Change in Core Symptoms from Ages 2 to 19 and Relative Stability from Ages 19 to 25 on the Adapted-ADOS, a New Instrument to Assess Minimally Verbal Adults

Background: Parent-reported ASD symptoms show different patterns of change across childhood and adolescence, depending on the individual’s age, language or IQ (e.g., Bal et al., 2018, McGovern & Sigman, 2005). Change in symptoms across childhood and adolescence has implications for diagnosis of ASD in adults. Additionally, identification of symptoms that remain stable across development may lead to a better understanding of mechanisms underlying observable impairments. To date, studies examining symptom trajectories are largely based on parent-report instruments or limited to comparisons of symptoms from direct observation measures, such as the ADOS-2, across childhood (e.g., Gotham et al., 2012). Longitudinal comparisons of symptoms captured on the ADOS-2 are hindered by a lack of appropriate instruments to assess older minimally verbal (MV) individuals. While there are different modules for verbally fluent children and adults (Modules 3 & 4), adaptations to the Modules 1 and 2, designed for use with children with less language, are needed to improve their utility with MV adults.

Objectives: To describe the Adapted-ADOS, a new tool designed for use with MV adolescents and adults and use A-ADOS scores to explore changes in the manifestation of ASD in young adulthood.

Methods: Diagnostic algorithms were derived from 160 adolescent and adult assessments [124 ASD, 36 Non-ASD, Mean age=19.18 (3.06) years]. McNemar’s test assessed change in the presence or absence of impairment of symptoms at ages 2 and 19 for 60 participants with ASD. Paired-samples T-tests compared A-ADOS totals in 35 participants seen at ages 19 and 26.

Results: Sensitivity was above .8 for all algorithms; specificity ranged from .56 (M1-No Words) to .88 (M1-Some Words). In the longitudinal sample, many adults demonstrated improvements in the areas of response to joint attention (X²=28.27, p<.001) and gestures (X²=13.89, p<.001), with significantly fewer participants exhibiting impairments on these items at age 19 compared to age 2 (RJA: 20% vs. 98%; gestures: 60% vs. 93%, respectively). In contrast, impairments in spontaneous initiation of joint attention and facial expressions appeared to persist into adulthood (Table 1). From ages 19 to 26, A-ADOS totals were relatively stable [t(34)=.90, p=.38, Figure 1]. Total scores remained stable or improved for 21 participants (Mdiff=3.10), but worsened for 14 adults (Mdiff=-3.4). The worsening group showed decreased engagement from 19 to 26, both with the examiner [t(13)=-2.69, p<.05] and the A-ADOS activities [t(13)=-2.48, p<.05].

Conclusions: The A-ADOS expands the repertoire of instruments to assess MV adolescents and adults and provides an alternative to the ADOS-2 modules 1 and 2 for assessment of older individuals. Consistent with parent report, adults exhibit both improvements (e.g., response to joint attention, gestures) and persistent impairments (e.g., facial expressions) in directly-observed symptoms considered to be core features of ASD in childhood. Preliminary analyses of a small sample suggest that symptoms appear relatively stable across young adulthood; however, a subset of MV adults may be at risk for decreased engagement during the transition years. Further analyses will explore factors that may explain the observed behavioral change (e.g., onset of depression; loss supports [see Taylor & Seltzer, 2010]).