Share this article on:

Commentary: What’s an Epidemiologist to Do?

McKeown, Robert E.

doi: 10.1097/EDE.0b013e31826d0d7c
Epidemiology & Society

From the Department of Epidemiology and Biostatistics, Arnold School of Public Health, University of South Carolina, Columbia, SC.

The author received no funding for writing this commentary.

Editors’ note: Related articles appear on pages 914, 919, and 927.

Correspondence: Robert E. McKeown, Department of Epidemiology and Biostatistics, Arnold School of Public Health, 800 Sumter St., Suite 205, University of South Carolina, Columbia, SC 29208. E-mail:

In 1942, editors of the American Journal of Public Health 1 invited letters “On the subject of what and who, and even why, is an epidemiologist.” Responses over the next 7 months make delightful reading for those with an interest in the evolution of the discipline. Now, 70 years later, we find ourselves still talking about what an epidemiologist is and does. It seems we continue to have questions about our proper role, despite our maturation as a discipline in the latter half of the 20th century.2

In this issue of EPIDEMIOLOGY, Dowdy and Pai3 propose a new model of epidemiologist: “Accountable Health Advocate” (AHA). It would be difficult to disagree with their arguments for increased funding of translational public health research, and there may be merit in their call for new publication venues and broader criteria for professional advancement in academic settings. However, creation of the AHA is not necessary for any of that and doing so implies that other epidemiologists do not have the responsibilities claimed for the AHA.

The value of the concrete actions or skills that Dowdy and Pai3 want to see enhanced cannot be denied: more funding for public health, including translation and evaluation; clearer communication; publication policies that value application and evaluation; and advancement opportunities for epidemiologists who focus on translating findings to clinics, communities, council chambers, and Congress. This does not mean we need to define the “Accountable Health Advocate” as a new type of epidemiologist.

Back to Top | Article Outline


Translation of epidemiologic evidence into practice is already supported by current trends—namely, the growth of multidisciplinary teams that marshal the joint expertise of persons skilled in health communication, behavior, intervention, implementation, systems management, environmental and occupational health, and policy formulation. These persons and others are natural collaborators in translating research findings into programs, policy, structural reform, and health-enhancing practices. Such multidisciplinary teams are more likely to develop innovative applications of epidemiologic research than epidemiologists alone, even if some of the epidemiologists were AHAs. The range of disciplines extends beyond traditional health sciences to include basic and social sciences, perhaps even art and humanities. Much as we may need to enhance our own skills, we already have around us colleagues who have the requisite skills, if we will but avail ourselves of what they have to offer (and teach) us.

The practice of epidemiology itself is multifaceted and multidimensional, and the roles we take on are just as varied. Some of us work in settings where the roles emphasized by Dowdy and Pai3 are part and parcel of our work, whereas others are in settings where such roles are difficult to assume, perhaps even forbidden. The nature of epidemiology and the complexity of health-related problems call for epidemiologists to use a breadth of knowledge, skills, methods, and applications that no single person could ever hope to master. Some of us engage more actively in translation or policy formation, whereas others find those activities beyond their expertise or outside the bounds of their position. Those limitations, however, do not relieve us of our professional obligations to conduct responsible (read “accountable”) research directed toward improving the health of people and communities.

In 1994, Nijhuis and van der Maesen4 explored the meanings of “public” and “health” with each on a continuum: “public” ranging from an aggregation of individuals to a more social construct, and “health” ranging from a mechanistic model to more holistic concepts of well-being. A decade later, Khushf5 expanded this approach to create a three-dimensional graphic using a dimension of well-being (from somatic function to environmental integration), a scale of intervention (from subsystems to policy), and system level (from individual to culture). The Figure draws on Khushf’s work to depict three of the multiple dimensions of the epidemiologist’s “workspace.” The placement is not static—our work goes back and forth on each axis (and others) as projects progress or we move among projects. (Labels should be considered suggestive rather than definitive or exhaustive.)



The horizontal dimension represents the spectrum of health research targets, from individual health to the health of small social groups (such as families) and to the health of society. The vertical dimension represents the focus of investigation ranging from molecular (genes, biomarkers) through traditional population-based studies and on to social epidemiology, with its emphasis on social and cultural determinants of health. Both dimensions are incorporated into multidimensional ecologic models of health that are now common,6 although more in conception than in realization.

The third dimension shows that advocacy and dissemination begin within the scientific community. One need only attend a scientific meeting to see evidence of advocacy of a scientific position that does not go beyond the circle of experts in the area. Epidemiologists and our colleagues in many other disciplines are engaged every day in the use of epidemiologic research as a basis for design and implementation of community intervention trials—one example of translating findings into the community. At the limit of the axis is policy formation (or structural reform and other upstream arenas) as just one example of how epidemiologic findings might constitute the basis for informed policy with potential to enhance population health. Things come full circle when epidemiologic methods are then used to evaluate the impact of that policy on population health. At no point in this three-dimensional space does one suddenly become an “Accountable Health Advocate,” and virtually any point in this space can be characterized as accountable health advocacy.

Back to Top | Article Outline


Although transparency and communication are important aspects of accountability, there is no account of accountability that demonstrates why these and other characteristics of it would not apply to all practicing epidemiologists. Accountability goes beyond openness and communication and includes taking responsibility and being responsible for something.7 To say the AHA is accountable to society is only part of the story. We are all accountable to the profession and to the community of scholars, to institutions for which we work, to funders, editors, reviewers, and authors, and to the public at large that supports the work we do. Indeed, such accountability is, in my view, one of the three essential characteristics of a profession. (The other two are commitment to expertise in a discipline and excellence in its practice, and commitment to a shared set of values.) Accountability is essential for every epidemiologist, regardless of the degree to which he or she focuses on etiologic research or application of discoveries.7,8 Our accountability is not simply payback but is a reflection of a fundamental ethical obligation and commitment to enhance population health. Accountability is part of responsible epidemiologic practice in general.

