Samiran Ghosh, Ph.D., assistant professor of Family Medicine and Public Health Sciences and of the Center for Molecular Medicine and Genetics for the Wayne State University School of Medicine, has been awarded a grant by the Patient-Centered Outcomes Research Institute to study novel statistical research methods meant to improve and produce medical study design.
The PCORI awarded Dr. Ghosh a three-year, $832,703 grant for his project, "Developing Bayesian Methods for Noninferiority Trial in Comparative Effectiveness Research." The study is one of 46 funded by the institute to advance the field of comparative clinical effectiveness research and provide patients, health care providers and other clinical decision-makers with information that will help them make better-informed choices.
The project, Dr. Ghosh explained, is primarily a statistical methodology development study focusing on "non-inferiority" trials, a specific type of randomized controlled trial, or RCT, used to compare two or more active interventions. Non-inferiority trials are often employed in comparative effectiveness research when the clear superiority of one intervention is not evident.
"Our goal is to use past patient and disease specific information in the randomized control trial design to produce better and improved study design," he said. "This improvement may yield significant cost savings, reduce trial durations and may result in more ethical and patient-centric trials. We will primarily focus on mental health and cardiovascular disease trials, though the general statistical method will be applicable to other disease areas."
Two paradigms or philosophies exist in statistics, the Classical/Frequentists and Bayesian. A statistician uses probability as a measure of uncertainty of a random event, but Bayesian philosophy uses information from two sources: prior or historical information about the point of interest coupled with the likelihood or probability of the event of interest obtained using data collected from a current set of experiments. Classical/Frequentists philosophy uses only the information collected from the current experiments. When the prior information is weak or vague, the answers produced by the two approaches are nearly identical. However, if the prior information is substantial and informative, the Bayesian approach has significant advantage in uncertainty quantification, Dr. Ghosh said.
In comparative effectiveness research, researchers often compare multiple treatment options, some or all of which have significant historical use. For his study, Dr. Ghosh proposes to use prior patient-specific data and similar historical trials as sources of prior information.
"Modern medical research is emphasizing the unprecedented importance of translational medicine, where we want to expedite the discovery of new interventions. RCT is a gold standard for establishing the efficacy of drugs, biologics and medical devices," Dr. Ghosh said. "RCTs also play a central role to comparative effectiveness research because of its lack of selection bias and tight internal validity, often resulting in the best quality comparative effectiveness data. Any possible improvement of RCT design will have a significant overarching impact that cuts across almost all areas of medicine.
"As a biostatistician, I believe we can play a central role in that direction," he added. "Our proposed research will yield new statistical methods that will be patient-centric and will make better usage of available information, thus improving trial design and diversifying available treatment choices. This will result in improved patient outcome and enhance patient satisfaction."
PCORI is an independent, non-profit organization authorized by Congress in 2010 to fund comparative clinical effectiveness research that will provide patients, their caregivers, and clinicians with the evidence-based information needed to make better-informed health and healthcare decisions. PCORI is committed to seeking input from a broad range of stakeholders to guide its work.