Applying the Social and Behavioral Sciences to Policy and Practice

Health Care That’s Not for Dummies

KENNETH PREWITT

ROBERT HAUSER

Applying the Social and Behavioral Sciences to Policy and Practice

In areas ranging from health care to national security, understanding the human dimension is essential to achieving national goals.

Smart electric meters and the smart grid are innovations in the delivery of electrical services to homes and businesses that can, in principle, permit both consumers and suppliers to save energy, exercise greater control over the uses of energy, and in some cases feed locally generated electricity back into the grid. Europe is adopting this technology, which can play a vital role in achieving environmental goals. In the United States, however, there is resistance prompted by understandable concerns about data privacy and physical security as well as fears about health effects of wireless telecommunication. Behavioral research has helped overcome this resistance through new methods of data collection and distribution that protect individual privacy while preserving the energysaving and informational value of smart metering.

Helping people and their technologies work well together is only one of many ways in which behavioral and social sciences contribute to national well-being. Researchers in these sciences have contributed to public policies in areas ranging from defense and national security to health care and education. However, these contributions are not always apparent to the broader science community and the public or in policy settings.

This lack of visibility risks underutilizing the nation’s scientific capabilities as new challenges emerge. For example, as the elderly grow in their proportion of the population — a demographic trajectory we can foresee from surveys designed by social scientists—there will be demand for changes in work arrangements, retirement systems, and health care services, as well as in the design of transportation systems, homes, and neighborhoods. Effective, efficient, and satisfying accommodations to these changes can benefit from the application of the tools of social and behavioral science.

There are also many examples of contributions where the social and behavioral sciences intersect with the physical sciences and engineering. When there is a bridge to be built, we turn to engineers; when there is a vaccine needed, we turn to biochemists. But in determining the relative efficiency of paying for the bridge with a toll or a tax, we turn to economists; and in understanding public acceptance of the vaccine, we turn to social psychologists and public health experts. Society has many challenges that require attention by engineering and the physical and biological sciences, a large number of which also have a human and social dimension.

It is within the reach of political science, decision theory, cognitive psychology, behavioral economics, and systems theory to focus analytically on the “use of science as evidence in public policy.”

In recognition of the role of sciences that investigate these dimensions, the National Research Council has launched a new initiative, Social and Behavioral Sciences in Action (SBSIA). This initiative presents key contributions that the social and behavioral sciences have made to policy and society; it will also highlight emerging areas requiring fresh attention. This article, which draws from presentations at the first SBSIA symposium in September 2012, provides details about past successes and future challenges.

Better health care

One of the most socially beneficial but often unnoticed roles the social and behavioral sciences play is facilitating health care research and practice. For example, demographers and other social scientists were instrumental in developing and introducing modern family planning methods and practices around the globe; for example, by illuminating the key role social networks play in people’s decisions to use contraception. Dartmouth’s Atlas of Health Care has collected, analyzed, and publicized geographic variations in health care expenditures in the United States. Atlas researchers found that the rate of hip replacements was four times higher in some U.S. regions than in others and that the rate of shoulder replacements was 10 times higher in some regions. Such data helped policymakers and health care leaders identify opportunities to reduce health care spending while improving the quality of care.

Cultural as well as economic barriers impede health improvements. Rita Colwell, former director of the U.S. National Science Foundation, led a three-year study to evaluate a new method for combating cholera in Bangladesh. That nation faces cholera epidemics every spring and fall, and the disease spreads when people drink contaminated water. Colwell’s team sought to evaluate whether villagers could lower the incidence of cholera by pouring their drinking and cooking water through old folded sari cloth to filter out contaminants. With the help of statisticians and social scientists, they conducted a study that involved 150,000 individuals in 50 villages, including a comparison group that did not use the sari filters. The researchers found that the group that used the filters decreased their rate of cholera by 50%, and that those who did develop the disease generally got milder cases.

The social sciences’ important role became clear early in the study, when they helped the researchers scale an initial hurdle, said Colwell. When the team first submitted the proposal for the study, one of the reviewers rejected it, saying that Indian men would never drink water that had been filtered through old “unclean” sari cloth. The research team, however, found that in fact the men were already using sari cloth to filter flies out of their beer. “We really needed to understand the cultural practices,” explained Colwell.

