Patient and Provider Factors Influence Completion Rates of Genetic Testing for Autism Spectrum Disorders

Background: Genetic testing is standard medical care for children with autism spectrum disorder (ASD) as results can assist with providing diagnostic certainty, guide medical decision-making, influence family planning and help prognosticate on other medical issues in the affected child. Despite the American Academy of Pediatrics, American Academy of Neurology, American College of Medical Genetics and Genomics, and American Academy of Child and Adolescent Psychiatry writing consensus statements that chromosomal microarray (CMA) should be performed as first-line testing for all patients with non-syndromic ASD, previous survey-based research has identified low rates of testing, especially CMA completion (15-20%). While patient factors like intellectual disability and seizures are associated with increased yield of clinically significant results on CMA testing, it is unclear if this knowledge may affect ordering practices. Providers also have different approaches to genetic counseling in autism and it remains unknown how the provider’s perspective influences completion of CMA testing.

Objectives: To evaluate completion of CMA testing using a prospective cohort of newly diagnosed children with ASD and investigate factors contributing to test completion . Patient factors include age, gender and medical comorbidities. Provider factors include specialty and CMA recommendation practice.

Methods: 343 patients with a new diagnosis of ASD made between 2/1/15 and 1/31/16 were identified. Cases were selected using a mandatory online institutional quality improvement questionnaire completed by providers at the time of initial diagnosis. Laboratory data were analyzed through 8/1/17, allowing for a minimum of 18 months for genetic testing to be performed. Patient and provider factors were extracted from billing data and chart review.

Results: The sample had 75% males and a mean age at diagnosis of 49 months (range: 16 months to 17.5 years). Epilepsy occurred in 3% and global developmental delay or intellectual disability (GDD/ID) occurred in 38%. Completion rate of CMAs was 40.2%. Higher rates of CMA completion were seen in individuals with GDD/ID (52% with v. 33% without, p<0.001). There was a negative correlation with age at time of diagnosis (r = -0.27, p<0.01). There was no effect of gender (48% males v. 38% females, p=0.14) or comorbid epilepsy (45% with v. 40% without, p=0.72).

Providers recommended genetic testing in 86% of initial diagnostic visits and also referred to Genetics in 6.4%. There was a lower overall CMA completion rate when providers did not recommend genetic testing at the first visit (15%) compared to when recommended by the provider (44%) or when recommended and referred to Genetics (45%, p < 0.01). Patients seen by providers in developmental medicine were more likely to have completed CMAs compared to patients seen by providers in neurology (46% v. 27%, p=0.02).

Conclusions: In this prospective cohort at a single center with institutional guidelines recommending routine CMA testing for all individuals with non-syndromic ASD, the CMA completion rate remains low. Certain patient-related factors (e.g., GDD/ID and age) and provider-related factors (e.g., specialty and recommendation made at first visit) significantly impact CMA completion rates. Further investigation is needed into specific provider practices, additional patient variables and family attitudes.