Introduction

People with autism spectrum disorders (ASD) are assumed to be especially vulnerable to developing psychiatric disorders. As in the general population, the distribution of prevalences indicates depression and anxiety as the most frequent disorders reported, while obsessive-compulsive disorder (OCD) and psychoses are less frequently reported (Bakken et al., 2010; Skokauskas and Gallagher, 2010; Helverschou et al., 2011; Steensel et al., 2011; Simonoff et al., 2012; Bakken, 2014). However, the estimates of mental illness vary considerably; some studies report estimates up to 84% (Howlin and Moss, 2012). The variation may be explained by methodological issues, such as biased samples, small sample sizes, differences in the disorders targeted and the populations’ characteristics (clinical samples vs. representative samples), differences in assessment methods used and the clinical experience of interviewers, and whether the measures have been validated for participants with autism. Futhermore, the majority of studies encompass children and adolescents, and individuals with IQ within the normal range (Matson and Cervantes, 2014).

In this chapter, we use the shorter term “autism” for ASD; likewise we use “ID only” instead of “intellectual disabilities” without additional ASD.

Autism as a risk for developing mental illness

Autism persistently disturbs the individual’s ability to understand life events and to communicate and interact with others. Communication impairments might include echolalia, repetitions of words or phrases, detailed language when explaining, idiosyncratic use of words or expressions, and literal interpretation. Autistic idiosyncrasies may cause communicational misunderstandings. For example, a woman with autism and mild ID had an idiosyncratic way of rating significant others by the size of their kidneys (Bakken et al., 2009). Autism also impacts the ability to regulate emotions in a suitable way, which may elicit negative responses from others. Emotion dysregulation is found to be significantly higher compared to typical controls (Samson et al., 2014).

The features that characterize people with autism and the problems those with autism often experience are associated with a number of problems that are considered as general risk factors when they occur in the general population (Reese et al., 2005). Thus, autism seems to imply a significant vulnerability for developing adjustment problems and mental illness (Simonoff et al., 2013).

The environment is often unpredictable and confusing for people with autism due to their autistic comprehension problems. They may experience daily activities as meaningless, have problems with interpreting the activities of others, and not know what is going to happen during the day. Each day they are confronted with challenges they are unable to cope with, and often experience catastrophic reactions and conflicts caused by comprehension difficulties, cognitive inflexibility, and difficulties in distinguishing between importance and non-importance. Three particular problems occur especially often among people with autism. They experience severe stress, are hypersensitive to sensory stimulation, and have large fluctuations in daily functioning. These problems are not part of the syndrome and not exclusively related to autism, but may nevertheless be named central autism problems. The magnitude of stress problems has, for example, raised the question: “Is autism a stress disorder?” (Morgan, 2006). These problems contribute to the daily burden experienced by those with autism and may give rise to cognitive overload (Helverschou, 2010). Cognitive overload may arise over time in people with autism when they experience stressful situations related to task solving, sensory difficulties, or demanding social settings (Hill and Frith, 2003; Colvin and Sheehan, 2014). Negative reactions to sensory stimuli, such as noise, light, and smell may exacerbate the stress of everyday life. Sensory disintegration (difficulty with processing sensory information) includes both over- and under-sensitivity (Lane et al., 2010) and occurs especially often in those with autism; prevalences up to 95% are found in children with autism (Schaaf et al., 2014). Anxiety reactions may be attributed to sensory over-responsivity (Green et al., 2012).

Confounding between autism and mental illness

The large variation in prevalences of mental illness in people with autism is probably explained mainly by the problems of delineation between autism and mental illness (Helverschou, 2010). The considerable overlap between autism and psychiatric constructs result in symptom overlap and problems in distinguishing between autism and psychiatric disorders (Helverschou et al., 2008). The clearest symptom overlap is found for autism and schizophrenia, and for autism and OCD. Lack of social interaction is both a feature of autism and a symptom of schizophrenia (Bakken and Høidal, 2013). Odd and unusual features in people with autism and idiosyncratic preoccupations can be mistaken for delusions or other positive signs of schizophrenia, and language problems, like literal comprehension, in people with autism may be confused with thought disorder (Bakken and Høidal, 2013). The ritualistic and repetitive behaviors that are among the core characteristics of autism may also be symptoms of OCD, but the compulsion-driven quality that characterizes OCD goes beyond the core features of autism (Scahill et al., 2006; Helverschou et al., 2013). Nervousness and anxiety symptoms were included in the first descriptions of autism, and despite the fact that anxiety symptoms are not included in the features that define autism, anxiety has been suggested as an integral component of the disorder (Helverschou and Martinsen, 2011). Symptoms of depression may be overlooked in people with autism due to the difficulties of observing mood changes, which are among the main symptoms of depression. Symptoms of depression may also be overshadowed by and wrongly attributed to autism, since social withdrawal, limited facial expression, and flattened affect may be indicators of both disorders. Similarly, sleeping and eating problems may be interpreted as related to autism or as symptoms of depression (Skokaukas and Gallagher, 2010; Helverschou et al., 2011).