Sociologists guided the team’s introduction into local communities, advising them on how to present the study. The sociologists also helped design the questionnaire, ensuring that specific questions were framed in culturally acceptable ways. What would have happened if the research team had not included social scientists? “I wouldn’t have had the entre to the villages on such a grand scale—150,000 individuals in 50 villages,” said Colwell. “It would not have been possible. And it would’ve been tragic, because this is an opportunity to take very advanced technology, science, engineering, and [use] those findings in a very practical way to help people.”

An area of health care where the social and behavioral sciences are badly needed but have not yet been tapped is in reforming the culture of medical practice, according to Lucian Leape of the Harvard School of Public Health. In particular, Leape argued, a dysfunctional medical culture is undermining patient safety; a field that for many in the profession really began with the Institute of Medicine’s (IOM) 1999 report To Err is Human. Before that report, patient safety was seen as an individual performance issue. “If you didn’t perform well, then it was because you were lazy or careless or ignorant, and if you made mistakes we would punish you,” said Leape. “The Institute of Medicine said, it’s time to change the paradigm, to recognize that errors are caused by bad systems, not bad people, and let’s get to work changing the systems.”

Although considerable work has been done since, with progress in fits and starts, it is hard to prove that efforts to reduce errors have had more than marginal impact, said Leape. It’s apparent that changing systems is very difficult, he continued, and that health care is not a learning culture. Medical education emphasizes individual performance rather than teamwork, and medical culture is hierarchical and siloed. “Changing that culture has got to be the ultimate social science challenge,” he said. For example, a study in eight hospitals around the world found that the use of surgical checklists reduced complications by 40% and deaths by 50%. But it has been hard to achieve these good results on a broader scale; in every hospital there is at least one physician who resists such checklists, feeling individually exempt. ”The simple way to say it is that we don’t do teams very well,” said Leape, “and that’s clearly a social science issue.”

Of course, we also know that the origins of many health conditions are social in nature. In explanations of premature death in the United States, observes Steven Schroeder, social, behavioral, and environmental factors far outweigh genetic factors or problems within the health care system. A 2013 report of the National Research Council (NRC) and IOM has shown that the United States lags well behind other developed nations in morbidity and mortality in almost every respect for individuals less than 75 years old and that social, environmental, and behavioral factors are largely responsible for this disparity. Schroeder found that 60% of premature deaths are attributable to social circumstances, environmental exposures, or behavior patterns, whereas 30% are due to genetic defects and only 10% to the health care system. Innovative longitudinal biosocial surveys of health and aging in the United States and other nations promise to add substantially to knowledge about social and behavioral factors in health and longevity.

Enhanced national security and public safety

Another context in which a better understanding of human behavior can save lives is presented in the work of national security psychologist Robert Fein, who has studied the preattack behavior of assassins and school shooters. In the late 1980s, after four cases in which individuals had tried to attack someone under Secret Service protection, the agency’s director asked Fein and a Secret Service agent to conduct a study that would provide relevant evidence to Secret Service agents about would-be attackers. The researchers examined the pre-attack thinking and behavior of 83 people who had tried to attack a public target.

What they found debunked some commonly held myths about would-be assassins, especially the misleading assumption that there is a standard profile of an assassin. In fact, attackers vary by age, personal background, and other characteristics. Another common myth is that mental illness is a direct cause of assassination attempts. In reality it rarely plays a critical role; when agents try to assess a threat, it may be far more important to ask how organized a suspect is than to ask whether he or she is mentally ill, the researchers concluded. They also found that those who pose threats may not make threats. “No presidential attacker has ever communicated a direct threat to the president, to law enforcement, or to the Secret Service,” said Fein. His findings were disseminated to Secret Service agents and law enforcement agencies around the country and led to changes in investigative activities designed to prevent attacks.

After several high-profile school shootings rattled the nation in the 1990s, the Secretary of Education asked the Secret Service to conduct a similar study of the shooters’ pre-attack behavior, and the agency once again turned to Fein and his colleagues. Using an approach similar to their study on assassins, they examined the thinking and behavior of 41 young people involved in attacks at school. As with the assassins, they found that there is no accurate, useful profile that can predict school violence. However, they did find that most attackers had come to the attention of adults before the attack for problems they were having;plus, many attackers felt bullied, persecuted, or injured by others. In addition, most attackers let others—usually other kids—know that they were planning attacks, but those who knew rarely told adults. These and other findings had a broad impact when Fein and his colleagues developed a series of video teaching tapes and presented them at meetings of school and law enforcement professionals across the country.

