Keywords

Autism spectrum disorder; anxiety; emotion; mindfulness; acceptance; adult

Introduction

Although Autism Spectrum Disorder (ASD) is most often identified during early childhood, it is becoming increasingly apparent that recognition can be delayed and that initial diagnosis of ASD can occur during adulthood. As Lai and Baron-Cohen (2015) proposed, adult identification may be due to an increasing public awareness of ASD as well as a broadening of the diagnostic criteria to include higher functioning individuals who would likely not be seen as having sufficiently severe symptoms during childhood to warrant a diagnosis. Regardless of the underlying causes for the increased, and still increasing, number of adults with ASD and heightened recognition of the disorder in adulthood, it is abundantly clear that more research is needed to inform effective treatment of co-occurring mental health conditions in adults with ASD, including anxiety disorders. In this chapter, we describe the presentation and prevalence of anxiety disorders in adults with ASD and evidence-informed approaches to assess and treat anxiety in this population.

Prevalence and Clinical Presentation

It is only in recent years that researchers have begun to investigate adult outcomes among people diagnosed with ASD as children. Largely, the extant research has focused on stability of the ASD diagnosis and severity of ASD over time (Louwerse et al., 2015; Woolfenden et al., 2012) and quality of life in adulthood and variables associated with adult outcomes (Howlin, 2000). Recently, however, data on rates of psychiatric comorbidity in adults with ASD have been reported (e.g., Lai and Baron-Cohen, 2015). Although this research is fairly young and epidemiological research is very limited, it is apparent that adults diagnosed with ASD have more co-occurring medical and mental health diagnoses than do adults without ASD. This finding has been reported in international samples and when psychopathology is viewed diagnostically as well as dimensionally.

For example, a large, case controlled study of health records has shown that rates of all the major mental health disorders, including anxiety disorders, are elevated among adults with ASD relative to age- and sex-matched nonASD adults (Croen et al., 2015). In addition, Bruggink and colleagues (2016) found that adults with ASD self-reported significantly more symptoms anxiety and depression, compared to an age- and gender-matched control sample of adults and Croen and colleagues (2015) found that more than half of their sample (n=1507) of adults with ASD were diagnosed with at least one additional mental health condition. Anxiety disorders were more commonly diagnosed than any other mental health diagnosis. Almost one-third (29%) of the sample had anxiety disorder diagnoses (not including obsessive-compulsive disorder [OCD], which was an additional 8%). By comparison, among the age- and sex-matched controls (10 controls for every proband), 9% had an anxiety disorder. Hofvander and colleagues (2009) found that anxiety disorders were second only to mood disorders in terms of prevalence. Half of their clinical sample of 122 adults met criteria for at least one anxiety disorder, with generalized anxiety disorder (GAD) being the most frequently diagnosed anxiety disorder (15%), followed closely by social anxiety disorder (SAD; 13%).

In a follow-up study with adults who were diagnosed with ASD as children and who had at least average cognitive ability, 56% of the sample (n=58, mean age=44 years) experienced mental health problems in adulthood (Moss et al., 2015). Although specific disorders were not diagnosed, 10% of the sample surpassed the clinical cut-off on a self-reported anxiety scale (Beck Anxiety Inventory: BAI; Beck and Steer, 1990) and 29% surpassed the cut-off for OCD (based on Y-BOCS (Goodman et al., 1989) with items added from the Children’s Yale-Brown Obsessive-Compulsive Scale-Pervasive Developmental Disorders: CY-BOCS-PDD; Scahill et al., 2006).

There is a slowly emerging picture of the onset and course of co-occurring anxiety problems in ASD among adults. Although problems with anxiety can emerge during adulthood, for the majority of adults with ASD, it is more likely that anxiety onset is during childhood. In a retrospective study of adults diagnosed with ASD without intellectual disability, Maddox and White (2015) found that, among adults with ASD and SAD, onset of impairing social anxiety was most often during middle school.