The considerable overlap in symptoms may explain both why a complex autistic condition may be diagnosed as mental illness, and why mental illness frequently is attributed to the autism diagnosis (Helverschou et al., 2011). The large variation in prevalences, thus, indicates that mental illness is both under- and overdiagnosed in people with autism.

There are two main pitfalls. The first is to interpret features of autism as mental illness. By using measurements not adjusted for people with autism, such as instruments for the general population or for people with ID only, autism features are likely to be misinterpreted as psychiatric symptoms. This may be the case for the studies reporting the highest prevalence rates (e.g., Simonoff, et al., 2008). The second pitfall is to overlook mental illness in people with autism, which may explain the lower rates of depression and anxiety reported (e.g., Tsakanikos et al., 2006; Melville et al., 2008). Likewise, when some studies report that people with ID-only more often suffer from psychiatric disorders compared to those with both autism and ID, these reports may be explained by psychiatric symptoms in those with autism being overshadowed by autism features.

Identifying mental illness in people with autism and ID

Differentiating between autism and mental illness may entail more valid psychiatric diagnoses. A previous study demonstrated that clinicians were able to identify symptoms of four groups of mental illness that do not overlap with the core characteristics of autism (Helverschou et al., 2008). The items scored were included in a carer-completed screening checklist, the Psychopathology in Autism Checklist (PAC). The first validation study of the PAC demonstrated that it is possible to differentiate between people with autism and ID diagnosed with and without mental illness (Helverschou et al., 2009). The results suggest that co-occurring mental illness can be identified by changes or deterioration in the patterns of behavior typical of autism (Ghaziuddin, 2005; Hutton et al., 2008).

Reduced capacity for introspection and problems in communicating personal state add to the complex process of identifying mental health problems in individuals with autism and ID.

In people with both autism and ID, the diagnostic process is further complicated by the combination of the comprehension and communication difficulties related to autism and the problems in self-reporting related to ID.

At present, there is no consensus about use of criteria diagnosing mental illness in people with autism. There are few instruments (Underwood et al., 2011). The diagnostical complexity indicates that standard diagnostic manuals may not be used unmodified. Thus, identification of mental illness in people with autism and ID requires comprehensive expertise in both autism and mental illness, and requires information comparing the person’s changes in behavior and mood over time (Helverschou, 2010; Helverschou et al., 2011).

Anxiety: overlooked and undertreated?

A high prevalence of anxiety symptoms and disorders has been reported in the last decade (Helverschou, 2010; Helverschou et al., 2011). This may have contributed to increased awareness about interventions (Matson and Nebel-Schwalm, 2007; Helverschou et al., 2013; White et al., 2013). Individuals with autism seem to be especially vulnerable to developing anxiety related to problems associated with autism (Gillott and Stranden, 2007). Cognitive comprehension difficulties may lead to confusion and stress-coping difficulties. Features that characterize autism, like rituals and repetitive behavior, have been considered as related to anxiety or as strategies for coping with anxiety (Helverschou, 2010). Thus, anxiety in individuals with autism has been interpreted as an effect of having autism as well as a cause of some of the characteristics of autism.

Comparisons across autism subtypes have yielded conflicting results, but the majority of studies suggest an interaction effect between higher levels of anxiety and higher levels of cognitive capacity (MacNeil et al., 2009; White et al., 2013). However, this assumption may be due to the difficulties related to anxiety recognition in people with autism and ID (MacNeil et al., 2009). For example, in a comparison between checklist assessment and individual clinical assessment, it was found that physiological arousal may be difficult to observe in people with autism and ID (Helverschou and Martinsen, 2011). Anxiety reactions, both usual and idiosyncratic reactions, seem more easily recognized. Thus, anxiety disorders may be identified in individuals with autism and ID with the same or similar symptoms as in non-autistic individuals. To be able to identify the anxiety symptoms, further psychiatric examinations are recommended, using both checklists and systematic observations, as well as signs of general adjustment problems (Helverschou, 2010).

