Project AIM (Autism Intervention Meta-analysis), led by Dr. Micheal Paige Sandbank, is a large-scale, ongoing meta-analysis that examines early interventions for young children on the autism spectrum. Unlike studies that focus on a single therapy or a single outcome, Project AIM reviews all non-pharmacological interventions for children under eight. The research evaluates interventions such as behavioral therapies and other psychosocial supports, aiming to estimate the effects of each approach and describe the amount and quality of evidence supporting them. Looking across the full landscape of early autism interventions, Project AIM provides a comprehensive picture of what currently is and is not known about the effects of different therapies provided to young children on the autism spectrum.
Project AIM uses meta-analysis to gather and analyze data from multiple existing studies to draw broader conclusions. Instead of conducting new experiments, researchers like Sandbank review the work of others, looking for patterns and trends across many studies. This approach allows them to see a bigger picture that would not be apparent in a single study. The project is ongoing and constantly updated as more research and data become available.
Prior to 2015, conducting a meta-analysis of all early autism interventions together wasn’t feasible. However, the development of specialized mathematical techniques around that time made it possible, allowing the initiative to officially begin in 2017. The team published their first paper in 2020, followed by a major update in 2023 in the BMJ, one of the highest-impact medical journals.
The current dataset includes 289 studies, and the team is now conducting an update that includes research published since 2021. The update begins with a systematic search that initially identified around 8,000 abstracts, each of which must be screened for eligibility. Selected studies then move into the coding phase. Sandbank serves as the primary coder, while a group of reliability coders –based at UNC, with partner labs at Vanderbilt, Boston College, Florida State, and soon Boys Town in Nebraska— crosscheck the work for accuracy. Many coders have been with Project AIM since their doctoral or even undergraduate years, supporting the project’s growth across multiple institutions.
Project AIM organizes data by intervention type, intensity, and other key features. With such a large dataset, the team can examine both broad patterns and the effects of specific interventions, including which approaches are effective and which children benefit most. Insights that would be invisible in a single study become detectable across this large body of evidence. The team has recently examined whether children’s benefit from intervention depends on their initial cognitive and language scores in a forthcoming paper published in the Journal of Speech, Language, and Hearing Research.
Since 2021, the number of eligible studies has doubled, reflecting both the pace of autism research and the importance of regular updates. At least 200 more studies are predicted to be added to the dataset, expanding it to include approximately 500 studies in total; the current Project AIM dataset features data from over 13,000 participants, and with those 200 additional studies they expect that number to rise significantly. Sandbank says that this increase highlights the rapid growth of research in the field. With continued growth of the dataset, the field gains a clearer understanding of which interventions have consistent, high-quality evidence behind them.
Sandbank also regularly collaborates with researchers across the globe, working with teams in France, the United Kingdom, and Italy. She recently co-authored the largest-ever umbrella review examining the effects of complementary, alternative, and integrative medicine approaches for autistic individuals with researchers from France and the United Kingdom, published in Nature Human Behavior. The team found that alternative treatments for autism, such as mind-body medicine and natural product-based therapies, have little evidentiary support. Recently, the Project AIM team began conducting a rapid meta-analysis in response to new recommendations about folinic acid (also known as folate or leucovorin) for autism that was raised during a recent press conference. They are collaborating with another meta-analysis team, including scholars in Italy and France, to review the few studies that currently exist on this topic.
Sandbank works to disseminate the findings of her research directly to clinicians that support young children on the autism spectrum through other mediums as well. In March 2025, she delivered a keynote address at the Elevate Your Impact Conference, a clinician-focused conference hosted by Catalight, one of the largest behavioral healthcare networks in the nation. She followed this address up with an appearance on the “What’s Up With Catalight!” podcast. These clinician-facing speaking engagements allow Sandbank to ensure that early intervention therapists continue to learn about the ever-evolving evidence in this field.
When it comes to non-pharmacological treatments, they often are not given as much scrutiny as medications or medical devices. Because these approaches are not thought to directly impact a patient’s physical health, providers, patients, and families may assume they are safe and beneficial, even if little research is available to support that conclusion. However, this can be unfair to patients and families, as it may lead them toward treatments that don’t have strong or consistent evidence supporting them. Everyone deserves access to interventions that are clinically proven to help, which is why Sandbank recently co-authored a commentary with 25 other autism researchers calling for increased quality and reporting standards in autism research, in the journal Autism Research. Sandbank says that the quality of available evidence increases with each update of Project AIM. “As more high-quality randomized controlled trials are published, we are able to draw more confident conclusions about the efficacy of early childhood interventions, and give more precise and helpful recommendations that families need.”