations, this study makes important contributions to our understanding of accommodation of anxiety among youth with ASD. Results highlight the key role that accommodation plays in impacting treatment outcomes and suggests that clinicians should consider the ways in which accommodation is provided in response to internalizing, externalizing, and ASD symptoms. In many ways, accommodation of anxiety appears to function similarly among youth with and without ASD and should be addressed accordingly in treatment.

Acknowledgments

Funding: Funding for this study was provided from the US National Institutes of Health awarded to Dr. Storch (R01HD080096), Dr. Kendall (R01HD080097), Dr. Wood (R01HD080098) and Ms. Frank (F31MH11221).

Footnotes

1

All linear regression and ANOVA analyses were re-run using mixed effects models to account for nesting of participants within sites. All outcomes were reproduced, suggesting comparable outcomes using both analytic approaches.

Disclosure Statement: Dr. Kendall receives royalties from the sales of materials related to the treatment of anxiety in youth. All other authors have no conflicts of interest.

References

  1. Achenbach TM, & Rescorla LA (2001). Manual for the ASEBA School-Age Forms and Profiles. Burlington, VT: University of Vermont Research Center for Children, Youth, and Families. [Google Scholar]
  2. Benito KG, Caporino NE, Frank HE, Ramanujam K, Garcia A, Freeman J, … Storch EA (2015). Development of the pediatric accommodation scale: reliability and validity of clinician- and parent-report measures. J Anxiety Disord, 29, 14–24. doi: 10.1016/j.janxdis.2014.10.004 [DOI] [PubMed] [Google Scholar]
  3. Constantino JN, Davis SA, & Todd RD (2003). Validation of a Brief Quantitative Measure of Autistic Traits: Comparison of the Social Responsiveness Scale with the Autism Diagnostic Interview-Revised. Journal of Autism & Developmental Disorders, 33(4), 427–433. [DOI] [PubMed] [Google Scholar]
  4. Feldman I, Koller J, Lebowitz ER, Shulman C, Ben Itzchak E, & Zachor DA (2019). Family Accommodation in Autism Spectrum Disorder. Journal of Autism and Developmental Disorders, 49(9), 3602–3610. doi: 10.1007/s10803-019-04078-x [DOI] [PubMed] [Google Scholar]
  5. Flessner CA, Freeman JB, Sapyta J, Garcia A, Franklin ME, March JS, & Foa E (2011). Predictors of parental accommodation in pediatric obsessive-compulsive disorder: findings from the Pediatric Obsessive-Compulsive Disorder Treatment Study (POTS) trial. J Am Acad Child Adolesc Psychiatry, 50(7), 716–725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Garcia AM, Sapyta JJ, Moore PS, Freeman JB, Franklin ME, March JS, & Foa EB (2010). Predictors and moderators of treatment outcome in the Pediatric Obsessive Compulsive Treatment Study (POTS I). Journal of the American Academy of Child & Adolescent Psychiatry, 49(10), 1024–1033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Gotham K, Risi S, Pickles A, & Lord C (2006). The Autism Diagnostic Observation Schedule: Revised Algorithms for Improved Diagnostic Validity. Journal of Autism and Developmental Disorders, 37(4), 613. doi: 10.1007/s10803-006-0280-1 [DOI] [PubMed] [Google Scholar]
  8. Griffiths DL, Farrell LJ, Waters AM, & White SW (2017a). ASD Traits Among Youth with Obsessive–Compulsive Disorder. Child Psychiatry & Human Development, 48(6), 911–921. doi: 10.1007/s10578-017-0714-3 [DOI] [PubMed] [Google Scholar]
  9. Griffiths DL, Farrell LJ, Waters AM, & White SW (2017b). Clinical correlates of obsessive compulsive disorder and comorbid autism spectrum disorder in youth. J Obsessive Compuls Relat Disord, 14, 90–98. doi:doi.org/ 10.1016/j.jocrd.2017.06.006 [DOI] [Google Scholar]
  10. Guy W (1976). Clinical global impressions. In ECDEU Asessment Manual for Psychopharmacology (pp. 218–222). Rockville, MD: National Institute for Mental Health. [Google Scholar]
