Smarter health care

Robert Saunders and Mark D. Smith (“The Path to Continuously Learning Health Care,” Issues, Spring 2013) describe a path for building a smart health care system that provides care at a lower cost. Their vision is akin to the “learning health care system,” defined by the Institute of Medicine (IOM) as a place where “each patient care experience naturally reflects the best available evidence and, in turn, adds seamlessly to learning what works best in different circumstances.”

This goal requires close relationships between research and care delivery, creating bidirectional learning where (1) research drives practice— including the organization of care; and (2) practice, steeped in the challenges of achieving the triple aim, drives research. The implication? Research must increasingly occur within real-world delivery systems. We at Group Health are among several health plans, research organizations, and funders working nationally to build such systems.

One key lesson to date is that learning systems require true partnerships among researchers, providers, and the communities they serve. As such, activities must focus on issues that are relevant and valuable to all. The Patient Centered Outcomes Research Institute, a new federally funded nonprofit aimed at finding evidence for health care decisionmaking, wisely exemplifies this value by requiring patient and provider involvement in defining and designing projects. Such engagement encourages research that results in evidence meaningful to patients and their care. One approach we have used successfully for the past five years is our Partnership for Innovation Program, which solicits ideas from clinicians who are then paired with investigators to collaborate on pilot studies. Ideas are selected based on their potential to improve quality, reduce cost, and be spread. To date, we have funded 31 individual projects based on innovations proposed by front-line providers, including planned concurrent evaluations to judge value and potential for spread.

Another key lesson is that learning systems must be nimble. Especially with advances in information technology, including the production of continuously generated and increasingly granular data, research must adapt its stride. The traditional pace of developing ideas, writing and executing grants, writing papers, and moving on to the next project, with too little attention to dissemination, will no longer do. A new model develops evidence synchronous with delivery systems, often working in the virtuous cycle of design, implement, evaluate, and adjust. This approach can be applied to system innovation and the adoption of new technology and treatments, and operate in real time. Rapid deployment with concurrent evaluation also helps learning systems abandon hoped-for innovations that do not pan out, a problem endemic in U.S. health care.

As Saunders and Smith point out, advances in information technology may raise increased privacy concerns. An ethical approach requires that we balance regulatory requirements for obtaining individual consent for the use of de-identified, continuously generated data with the benefits that the public will reap from generalizable knowledge that results from research based on reasonable access to such information sources. Resolving uncertainty in this domain is urgent.

Working together in authentic partnerships, we can create the important changes needed to advance learning health care systems aimed at discovering ways to provide the best care at lower cost.


Executive Director

Group Health Research Institute

Seattle, Washington

Disturbing health data

Tom Plewes’ Real Numbers feature (“Shorter Lives, Poorer Health,” Issues, Spring 2013) offers a snapshot of the data compiled by a prestigious National Research Council panel. The data are so much more than statistics and numbers. They represent real human suffering, loss, and trauma. Whether from the gun violence that dominates the lives of too many communities today, or a result of the outrageous monthly costs of paying for medications to manage diabetes or hypertension for decades, or from the pain of grieving after the loss of a limb or a loved one; these data should trigger emotions that mobilize us to action. Unfortunately, emotions can paralyze more often than they mobilize. When I face our nation’s health disadvantage, I feel anger and extreme disappointment in our refusal to face the fact that this degree of suffering is needless. As a nation, we could shift from an overmedicalized culture to one that promotes health and well-being in our communities by deciding to create the “social conditions or determinants of health.” Becoming a nation that experiences better health outcomes would require ensuring access to quality, nutritious, affordable food for all, as well as more affordable housing, transportation, and education. Health would be supported by more equitable employment opportunities that pay a living wage. Reducing inequality would reduce associated stress and allostatic loads, thereby lowering the risk of many stress-related diseases.

The data alone should serve as a wake-up call for policymakers and for the public about the future health and viability of our nation. The fact that we spend more on medical care than our peer nations and still have worse health outcomes should prompt a re-examination of U.S. health-related policies and government expenditures. We spend less on social policies that support overall well-being than our peer nations who have significantly better health outcomes. The irony is that most Americans still believe that we have the best health care in the world. The idea of a “U.S. health disadvantage” is not widely known.

“Shorter Lives, Poorer Health” makes it clear that the U.S. health disadvantage is not limited to low-income or to minority communities alone. This nation, as a whole, is facing a major crisis in the premature loss of life and the disability of too many of our citizens under age 50. What does losing so many productive days and years of life mean to our future economy, to our national security, and to the U.S. capacity to compete on a global scale?

The full report calls for a comprehensive national communications campaign to raise the public’s awareness about our nation’s health disadvantage. Tom Plewes’ article is an important step toward broader communication of the report’s findings. If we fail to move toward creating healthier communities and providing the social infrastructures needed to promote and sustain health, the demands for health care will far outpace our care delivery systems’ capacity to respond. That is a frightening inevitability, indeed.


