Systems Science

Systems Science and Public Investments in Systems Science Research

Systems science is a collection of analytic approaches that share in common the capacity to examine the big picture, that is, a problem and the context in which it is embedded, as well as the individual pieces that make up the larger picture (Midgely, 2003). In addition, these methods are designed to address complexity, that is, change over time, nonlinear relationships, bidirectional relationship (feedback loops), time-delayed effects, and emergent properties of the system—phenomena that are observed at the system level but cannot be causally linked to a specific individual component of the system (Mabry, Marcus, Clark, Leischow, & Mendez, 2010). Systems science methods do not substitute for traditional research methods, such as regression-based methods and reductionist approaches. For most research questions pursued in the behavioral and social sciences, a combination of traditional and systems science approaches is ideal.

Created by the U.S. Congress, OBSSR opened officially on July 1, 1995, with the mandate to stimulate behavioral and social sciences research throughout NIH and to integrate these areas of research more fully into other areas of the NIH health research enterprise, thereby improving our understanding, treatment, and prevention of disease. Since the formation of OBSSR, the field of behavioral and social science research has become increasingly aware of the complexity of the problems it is trying to address (Glass & McAtee, 2006). Simultaneously, due to technological advances and improvements in computational speed and power of desktop computers, modeling and simulation methods became more accessible to the average investigator. It was in this context that OBSSR sought to bring the methods for dealing with complexity that had their roots in physics, computer science, engineering, and other mathematically oriented disciplines to the behavioral and social sciences. Systems science was identified as one of four core themes for the OBSSR strategic plan (Mabry, Olster, Morgan, & Abrams, 2008).

Meanwhile, NIH was already making several large investments in systems science for cancer control (i.e., the Cancer Intervention and Surveillance Modeling Network, CISNET; http://cisnet.cancer.gov), infectious disease (MIDAS; http://www.nigms.nih.gov/Research/FeaturedPrograms/MIDAS), clinical and biomedical applications (Interagency Modeling and Analysis Group, IMAG; http://www.nibib.nih.gov/Research/IMAG), and systems biology (National Centers for Systems Biology; http://www.nigms.nih.gov/Research/FeaturedPrograms/SysBio).

In 2006, the American Journal of Public Health (AJPH) published a theme issue, Systems Thinking. For many in public health, this was the first time they had encountered systems science by any name. Particularly influential was John Sterman’s (2006) article, “Learning from Evidence in a Complex World” (cited 239 times to date), which explained why human beings naturally are prone to error when intervening in complex systems, and how the use of systems science methods, in particular mathematical modeling and simulation, can help overcome some of our human shortcomings. The AJPH 2006 theme issue stimulated a great deal of interest in systems science among public health practitioners, researchers, and others in related fields.

At the same time that systems science was capturing OBSSR’s attention, the Centers for Disease Control and Prevention (CDC) was also investing in systems science, and the two agencies began collaborating. Specifically, two of the authors and Co-Guest Editors of this issue (Drs. Mabry and Milstein) joined forces and collaborated to nurture the nascent field of systems science and health. To raise awareness among behavioral and social science investigators, program officers, program review staff, journal editors, and journal reviewers, they produced the 2007 Symposium on Systems Science and Health. This was a four-part lecture series (8 hours of content) designed to serve as a primer on systems science and made publicly available on demand through NIH’s videocasting website. Interest has been high, with lectures viewed more than 25,000 times by people in more than 60 countries.

Another way to raise awareness of systems science in the behavioral and social research community has been through publication. Elsewhere, we have discussed the utility of systems science for addressing research questions across a broad range of behavioral and social sciences topics, including developmental science (Urban, Mabry, Osgood, Okamoto, & Lich, in press; Urban, Osgood, & Mabry, 2011), prevention science (Lich, Ginexi, Osgood, & Mabry, 2013), big data (Mabry, 2011), the study of epidemics (Salathé et al., 2012), and tobacco control (Marcus, Leischow, Mabry, & Clark, 2010). Additional topics on which we expect forthcoming publications include dissemination and implementation science, public health innovation, and mental health services research.

In addition to the applied work presented in this volume, the application of systems science methods to specific research questions has included the Co-Guest Editors’ own work on the use of system dynamics modeling for informing community level policy decisions to reduce cardiovascular disease (Homer et al., 2008; Homer et al., 2010; Loyo et al., 2013) and Markov modeling to inform tobacco control policy (Abrams, Graham, Levy, Mabry, & Orleans 2010; Levy, Graham, Mabry, Abrams, & Orleans, 2010; Levy, Mabry, Graham, Orleans, & Abrams 2010a, 2010b). We have also reviewed a variety of modeling applications in obesity (Levy et al., 2011).

Training and Cross-Disciplinary Fertilization

Investing in a future workforce capable of using systems science methods was recognized as essential. Therefore, OBSSR developed the Institute for Systems Science in Health (ISSH) course (led by Drs. Mabry and Milstein). ISSH was designed to provide a learning environment stocked with the best teachers and applied examples in public health from guest lecturers. Each year, from 2009 to 2012, 45 trainees received hands-on training in one of three methods (agent-based modeling, system dynamics modeling, and network analysis) through a week-long in-residence training course.