Cognitive and verbal ability may be predictive of problems with anxiety in this population. Gotham and colleagues (2015) found that verbal IQ was positively associated with anxiety among adults with ASD. ASD severity, gauged by self-report via the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001), has also been found to be positively associated with social anxiety (Bejerot et al., 2014). There is evidence that anxiety is more problematic among women than men with ASD (Croen et al., 2015) and that anxiety symptoms worsen more sharply during adolescence and early adulthood for females with ASD, compared to males with ASD (Gotham et al., 2015). This gender difference has not been found consistently. Lugnegård and colleagues (2011) found no gender differences with respect to rates of psychiatric comorbidity in their high-functioning adult sample. The potentially moderating role of gender is consistent with prior research showing that in childhood, girls are at greater risk for anxiety and mood problems than boys with ASD (May et al., 2013; Solomon et al., 2012), but divergent from the comparable rates of anxiety disorders across the genders seen in treatment seeking children with anxiety who do not have ASD (Kendall et al., 2010).

In one of the only qualitative studies on the phenomenology of anxiety in ASD during adulthood, Trembath and colleagues (2012) conducted focus groups with 11 diagnosed adults. They found the primary sources of anxiety identified by these individuals to be related to the environment (e.g., crowds), social interaction, concern for others (e.g., broad societal issues), fearful anticipation, and disappointment. These anxiety triggers are similar to the findings of Gillott and Standen (2007) who, using a quantitative approach, found that anxiety was associated with having to cope with change, anticipation, personal contact, and sensory stimulation among adults with ASD.

Based on this fairly limited research base on anxiety in adults with ASD and extrapolating from the more mature research base with children and adolescents, we can surmise that anxiety, of all types, is common among adults with ASD. Adults with ASD experience more anxiety, based on self- and other-report, than do neurotypical adults and adults with other types of developmental disabilities. Additionally, anxiety may be more problematic, or at least more often recognized, among adults who do not have co-occurring intellectual disability. However, this conclusion is tentative as the majority of prior studies have been with higher functioning, cognitive unimpaired samples. It is possible that reliance on verbalization of internal symptoms and self-report diminishes our ability to accurately assess for anxiety in adults with ASD who are less verbal or cognitively impaired, thus inflating the actual effect of cognitive ability. The limited research with adult samples with ASD and intellectual disability, however, suggests that anxiety is more problematic for those who have co-occurring ASD than for adults with intellectual disability without ASD (Cervantes and Matson, 2015).

Evidence-Informed Assessment of Anxiety in Adults With ASD

Historically, the degree to which any anxiety disorder can truly be “comorbid” alongside ASD has been unclear. Specifically, given the high prevalence of anxiety in the context of ASD it has been debated whether anxiety could be diagnosed as a comorbid disorder, or whether anxiety was a phenomenological manifestation, more core to the ASD itself (Kerns and Kendall, 2012). Most experts now agree that anxiety disorders, as clinical conditions distinct from ASD in terms of etiology and treatment, can and do frequently co-occur with ASD and, as such, can be separable diagnosable disorders (e.g., Lai and Baron-Cohen, 2015). Although the mechanisms underlying co-occurrence of anxiety in ASD are not fully understood, Lai and Baron-Cohen (2015) suggest that two probable pathways are shared causative factors (e.g., amygdala hyper-reactivity) and anxiety resulting, directly or indirectly, from the ASD itself (e.g., social alienation, need for sameness).

How anxiety is assessed (e.g., observational tools, self-reports) and conceptualized (e.g., as a separable condition that is the same as anxiety in neurotypical individuals or something related to ASD) greatly affect how the manifest symptoms are “counted” or categorized diagnostically. Accurate diagnostic assessment of apparent anxiety problems in adults with ASD is often challenging. Primary reasons for this challenge include limited insight into and reliance on verbal reporting for internal symptoms of anxiety (e.g., worries, self-doubt, fear of negative evaluation), which affect self-reporting. Alexithymia, or the inability to identify and label one’s own emotions, as well as impaired attention and executive function (e.g., inability to step back and contemplate what one is fearful about, outside of the moment or after the situation has passed) also contribute to this diagnostic quandary.

