Optimizing Smoking Interventions with Dynamic Modeling and Control Engineering

Project: Research project

Project Details


Optimizing Smoking Interventions with Dynamic Modeling and Control Engineering Optimizing Smoking Interventions with Dynamic Modeling and Control Engineering Approximately $193 billion in economic costs and 443,000 premature deaths are attributed to cigarette smoking in the United States annually. Recently, decades of decreases in U.S. smoking rates have stalled, with approximately 20% of adults continuing to smoke. Smokings status as the leading cause of premature deaths and continued significance as a public health issue is due in part to the limited success of cessation therapies. For example, counseling is documented as having a successful quit rate as low as a 2%, and pharmacological approaches have a maximum success rate of around 35%. Clearly, the forces influencing the success or failure of smoking cessation interventions need to be better understood so that more effective therapies can be designed. Mobile health (mHealth) technologies have emerged as a cost-effective tool for enhancing our understanding of smoking behavior and developing novel interventions: in addition to being able to collect intensive longitudinal data (ILD; behavioral data collected over short time intervals), mHealth entails administering individualized interventions adapted to an abstaining smokers specific biological factors, previous behavior, and changing environment; such a personalized and contextualized mHealth intervention may take the form of tailored dosage reminders, counseling support, and more, as determined by an individuals previous behavior, real-time measurements of behavioral outcomes, and a range of other factors that influence intervention success. Rapid and judicious treatment adaptation, though, requires comprehensive models of behaviors dynamic response to an intervention. Such dynamic phenomena is not captured in traditional statistical models. Developed by engineers primarily for industrial applications, dynamical systems modeling and system identification techniques can be applied to ILD to describe an interventions time-varying effect, quantifying characteristics like magnitude, speed, and periodicity of the behavioral response to quitting. This project draws from dynamical systems modeling and system identification principles to use ILD collected through a mobile technology in a smoking cessation study of bupropion and counseling treatments in order to produce rich descriptions of the smoking cessation process. As part of a long term goal of optimizing smoking cessation interventions, these dynamic descriptions are then used in conjunction with control systems engineering principles to propose an adaptive smoking cessation intervention that could utilize mHealth technologies.
Effective start/end date1/1/1312/31/14


  • HHS: National Institutes of Health (NIH): $66,033.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.