Vice President for Program Strategy

W. K. Kellogg Foundation

Battle Creek, Michigan


Fuel cell future

As pointed out by Noriko Behling in “Making Fuel Cells Work” (Issues, Spring 2013), “fuel cells are singularly remarkable in their potential” for efficiently converting chemical energy to electricity. Moreover, they are also unique in their ability to address all six of the key energy strategies of the recent U.S. Department of Energy (DOE) Quadrennial Technology Review. As such, they fully warrant the proposed National Fuel Cell Development Project to spur transformational innovation in this critical high-efficiency technology. However, if begun, the program must focus on driving the technology to higher system efficiency and lower capital cost with the currently available fueling infrastructure.

Unfortunately, fuel cells have been linked programmatically to a hydrogen economy, through the DOE’s EERE Hydrogen and Fuel Cells Program, creating three grand challenges: (1) develop reliable, low-cost, protonexchange—membrane fuel cells (PEMFCs); (2) develop portable H2 storage; and (3) develop and deploy an H2 production and distribution infrastructure. The tremendous infrastructural cost of creating the latter has perceptually relegated fuel cells to a “future technology” and resulted in a drastic reduction in funding by the DOE in favor of vehicle electrification.

In contrast, the lesser-known solid oxide fuel cells (SOFCs) are fuel-flexible, capable of operating on both conventional fuels (such as natural gas and gasoline) and future alternative fuels (such as H2 and biofuels) and thus only have one grand challenge: Reduce the operating temperature and cost. Unfortunately, as pointed out in the article, SOFCs have been programmatically relegated by the DOE Fossil Energy SECA Program to use in large-scale electric power production from coal, when in fact smaller-scale distributed generation with natural gas is a more attractive market for SOFC manufacturers.

This stovepiping within the DOE by fuel type and application has impeded technological progress and commercial success. The vast majority of funding has gone to address the three grand challenges of a hydrogen economy rather than focusing exclusively on advancing fuel cell technology itself. Moreover, although SOFCs are a technology that doesn’t require a hydrogen economy, the primary program that supports their development (SECA) had its budget for them zeroed out in the administration budget request for three years in a row.

Finally, the growing abundance of domestic natural gas provides tremendous potential for the U.S. economy, and it is incumbent on us to use this resource as efficiently as possible. Fuel cells are unique in this regard, and a National Fuel Cell Development Project should focus on advancing the technology that is most capable of operation on this game-changing fuel.


Director, University of Maryland Energy Research Center

William L. Crentz Centennial Chair in Energy Research

University of Maryland

College Park, Maryland

Climate engineering research

Jane C. S. Long and Dane Scott (“Vested Interests and Geoengineering Research,” Issues, Spring 2013) argue rightly that work must start now on the governance of research on climate engineering (CE), an issue that is likely to pose severe environmental and geopolitical challenges as the slow-motion debacle of climate change unfolds. After cataloging individual motives that may obstruct rational and impartial policymaking (and veering a little close to exhortations not to act on them), they advance several proposals for the governance of CE research. A few of these, such as transparency about research aims and results, public research funding, independent advisory bodies, and public consultation, are sensible and widely supported, albeit a little underspecified, when the devil is in the details. I focus on two proposals that hold more interest and more potential difficulty.

First, they propose that publicly funded research should generate no private intellectual property(IP) but do not propose prohibiting privately funded research, nor barring private IP arising from it. It is clear why they do not try to prohibit private funding: This would pose grave problems in defining and enforcing the boundary of what is prohibited. But their proposal could face serious difficulties if investors foresee large commercial opportunities in CE. Research might then split into two streams, with activities thought to promise valuable innovations funded privately and the rest funded publicly. Such a two-track research program, with strong selection pressure, may not advance the authors’ aims of keeping legitimate public control over key information about capabilities and risks and generating attendant public trust in decisions about whether, when, and how to develop or deploy CE capabilities.

Second, the authors struggle at a few points with the serious problem of how to organize research and craft incentives to attract the best minds but not tilt the playing field either to any particular approach or to a favorable bias toward CE overall. They worry, appropriately, that incentives will favor technical successes and positive assessments, when the societal need is for clear understanding of both efficacy and risks, in the context of the total response to climate change—including diligent efforts to root out every problem, limitation, and risk of any proposed CE approach.

They sketch two responses: adversarial assessment via some adaptation of the red team/blue team approach used for military technologies that could not be discussed in the open literature; and a shift in program structure, away from purely investigatordriven research toward a “collaborative, mission-driven” model. They advance this last point quite tentatively, expressing concern based on the legacy of environmental harms of past mission-driven research programs, particularly military ones.

Adversarial CE assessment appears to be a highly promising idea, but unlike its military antecedents is fully compatible with evaluating CE approaches and risks in the open literature. Once sufficiently well-posed questions or technical proposals can be teed up for such an exercise (a nontrivial matter), it only needs someone to frame the questions and fund and manage the exercise. Far from requiring secrecy, the combination of intensive debate within the exercise and wide expert and public review outside it is likely to both improve the results and help build public trust.