The research base on anxiety, and its treatment, in adults with ASD has primarily relied on measures developed for evaluating anxiety in typical, nonASD adults. Self-report measures, such as the BAI (Beck and Steer, 1990) and the Symptom Check List (SCL-90; Derogatis, 1977), have been used (e.g., Bruggink et al., 2016; Moss et al., 2015). Researchers (e.g., Maddox and White, 2015) have also used semi-structured diagnostic interviews, such as the Anxiety Disorders Interview Schedule for DSM-IV (ADIS-IV-C/P; Brown et al., 1994), with cognitively unimpaired adults. With respect to measurement of specific types of anxiety disorder symptoms, arguably the most research has occurred in the area of social anxiety. The Social Anxiety Scale (SAS; La Grecaand Lopez, 1998) has been used as both self- and parent-report (e.g., Swain et al., 2015). There has been very little development of tools for assessing anxiety specifically in adults with ASD. The Social Anxiety Scale for People with ASD (SASPA; Kreiser and White, 2011) is one exception. The SASPA is a self-report measure of social anxiety symptoms that are not confounded by core ASD symptoms, such as behavioral avoidance owing to lack of social interest. Unfortunately, there is very little rigorous evaluation of the sensitivity or validity of most of these measures, including the SASPA. Additionally, there is a dearth of measures that are appropriate for adults yet available as both caregiver- and self-report, which poses a considerable challenge for service providers attempting to assess adults with ASD who may also have cognitive limitations or be functionally delayed (e.g., living with parents, no employment). In this situation, the evaluator typically must decide between using a measure for children or adolescents, that might better lend itself to parental report (e.g., rely on behavioral indicators), or asking the caregiver to try to intuit about symptoms that are difficult for any third party to ascertain in a valid way (e.g., feelings of despair, negative interpretations).

Clinically, it can be helpful to set up situations, or presses, that are likely to trigger the symptoms of interest in order to assess them “in the moment.” For instance, the clinician can observe how the client responds when a new student introduces herself (social stress) or when there is an unanticipated event, such as changing to a different, unfamiliar assessment room. Sequencing symptoms, and associated impairment, is also helpful. Although impairments in socialization and interpersonal communication must be present (even if not identified as such) early in childhood for diagnosis of ASD, new symptoms in this domain or their worsening during late adolescence or in adulthood may be suggestive of anxiety in addition to the ASD. Caution must be taken, however, as core and secondary symptoms of ASD often change over the course of development. For example, whereas behavioral avoidance of social situations tends to decline during adolescence among typically developing youth, behavioral avoidance as well as fear of negative evaluation has been found to be higher among adolescents with ASD relative to children (under 12 years) with ASD, based on cross-sectional data (Kuusikko et al., 2008). As such, a carefully illustrated timeline and an informed understanding of the heterotypic continuity within ASD can inform the clinician’s differential diagnosis, as well as help the client with reporting on events and symptom onset.

The practice demand for empirically supported tools for the screening and diagnosis of anxiety disorders in people with ASD is great. Indeed, many adults are referred for additional diagnostic evaluation due to problems with mood and anxiety (e.g., Hofvander et al., 2009). Evidence-based assessment of co-occurring conditions, including anxiety disorders, in adult ASD warrants further study. As this research base develops, given the substantial phenotypic overlap between many of the anxiety disorders and ASD symptoms, we suggest a nuanced assessment that goes beyond symptom count, duration, distress, and interference when determining if an anxiety disorder in comorbidity presents in a client with ASD.

The conceptual overlap between anxiety disorders and ASD is perhaps best exemplified by considering SAD and ASD. Social awkwardness and difficulty in social interaction, which are hallmark features of ASD, are often considered in the evaluation of possible SAD. This type of social difficulty can be due to impaired ability to infer others’ thoughts and feelings (theory of mind impairment), underdeveloped interaction skills, or awkward social behaviors (i.e., a fluency deficit). Social difficulty can also be due to anxiety and related social avoidance. In a person with ASD, all these factors may underlie the manifest social difficulty. In practice, when faced with a client who is socially avoidant and hyper-aroused in social interaction and performance situations, the clinician probes into the reasons for avoidance and variability in arousal. The client with ASD who does not have co-occurring SAD may, for instance, report feelings of discomfort when around others due to a variety of reasons (e.g., too loud, unsure of what to do or say socially), but deny any socio-evaluative fears or perhaps even thoughts about the repercussions of social missteps. If these factors are not evaluated, the SAD diagnosis may be misapplied. In this example, a consideration of the factors underlying the observed problem (e.g., social avoidance) will inform diagnosis and the treatment approach that is taken. In this case, for example, the clinician may target skill deficits primarily and general anxiety and sensory sensitivities adjunctively, but not engage in exposures targeting fear of social evaluation.

