Reinvigorating the Promise of the National Database for Autism Research (NDAR) to Advance Autism Knowledge.
NDAR’s red tape slashed usable autism files by ninety percent, so plan data-access time or pick a cleaner repository.
01Research in Context
What this study did
The team checked how well the National Database for Autism Research (NDAR) really works. They traced one planned big-data project from start to finish. The goal was to pool years of records for each child with autism.
What they found
Paperwork hurdles shrank the usable files from 3,850 kids to about 300. Less than ten percent had complete longitudinal data. The planned individual-participant meta-analysis stalled.
How this fits with other research
Cummings et al. (2024) describe SPARK and Simons Searchlight—two newer autism banks that let you re-contact families and boast cleaner pipelines. Their upbeat tone seems to clash with L et al.’s headache story, but the difference is timing and design: SPARK started with user-friendly software while NDAR still runs on older, stricter rules.
Leaf et al. (2021) warn that even perfect meta-analyses can mislead if the original studies are weak. L et al. show we can’t reach that worry because we can’t get the files at all. Together the papers flag both ends of the pipeline—access and quality.
Vassos et al. (2023) found that transition-studies also lack equity focus and cross-sector data. L et al. add a concrete reason: the main U.S. data vault is clogged. Fixing the vault could speed every gap review like M et al.’s.
Why it matters
If you hope to run large-scale autism research, know that NDAR’s red tape can gut your sample before you begin. Ask early: Are the data truly ready? Budget extra months for permissions or switch to SPARK if re-contact fits your question. Push your IRB and funders to streamline forms—your timeline depends on it.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Before you draft that big-data grant, email NDAR help desk for a realistic file count—then double the timeline.
02At a glance
03Original abstract
The National Institute of Mental Health created the National Database for Autism Research (NDAR) to accelerate autism knowledge through data sharing and collaboration. However, our experience using NDAR reveals systematic challenges across several aspects of data submission, selection, management, and analysis that limit utility of this resource. We describe our NDAR experience in an ongoing project examining autism intervention outcomes among marginalized racial, ethnic, and gender groups. For this study, we planned to gather data from NDAR to conduct an individual participant data meta-analysis. Eighteen studies met inclusion criteria and reported data on participants at more than one point in time on the Vineland Adaptive Behavior Scales (Vineland) and the Autism Diagnostic Observation Schedule (ADOS). The difficulties with submitting, selecting, downloading, and managing data from NDAR posed limitations on data availability and analysis. Of the 3,850 unique participants in the selected studies, data at multiple time points were available for 312 participants on the Vineland and 278 on the ADOS. No participants had data on all assessment domains. To accelerate autism research via data sharing and collaboration with NDAR necessitates improving the processes for submitting, selecting, and managing data.
Journal of autism and developmental disorders, 2025 · doi:10.1007/s40273-015-0331-6