Retention of autism spectrum disorder diagnosis: The role of co-occurring conditions in males and females

Highlights

• 91% of males and females with ASD diagnosis retained their ASD diagnosis.

• Co-occurring intellectual disability and sensory integration disorder were associated with diagnosis change from ASD to non-ASD in males.

• Delays in personal/social development and epilepsy were associated with diagnosis change from ASD to non-ASD in females.

• More ASD-related evaluations and less ASD-related impairment were associated with ASD being ruled out in both sexes.

Abstract

This study examined associations between ASD diagnosis retention and non-ASD co-occurring conditions (CoCs) by child sex. The sample included 7077 males and 1487 females who had an ASD diagnosis documented in their school or health records in a population-based ASD surveillance system for 8-year-old children. ASD diagnosis retention status was determined when an initial ASD diagnosis was not later ruled out by a community professional. We found that ASD diagnosis remains fairly stable, with only 9% of children who had an initial documented ASD diagnosis later being ruled-out. Although most of the associations between the ASD diagnosis retention status and CoCs are similar in both sexes, the co-occurrence of developmental diagnoses (e.g., intellectual disability or sensory integration disorder) was predictive of ASD diagnostic changes in males, whereas the co-occurrence of specific developmental (e.g., personal/social delay) and neurological diagnosis (e.g., epilepsy) was associated with ASD diagnostic change in females. More ASD-related evaluations and less ASD-related impairment were associated with later ASD rule outs in both sexes. Our findings highlight that CoCs can complicate the diagnostic picture and lead to an increased likelihood of ambiguity in ASD diagnosis. Using sensitive and appropriate measures in clinical practice is necessary for differential diagnosis, particularly when there are co-occurring developmental conditions.

Introduction

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by persistent deficits in social interaction and communication, and restricted, repetitive patterns of behaviors, interests or activities (American Psychiatric Association, 2013). The most recent prevalence estimate of ASD among 8-year-old children in the United States (US) is 14.7 per 1000 children (1 in 68 children) (Autism and Developmental Disabilities Monitoring Network Surveillance Year 2010 Principal Investigators, 2014). Persons with ASD were previously defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition − Text Revision (DSM-IV-TR) (American Psychiatric Association, 1994) as those diagnosed with Autistic Disorder (AD), Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS) or Asperger’s Disorder. All of these diagnoses are now subsumed under one label of ASD in the Fifth Edition (DSM-5) (American Psychiatric Association, 2013). Although ASD symptoms typically appear in the first 3 years of life, other developmental, psychiatric and neurologic disorders frequently co-occur with ASD (Close, Lee, Kaufmann, & Zimmerman, 2012; Levy et al., 2010) and differentiation between the core features of ASD and co-occurring conditions can be challenging and may have a significant impact on the accurate identification of children with ASD. As the number of children being diagnosed with ASD is growing, it is important to explore cases in which the diagnosis of ASD changed to non-ASD, as well as the role co-occurring non-ASD diagnoses play in changing diagnostic status in order to inform early identification and intervention.