Many other areas of the justice system and national security policy have benefited from greater application of behavioral and social science research. For example, two 2011 NRC reports showed that the adoption of behavioral and social science findings and methods could improve the quality and effective use of national intelligence analysis. Another NRC report demonstrates that widely used industrial methods of quality control could speed development, reduce costs, and increase the effectiveness of new defense systems. Experimental and observational research on eyewitness identification have identified unreliability and suggestibility in erroneous identification, and these findings are influencing the conduct of criminal investigations. A 2012 NRC report has identified ways in which new insights into individual development during adolescence may inform improvements in the juvenile justice system. At the same time, social scientific analysis has identified some areas in which it cannot provide definitive findings; for example, with respect to the deterrent effect of capital punishment.

Safer car travel

The goal of human factors engineering is developing technologies in a way that keeps humans at the center of the design process, “making the human-technology marriage work,” as human factors engineer John Lee, University of Wisconsin, Madison, described his field. Lee’s work focuses on finding ways to help people navigate a technological environment many enter every day: their cars. He studies driver distraction, an increasingly serious problem on U.S. roadways. In 2010, for example, 18% of injury-causing crashes, 3,092 deaths, and 416,000 injuries in the United States were attributed to distracted driving.

The difficulty of addressing this problem became evident in one of Lee’s studies, which tested a device designed to make it easier and less distracting to select songs on an MP3 player while driving. The study found that, contrary to expectations, drivers glanced away from the road even longer when selecting songs using the MP3 controller designed to reduce distraction than with the MP3 player alone; and the longer a driver glances away from the road, the more dangerous the situation becomes. “Engineers don’t always get it right,” Lee said.

How can the driver distraction problem be solved? One possible solution is attentive cars, which have sensors to spot obstacles in the environment and warn drivers about them, said Lee. Another option may be sensors that detect when drivers look away and cue them to get their eyes back on the road: “providing feedback so that drivers don’t have to die in order to realize that texting while driving is dangerous,” said Lee. By giving drivers reminders in the moment, he noted, attentive technology can complement the limits of regulations and education.

Human systems engineering has wide-ranging applications. Another context in which people may increasingly need help navigating technological interfaces is education, especially given the sudden emergence and rapid growth of massive open online courses (MOOCs). These courses, taught by professors at MIT, Stanford, and many other prestigious universities, are available free to any student anywhere in the world with Internet access, and they are radically extending the accessibility of higher education. What software interfaces are most useful for these courses? How should materials be presented in different areas of study for different populations? And how can valid methods of assessment be designed and implemented on a large scale with reliable security, given that MOOCs are increasingly taken for credit? Answering questions such as these will depend on findings and insights from the social and behavioral sciences.

The rest of the iceberg

These few though significant examples of the successful use of social and behavioral science to guide policies that work could easily be extended, reaching well into the multiple hundreds if we incorporate examples stretching back across more than a century: standardized testing that solved military recruitment challenges in World War I; National Accounts and sample surveys that helped to meet challenges facing the nation during the Great Depression; content analysis and area studies that contributed to national security in World War II and the Cold War; theoretical understanding of the demographic transition that was used to design global family planning that slowed population growth in the 1950s; large social experiments that revealed unintended consequences and weaknesses in social programs being implemented in the 1960s; and major research findings in early childhood learning that led to innovative parenting and schooling practices.

Despite these successes and opportunities, in one important regard the social and behavioral sciences are offering less guidance to the government and society than they could. There is not a robust understanding of the conditions that lead to the effective use of all of the sciences in the policy process: the physical and biological sciences and engineering as well as the behavioral and social sciences. Creating this understanding is a unique responsibility of the behavioral and social sciences. As documented in a recent NRC report, the use of science is itself a social phenomenon, one that should become a focus of scientific attention. Although researchers and policymakers have given much attention to randomized field trials, evidence-based-policy, and translational research, understanding of the uses of science is still too anecdotal. It is within the reach of political science, decision theory, cognitive psychology, behavioral economics, and systems theory to focus analytically on the “use of science as evidence in public policy.” This will require a much more ambitious program of research than has yet been mounted. Thus, improvement in the dissemination, communication, and use of scientific evidence looms large among the many challenges and opportunities that face the social and behavioral sciences.