Targeting Key Mechanisms

Determination of the processes that influence and give rise to co-occurring psychopathology can allow for the targeting of specific symptoms in treatment. Likewise, ascertainment of the mechanisms that mediate clinical changes (e.g., symptom reduction) as a result of effective treatment can facilitate treatment personalization (the choice of one treatment over another for a given client), improve treatment outcome, and augment adaptations to current and future treatments (Kazdin, 2007; Norcross and Wampold, 2011; Paul, 1967). Because the understanding of how, why, and for whom these treatments work is largely unknown, despite the number of empirically based treatments for anxiety in individuals with ASD (mostly for children and adolescents), study of the etiological and treatment mechanisms is an essential next step in our growing research base.

Basic Processes Underlying Anxiety in ASD

Several commonly seen deficits among individuals with ASD could underlie anxiety symptoms. The social deficits required for a diagnosis of ASD have been postulated to contribute to increased anxiety, especially social anxiety symptoms, at least for a subset of individuals with insight into their social deficits (White et al., 2014). Additionally, social motivation, the extent to which a person is interested in interacting with others, could play a role in anxiety among individuals with ASD. Swain and colleagues (2015) found that greater social motivation was associated with higher social anxiety among 69 young adults with ASD, who participated in a social skills intervention, suggesting that the desire to interact may also increase anxiety for adults with ASD.

Alternatively, there may be core processes that contribute independently to both symptoms of ASD and symptoms of anxiety. For instance, alexithymia may contribute to or exacerbate social deficits, as well as lead to anxiety among individuals with ASD. Up to 50% of individuals with ASD have been observed to experience difficulty in identifying and describing their emotions (Lombardo et al., 2007). Difficulties in labeling others’ emotional states likely contribute to theory of mind deficits, the ability to surmise another person’s perspective and emotional state, which compromise social interaction and may lead to discomfort or anxiety (Mazefsky and Herrington, 2014). Indeed, adults with ASD have been found to misinterpret happy face stimuli as negative, potentially indicative of a bias that could lead to anxiety (Eack et al., 2015). These results are consistent with a previous meta-analysis of emotion recognition research in ASD across the lifespan, finding that individuals with ASD are significantly less able to recognize emotions other than happiness in face stimuli (Uljarevic and Hamilton, 2012). In addition, individuals with ASD have been shown to have difficulties with their own emotion expression (Trubanova, 2015), which can also diminish the ability to interact with others.

The second cluster of symptoms in the diagnostic criteria for ASD, repetitive and restricted behaviors or interests, contains several behaviors that overlap with anxiety symptomatology. Symptoms that may be present for a diagnosis of ASD include resistance to changes in routine and in one’s environment, as well as sensory hypersensitivity or hyposensitivity (APA, 2013). These symptoms may lead individuals to experience their world as volatile and overwhelming, thus contributing to fear and anxiety (Mazefsky and Herrington, 2014). Similarly, intense focus or perseveration on specific topics of interest, although typically thought of as relating to topics that the individual enjoys, may signify an overall emotional and cognitive style characterized by difficulty shifting one’s attention (Mazefsky and Herrington, 2014; White et al., 2014). This cognitive “stickiness” can easily be re-construed as rumination if focused upon negative thoughts or emotions, and has been hypothesized to account for increased negative affect and emotional lability for individuals with ASD (Keehn et al., 2013).