Past research examining the retention rate of ASD diagnosis is inconsistent, and it is unclear what role early intervention and subsequent symptom improvement play in the developmental outcomes of youth with ASD. High rates of ASD diagnosis retention for overall spectrum of ASDs (91–100%) have been reported between 2–4 years of age (Banach et al., 2009; Chawarska, Klin, Paul, Macari, & Volkmar, 2009; Van Daalen et al., 2009) and 88–90% between 2 and 9 years of age (Kleinman et al., 2008, Lord et al., 2006), whereas the retention rate of the subtypes diagnoses within the autism spectrum (AD or PDD-NOS) has been more variable (54–88%) compared with the retention rate of entire category of autism spectrum (Chawarska et al., 2009, Daniels et al., 2011, Malhi and Singhi, 2011; Turner, Stone, Pozdol, & Coonrod, 2006; Van Daalen et al., 2009, Wiggins et al., 2012). Furthermore, recent studies showed that almost half of ASD siblings children (41–63%) who were not diagnosed at 18 or 24 months of age received an ASD diagnosis at 36 months of age (Ozonoff et al., 2015, Zwaigenbaum et al., 2015). Other research showed 12% of ASD siblings children may change from non-ASD diagnosis at age 3 to an ASD diagnosis in middle childhood (Brian et al., 2015). Although these findings suggest that some children with an initial diagnosis of ASD no longer meet ASD diagnostic criteria at a later time or vice versa, most studies collected information from small clinic-referred samples (Banach et al., 2009, Kleinman et al., 2008, Van Daalen et al., 2009, Wiggins et al., 2012), high-risk ASD siblings (Brian et al., 2015, Ozonoff et al., 2015, Zwaigenbaum et al., 2015) or provided estimates only for a specific spectrum of ASD (e.g. AD, PDD-NOS) (Chawarska et al., 2009, Malhi and Singhi, 2011, Turner et al., 2006). These differences in study design likely contribute to the varying estimates on diagnosis retention rate. Investigation of diagnosis retention for children with ASD in a large and population-based sample is essential to provide a clearer picture of the diagnostic pattern in ASD.

Previous studies consistently report a sex difference in the prevalence of ASD of 4.5:1 males to females (Autism and Developmental Disabilities Monitoring Network Surveillance Year 2010 Principal Investigators, 2014; Newschaffer et al., 2007). Research has also found discrepancies in clinical presentation: males with ASD are more likely to exhibit repetitive and stereotyped behaviors (Carter et al., 2007, Hartley and Sikora, 2009; Hattier, Matson, Tureck, & Horovitz, 2011; Rubenstein, Wiggins, & Lee, 2015), whereas females with ASD have higher rates of severe cognitive and developmental delays (intellectual quotient [IQ] ≤ 70) (Banach et al., 2009, Nicholas et al., 2008, Rubenstein et al., 2015; Volkmar, Szatmari, & Sparrow, 1993). A few studies also found a sex discrepancy in timing of ASD evaluation and diagnosis: females may be referred for evaluation of ASD at earlier ages than males; however girls had a longer delay between first evaluation and ASD diagnosis than boys (Wiggins, Baio, & Rice, 2006) and are ultimately diagnosed with ASD at later ages (Shattuck et al., 2009). Although previous studies suggest that the timing of ASD evaluation and diagnosis varies between the sexes, no epidemiologic studies have further explored possible sex differences in ASD diagnosis retention.

In addition, previous studies revealed that co-occurring conditions (CoCs) and other diagnostic-related factors are associated with the change in ASD classification from ASD to non-ASD (Close et al., 2012, Wiggins et al., 2012). Wiggins et al. (2012) reported that children receiving ASD diagnoses at young ages and those with co-occurring specific developmental delays are more likely to experience a change of ASD classification. Furthermore, presence of CoCs, such as learning disability, developmental delay, speech or hearing problem, anxiety and epilepsy, can distinguish children who had a previous and current ASD diagnosis from those children with a past but not current ASD diagnosis (Close et al., 2012). Although these studies support the relationship between CoCs and changes in classification from ASD to non-ASD, possible variation in the associations between ASD diagnosis retention and CoCs based on child sex was not investigated in previous studies. Whether child sex plays a role in the association between specific CoCs and ASD diagnosis retention warrants further investigation.

In this study, we examined a large sample of children from the Autism and Developmental Disabilities Monitoring (ADDM) Network, an ongoing, population-based, multisite surveillance program established by the Centers for Disease Control and Prevention (CDC) to estimate the prevalence of ASD among children aged 8 years. Our aim was to describe the nature of ASD diagnosis retention and change according to child sex. Specifically, we examined factors that were associated with ASD diagnostic change, including presence of CoCs and ASD diagnosis-related factors in male and female children who retained an ASD diagnosis (i.e., child with past ASD diagnosis that was not later ruled-out in health or education records) and who did not retain an ASD diagnosis (i.e. child with past ASD diagnosis that was later ruled-out in health or education records).