On Mechanisms vs. Foundations.

Before I begin, I want to take a moment to congratulate the entire School community for an unprecedented achievement: The School of Public Health now ranks among the top 10 public health schools in the country, according to the latest U.S. News & World Report graduate school rankings released last Tuesday. This new ranking represents the School’s third consecutive rise up the U.S. News list, which assesses the quality of schools accredited by the Council on Education for Public Health. I firmly believe that this achievement reflects both our commitment to be a global leader in public health scholarship and our deep commitment to train the next generation of the public health workforce with a real-world approach to public health. I am truly humbled to have joined this institution at such an ascendant time. This new recognition, of course, also encourages and challenges us to do better, to ever more clearly engage our responsibility as a leading school of public health, and to further innovate in our scholarship, education, and translation as we engage the critical public health issues of today, and of tomorrow.

On Mechanisms vs. Foundations

Today I wanted to comment on one of the challenges in public health scholarship that I have struggled with during my career. At core, the challenge is as follows: Should we, in academic public health, focus on the study of the foundational drivers of population health—on the factors that we know influence the conditions that make people healthy—or should we focus on the mechanisms that explain how these foundational drivers shape the health of populations? And, if the answer were to be “somewhere in the middle,” what is the relative weight we should be giving our scholarship on each of these two areas?

There are ample arguments to be made for a focus on either of these two approaches. Insofar as public health is concerned with the conditions that make people healthy, it is then rational that public health scholarship should be concerned with understanding those conditions. This has led to a burgeoning concern with these forces, sometimes called fundamental causes and including any number of factors ranging from poverty to income inequality, social structures, and policies. One of my books, Macrosocial Determinants of Population Health, includes chapters on governance, corporate practices, taxation, culture, patent law, migration, and the mass media, among other factors, that can be foundational determinants of the health of populations. Our research group has also, in other work, quantified the potential contribution of some of these foundational factors to population health, suggesting that approximately 245,000 deaths in the United States in 2000 were attributable to low education, 176,000 to racial segregation, 162,000 to low social support, 133,000 to individual-level poverty, 119,000 to income inequality, and 39,000 to area-level poverty. I shall leave the discussion of the methodological challenges behind this work to another comment, but here I note that a public health that is informed by an appreciation of the centrality of foundational determinants then might focus its attention on studying these factors, how they influence population health, and what we might do to mitigate their adverse consequences.

There is also plenty to agitate for public health scholarship that focuses on more “downstream” factors—on the study of the mechanisms that explain how these foundational factors get “under the skin.” At core, the relationship between large-scale social forces has to be mediated by more downstream factors. Saying that poverty is associated with poor health without thought to mechanistic intermediates is akin to saying poverty does not, in and of itself, result in the production of poor health directly. Poverty is associated with limited access to health care resources, with greater experience of a broad range of adversities including violence, with greater psychological burden that may modify immune vulnerability, and with more limited and fragmented social supports, all of which are more proximally related to the production of health. Understanding these mechanisms therefore provides us both with an appreciation of how foundational determinants influence health and can also provide guidance on easier ways to intervene. It is, perhaps, easier to ensure that all populations have access to health care than to change the profile of poverty in this country. Aiming to inform the science, we have, in our work, explored a range of biological mechanisms that explain how exogenous exposures “get under the skin.”

But is the observation that action on mechanistic factors is easier than work on fundamental factors sufficient justification for a focus on mechanisms? In some respects, the easy answer is that surely they both matter—that we should understand both core drivers and mechanisms. Certainly there is a large literature proposing conceptual frameworks that tie the “upstream” to the “downstream,” satisfyingly demonstrating how foundational factors and downstream factors all matter to the production of health. I find this particular illustration useful and simplifying as a comprehensive framework that brings various factors together.

That is all fine and good, but, in a world of finite resources, including our own bandwidth, is an academic public health focus on upstream factors equivalent to one of downstream factors? Does it all matter equally? Let us consider as an example the subject of a previous Dean’s Note: firearm violence in the United States. I have argued before that death by firearm is among the most central public health threats currently facing the US. One response to the threat has been to restrict firearms to individuals with mental illness. In fact, firearm-disqualifying mental health adjudications increased from 7 percent of firearm disqualifications in 2007 to 28 percent in 2013. When we step back from the US and compare across countries, however, we see that by comparison, Canada has a similar prevalence of psychiatric disorders compared with the US but a much lower overall rate of gun violence and a much smaller proportion of homicides and suicides committed with guns. The best evidence suggests that there is only a marginal association between mental illness and increased risk of violence. This implies that the dynamics that underlie the epidemic of firearm injury in the US require little focus on mental illness; it is instead the differences in firearm availability and gun culture that likely explain the differences in firearm injury across these two geographic contexts.

Therefore, while explaining the link between upstream determinants and population health can indeed be helpful to illuminate pathways, does such an approach in fact distract us from a focus on what matters most to improving the health of the public? There is indeed much to be said about the potential distractions inherent in approaches that focus on the mechanistic drivers of heath production. Arguably the recent surge in interest in precision medicine represents the apotheosis of this approach. Francis Collins, director of the National Institutes of Health, recently suggested that “personalized medicine, also referred to as precision medicine, is a promising area for improving health outcomes.” Precision medicine, of course, represents the ultimate study of “mechanisms,” aiming to help us identify the causal processes that produce health. The precision medicine juggernaut is suggesting, however, that this focus is a si ne qua non for the improvement of health. Is it? I will, in a future Dean’s Note, comment on precision medicine and its role in public health, but here bring it up only to wonder whether a focus on mechanisms indeed detracts from a focus on the conditions that produce healthy populations. At the end of the day, the focus we choose for our work is not cost free—our energies, our resources, our attention are all limited, and the areas we choose to focus on and publish about contribute to a production of understanding that informs policy changes that impact health. None of this suggests that mechanistic work is not immensely valuable for our understanding of the science, but it does prod us to confront why it is that we are doing what we do, and what it is we should be focusing on. I shall talk further about this issue when I present on “What Matters Most?” at the School’s Public Health Forum on March 18.

I hope everyone has a terrific week. Until next week.

Warm regards,

Sandro

Sandro Galea, MD, DrPH
Dean and Professor, Boston University School of Public Health
@sandrogalea