Stability of Autism Diagnosis in Children Under 24 Months

Background: The early diagnosis of autism spectrum disorder (ASD) has become an important clinical and public health goal. While it has been well established that an ASD diagnosis made between 2 and 3 years of age is stable, less is known of the stability of an ASD diagnosis in younger toddlers.

Objectives: To prospectively examine the stability of an ASD diagnosis made by age 24 months (mo) and to examine initial clinical and demographic factors associated with retention of the diagnosis.

Methods: 108 children were diagnosed with ASD by age 24 mo at a University-affiliated inner-city diagnostic Early Intervention program between 1998 and 2009. Diagnoses were made by a multidisciplinary team (developmental pediatrics, psychology, speech and language pathology), based on the DSM-IV Pervasive Developmental Disorder criteria, Childhood Autism Rating Scale (CARS), and, in some, the Autism Diagnostic Observation Schedule (ADOS).  At least one year after initial diagnosis, families were invited to return for updated developmental evaluation and ADOS to confirm ASD diagnosis. Data collected included initial assessment (demographics, cognitive level, and autistic features) and follow up information (interventions received and current educational placement). Statistical analysis included chi-square, t-test and nonparametrics.

Results: The mean age at initial diagnosis was 20±3mo (range:12-24mo),75% male, mean initial CARS score of 35±4, developmental quotient <70 in 57%. Autism was diagnosed in 25% (27/108) and ASD/Pervasive Developmental Disorder Not Otherwise Specified (PDD-NOS) in 75% (81/108). Of the 108 children, 77 returned for follow up developmental evaluation at least one year after initial diagnosis. There were no differences in age at diagnosis, ASD severity, or demographic characteristics between those who returned for follow up (N=77) and those who did not  (N=31). The mean age of the children who returned for follow up was 54.4±29.9mo. All 22 of the children initially diagnosed with autism and 82% (45/55) of those diagnosed with ASD/PDD-NOS continued to meet criteria for ASD based on ADOS (available in 40/77), CARS, and/or autism-specific school placement. Ten children did not meet criteria for Autism or ASD at follow up. These children had significantly lower initial CARS scores (32.2 ±1.4 vs. 35.3 ± 4.8, p< 0.001) and were less likely to have presented with mannerisms at initial diagnosis (60% vs. 91%; p=0.006), without significant differences in demographics, cognition, length of time between initial diagnosis and follow up, or types of intervention as compared to those who retained an ASD diagnosis.  Specifically, children who continued to meet criteria for ASD at follow up were more likely to have presented with motor mannerisms (77.6% vs. 20%; p<0.001) before age 24mo.

Conclusions: In this sample, an ASD diagnosis made by experienced clinicians in toddlers 12 to 24mo of age was quite stable. The diagnosis of Autism was more stable than PDD-NOS. The presence of motor mannerisms at initial diagnosis was associated with retention of ASD diagnosis at follow-up.