Grappling with data to determine how adults with autism fare in the US
by Jessica Rast
February 9, 2016
What is the national high school graduation rate of youth on the autism spectrum?
What percentage of individuals on the autism spectrum need services to facilitate employment? How many receive them?
What percentage of youth with autism receive comprehensive transition services when leaving high school?
At the Life Course Outcomes Research Program, we often try to answer questions like these by looking at national, taxpayer-funded surveys which include data about people with autism. Although we have a wealth of information available to us, wading through survey and other data to answer broad questions is trickier than one might imagine.
One of the first tasks we set out to do was to catalog what we currently know about how young adults with autism are faring. In April 2015, we began by producing a series of National Autism Indicators Reports – the first of which focused on the Transition into Young Adulthood. Data for our study primarily came from the NLTS2 -- a national survey initiated in 2000 by the U.S. Department of Education to capture the experiences of youth who received special education during high school and follow them forward into adulthood.
We have used this and other data extensively to report baseline indicators on how well the needs of young adults on the autism spectrum are being met in the U.S. Putting this report together made us stop and ask ourselves: What are the issues faced by researchers working to understand the experiences of people with autism in the U.S.? What needs to be done in order to further our research agenda? Here are some of our recommendations:
Increase researcher awareness of datasets
We have a lot of data about autism. We aspire to increase the use of national indicators data by creating the National Autism Data Center (NADC), a collection of data sources that have information about individuals on the autism spectrum in one location. The NADC – effectively the “census bureau” of autism data – will allow for data analysis and effective partnerships with those who also need access to this information.
We need to build new surveys with existing surveys in mind. More efficient data collection using the same questions across years allows us to see how people’s answers to the same question change over time, letting us know whether we are making any progress on the problems at hand. The National Longitudinal Transition Study-2 (NLTS2) conducted five waves of data collection over 10 years and was a good example of thoughtful data collection. The waves included questions that were consistent across years and incorporated questions from other national surveys, allowing for comparisons to other populations.
We also need to increase alignment of efforts to collect data. This would include collaborating across organizations who conduct health and education surveys or asking questions in the same way to the same groups.
Streamline websites to make key information accessible
Websites housing national data often serve several audiences and functions at the same time and include vast amounts of information that make them hard to navigate. Clear, easy to navigate websites are crucial to researchers. The Data Resource Center for Child & Adolescent Health, for example, does a great job of displaying information in a way that is accessible for different audiences including researchers seeking national level data.
Fund analysis of existing data
Some federal agencies fund data collection via large population-based surveys that often include people on the autism spectrum, but are not created specifically for autism research. The funding of large data collection efforts that are not autism-specific rarely also includes funding for extensive analyses of the resulting datasets. Researchers who want to use these datasets to study autism must compete for funds through other funders like the National Institutes of Health. Studies of autism that use these large population datasets are essential for understanding the unmet needs and service system experiences of people with autism and their families. They provide the benchmark statistics and indicators that tell us where things stand so we can be in a position to literally “move the needle” on outcomes. However, relatively little autism-specific funding goes towards this kind of work, particularly for older youth and adults. According to the federal Interagency Autism Coordinating Committee’s 2011-12 Portfolio Analysis, $1.5 billion was spent on autism research in the U.S. between 2008 and 2012. We estimate that $3.3 million was spent over the same period to fund studies that examined outcomes of youth and young adults using population-level data.*
Are we making progress as a nation in helping youth affected by autism? To answer this question, we need usable, accessible data to track progress over time across disciplines, agencies, and entities. The National Autism Data Center is taking steps to centralize this type of information, but we need policy and regulatory changes at the national level, such as the ones suggested above, to ensure efficient and effective data collection and analysis. This is one necessary step toward improving services and outcomes for those on the autism spectrum.
*To estimate research funding for analysis of existing secondary data, we searched the IACC Autism Research Database for years 2008-2009 using the keywords “secondary,” “population,” and “nationally representative” in successive searches. Titles from each of these searches were read to identify projects targeting youth, young adults, or adults. The abstract and any additional information provided by the funder for each identified project were read to see if secondary data analysis of existing population-level data was used. The following projects were identified:
- OAR, 2008-2010
- HRSA R40 MC 15595, 2009
- NIH 1R01MH086489-01, 5R01MH086489-02, 5R01MH086489-03, 5R01MH086489-04, 2009-2012
- HRSA R40 MC 22646, 2011
- HRSA R40 MC 22647, 2011
- DOE R324A120012, 2012
- NIH 1R01MH093522-01A1, 2012
- NSF 1130088, 2012
- OAR, 2012
Jessica Rast received her MPH from Drexel in 2014 and has been working since then on the Life Course Outcomes team. She is a researcher, data analyst, and an idealist.