Prevalence of mental illness in people with ID and autism

A representative study of mental illness among adolescents and adults in Norway screened prevalence of mental illness in people with ID and autism – the “autism group,” compared to people with “ID only” (Bakken et al., 2010). Sixty-two autism and 132 ID-only participants were screened for mental illness with the PAC (Helverschou et al., 2009). The PAC encompasses 42 items: 30 items representing four diagnostic groups (psychosis, depression, anxiety, and OCD) and 12 items representing general adjustment problems – GAP (passivity, unrest, sleeping and eating problems, social avoidance, aggression, and self-harm).

The screening procedure required severe GAP scores concomitantly with above cutoff scores for one of the four mental illness categories to screen for a psychiatric disorder. General adjustment problems also occur concomitantly with symptoms of mental illness in the general population. In people with autism and ID who suffer from mental illness, such symptoms are found to be prominently present (Helverschou et al., 2009).

Severe GAP prevalence was found to be high both in the autism group – 56.5%, and ID-only group – 22.7%. The prevalence of severe adjustment problems was very high in the autism group and markedly lower in the ID-only group. Most of the individuals with severe GAP were also found to have a psychiatric disorder. The screened prevalence of the four diagnostic groups varied markedly in both the autism and the ID-only group. The relative frequency of depression was highest in the ID-only group – 87% of the participants with severe GAP scores, compared to 70% in the autism group (not statistically significant). The lowest prevalence found in both groups was for OCD; about 24% of those with autism and 17% of those without. More than a quarter of the participants with autism screened for psychosis, compared to about 9% in the ID-only group. About 64% in the autism group screened for anxiety.

Generally, those screened with one of the psychiatric diagnoses also had high scores corresponding to the other disorders. Thus, it seems that signs of all the psychiatric disorders were widespread among those who scored for at least one disorder. In the total group screened with mental illness, the differences between the psychosis and the depression scores on one hand, and the anxiety and the OCD scores on the other hand, were statistically significant.

The results from the study suggested an interaction effect between autism/ID and mental illness. Especially the prevalence of anxiety seemed to be relatively higher in the autism than in the ID-only group, though all four diagnostic groups had a substantially higher screened prevalence in the autism than in the ID-only group. The result suggested that having anxiety problems is an important characteristic of the adult autism population, and anxiety problems are probably involved in the development of psychiatric disorders in this population (Helverschou et al., 2011). It is suggested that the occurrence of psychiatric disorders in autism constitutes an intrinsic feature of an autistic liability, due to the poor association between psychiatric disorders and language skills or level of IQ (Hutton et al., 2008).

Implications for clinical practice

Mental illness in people with autism and ID is linked to the risk factors of adjustment problems, anxiety, and breaks of continuity. Adjustment problems are directly and indirectly caused by main characteristics of autism. Most basic is an inadequate understanding of the social context; for example, deficient social understanding and social skills, failure to know how to perform some ordinary tasks, and an inability to meet the expectations of others. There is an indirect link with structure dependency. ID and mental illness strengthen and add to the adjustment problems, which may increase their autistic traits in number and severity.

Diagnostic assessment

Research indicates that, even in countries where research addresses mental illness in people with autism, there is probably extensive underdiagnosis of mental illness in this group (Bakken et al., 2010; Bradley et. al., 2011). However, recent studies showing an increased prevalence indicate that more people with autism are being acknowledged as suffering from mental illness (Joshi et al., 2010; Helverschou et al., 2011).

Understanding the patients’ symptoms, colored by idiosyncratic speech and behavior, is a prerequisite for valid diagnoses. For example, to distinguish between idiosyncratic ways of expressing feelings and psychotic delusions, information about characteristics and symptoms must be collected from preschool childhood onwards (Bakken and Høidal, 2013). Information including psychometrics, interviews with formal and informal caregivers, observations, and physiological measures is crucial (MacNeil et al., 2009; Bakken and Høidal, 2013).

Confounding between autism and mental illness may cause diagnostic shadowing both ways; mental illness may be overlooked when symptoms are attributed to impairments caused by autism, and autism may be overlooked by mental illness symptoms overshadowing characteristics of autism (Geurts and Jansen, 2011; Cholemkery et al., 2014).