A Glimpse into the Future of Multiphase Optimization Strategy (MOST): A conversation with Dr. Jillian Strayhorn

Yohansa Fernández, PhD, Optimization of Behavioral and Biobehavioral Interventions (OBBI) SIG

The Multiphase Optimization Strategy (MOST) is an innovative multi-phase (preparation, optimization and evaluation) framework that uses highly efficient experiments to efficiently and systematically evaluate intervention components’, or component levels’, individual and combined effects. Dr. Jillian C. Strayhorn is a quantitative methodologist and Assistant Professor in the Department of Social and Behavioral Sciences at New York University School of Global Public Health. Her research focuses on developing and applying methods for complex multicriteria decision-making based on empirical data from optimization randomized control trials. Her professional mission is to facilitate more successful identification and advancement of optimized interventions that accomplish investigators' objectives. She holds a Ph.D. from Pennsylvania State University and a BA from Cornell University. Below, she shares her experiences and vision for MOST’s future:

What problem in optimization methodology is your research addressing?

Dr. Strayhorn: My research addresses what could be considered a "problem of opportunity" because though the empirical richness of the optimization trial opens so many possibilities for strategic, evidence-based decision-making, that decision-making can be challenging. I am working to advance methods that make decision-making in intervention optimization easier and more successful, such that the interventions that get selected as optimized interventions really are optimal (i.e., really are effective, really are implementable, and really do accomplish the strategic balance that matches the investigator's objective!).

What are you working on now related to MOST?

Dr. Strayhorn: I am really excited about our new work to incorporate equitability as a criterion in the selection of optimized interventions. This work acknowledges the important possibility that different alternative versions of an intervention (e.g., containing different components) will perform differently in terms of equitability--for some interventions, benefits might be concentrated among those who are already advantaged, whereas for other interventions, benefits may be more evenly distributed. Our goal is to advance methods that make it possible for empirical information about equitability to be carefully balanced with other criteria (like overall effectiveness and efficiency) in selecting an optimized intervention.

What are some common misconceptions regarding decision-making with MOST?

Dr. Strayhorn: 1) That decision-making has to be based on a single primary outcome. Our latest methodological advancements make it possible (and relatively straightforward, even) to base the selection of an optimized intervention on more than one outcome variable of interest. If performance on more than one outcome variable matters (to the investigator or other interested parties), then more than one outcome variable should be used for decision-making.

2) That incorporating cost in optimization decision-making has to be really hard. A great first step is to simply measure how much it costs to deliver each alternative version of the intervention (something that tends to happen anyway in an optimization trial!), then look at the balance between expected outcomes and delivery costs to see which interventions appear to be value efficient; we model a few different ways of going about this in our latest paper, out now in Health Psychology^1,2 .

3) The selection of an optimized intervention can be done in an hour-long meeting. Strategic balancing takes careful thought and time.

Where do you see optimization in the next 5 to 10 years?

Dr. Strayhorn: I see optimization becoming standard practice for the development of multicomponent interventions.

References:

1. Strayhorn, J. C., Cleland, C. M., Vanness, D. J., Wilton, L., Gwadz, M., & Collins, L. M. (2023, August 3). Using Decision Analysis for Intervention Value Efficiency to Select Optimized Interventions in the Multiphase Optimization Strategy. Health Psychology. Advance online publication. https://dx.doi.org/10.1037/hea0001318
2. Strayhorn, J. C., Collins, L. M., & Vanness, D. J. (2023, April 13). A Posterior Expected Value Approach to Decision-Making in the Multiphase Optimization Strategy for Intervention Science. Psychological Methods. Advance online publication. https://dx.doi.org/10.1037/met0000569