Emotion Regulation

Emotion regulation (ER) can be defined as one’s attempts to monitor and modulate their emotional experience (Gross and Thompson, 2007). Traditionally, research has categorized ER strategies as typically adaptive or maladaptive, although whether ER is adaptive for the individual is dependent upon context (Aldao and Nolen-Hoeksema, 2012). Among typically developing individuals, ER impairment or over-reliance on maladaptive ER strategies has been posited to be associated with many forms of psychopathology (Aldao et al., 2010; Gross, 2002). Similarly, ER deficits have been studied as a potential underlying factor for many of the associated behavioral problems seen in ASD, such as aggression, irritability, and anxiety (Mazefsky and White, 2014; Mazefsky et al., 2013; Samson et al., 2015; Weiss, 2014; White et al., 2014). White and colleagues (2014) proposed a model of how various social-cognitive, neurological, and physiological mechanisms contribute to ER impairments, thus leading to heightened levels of anxiety among individuals with ASD. They posited, based on extant research on typically developing individuals, that ER difficulties in ASD arise in part from ASD-associated symptomatology, such as challenges with emotion recognition and expression, difficulty with attentional processes such as shifting one’s attention, neural hyper- and hypoconnectivity, physiological overarousal, and avoidance (White et al., 2014). Further empirical research is needed to better elucidate the relationship of maladaptive ER and ASD.

In terms of intervention, targeting maladaptive ER has also been posited as a means to address the underlying factors that contribute to anxiety disorders (Weiss, 2014). As transdiagnostic treatments, such as Barlow and colleagues’ Unified Protocol (Barlow et al., 2004), have been used for the treatment of emotional disorders in individuals without ASD, such an approach has been suggested for ER difficulties, as they may underlie not only anxiety, but other common co-occurring difficulties in ASD such as irritability and depressive symptoms (Weiss, 2014).

Mindfulness and Acceptance

Another potential mechanism for anxiety treatment is mindfulness and acceptance. As previously mentioned, adults with ASD often seem to demonstrate lack of insight into their and others’ emotional states (Lombardo et al., 2007). Mindful awareness practices often ask individuals to attend to their thoughts, bodily sensations, and breathing, which can serve to improve one’s ability to notice their own emotions (Segal et al., 2002). Similarly, cognitive inflexibility, which is often observed among adults with ASD (Eack et al., 2013), can also be facilitated by mindfulness/acceptance-based treatments, as the focus upon accepting negative thoughts and emotions when they occur can be applied towards rigidity, or the difficulties that arise from routines being broken (Lee and Orsillo, 2014).

A treatment study of adapted mindfulness-based cognitive therapy (MBCT) for adults with ASD looked at rumination as a potential mechanism of treatment for anxiety and depressive symptoms (Spek et al., 2013), and found that the treatment did decrease affective symptoms partially via reduction of rumination. Given the chronicity of ASD and the proposed relationship between ASD and anxiety (White et al., 2014), utilizing mindfulness- and acceptance-based interventions (MABIs) may be especially useful to target anxiety in this population. MABIs focus upon redirecting one’s attention and one’s relationship to negative or maladaptive thoughts and emotions such as worries and anxiety, rather than the traditional cognitive-behavioral therapy (CBT) focus upon identifying and changing maladaptive cognitions.

Behavioral Treatments

Extant Research in Adults With ASD

Although the majority of work in this field has focused on children and adolescents with ASD (see Chapter 8: Group Cognitive Behavior Therapy for Children and Adolescents with Anxiety and Autism Spectrum Disorders and this chapter), an emerging body of research suggests that cognitive-behavioral interventions—including CBT and mindfulness-based techniques—can effectively reduce anxiety in adults on the autism spectrum (Kiep et al., 2015; Spain et al., 2015). Cardaciotto and Herbert (2004) reported a case study of treating anxiety in an adult with ASD. A 23-year-old male with Asperger’s Disorder and SAD demonstrated a consistent decrease in symptoms of social anxiety and comorbid depression, as determined by self-report measures, throughout the 14-week individual CBT program. At the 2-month follow-up assessment, the young man no longer met diagnostic criteria for SAD.

