Control Engineering and Related Systems Approaches for Improving Behavioral Health

*Presentations and Links to Additional Information Are Now Available*

A Special Session held as part of the 2009 American Control Conference (ACC 2009), Hyatt Regency St. Louis Riverfront Hotel, June 10-12, 2009;

Organizers:

Daniel E. Rivera, Department of Chemical Engineering, Arizona State University; Fahmida Chowdhury, Cross-Directorate Activities Social, Behavioral, and Economic Sciences, National Science Foundation

Time: Thursday June 11, 2009; 6:30 - 8:00 p.m.
Location: Mills Studio 3 (Hyatt Regency St. Louis Riverfront Hotel)

The goal of this special session is to describe emerging approaches and research opportunities for control engineering in a developing research topic of important societal significance. Specifically, we explore how control systems and related approaches from systems science can be applied to the prevention and treatment of chronic behavioral disorders; these include drug and alcohol abuse, depression, HIV/AIDS, cancer, diabetes, obesity, cardiovascular health, and aging. Effective management of chronic behavioral disorders has major impact on public health, requiring hierarchical, multi-stage decision-making of prevention and treatment components over time. Conceptually, such time-varying interventions represent forms of closed-loop control systems where intervention dosages (i.e., manipulated variables) are determined by decision rules (i.e., controllers) based on the values of a participant's key characteristics (i.e., tailoring variables or controlled variables). Consequently, drawing from control engineering has the potential to significantly inform the analysis, design, and implementation of novel behavioral interventions, leading to improved adherence, better management of limited resources, a reduction of negative effects, and overall more effective interventions. Advances in this field involve significant modeling and computational challenges that need to be addressed. Novel decision rules will draw not only from control engineering, but also from the fields of artificial intelligence, statistics, and computer science. Practical solutions will involve individuals from diverse disciplines (e.g., psychologists, physicians, statisticians, computer scientists, applied mathematicians, and engineers). The session brings together a control engineer (Rivera), a quantitative psychologist (Collins), a statistician (Murphy) and a computer scientist (Pineau) with relevant program officers from NSF (Chowdhury) and NIH (Mabry) to address challenges and opportunities in this field. The paper titles and authors are summarized below: (* denotes the corresponding author and presenter)

  1. Engineering Control Approaches for the Design and Analysis of Adaptive Behavioral Interventions, Daniel E. Rivera* (ASU) and Linda M. Collins (Penn State).
  2. Using Clinical Trial Data to Construct Behavioral and Medication Policies, Susan A. Murphy* (Michigan) and Joelle Pineau (McGill)
  3. Systems Science and Health at NIH and Beyond: Areas of Interest and Funding Opportunities, Patty Mabry* (NIH)
  4. Discussion session (led by Fahmida Chowdhury, NSF)

Paper Synopses

  1. Engineering Control Approaches for the Design and Analysis of Adaptive Behavioral Interventions

    Daniel E. Rivera
    Department of Chemical Engineering
    Arizona State University

    Linda M. Collins
    The Methodology Center and Department of Human Development and Family Studies
    Penn State University

    The talk will discuss how control engineering concepts can be applied to optimize adaptive interventions for prevention and treatment of chronic, relapsing behavioral disorders such as substance abuse, mental illness, and obesity. Adaptive interventions are feedback systems that individualize therapy via decision rules that assign dosages and forms of treatment over time; consequently drawing from principles in control engineering can significantly inform the analysis, design, and implementation of adaptive interventions, leading to improved adherence, better management of limited resources, a reduction of negative effects, and overall more effective interventions. The application of concepts from Internal Model Control, Model Predictive Control, and system identification will be discussed. A simulated example based on Fast Track, a real-life preventive intervention designed to reduce conduct disorder in at-risk children, will be presented as an illustration.

    Collins, L.M., S.A. Murphy, and K.L. Bierman, “A conceptual framework for adaptive preventive interventions,” Prevention Science, 5, No. 3, pgs. 185-196, Sept., 2004.

    Rivera, D.E., M.D. Pew, and L.M. Collins, “Using engineering control principles to inform the design of adaptive interventions: a conceptual introduction,” Drug and Alcohol Dependence, Special Issue on Adaptive Treatment Strategies, Vol. 88, Supplement 2, May 2007, Pages S31-S40.

    Rivera, D.E., M.D. Pew, L.M. Collins, and S.A. Murphy, “Engineering control approaches for the design and analysis of adaptive, time-varying interventions,” Technical Report 05-73, The Methodology Center, Penn State University; available electronically from

  2. Using Clinical Trial Data to Construct Policies for Guiding Clinical Decision Making

    Presenter: Susan Murphy
    Departments of Statistics and Psychiatry
    University of Michigan

    Joelle Pineau
    Department of Computer Science
    McGill University

    Constructing policies for managing behavioral and mental disorders presents a number of challenges to control engineering. First the clinical trial data sets are quite small, second the system dynamics are incompletely understood and third clinically acceptable policies should avoid falsely specifying one action as best when in reality there is little evidence in the data to select one action over another. These challenges motivate the development of algorithms and methods that are robust to the incomplete knowledge of the system dynamics, and that provide measures of confidence for the value of estimated policies. We discuss present approaches to addressing these challenges.

    SAMSI Summer 2007 Program on Challenges in Dynamic Treatment Regimes and Multistage Decision-Making, June 18-29, 2007

    Drug and Alcohol Dependence Supplemental Issue on Customizing Treatment to the Patient: Adaptive Treatment Strategies

  3. Systems Science and Health at NIH and Beyond: Areas of Interest and Funding Opportunities.

    Patricia L. Mabry, Ph.D.
    Office of Behavioral and Social Sciences Research (OBSSR)
    National Institutes of Health (NIH)

    Located in the Office of the Director at NIH, the Office of Behavioral and Social Sciences Research (OBSSR) at the National Institutes of Health (NIH) is well situated to work across the 27 Institutes and Centers that comprise NIH and with other federal agencies (especially the Centers for Disease Control and Prevention) to stimulate and nurture an under explored field of inquiry: the area at the intersection of behavioral and social sciences with systems science and health. The presentation will provide a brief history of how OBSSR came to see the value of systems science, a description of some of the complex problems that threaten the public's health and examples of how some of these have been addressed with systems science methodologies. Specific funding opportunities that feature systems science methodologies will be presented along with a look at the future of this new and growing area.

    2009 SSHD Biennial Meeting: October 18 - 20, 2009 Call for Proposals

    Videocasts of the 2007 Symposia Series on Systems Science and Health

    To stay apprised of new Funding Opportunity Announcements, join the Behavioral and Social Science-Systems Science Listserv. Send email to Patty Mabry to join.