Abstract
Objectives:
The purpose of this study was to investigate the prevalence of depression diagnoses and related clinical data in an outpatient sample of youth with autistic disorder.
Methods:
Records of 123 psychiatrically referred children and adolescents with a Diagnostic and Statistical Manual of Mental Disorders, 4th ed., Text Revision (DSM-IV-TR) diagnosis of autistic disorder were examined. Mood disorder diagnoses and chief complaints along with family mood disorder history were the primary variables analyzed.
Results:
Four subjects (3%) presented with depressed mood. Irritability complaints were frequent (n=78, 63%). Six subjects (5%) received a mood disorder diagnosis; all with mood disorder, not otherwise specified. No subjects received a depressive disorder diagnosis. Family history of mood disorders was common.
Conclusions:
Findings raise questions about the appropriate characterization and potential misdiagnoses of depression in youth with autistic disorder.
Introduction
D
Such difficulties are especially relevant with the diagnosis of autistic disorder in which language deficits are a defining feature, and with children, as depression presents atypically independent of ASD symptoms (Birmaher et al. 1996). Published frequencies for concurrent depression in youth with autistic disorder have ranged from 0.9% to 50% (Chung et al. 1990; Kim et al. 2000; Vickerstaff et al. 2007; Simonoff et al. 2008; Mattila et al. 2010; Gjevik et al. 2011; Strang et al. 2012) with lifetime rates of 10.1–53% (Wozniak et al. 1997; Bradley and Bolton 2006; Leyfer et al. 2006; Mazefsky et al. 2010). These differing rates may relate to variations in study designs but are likely affected by diagnostic challenges.
None of these investigations involved chart reviews of diagnosed depression rates in a naturalistic clinical setting. Reporting on real world diagnostic practices has possible face validity that is of value given the lack of clear assessment standards. On the other hand, analyzing the findings in the context of associated clinical data may expose possible misdiagnoses. Only two prior studies of this type were identified. Both are limited to adults with ASDs (Morgan et al. 2003; Tsakanikos et al. 2006). The purpose of this study is to report on the prevalence of depression diagnoses and related clinical data in an outpatient setting of children and adolescents with a Diagnostic and Statistic Manual of Mental Disorders, 4th ed., Text Revision (DSM-IV-TR) (American Psychiatric Association 2000) diagnosis of autistic disorder.
Methods
Participants
Medical records for psychiatrically referred outpatients between 2004 and 2012 from an academic clinic specializing in ASD were reviewed. Patients ≤18 years of age with DSM-IV-TR autistic disorder as determined by the evaluating clinician were included. One hundred twenty-three subjects were identified. DSM-IV-TR-based depression (major depressive disorder, dysthymic disorder, bipolar disorder – most recent episode depressed, or depressive disorder - not otherwise specified [NOS]) or other mood disorder diagnoses at the time of the initial assessment were recorded. All clinicians were either board certified or board eligible child and adolescent psychiatrists.
Chief complaints, prior mood disorder diagnoses, family history, and Global Assessment of Functioning (GAF) (American Psychiatric Association 1994) score at time of evaluation were noted. For subjects with whom a GAF score was not assigned, functioning level was determined by the record examiners. Cognitive testing was not recorded, as not all subjects had available test results, and existing evaluations were performed by outside clinicians with no uniformity. Severity of the presentation was determined at the time of review by the Clinical Global Impressions - Severity Scale (CGI-S) (NIMH 1985). Charts were examined separately by two of the authors, a board certified child and adolescent psychiatrist (C.H.) and a licensed clinical psychologist (L.N.). The Institutional Review Board overseeing research at the clinic approved the study.
Chief complaints were clustered into categories of depressed mood, irritability, anxiety, attention-deficit/hyperactivity disorder (ADHD) symptoms, sleep abnormalities, developmental complaints, and “other” (catatonia, tics, anorexia, paranoia, and sensory sensitivity). Specific irritability complaints were clustered based on the Aberrant Behavior Checklist irritability subscale (Aman et al. 1985) and included chart references to aggressions, self-injurious behaviors (SIB), disruptive behaviors, irritability, and mood lability. Anxiety symptoms referred to complaints of general anxiety, obsessive-compulsive/perseverative symptoms, and stereotypies. ADHD symptoms included hyperactivity, impulsivity, and attention symptoms. Developmental complaints encompassed delays/regressions in language, social, nonverbal communication, motor skills, and play skills. Sleep abnormalities listed were either sleepiness or insomnia. Depressed mood complaints referenced only the specific complaint of depressed mood.
Mean and standard deviation were calculated when relevant. Chi-square testing was used to determine associations between mood-related chief complaints (depressed mood and irritability) and family history of depression, bipolar disorder (including bipolar disorder - NOS), or any mood disorder (depression, any bipolar disorder, mood disorder - NOS, and seasonal affective disorder).
Results
One hundred and two (83%) of the subjects were male and 21 (17%) were female, with an average age of 9.5 years (SD=4.2 years) and an age range of 2–18 years (Table 1). Thirty subjects (24%) were adolescents (ages 13–18 years). The mean CGI-S score was 4.9 (SD=0.6), indicating an overall markedly ill sample with a GAF average of 39.4, depicting major functional impairment. Psychiatric medication treatments were common, with several subjects taking either selective serotonin reuptake inhibitors (SSRIs) or mood stabilizers.
CGI, Clinical Global Impressions; GAF, Global Assessment of Functioning; NOS, not otherwise specified; SSRIs, selective serotonin reuptake inhibitors; ADHD, attention-deficit/hyperactivity disorder.