In October 2007, the NIH Behavioral and Social Science Research Listserv (NIH BSSR-Systems Science Listserv) was created by Dr. Mabry (listowner) to inform people from diverse disciplinary backgrounds about current developments and has steadily grown to more than 1,000 subscribers. This has been an effective way to share relevant materials such as articles of interest, new funding opportunities, training and conference announcements, and job postings. Moreover, it has led to a common body of knowledge and understanding among researchers with greater and less familiarity with systems science methods and among scholars hailing from very different disciplines. For example, it is helpful for behavioral and social science investigators just learning about systems science as well as established engineers, physicists, and other systems scientists who may have recently turned their methodological skills to public health as an application area.

Establishing collaborations between formally trained systems scientists and behavioral and social scientists has been a formidable challenge. Overcoming it is crucial to the future growth and health of the field. In addition to the listserv, OBSSR has worked to establish cross-disciplinary connections through a variety of in-person meeting such as ISSH and by hosting the first International Conference on Social Computing, Behavioral Modeling, and Prediction (SBP10; http://sbp.asu.edu/sbp2010/sbp10.html), which included a special session devoted to cross-disciplinary relationship building.

Funding Initiatives

In addition to training and the listserv, funding opportunities were also crucial to systems science gaining a foothold in the behavioral and social sciences health community. In 2008, OBSSR, with participation from 13 other NIH Institutes, Centers, and Offices, released the first funding opportunity announcement (FOA) expressly requiring systems science methods to address behavioral and social research questions in health. Between 2009 and 2013, Using Systems Science Methodologies to Protect and Improve Population Health resulted in the funding of 32 grant applications totaling $6.3 million.

OBSSR has continued to stimulate the application of systems science to public health problems. In 2010, with participation from 10 other NIH Institutes and Centers, OBSSR issued, Social Network Analysis and Health (R01, R21). This FOA was a call to encourage basic research on social network analysis. Grant applications under this announcement were also reviewed by a special emphasis panel and resulted in 25 R01 applications and 24 R21 applications being funded ($16 million) to date.

In 2011, OBSSR issued a broader call for research projects, Systems Science and Health in the Behavioral and Social Sciences (R01, R21). With participation from 11 NIH Institutes and Centers, this FOA was intended to subsume and go beyond previous announcements by encouraging both basic and applied research. The first applications were funded in 2012, and six projects, totaling $1.8 million, have been funded to date. NIH intends to spend an additional $9.5 million in 2013 on the aforementioned systems science focused announcements, bringing the grand total of NIH investments in these FOAs to $35.4 million for 91 projects. This growth in funding is noteworthy, considering that prior to 2008 there were no NIH funding announcements for the behavioral and social sciences that were focused on systems science.

Other NIH efforts are worth noting. To name but a few, in 2007, OBSSR issued Behavioral and Social Science Research on Understanding and Reducing Health Disparities (R01, R21). This FOA encourages the application of systems science methods to addressing health disparities and was reissued in 2010 and 2013. In 2011, the National Institute on Alcohol Abuse and Alcoholism (NIAAA) issued Research on Alcohol-Related Public Policies such as Those Detailed in the Alcohol Policy Information System (R01, R03, R21), encouraging the application of systems science methods to understand and improve alcohol policy. In 2012, the National Institute of General Medical Sciences (NIGMS) issued a request for applications on Modeling Social Behavior (R01).

Under the auspices of the National Collaborative on Childhood Obesity Research (NCCOR), NIH, CDC, the U.S. Department of Agriculture (USDA), and the Robert Wood Johnson Foundation have invested in Envision. Led by OBSSR and the Eunice Kennedy Shriver National Institute on Child Health and Human Development (NICHD), Envision is a network of 11 teams formed in 2010 to facilitate collaboration and cross-fertilization among researchers developing mathematical and statistical models to address obesity (www.nccor.org/envision). NICHD and OBSSR have also made a 5-year, $16-million commitment to fund the John Hopkins Global Center on Childhood Obesity (JHGCCO; www.jhgcco.org), which focuses on the application of systems science methods to address obesity. OBSSR has also provided $3.8 million since 2009 to fund the Network on Inequality, Complexity, and Health (NICH; http://obssr.od.nih.gov/scientific_areas/social_culture_factors_in_health/health_disparities/nich.aspx), which will showcase its work at a national conference in February 2014.

This HE&B Supplement and Looking Forward

It takes time for the kind of investments made by NIH and CDC to pay dividends. Thus, this supplement issue of HE&B is appearing just as the field has begun to see the first grants completed and the trainees of the ISSH beginning to hit their stride. The field has gained a foothold in terms of NIH funding, and journals, such as HE&B and others, have become interested in systems science as a topic for their readership. But the field is also still in its infancy. As publications come out, readers will judge, many of them for the first time, the utility of the methods.

In conclusion, OBSSR is committed to continued support for systems science research in the future. OBSSR and 10 NIH Institutes and Centers recently released Short Courses on Innovative Methodologies for the Behavioral and Social Sciences (R25) in order to create additional training opportunities in systems science and other state-of-the-science research methods. We also look to institutions of higher education to build systems science programs. Early leaders in this area are emerging and include: the University of Michigan, Columbia University, the University of Pittsburgh, Washington University in St. Louis, Virginia Tech, University of Saskatchewan, Johns Hopkins University, and the University of Southern California. We hope that the systems science training available at institutions focusing on technology, like those of MIT, Stevens Institute of Technology, and Worchester Polytechnic Institute, will partner will schools of public health to produce research and training that features a stronger integration of systems science and health. Finally, we also hope that the funders and the field at large take the long view and that they do not dismiss systems science too soon or cease funding of systems science research before the work has fully matured and its promise fully realized.