The Fred Hutch Biostatistics Program hosts seminars featuring presentations by Hutch and outside scientists to share their latest developments and recent research. Each seminar includes an hour-long presentation and discussion during which speakers showcase their work and findings.
This seminar will be held on Zoom due to the COVID-19 pandemic.
A joint presentation by the Biostatistics Program and the Clinical Trials Affinity Group
“Introduction to SMART design for large and small N”
Sequential, multiple assignment, randomized trials (SMARTs) have been implemented in oncology, drug abuse, obesity, insomnia, mental health, smoking cessation and other areas. A SMART is a multi-stage trial design that allows for individuals to be randomized at two or more stages based on intermediate outcomes. SMARTs have primarily focused on informing the construction of dynamic treatment regimens (DTRs) or adaptive interventions. DTRs are evidence-based treatment guidelines where treatment can be altered over time based on the individual. Most SMARTs are conducted in large samples and analyzed using frequentist methods to explore potential delayed effects and treatment interactions over time to estimate and compare DTRs. More recently, Bayesian methods have been developed to apply SMART designs in rare diseases, or more generally, small samples to find the best overall treatment sharing information across stages. This talk will introduce SMART design for both large and small samples.