Complaints of irritability were most common (n=78, 63%), followed by anxiety (n=58, 47%), ADHD symptoms (n=53, 43%), sleep abnormalities (n=10, 8%), and developmental concerns (n=7, 6%). Four subjects (3%) had a chief complaint of depressed mood. Three of the four subjects were adolescents, with the fourth being 12 years of age.
Six subjects (5%) received a mood disorder diagnosis: All with mood disorder, NOS (Table 2). Two of these subjects had a chief complaint of depressed mood, and four had irritability, with an age range of 10–15 years. Three subjects had a prior mood disorder diagnosis, two with mood disorder NOS and one with bipolar disorder. One of these subjects received a mood disorder diagnosis. No subjects were given a depressive disorder diagnosis.
NOS, not otherwise specified; dx, diagnosis; ADHD, attention-deficit/hyperactivity disorder.
Forty-two subjects (34%) had a family history of a depressive disorder with 25 (20%) having a familial bipolar diagnosis and 61 (50%) having any family mood diagnoses. Chi square testing showed a trend association between irritability complaints and a family history of any mood disorder (χ2 3.25, p=0.07). Testing for a depressed mood complaint and family mood disorder history could not be performed because of the limited number of subjects.
Discussion
No subjects were diagnosed with a DSM-IV-TR depressive disorder, and chief complaints of depressed mood were low (3%) in our sample. A low comorbidity rate of 0.9% for major depression and of 0.5% for dysthymia was found in a pool of ASD youth using a schedule that emphasizes behavioral descriptions. The measure lessens double symptom coding and might have decreased depression diagnoses (Simonoff et al. 2008). In a study of children with a Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., Revised. (DSM III-R) diagnosis of autism, 2.9% of subjects were diagnosed with major depression (American Psychiatric Association 1987; Ghaziuddin 1992). As in our investigation, higher functioning ASD diagnoses were excluded. This may have affected the results, as those with more severe ASD symptoms may be less likely to be diagnosed with depression (Vickerstaff et al. 2007; Sterling et al. 2008), which is possibly a remnant of the diagnostic language limitations (Lainhart and Folstein 1994).
Contrasting studies have noted depression rates in the 30–50% range for youth with ASDs (Bradley et al. 2004; Vickerstaff et al. 2007; Strang et al. 2012). Two studies involved higher functioning subjects without significant intellectual impairments (Vickerstaff et al. 2007; Strang et al. 2012). The cognitive status of the subjects might have been an additional factor in the depression rate, as a higher intelligence quotient (IQ) may increase depression risk in those with autistic disorder (Sterling et al. 2008).
Of the two adult chart reviews, only one reported on concurrent depression diagnoses, noting a frequency of 6.4% in patients with autistic disorder and an IQ <70 (Tsakanikos et al. 2006). Children have a lower prevalence of depression, although adolescents may have a similar rate as adults (Birmaher et al. 1996). Limiting our sample to adolescents, 10% (3 of 30 total subjects) had a chief complaint of depressed mood, with 7% receiving a mood disorder-NOS diagnosis, findings perhaps similar to those of the adult investigation.
The high frequency of irritability complaints is consistent with a review of depression reports in autism (Lainhart and Folstein 1994). Our observed trend association between irritability and a family history of mood disorders might suggest a predisposition. A relationship between depression in autistic disorder and a family history of mood disorders has been reported (Mazefsky et al. 2010), although elevated family mood disorder rates have been noted in autistic disorder independent of depression in probands (Piven and Palmer 1999).
Taking our results at face value, a true low depression rate in children with severe ASD might reflect limitations in self-awareness of the sort that distresses those who are higher functioning. Such an effect has been noted in other clinical populations (Amador et al. 1996). It is also of note that there has been past speculation that differences in affect vulnerability might be inherent in autism (Eisenberg 1956; Fein et al. 1986). As a “disorder of affective contact” (Kanner 1943), social emotions could be centrally dysfunctional. The ability to relate to feelings such as depression might be important in the capacity for “affective” interpersonal connections (Watson and Andrews 2002). It is of relevance that in a sample of adults with autism, those with greater social dysfunction were found to have less ability to recognize sadness (as opposed to happiness, anger, and fear) in facial representations and abstract animation (Boraston et al. 2007). Elevated peripheral serotonin levels (Anderson 2002) and the atypical response to SSRIs (Doyle and McDougle 2012) may be noteworthy from this perspective.
Regardless of these considerations, our findings may indicate an under/misdiagnosis of depressive disorders. Although there is no specific standard for diagnosing depression in those with ASDs, without the use of assessment instruments the results are limited by the potential diagnostic biases of individual clinicians/treatment centers and are difficult to compare with existing investigations. This limitation affects the reliability of the autistic disorder diagnosis as well. Further, incomplete recording by clinicians may have influenced chief complaints and family history data. Many of the subjects were receiving antidepressants and mood stabilizers, which might have lessened depressive symptoms, although there are minimal published data regarding the effectiveness of pharmacologic treatment for such symptoms in those with autistic disorder (Doyle and McDougle 2012). Examining ASD subtypes would have been of interest, given questions about greater depression rates in those subjects who are higher functioning. Of note, with the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-V), such individuals may no longer meet criteria for the new ASD diagnosis (American Psychiatric Association 2013).
Conclusion
Few complaints of depressed mood, and no depressive disorder diagnoses, were observed in this chart review of youth with autistic disorder. Findings were likely affected by complications in the diagnosing of depression in this population. Understanding depressive states in ASDs will require additional thought and investigation.
Clinical Significance
Potential depressive symptoms in children with autism should be analyzed in light of common overlapping symptoms and communication limitations. Depression may be underdiagnosed, and may primarily present with irritability complaints.
Footnotes
Acknowledgment
The authors would like to thank Dr. Christopher J. McDougle for his contribution to the manuscript.
Disclosures
No competing financial interests exist.