Skip to main content

Transforming the understanding
and treatment of mental illnesses.

Celebrating 75 Years! Learn More >>

Early Screening for Autism Spectrum


Lisa Gilotty, Ph.D.
Division of Translational Research


The primary goal of this initiative is to encourage research that would develop and validate new screening methods for autism spectrum disorders (ASD) that can be used in infancy (0-12 months of age).


Evidence from studies of infants at high risk for ASD reveals that although the age of onset of ASD clinical features is variable, subtle signs can already be detected in the first year of life. These include brain volume changes detectable with neuroimaging, as well as behavioral changes detectable with eye-tracking technology. Despite these findings, the average age of diagnosis for ASD remains at approximately four years of age. In addition, studies have established that children from racial and ethnic minority groups are often diagnosed and engaged in treatment at later ages. Delays in diagnosis have profound cascading effects, since early intervention has been demonstrated to improve cognitive and behavioral outcomes for young children with ASD.

Universal screening is one approach to reducing the age of diagnosis and improving outcomes. Well-validated instruments exist to screen toddlers for ASD between 18-24 months of age, and research in high-risk populations has shown that screening at age 18-24 months is a successful strategy to detect risk for ASD. This concept is in response to strong evidence that many infants at risk for ASD show differences in the way social attention and early forms of communication develop over the first year of life. These early signs correspond with other research findings that genetic and environmental risks for ASD appear to act before birth to alter very early stages of brain development. NIH-funded research has identified risk markers within the first 12 months of age, yet a critical gap exists in translating these methods into practical screening tools that could be implemented in the general population, within community settings. This concept would encourage applications to translate these methods and findings of ASD risk into efficient, inexpensive screening tools for use in the general population. 

Projects responsive to this concept would:

  • Evaluate the sensitivity, specificity and other psychometric properties of tools and methods to identify ASD risk in infants (0-12 months of age).
  • Conduct early stages of testing and development in laboratory settings, and demonstrate feasibility for future implementation in the general population and within the general pediatric clinical care setting.
  • Utilize multi-modal methods of determining ASD risk, such as measures of behavior (e.g., eye tracking, social responsivity), medical history (e.g., prenatal/developmental/ family history), genetics (e.g., whole exome sequencing from umbilical cord blood), biomarkers (e.g., immune profile in newborn blood spots) and/or other measures.
  • Construct risk algorithms using single or multiple parameters using computational approaches.