“The imperative of modern society demands more attention to the social and behavioral sciences,” said National Academy of Sciences President Ralph Cicerone at the September conference. “The ambition is there, the methods are getting better, the successes are there, but not many people are hearing about it.”

Recommended reading

H. Russell Bernard, “The Science in Social Science,” Proceedings of the National Academy of Sciences 109 (2012): 20796–20799.

Axel Börsch-Supan, Karsten Hank, and Hendrik Jürges,.. “A New Comprehensive and International View on Ageing: Introducing the ‘Survey of Health, Ageing and Retirement in Europe’,” European Journal of Ageing 2 (2005): 245–253.

Committee on Assessing Juvenile Justice Reform, Committee on Law and Justice, Division of Behavioral and Social Sciences and Education, and National Research Council, Reforming Juvenile Justice: A Developmental Approach, R. J. Bonnie, B. M. Chemers, and J. Schuck, Eds. (Washington, DC: National Academies Press, 2012).

Committee on Behavioral and Social Science Research to Improve Intelligence Analysis for National Security and National Research Council, Intelligence Analysis for Tomorrow: Advances from the Behavioral and Social Sciences (Washington, DC: National Academies Press, 2011).

Intelligence Analysis: Behavioral and Social Scientific Foundations, B. Fischhoff and C. Chauvin, Eds. (Washington, DC: National Academies Press, 2011).

Committee on Deterrence and the Death Penalty, Committee on Law and Justice, Division of Behavioral and Social Sciences and Education, and National Research Council, Deterrence and the Death Penalty, D. Nagin and J. Pepper, Eds. (Washington, DC: National Academies Press, 2012).

Committee on National Statistics, Board on Army Science and Technology, and National Research Council, Industrial Methods for the Effective Development and Testing of Defense Systems (Washington, DC: National Academies Press, 2012).

Committee on the Use of Social Science Knowledge in Public Policy, Division of Behavioral and Social Sciences and Education, and National Research Council, Using Science as Evidence in Public Policy, K. Prewitt, T. A. Schwandt, and M. L. Straf, Eds. (Washington, DC: National Academies Press, 2012).

Elliot S. Fisher, J. E. Bell, I. M. Tomek, A. R. Esty, D. C. Goodman, and K. Bronner, “Trends and Regional Variation in Hip, Knee, and Shoulder Replacement” in Dartmouth Atlas Surgery Report (Hanover, NH: Dartmouth Institute for Health Policy and Clinical Practice, 2010).

Ronald Freedman, “The Contribution of Social Science Research to Population Policy and Family Planning Program Effectiveness,” Studies in Family Planning (1987): 57–82.

Robert M. Hauser, and David Weir, “Recent Developments in Longitudinal Studies of Aging in the United States,” Demography 47 (2010): 111–130.

Beth Karlin, “Public Acceptance of Smart Meters: Integrating Psychology and Practice,” in ACEEE Summer Study on Energy Efficiency in Buildings (Pacific Grove, CA, 2012): 7-102–7-113.

Elizabeth F. Loftus, Eyewitness Testimony (Cambridge, MA: Harvard University Press, 1996).

Patrick McDaniel, and Stephen McLaughlin, “Security and Privacy Challenges in the Smart Grid,” Security & Privacy, IEEE 7 (2009): 75–77.

Panel on Understanding Cross-National Health Differences Among High-Income Countries, Division of Behavioral and Social Sciences and Education, Institute of Medicine, and National Research Council, U.S. Health in International Perspective: Shorter Lives, Poorer Health, S. H. Woolf and L. Aron, Eds. (Washington, DC: National Academies Press, 2013).

Steven A. Schroeder, “We Can Do Better—Improving the Health of the American People,” New England Journal of Medicine 357 (2007): 1221–1228.

Philipp Wunderlich, Daniel Veit, and Saonee Sarker, “Examination of the Determinants of Smart Meter Adoption: A User Perspective,” in ICIS 2012 Conference Proceedings (2012).

Kenneth Prewitt, the Carnegie Professor of Public Affairs at Columbia University, is chair of the advisory committee and Robert Hauser () is executive director of the National Research Council’s Division of Social and Behavioral Sciences and Education.