Weiss and Lunsky (2010) conducted a case series of group-based, manualized CBT for three adults with Asperger’s and co-occurring anxiety or mood disorders. One male participant in his mid-50s had a diagnosis of panic disorder with agoraphobia, along with impairing symptoms of SAD. Although his self-report of anxiety symptoms did not significantly decrease across the 12 weekly sessions, he showed notable gains behaviorally in the group setting (e.g., less flushed while speaking, reduced tremble in his voice, increased eye contact with others). The other two group members (one male with depression and one female with depression and post-traumatic stress disorder) showed a decrease in self-reported anxiety symptoms over the course of treatment. Another study of a group-based CBT program with 32 older adolescents and young adults with ASD (23 males; age 15–25; mean age=20.6 years) found significant reductions in self-reported depression and stress, but not anxiety, for the treatment group, relative to the waitlist control group (McGillivray and Evert, 2014). The improvements were maintained at 3- and 9-month follow-up assessments.

Promising results have also been reported regarding the treatment of adults with ASD and co-occurring OCD (Russell et al., 2009, 2013). The first published study (Russell et al., 2009) was a nonrandomized trial comparing CBT for OCD to treatment as usual (TAU) in 24 adults with ASD and co-occurring OCD (21 males; mean age=28 years). All participants had average cognitive abilities. OCD symptoms, as measured by the Y-BOCS (Goodman et al., 1989) total severity score, significantly decreased pre- to post-treatment for the CBT group (d=1.01), but not the TAU group. However, in 50% of the CBT cases, the Y-BOCS was completed by the treating therapist, which is a potential confound. In addition, the CBT group had more severe OCD symptoms at baseline, relative to the TAU group. To improve upon this initial pilot study, Russell and colleagues (2013) conducted a RCT of CBT for OCD with 46 cognitively unimpaired adolescents and adults with ASD (35 males; mean age=27 years). The control group received an equivalent number of sessions of anxiety management (AM), which included psychoeducation about anxiety and general anxiety reduction techniques (e.g., diaphragmatic breathing, progressive muscle relaxation, problem solving training). The AM manual did not include any of the active ingredients of CBT (e.g., exposure–response prevention) for OCD or any cognitive techniques addressing OCD-related beliefs. Study assessors were blind to treatment assignment. Both groups demonstrated a significant reduction in OCD symptoms, based on Y-BOCS total severity score (within-group effect sizes of 1.01 for the CBT group and 0.6 for the AM group). The groups did not significantly differ in OCD symptoms at post-treatment, although the CBT group had more responders (45% versus 20%, with treatment response defined as >25% reduction in Y-BOCS total severity score). In addition, the CBT group (but not the AM group) completed a one-year follow-up assessment, and the treatment gains were sustained over this period (Russell et al., 2013).

In a RCT conducted by Spek et al. (2013), the results demonstrated that adults with ASD and co-occurring anxiety can benefit from modified mindfulness-based therapy (MBT). Their treatment was based on MBCT for depression in adults without ASD (Segal et al., 2002), with the cognitive elements (e.g., examining the content of one’s thoughts) omitted, given the difficulties with information processing that are characteristic of ASD. Primary program content included mindful meditation exercises (e.g., eating, walking, breathing), along with psychoeducation about physical reactions to stress and ruminative thoughts. Forty-two cognitively unimpaired participants with ASD (27 males; mean age=42 years) were randomized to the 9-week group-based MBT program (with each session lasting 2.5 hours) or a waitlist control group. At post-treatment, the intervention group demonstrated a significant decrease in self-reported anxiety and a significant increase in self-reported positive effect, whereas the control group did not. Kiep and colleagues (2015) examined the same MBT protocol in 50 cognitively unimpaired adults with ASD (34 males; mean age=40 years) and found similar results. The reduction in anxiety symptoms remained stable at a 9-week follow-up.

In addition to preliminary evidence for treatment efficacy for traditional CBT and MBT approaches, these studies document treatment feasibility and acceptability with adults who have ASD (e.g., high attendance rate, high homework completion rate). The rule-bound, analytical style of thinking that is characteristic of people with ASD may be particularly well-suited to a CBT approach (Gaus, 2011). Case reports of adults receiving CBT suggest that the clients liked the structure and predictability of the treatment sessions, along with the scientific aspects of the CBT model (Weiss and Lunsky, 2010). However, clinicians can also face significant challenges when conducting CBT with adults who have ASD (Kerns et al., 2016). For example, parental involvement is a key component of effective CBT for children with ASD and anxiety, but including parents in treatment of adults may interfere with the adult client’s independence or may not be possible due to practical considerations. Without a parent or other support person to help with treatment engagement and the generalization of skills, some adult clients may struggle to complete in-session or between-session therapeutic activities due to limited inherent motivation or executive functioning difficulties.

