Research

Intercurrent events that occur post randomization in randomized control trials (RCT), can alter the course of these trials and jeopardize comparative effectiveness evaluation and consequently decision making in regulatory science. The standard intention-to-treat analysis used for RCT does not account for intercurrent events, thus fails to capture treatment effect, heterogeneity, and causal mechanisms.

The FDA’s OCE launched 'Project Optimus' to emphasize the need for dosage optimization prior to drug approval. In collaboration with Project Optimus and the Oncology Real World Evidence program, this project harnesses the capabilities of statistical modeling, advanced data methods, and causal inference approaches using real-world data sourced from UNC-CH.

In the U.S., opioid overdose deaths have increased for those living in rural communities. Access to support services for opioid misuse is lacking for these geographic regions and populations. This project aims to characterize the role community pharmacies play in decreasing harm from opioids by supporting safe use of opioid medications and access to medications for opioid use disorder treatment and overdose reversal.

The unregulated drug supply is an ever-changing chemical landscape with novel additives that are often unknown to end users, clinicians and public health officials. One such compound, xylazine, is now detected in the unregulated drug supply at alarming levels. This project uses pharmacological approaches to examine how xylazine and dexmedetomidine interact with receptors in the body, how they affect brain function, and influence fentanyl’s effects.

Opioid Use Disorder (OUD) affects 2.7 million people in the U.S. and is responsible for 75% of opioid overdose deaths. Our project will generate a scientific plan to use digital health technologies (DHTs) and the data they collect to predict relapse and to inform early-intervention strategies for people living with OUD with the care they need, when they need it most.

Patient-reported outcome (PRO) questionnaires ask patients about their symptoms and functions as input for medical device safety and effectiveness evaluation. The timeframe in these questionnaires needs to be long enough for completeness of recollections but also short enough for it to be easy to remember. This project aims to find the best ways to measure PRO for temporomandibular disorders, that present episodically or vary in frequency and intensity over time.

Buprenorphine is a proven and widely prescribed treatment for patients with opioid use disorder. However, many patients may not receive adequate dosing in current practice. To address this, the study will use linked healthcare, prescription, and mortality data and apply advanced epidemiologic and statistical methods to assess the relationship between different dosing levels and patient outcomes thereby generating evidence that informs clinical guidelines, regulatory decisions, and prescribing policies.

Breaking down complex scientific concepts and methods into clear, accessible training materials that guide users in collecting and using patient experience data enhances how stakeholders integrate patient input into medical product development. This project strengthens regulatory science by equipping stakeholders with practical tools to apply the FDA’s Patient-Focused Drug Development (PFDD) guidance.

Generating real-world evidence (RWE) on the safety and effectiveness of IDH inhibitors used for treatment for the approximately 20,000 Americans who are diagnosed with acute myeloid leukemia (AML) is important to address evidence gaps left by pivotal trials that did not fully reflect real-world treatment settings. Advancing the use of real-world data (RWD) in regulatory decision-making contributes to a better understanding of treatment outcomes and safety in clinical practice.

The choice of index date (point when outcome tracking begins) plays a critical role in externally controlled trials. It influences which patients are included in the comparator group, what baseline characteristics are measured, which outcomes are captured, and how long follow-up continues. Index date strategies can introduce or reduce bias, affect comparability between groups, and ultimately change the interpretation of treatment effectiveness, even when the actual treatment effect is constant.