Response to ARPA-H’s Ethical, Legal, and Social Implications RFI

Building H co-founders Steve Downs and Thomas Goetz respond to ARPA-H’s request for ideas for their ELSI initiative.

ARPA-H is to be commended for launching an Ethical, Legal, and Social Implications (ELSI) initiative and for its understanding that in order to avoid or mitigate unforeseen consequences, technological innovation must take place within a strong ELSI framework. Our response to this RFI focuses less on a process or framework for ensuring that individual technological innovations do not result in unexpected harms but rather on the societal implications of an advanced research agenda — as a whole — that focuses primarily on advancing technology and biomedical innovation.

Our starting point for this response is recognition that it is hard to be healthy in modern American society. We are living in an era of chronic disease epidemics — marked by record high levels of diabetes, obesity and mental health conditions. These conditions are largely driven by behaviors and exacerbated by socioeconomic status. Their rise, over several decades, has been accompanied by worsening diets, sedentary lifestyles, decreasing sleep, increasing social isolation and increasing inequality. (See Downs and Volpp, “Improving Health Outcomes in the US: Let’s Stop Relying on People Swimming Upstream” for an overview of these changes.)

Health behaviors take place in contexts. Decisions on what to eat, how to transport oneself, when to sleep and how to spend one’s time are constrained. Constrained by finances, constrained by time, constrained by the conditions of the neighborhood in which one lives, and constrained by the choices that are available. The products and services of everyday life — from industries such as food, housing, transportation and entertainment — collectively form what we refer to as the “product environment.” The product environment is a major driver of health behavior because it determines what is available, accessible, convenient and culturally normal. Successive technological innovations have brought us an automobile-centric transportation culture, a diet high in calories and ultra processed foods, “always on” 24–7 entertainment, and the increasing ability to perform many of life’s tasks without ever leaving the isolation of one’s home. While there is no arguing the benefits that have resulted from many of these innovations, they have also created a product environment that — in order for people to live healthy lives — must be overcome.

Health interventions can be conceptually located along a continuum from downstream to upstream. Most biomedical innovations are found downstream — seeking to cure or to treat diseases. Some, such as those aimed at early detection or vaccines to prevent infections, are further upstream. Innovations that aim to change the behaviors of individuals are even further upstream, yet still far from the source of many health behaviors, which is the product environment. Put quite simply, given all of the attention and resources paid to mitigating the consequences of an unhealthy environment and to helping individuals overcome that environment, why would we not seek to make that environment less harmful?

One consequence of a downstream, biomedical approach to improving public health is what the evolutionary biologist Daniel Lieberman has termed “dysevolution.” Dysevolution, explains Lieberman, is characterized by a vicious circle which starts from “an evolutionary mismatch that results from being inadequately adapted to a change in the body’s environment, either from too much, too little, or too novel a stimulus. Although we often treat the disease’s symptoms with varying degrees of success, we fail to or choose not to prevent the disease’s causes. When we pass on those environmental conditions to our children, we set in motion a feedback loop that allows the disease to persist and perhaps increase in prevalence from one generation to the next.”

Lieberman gives the example of Type II diabetes, which, through medical advances, we have turned into a manageable condition, albeit with much suffering and premature death. The success in making the disease tolerable has reduced societal pressure to tackle the problem at its roots — to address the unhealthy diets and sedentary lifestyles that have led to its explosion in prevalence. Instead, these behaviors have become more deeply rooted in our culture over time and thus harder to dislodge.

A focus on biomedical innovation can also reinforce a deeply embedded cultural mindset that hinders opportunities for upstream solutions. In their commentary, “Understanding the US Health Care Industry as a Commercial Determinant of Health,” Galea and van Schalkwyk note that “[e]ven though modern health care brings unprecedented benefits, the promotion of pharmacological and other individualized interventions entrenches a medicalized view of health that can undermine efforts to promote public health and equity.” This medicalized view locates responsibility for health behaviors within individuals and champions medical solutions. Chater and Lowenstein make an analogous argument in their critique of behaviorally-focused interventions, noting that the “i-frame” focus on individuals has deflected attention and support away from “s-frame” systemic solutions and effectively left the corporations whose products contribute to an unhealthy environment off the hook. This cultural mindset limits our imagination and results in an acceptance of the product environment — even of its progressive worsening.

A purposeful focus on upstream challenges and on the product environment in particular would create multiple opportunities for research and development, ranging from public health surveillance to systems science, and for a framework for understanding the potential impacts of technological innovations on lifestyles, and thus health.

• A national dashboard on chronic conditions and associated behaviors. An elevated focus on reversing the trend of increasing chronic conditions such as diabetes, obesity, cardiovascular disease and depression would require greater and more consistent measurement of their prevalence and of the common behaviors, including diet, physical activity, sleep and social engagement, that are known to be their drivers.
• Additional research and research synthesis on the links between health behaviors and diseases. Many of the links between certain behaviors and diseases have been well-established but additional research to develop more precise understandings of the dose-response relationships and mitigating factors would enable more sophisticated modeling of the societal costs and benefits of population-level shifts (positive or negative) in health behaviors.
• Research to understand the dynamics of the system of behavior and health. Health behaviors and diseases act in a dynamic system. Different behaviors can interact with one another: poor sleep, for example, can lead to unhealthy dietary choices; physical activity can result in better sleep. Likewise, conditions resulting from certain behaviors can influence other behaviors, as in the case of weight gain associated with poor diet leading to limitations on physical activity. Understanding the dynamics of this system and how they interact with other factors, such as genes and environment could lead to identification of promising leverage points.
• A framework for understanding the impacts of technological innovations on health. The public health community has a technique, known as Health Impact Assessment, that is frequently applied to proposed public policies or large public works projects. In it, the policy or project is analyzed from the perspective of how, through many different pathways, the health of the affected public could be affected — and, crucially — how they could be adapted to yield more positive health impact. There is an opportunity to build on this approach and apply it to different technological and biomedical pursuits, starting with explorations of the knock-on effects that could lead to changes in people’s health behaviors.

As we advocate for greater attention, focus and resources on the upstream factors — and principally the interplay between environment and health behaviors — we do not seek to denigrate the pursuit of biomedical advances. Upstream interventions are intrinsically no more moral or more practical than downstream interventions. Indeed, upstream conditions are slow to change and the reality is that downstream intervention will always be necessary to prevent suffering and premature loss of life. Examining the question of how to advance population health from an ELSI perspective leads to the recognition of the value of a society that lives under health-supporting conditions versus one in which the consequences of unhealthy conditions are mitigated by our technological brilliance. And so the question is one of emphasis: as we necessarily pursue advanced biomedical solutions, we must also dedicate our imagination and our resources to the endeavor of creating better, more equitable conditions for health.