The institute’s work on the Autism Data Science Initiative aims to advance our understanding of what positive autism care outcomes look like and how they can be achieved.
According to the latest CDC data, the prevalence of autism spectrum disorder (ASD) in today’s eight-year-olds is about one in 31. But despite its prevalence, many families trying to support children diagnosed with ASD continue to face complexity and uncertainty.
Part of the challenge is that autism spectrum disorder itself is extremely broad. Combine that with numerous treatment modalities and a patchwork of national, state, and provider-specific approaches, and it becomes extraordinarily difficult to reach a general consensus about what is working and why.
To address the information vacuum surrounding autism, the National Institutes of Health is funding the Autism Data Science Initiative (ADSI), a new $50 million program that will explore and learn from the vast amount of autism-related data that exists within government and healthcare systems.
Co-leading an ADSI project housed at the Oregon Health & Science University (OHSU), SRI will play a key role in turning massive datasets into clear findings that can impact the way we provide services to individuals with autism.
Turning autism data into new insights
The project aims to create a data-driven framework that can predict, based on objective factors like insurance status, location, service use, and co-existing conditions, the likelihood of optimal and suboptimal outcomes for toddlers and children with autism. Ultimately, this Autism Quality Index (AQI) will guide future research and help states and healthcare systems improve concrete outcomes across the autism spectrum.
To serve this mission, SRI will contribute extensive experience with state early intervention and special education systems to identify state and local system-level variables. SRI will also clean and manage early intervention data from participating states. These data will be aggregated with data from other sources, including Medicaid claims, the National Survey of Children’s Health, the Census Bureau, the Child Opportunity Index (COI) 3.0, and state health and education systems. Leveraging clean, high-quality data provided by SRI, biostatisticians and other collaborators at OHSU and project partner Northern Arizona University will then apply machine learning and other techniques to build an Autism Quality Index that can reliably predict care outcomes.
“We really believe that it’s important that the people who are most affected by your work have a hand in designing and implementing the solutions. So we’re excited that this project will also convene individuals with autism and family members to identify the outcomes that are important to the community.” — Margaret Gillis
SRI’s work on the ADSI project will take advantage of the institute’s deep experience distilling insights from educational and public health datasets. The SRI-managed DaSy Center, for example, has become a nationally recognized lynchpin in the effort to help states build high quality data systems that improve results for young children with disabilities and their families. Furthermore, SRI’s ADSI work will take advantage of data and learnings collected through two other NIH-funded projects: ASCEND and ASD3.
“It’s about knowing where the most important data lives, but also understanding what’s most relevant and then handling the data appropriately,” explains Margaret Gillis, a principal researcher with SRI Education and principal investigator on the project. “We have experts on our team who have worked in state agencies and have significant experience with data sharing and data use agreements. That helps expedite the process. It’s also about understanding the nuances of the state-level variables that we might be able to look at. And then there’s the security piece. Because of SRI’s experience working with sensitive government data, state and federal stakeholders trust us to maintain best-in-class data security practices for these sorts of projects.”
A collaborative approach to impact
In addition to predicting outcomes, Gillis emphasizes, there’s another side of this project that’s equally important: working directly with the autism community to hone our understanding of what counts as an optimal outcome.
“One of the cool things about this opportunity is the recognition that you need to have people with autism involved in these projects,” Gillis adds. “At SRI Education, we really believe that it’s important that the people who are most affected by your work have a hand in designing and implementing the solutions. So we’re excited that this project will also convene individuals with autism and family members to identify the outcomes that are important to the community. And at the end of the project, we’re going to be convening these folks to think about policy recommendations and other long-term implications of this research.”
Learn more about SRI’s critical work in education and health.
