Big Data

Artificial intelligence (AI) tools, particularly those that use natural language processing (NLP), are becoming increasingly prominent in healthcare delivery. For these systems to operate safely and reliably in clinical settings, they require high-quality training datasets drawn from multiple sources, including electronic health record (EHR) data, that closely reflect real-world conditions.

The United States (U.S.) faces urgent health challenges, including poor maternal and childhood health outcomes, childhood obesity, opioid use disorder, cancer, and chronic conditions, which require personalized care tailored to each patient’s unique circumstances.

This report is to serve as a data source guide that helps policymakers and researchers understand the current data infrastructure landscape for SCD-focused health outcomes research. Specifically, the goals are to identify existing SCD data sources and data dashboards as well as key considerations to enhance data collection and use.

RAND identified use cases on identifying frailty using electronic health record (EHR) data in health systems in the US and examples from other countries, which demonstrate applications in both primary and specialist care. The final EHR implementation guide summarizes the learnings from participants in the EHR Learning Network and the identified use cases.

The Synthetic Health Data Generation to Accelerate PCOR project was launched in 2019 by the Office of the National Coordinator for Health Information Technology (ONC).

ASPE recently published a narrative review in PLOS Digital Health exploring how synthetic data are being used. Researchers searched published literature and known, publicly available synthetic datasets.

Using a national Medicaid database, the report shows significant racial/ethnic disparities in mental health service use among children during the COVID-19 pandemic.

In recent years several researchers and child welfare agencies have begun developing predictive risk models to support child welfare decision-making. Predictive analytics is a sophisticated form of risk modeling that uses historical data to understand relationships between myriad factors to estimate a probability score for the outcome of interest.

This brief analyzes information on administrative data resources collected by the Administration for Children and Families (ACF). It explores how the data can be leveraged to improve evidence and research on ACF programs and beneficiaries. Key highlights include:

National health surveillance instruments are intended to monitor important health issues and health status of all populations in the United States. Several population subgroups present with disparities in health conditions and health care. To effectively create programs and policies to address these issues requires accurate identification of key population subgroups.