Daniella Meeker, PhD

My research program is centered on the design, evaluation, and responsible deployment of data-driven systems that improve healthcare delivery at scale. Situated at the intersection of behavioral economics, biomedical informatics, and artificial intelligence, my work addresses a persistent challenge in health research: how to generate evidence that is actionable, equitable, and sustainable within real-world clinical environments. Across my career, I have pursued a coherent scientific trajectory that has evolved from behavioral intervention trials to informatics infrastructure, federated data platforms, and, most recently, AI-enabled learning health systems governed by strong ethical and institutional safeguards.

Phase 1: Behavioral Interventions to Improve Care and Uptake

My early research focused on applying behavioral economics and social psychology to improve clinician and patient decision-making. Through large, pragmatic randomized trials funded by NIH, AHRQ, and PCORI, I evaluated EHR-embedded interventions designed to reduce low-value care, improve antibiotic stewardship, enhance opioid prescribing safety, and increase uptake of evidence-based practices, including vaccination in underserved populations.

These studies demonstrated that subtle, theory-driven interventions could produce durable improvements in care quality without restricting clinician autonomy. They also established my expertise in pragmatic trial design, real-world evaluation, and implementation across complex health systems. Importantly, this work generated some of the earliest evidence that informatics-embedded behavioral interventions could be deployed at national scale.

Phase 2: Identifying Infrastructure as the Binding Constraint

As these behavioral trials expanded across institutions, a consistent pattern emerged: the primary barrier to translation was not behavioral theory, but data infrastructure. Challenges related to phenotyping, data quality, governance, interoperability, and analytic reproducibility limited both speed and generalizability.

This insight catalyzed a deliberate shift in my research agenda. Rather than treating infrastructure as a background condition, I began to study it as a scientific object in its own right, focusing on how data systems could be designed to support rigorous, ethical, and scalable research. This transition marked the beginning of my sustained contributions to biomedical informatics and learning health system science.

Phase 3: Federated Data Platforms and Inclusive Research Infrastructure

Building on this shift, my research expanded to include the design and evaluation of federated data networks and inclusive research repositories. Through leadership roles in NIH-, PCORI-, and NCATS-funded consortia, I contributed to national efforts to enable cross-institutional learning while preserving local control, patient privacy, and trust.

Across grants spanning R01, R33, U-series, CTSA, and Roybal Center mechanisms, my work advanced governance-first approaches to data sharing, standardized data representations, and privacy-preserving analytics. A central emphasis of this phase was ensuring that safety-net institutions and underrepresented populations could participate meaningfully in large-scale research, rather than being excluded by technical or governance barriers. These efforts supported pragmatic trials, public health surveillance, and real-world evidence generation across diverse care settings.

Phase 4: Responsible AI in Learning Health Systems

More recently, my research has incorporated machine learning and AI methods to support real-time phenotyping, adaptive decision support, and learning health system functions. This evolution reflects both advances in analytic capability and my longstanding commitment to responsible data use.

Rather than pursuing AI in isolation, my work integrates AI within the governed informatics frameworks developed in earlier phases of my research. Current NIH-funded projects address how AI-enabled tools can be deployed safely and transparently in clinical and research environments, with explicit attention to interpretability, bias, representativeness, and institutional oversight. This approach ensures that expanding analytic power is matched by safeguards that protect patients, clinicians, and health systems.

Across this work, AI is treated not as a replacement for clinical judgment, but as a decision-support and learning tool embedded within socio-technical systems. This framing distinguishes my research within biomedical informatics and implementation science and aligns with national priorities for trustworthy and ethical AI.

Yale: Translating Scholarship into Institutional Capability

Since joining Yale School of Medicine in 2023, my research agenda has entered a new phase focused on institutional translation of informatics scholarship. As Chief Research Informatics Officer, I founded and lead the Research Informatics Office, which serves as a research-enabling platform that operationalizes principles developed through my federally funded work.

In close partnership with Yale Biomedical Informatics and Data Science (YBIC) and Yale New Haven Health System, my research informs the design of data systems that make high-quality clinical data more accessible for research while strengthening privacy protections and governance. This includes secure, tiered access models; standardized pipelines linking Epic-based data to research environments; and oversight processes for evaluating advanced analytics and AI use cases.

These systems support hundreds of active studies annually, including NIH-funded pragmatic trials and observational research, and provide a foundation for responsible AI deployment across the research enterprise. Importantly, Yale serves not merely as an implementation site but as a testbed for studying how data integration, governance, and AI function at institutional scale, generating insights that inform national efforts.

Ethical Stewardship, Equity, and Trust

Ethical stewardship of data and AI is a core throughline of my research. My work explicitly addresses algorithmic bias, representation, and fairness, including integration of social and environmental determinants of health into analytic frameworks. I have contributed to national standards and governance initiatives through leadership in federally funded consortia and advisory roles.

This emphasis on responsible innovation has been recognized through election as a Fellow of the American College of Medical Informatics and receipt of the Journal of Clinical and Translational Science Publication Award, reflecting the impact of my work in advancing both analytic capability and trustworthiness in biomedical data science.

Future Directions

Looking ahead, my research will continue to advance portable, privacy-preserving, and governance-aligned AI-enabled data systems that support equitable learning health systems. I am particularly focused on extending these approaches to complex, high-need populations and on developing frameworks that allow institutions to adopt AI responsibly, transparently, and sustainably. Through sustained federal funding, national collaboration, and integration with Yale’s research enterprise, my work aims to strengthen both scientific discovery and public trust in data-driven healthcare.