More research on behavioral treatments for anxiety in adults with ASD is clearly needed. As highlighted here, only a handful of fairly small n studies have addressed this common comorbidity, with most lacking a rigorous design. In addition, these studies have focused exclusively on cognitively unimpaired adults with ASD, leaving important questions about the appropriateness of cognitive-behavioral approaches for adults with intellectual disability and anxiety unaddressed.

Primary Modifications to Behavioral Treatments for Adults With ASD and Anxiety

Research suggests that behavioral treatments demonstrated to be effective with clients who do not have ASD are applicable to people with ASD, with some adjustments. Adapting the structure, content, and process of cognitive-behavioral approaches may be important due to differences in learning styles between adults with and without ASD, core ASD symptoms such as social communication impairments that affect therapeutic rapport, and common difficulties with emotion recognition and executive function (Gaus, 2007, 2011). These modifications to improve engagement, acceptability, and utility of CBT have mostly been evaluated in anxiety treatment studies for youth with ASD (Lang et al., 2010).

Several studies for adults with ASD have described adaptations to CBT or mindfulness techniques, and these adaptations often include an increased number of sessions with more frequent practice of cognitive restructuring and exposure exercises, emphasis on social skills training, additional information using concrete examples to enhance the client’s understanding of emotions, inclusion of visual aids and written materials, avoidance of colloquialisms or metaphors, reliance on a more directive (rather than Socratic) therapeutic style, and incorporation of the client’s circumscribed interests (Ekman and Hiltunen, 2015; Spain et al., 2015). In the two published studies of MBT for anxiety in adults with ASD (Kiep et al., 2015; Spek et al., 2013), the cognitive elements (e.g., exercises examining the content of one’s thoughts) were omitted, the original program length was increased from 8 weekly sessions to 9 weekly sessions, and the original three-minute breathing exercise was extended to five minutes.

Medical/Pharmacological Treatments

Primary Modifications to Medical Treatments of Anxiety in ASD

No guidelines for medication usage exist for treating anxiety in adults with ASD, although clinical case studies and reviews of extant literature provide some instruction. Handen and colleagues (2011) recommend using lower doses than typically given, as adults with ASD often seem to have stronger negative reactions to medications. In addition, they suggest slower dosage titration due to these difficulties (Handen et al., 2011). Propper and Orlik (2014) recommend combined treatments (both psychological and pharmacological) for anxiety or repetitive behaviors in this population. Similar to typically developing individuals with anxiety disorders, pharmacological treatment of anxiety is recommended when comorbid conditions are present, in cases of treatment resistant symptoms, or when treating severe psychopathology that prevents psychological interventions (Soorya et al., 2008). Furthermore, careful monitoring of the medication’s effectiveness in treating anxiety symptoms and potential side effects is vital (Propper and Orlik, 2014).

Looking Forward: Next Steps in Research and Practice

The extant research suggests that anxiety is endemic in ASD. Anxiety is a common experience for adults who have ASD, and often the impetus for treatment referral. Within-person factors such as sex, cognitive/verbal ability, and insight may moderate the experience and severity of anxiety, though more research on factors that affect risk is needed. The adverse impact of anxiety disorders on psychosocial functioning and quality of life in adult samples without ASD has been well-documented (Mendlowicz and Stein, 2000; Olatunji et al., 2007). Although anxiety has not yet been explored as a mechanism that is predictive of, or causally related to, the adverse outcomes often documented among even cognitively more able adults with ASD (Howlin et al., 2004; Taylor and Seltzer, 2011), it is reasonable to expect that untreated anxiety adversely affects quality of life and other outcomes. For instance, anxiety can diminish social relationships, limit willingness to take on new challenges or risks, and decrease enjoyment in daily activities.

Our empirical understanding of the experience of anxiety in adults with ASD has been informed by correlational and cross-sectional research. To move the field forward, longitudinal research is needed, as well as treatment research that is sufficiently powered to detect moderators and identify mechanisms of action that underlie change in observed symptoms.