  11. Kagan ER, Frank HE, & Kendall PC (2017). Accommodation in Youth With OCD and Anxiety. Clinical Psychology: Science and Practice, 24(1), 78–98. doi: 10.1111/cpsp.12186 [DOI] [Google Scholar]
  12. Kagan ER, Peterman JS, Carper MM, & Kendall PC (2016). Accommodation and treatment of anxious youth. Depression and Anxiety, 33(9), 840–847. doi: 10.1002/da.22520 [DOI] [PubMed] [Google Scholar]
  13. Kendall PC, & Hedtke KA (2006). Cognitive-behavioral therapy for anxious children: Therapist Manual (W. Publishing; Ed. 3 ed.). Ardmore, PA. [Google Scholar]
  14. Kerns CM, Kendall P, Berry L, Souders M, Franklin M, Schultz R, … Herrington J (2014). Traditional and Atypical Presentations of Anxiety in Youth with Autism Spectrum Disorder. Journal of Autism & Developmental Disorders, 44(11), 2851–2861. doi: 10.1007/s10803-014-2141-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kerns CM, Kendall PC, Zickgraf H, Franklin ME, Miller J, & Herrington J (2015). Not to Be Overshadowed or Overlooked: Functional Impairments Associated With Comorbid Anxiety Disorders in Youth With ASD. Behav Ther, 46(1), 29–39. doi:doi.org/ 10.1016/j.beth.2014.03.005 [DOI] [PubMed] [Google Scholar]
  16. Kerns CM, Winder-Patel B, Iosif AM, Nordahl CW, Heath B, Solomon M, & Amaral DG (2020). Clinically Significant Anxiety in Children with Autism Spectrum Disorder and Varied Intellectual Functioning. Journal of Clinical Child & Adolescent Psychology, 1–16. doi: 10.1080/15374416.2019.1703712 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kerns CM, Wood JJ, Kendall PC, Renno P, Crawford EA, Mercado RJ, … Storch EA (2016). The Treatment of Anxiety in Autism Spectrum Disorder (TAASD) Study: Rationale, Design and Methods. Journal of Child and Family Studies, 25(6), 1889–1902. doi: 10.1007/s10826-016-0372-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lai M-C, Kassee C, Besney R, Bonato S, Hull L, Mandy W, … Ameis SH (2019). Prevalence of co-occurring mental health diagnoses in the autism population: a systematic review and meta-analysis. The Lancet Psychiatry, 6(10), 819–829. [DOI] [PubMed] [Google Scholar]
  19. Lebowitz ER, Scharfstein LA, & Jones J (2014). Comparing family accommodation in pediatric obsessive-compulsive disorder, anxiety disorders, and nonanxious children. Depression and Anxiety, 31(12), 1018–1025. doi: 10.1002/da.22251 [DOI] [PubMed] [Google Scholar]
  20. Lebowitz ER, Woolston J, Bar-Haim Y, Calvocoressi L, Dauser C, Warnick E, … Leckman JF (2013). Family accommodation in pediatric anxiety disorders. Depress Anxiety, 30(1), 47–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lord C, Rutter M, DiLavore PC, Risi S, Gotham K, & Bishop SL (2012). Autism diagnostic observation schedule (ADOS-2) manual (Part 1): Modules 1–4. (2nd ed.).
  22. March JS, Parker JDA, Sullivan K, Stallings P, & Conners CK (1997). The Multidimensional Anxiety Scale for Children (MASC): Factor Structure, Reliability, and Validity. Journal of the American Academy of Child & Adolescent Psychiatry, 36(4), 554–565. doi:doi.org/ 10.1097/00004583-199704000-00019 [DOI] [PubMed] [Google Scholar]
  23. Merlo LJ, Lehmkuhl HD, Geffken GR, & Storch EA (2009). Decreased family accommodation associated with improved therapy outcome in pediatric obsessive-compulsive disorder. J Consult Clin Psychol, 77(2), 355–360. doi: 10.1037/a0012652 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pandolfi V, Magyar CI, & Dill CA (2012). An initial psychometric evaluation of the CBCL 6–18 in a sample of youth with autism spectrum disorders. Research in Autism Spectrum Disorders, 6(1), 96–108. doi: 10.1016/j.rasd.2011.03.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Piacentini J, Bergman RL, Chang S, Langley A, Peris T, Wood JJ, & McCracken J (2011). Controlled comparison of family cognitive behavioral therapy and psychoeducation/relaxation training for child obsessive-compulsive disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 50(11), 1149–1161. doi: 10.1016/j.jaac.2011.08.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Research Units on Pediatric Pharmacology Anxiety Study Group. (2002). The Pediatric Anxiety Rating Scale (PARS): Development and psychometric properties. J Am Acad Child Adolesc Psychiatry, 41 1061–1069. [DOI] [PubMed] [Google Scholar]
