Loading...
When and How Does it Work? Building Better Evidence for the Design of Mobile Health Interventions
Private Location (sign in to display)
View MapDetails
Pedja Klasnja
Associate Professor, School of Information
University of Michigan
Abstract:
In spite of the rapid growth in research in mobile health (mHealth), the evidence for the effectiveness of mHealth interventions is mixed and has not improved substantially over the last decade. In this talk, I propose that an important reason for this lack of progress lies in how mHealth interventions are evaluated. In both health sciences and HCI, mHealth interventions are typically evaluated as a package, making it difficult to understand how individual components and their designs are contributing to the system’s overall effectiveness. Consequently, it’s hard to know how to improve intervention design in the future. I will present several examples from our work of an alternative approach that focuses on gathering granular evidence for the operation of individual intervention components, with a particular emphasis on understanding how the design of those components impacts how—and how effectively—they work. I argue that such studies generate evidence that more directly supports decisions on how to design new interventions and enables the development of data-driven best practices for intervention design. Just as importantly, such studies can inform our theoretical understanding of the underlying processes through which mHealth interventions operate.
Bio:
Predrag "Pedja" Klasnja is an Associate Professor in the School of Information at the University of Michigan. He focuses on the design and optimization of novel mHealth technologies for health behavior change. He is particularly interested in the design and evaluation of just-in-time adaptive interventions (JITAIs), interventions that continuously adapt their functioning to provide optimal support to individuals as their needs and circumstances change. In addition to his intervention development work, Dr. Klasnja develops optimization methods for implementation science, with an emphasis on causal modeling of processes hypothesized to underlie the functioning of implementation strategies.