Evaluating the utility of negative controls in drug safety and effectiveness studies using real-world data
CERSI Collaborators: Zafar Zafari, M.Sc., PhD (Principal Investigator), Susan dosReis, PhD (co-Investigator), Chintal H. Shah, MS, Jeong-eun Park, MPH, Emily Gorman, MLIS, AHIP, University of Maryland Baltimore
FDA Collaborators: Fang Tian, PhD, MPH, MHS, Wei Hua, MD, PhD, Rita Ouellet-Hellstrom, PhD, MPH, Yong Ma, PhD, MS
Project Start Date: June 1, 2020
Regulatory Science Challenge
Randomized controlled trials (RCTs) are the gold standard for evaluating the efficacy and safety of a treatment. However, the high costs of conducting RCTs coupled with the limited applicability of the findings of RCTs to real-world settings have made the use of real-world data in long-term drug safety and effectiveness studies more popular. Nevertheless, a major concern in using real-world data is the presence of unaccounted biases (e.g., selection bias, information bias) and confounding (influences from multiple sources that cannot be separated) in the data, which may threaten the internal validity of the study results.
Project Description and Goals
Negative controls, defined as negative control exposures (or outcomes) that share the same potential source of bias with the primary exposure (or outcome) but are not causally related to the outcome (or exposure) of interest, are useful tools to address issues involving confounding and other unaccounted biases (e.g., selection bias, information bias), with potential for broad application.
In this project, researchers will:
- Conduct a full literature review for the use of negative controls in epidemiological studies on drug safety and effectiveness, and
- Evaluate assumptions behind the use of negative controls in these studies.
The main goal of this project is to build a methodological framework for evaluating the use of negative controls in drug safety and effectiveness studies using real-world data.