  27. Selles RR, Rowa K, McCabe R, Purdon C, & Storch EA (2013). Prevalence and clinical correlates of treatment concerns in a sample of treatment seeking youth with obsessive–compulsive disorder. J Obsessive Compuls Relat Disord, 2(3), 286–291. doi: 10.1016/j.jocrd.2013.03.001 [DOI] [Google Scholar]
  28. Storch EA, Arnold EB, Lewin AB, Nadeau JM, Jones AM, De Nadai AS, … Murphy TK (2013). The Effect of Cognitive-Behavioral Therapy Versus Treatment as Usual for Anxiety in Children With Autism Spectrum Disorders: A Randomized, Controlled Trial. Journal of the American Academy of Child & Adolescent Psychiatry, 52(2), 132–142.e132. doi:doi.org/ 10.1016/j.jaac.2012.11.007 [DOI] [PubMed] [Google Scholar]
  29. Storch EA, Wood J, Ehrenreich-May J, Jones A, Park J, Lewin A, & Murphy T (2012). Convergent and Discriminant Validity and Reliability of the Pediatric Anxiety Rating Scale in Youth with Autism Spectrum Disorders. Journal of Autism & Developmental Disorders, 42(11), 2374–2382. doi: 10.1007/s10803-012-1489-9 [DOI] [PubMed] [Google Scholar]
  30. Storch EA, Zavrou S, Collier AB, Ung D, Arnold EB, Mutch PJ, … Murphy TK (2015). Preliminary study of family accommodation in youth with autism spectrum disorders and anxiety: Incidence, clinical correlates, and behavioral treatment response. J Anxiety Disord, 34, 94–99. doi: 10.1016/j.janxdis.2015.06.007 [DOI] [PubMed] [Google Scholar]
  31. Sukhodolsky DG, Bloch MH, Panza KE, & 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 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. van Steensel FJA, Bögels SM, & Perrin S (2011). Anxiety Disorders in Children and Adolescents with Autistic Spectrum Disorders: A Meta-Analysis. Clinical Child and Family Psychology Review, 14(3), 302. doi: 10.1007/s10567-011-0097-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Villabo M, Gere M, Torgersen S, March JS, & Kendall PC (2012). Diagnostic efficiency of the child and parent versions of the Multidimensional Anxiety Scale for Children. J Clin Child Adolesc Psychol, 41(1), 75–85. doi: 10.1080/15374416.2012.632350 [DOI] [PubMed] [Google Scholar]
  34. Walkup JT, Albano AM, Piacentini J, Birmaher B, Compton SN, Sherrill JT, … Kendall PC (2008). Cognitive behavioral therapy, sertraline, or a combination in childhood anxiety. N Engl J Med, 359(26), 2753–2766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. White SW, Lerner MD, McLeod BD, Wood JJ, Ginsburg GS, Kerns C, … Compton S (2015). Anxiety in Youth With and Without Autism Spectrum Disorder: Examination of Factorial Equivalence. Behav Ther, 46(1), 40–53. doi: 10.1016/j.beth.2014.05.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. White SW, Oswald D, Ollendick T, & Scahill L (2009). Anxiety in children and adolescents with autism spectrum disorders. Clinical Psychology Review, 29(3), 216–229. doi: 10.1016/j.cpr.2009.01.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wood JJ, Drahota A, Sze K, Har K, Chiu A, & Langer DA (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 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wood JJ, Kendall PC, Wood KS, Kerns CM, Seltzer M, Small BJ, … & Storch EA (2019). Cognitive 910 behavioral treatments for anxiety in youth with Autism Spectrum Disorder: A randomized clinical trial. JAMA Psychiatry. 10.1001/jamapsychiatry.2019.4160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zaider TI, Heimberg RG, Fresco DM, Schneier FR, & Liebowitz MR (2003). Evaluation of the Clinical Global Impression Scale among individuals with social anxiety disorder. Psychological Medicine, 33(4), 611–622. doi: 10.1017/S0033291703007414 [DOI] [PubMed] [Google Scholar]

RESOURCES