The question of investigator- versus mission-driven research is more difficult, as the authors’ evocative but imprecise term “collaborative, missiondriven” suggests they recognize. Criticizing past mission-driven programs for environmental damage does not quite capture the problems, because those programs’ missions were defined exclusively in terms of technical performance and national security. Their environmental harms do not mean that they failed, but that their mission failed to capture all that was at stake in what they were doing; showing the importance, and difficulty, of properly defining the mission.

What would be the mission be for a CE research program, and how should it be defined? It clearly would include advancing understanding of the effectiveness and risks of identified CE approaches, but should it also include improving CE methods to make them more effective and less risky? This is not as obvious as it sounds. It is clearly the case under some conditions: for example, if climate-change risks were so severe and imminent that the need to deploy effective CE was widely agreed on. But under other conditions, “improving” CE capabilities could be destructive; for example, if it were clear that an advance in capability would sharply undermine prospects for reaching an effective global mitigation agreement (but how would we know that?), or if advances in CE capability were to allow precise regional tuning of effects, risking increased international tensions or suspicions.

One clear implication is that the “mission” of a CE research program cannot be defined without high-level political and democratic input (and hopefully some wisdom), and that any such definition would have to iterate, perhaps repeatedly, between political and scientific inputs. In the absence of some political convergence on goals for climate policy overall, attempts to define a mission for CE research may be not just premature, but risky.


Dan and Rae Emmett Professor of Environmental Law

Faculty Co-Director, Emmett Center on Climate Change and the Environment

University of California Los Angeles School of Law

Los Angeles, California

Basic/applied research dichotomy

In his reaction to our paper “RIP: The Basic/Applied Research Dichotomy” (Issues, Winter 2013), Neal Lane (Forum, Spring 2013) raised a few important issues to which we are pleased to respond. Lane’s thoughtful reading of our paper resulted in his agreement with our critique of the linear model and the discontinuity it represents with the actual practice of research. At the same time, he is rightly cautious about possible misinterpretations of our arguments for the restructuring of long-standing federal policy. We agree with Lane about the need for caution and provide two suggestions on how the discovery-invention model might be properly used in improving current federal policy.

We are in agreement with Lane about the necessity of funding fundamental studies in the natural and social sciences. The funding of such work is fully compatible with our understanding of the public interest, and such research exemplifies the best of the long-term viewpoint we advocate. It is for this reason that “discovery” is placed on an equal footing with “invention” in our model. Indeed, our empirical example of the 1998 Nobel Prize in Physics given for the fractional quantum Hall effect exemplifies this broader understanding of the public interest. It is our hope that a fuller consideration of discovery and invention will actually lead to increased funding, in both the natural sciences and engineering, than is currently the case.

Lane’s characterization of the discovery-invention model as an outcomesoriented model does not fully reflect our arguments. In our view, one of the primary failings of the basic/applied model is its assumed linearity that creates a false hierarchy between so-called basic and applied research. The discovery-invention cycle provides a much needed correction. As presented in our examples, discovery-research is about creating new knowledge about the natural world. Invention-research is about constructing new artifacts. Our model places both types of activity on an equal footing, thereby incorporating the inherent bidirectionality in the flow and development of knowledge. When it comes to setting priorities in federal science and technology (S&T) policy, use of the discovery-invention model should broaden the relevant decision matrix to incorporate possible future technologies and the development of novel processes and engineered materials, in addition to the importance of increasing knowledge about nature. It is our hope that the use of the discovery-invention cycle will lead to more integrative decisionmaking in which motivational goals are but one of many considerations.

We hope that consideration of our model will lead to the kind of initial experimentation that Lane calls for and that the uptake of some of these ideas will lead to a rethink of many of the dysfunctional elements of the current national S&T infrastructure, while preserving the very best aspects of publically funded science and engineering.


Benjamin Peirce Professor of Technology
and Public Policy


Postdoctoral Research Fellow

John F. Kennedy School of Government

Harvard University

Cambridge, Massachusetts

Renewable energy puzzle

Here in New Mexico in the 1970s, before the expansion of government programs and subsidies for renewable energy, there was great interest in the Sun and in the adobe brick used in construction for passive heating and cooling.

Unfortunately, we govern ourselves with great concern for financial elites but little for the public. The elites show less and less interest in the many now almost extinct nonelectric uses of the Sun. We have few clotheslines and need electric lights on all day in our shopping centers despite our usually sunny weather.

While ignoring traditional nonelectric uses of the Sun, we continue subsidizing solar power plants and wind generators. I wonder if Michael Levi (“The Hidden Risks of Energy Innovation,” Issues, Winter 2013) doesn’t find this same thing where he lives.


Zomeworks Corporation

Albuquerque, New Mexico

Cite this Article

“Forum.” Issues in Science and Technology 29, no. 4 (Summer 2013).

Vol. XXIX, No. 4, Summer 2013