The ability to communicate clearly is essential for epidemiologists. The Plain Writing Act of 2010 requires government documents to use “writing that is clear, concise, well-organized, and … appropriate to the subject or field and intended audience” (PL 111–274; October 13, 2010). The act also requires agencies to train staff in “plain writing.” Thus, one of the needs described by Dowdy and Pai is being addressed through government training programs in those settings.6 Experts from government agencies can be enlisted to provide workshops for professional organizations. Many academic institutions offer resources to assist researchers in improving communication skills. There are increasing opportunities to develop communication skills, without resorting to a new type of epidemiologist who relieves the rest of us of responsibility for translating our findings to communities most affected by them.

The AHA concept emphasizes “health” without an exposition of what it means, especially in the public health arena. The claim that health has value because it affects human lives is one that requires further development. What is health and why does it have such value? Is the value of individual health distinct from the value of population health? There is an implicit moral weight given to it, but it is never clear whether the moral value of health is instrumental or intrinsic or both. A careful examination of the foundations of the concepts of health being used and the nature of their value would be helpful.

Similarly, the nature of the advocacy proposed is not clear. Concerns about epidemiologists as advocates have been related to lack of objectivity and inadequate training.7 Only the latter is addressed. Greenland9 rightly argues that objectivity is not attainable, nor is it desirable. But neutrality before the data, circumspect humility, and awareness of fallibility demand a degree of caution and disclosure of the values that shape our judgment. What is troubling is the tendency of the language of the piece to set the role of epidemiologist as researcher (“knowledge generation”) against the role of epidemiologist as translator of research and participant in shaping programs, practice, and policy, when most of us play these roles to varying degrees at different points (Fig.).

The case for greater funding of public health translational research requires tackling a range of issues, including the problems of credibility of epidemiologic research and ethical issues of resource allocation and distributive and restorative justice (including the “prevention paradox” of Rose10). A probing exploration of these issues would be of value, but it would require asking why the movement from discovery to meaningful implementation and impact is so hard and takes so long and is often so limited or selective. The AHA proposal does not show us how that problem is alleviated in concrete ways that are novel.

What is needed is a model of translation for epidemiologic research to public health practice that is more nuanced and sophisticated than the diagram from Dowdy and Pai. It must account for uncertainty and balancing of goods—both outcomes and resources—and must consider the multileveled determinants of health and complex factors that affect policy formation and program implementation. Furthermore, to suggest that there needs to be a shift in scientific effort from discovery (or knowledge generation) to translation is misguided. The scientific process does not operate that way. Translation and application move alongside discovery but often at different rates. Sometimes, discovery far outpaces the realization of application, whereas at other times, the demand for an application is unmet because the underlying scientific knowledge does not yet support what is needed.

Back to Top | Article Outline


MacIntyre11 defines practice as a “coherent and complex form of socially established cooperative human activity through which goods internal to that form of activity are realized.” At its core epidemiology is a conceptual approach to defining and investigating problems that affect the health and well-being of populations, with a continually developing methodological toolbox and rapidly growing substantive knowledge. Public health in its quest for healthy people and healthy communities defines a common goal that unifies otherwise disparate disciplines. Policymaking is one possible end toward which epidemiologic research is directed, but only as an intermediate end, with the ultimate goal being the people’s health.12

If we value epidemiologists who emphasize translation and application to practice, then we can simply say that some of us are translational epidemiologists or applied epidemiologists, although for many of us, those are roles we assume in addition to research and discovery. There are already numerous examples of epidemiologists who not only have expertise in scientific content and epidemiologic methods but also actively engage in synthesizing knowledge, translating it to practical application in program design and policy formulation, design and implement interventions, and disseminate findings. Such experts are not confined to public health agencies, where that role is surely the norm rather than the exception.

Much as I might agree with and applaud some of the goals of the argument, I do not see the need for a new type of epidemiology—if for no other reason than because it could imply that the rest of us are thereby relieved of our responsibility to be accountable to society, to pursue health for the people, and to advocate for those things that make for excellence in epidemiologic practice and health and well-being in our communities.

Back to Top | Article Outline


ROBERT MCKEOWN is Distinguished Professor Emeritus of Epidemiology at the University of South Carolina. He is immediate past president of the American College of Epidemiology and a past chair of the Epidemiology Section of the American Public Health Association. The writings of philosopher of science Michael Polanyi have influenced his thought. His research focuses on psychiatric epidemiology and public health ethics.

Back to Top | Article Outline


1. . What and Who is an Epidemiologist? Am J Public Health. 1942;32:414–415
2. Susser M. Epidemiology in the United States after World War II: the evolution of technique. Epidemiol Rev. 1985;7:147–177
3. Dowdy DW, Pai M. Bridging the gap between knowledge and health: the epidemiologist as Accountable Health Advocate (“AHA!”). Epidemiology. 2012;23:914–918
4. Nijhuis HG, van der Maesen LJ. The philosophical foundations of public health: an invitation to debate. J Epidemiol Community Health. 1994;48:1–3
5. Khushf G. Systems theory and the ethics of human enhancement: a framework for NBIC convergence. Ann N Y Acad Sci. 2004;1013:124–149
6. Institute of Medicine, Committee on Assuring the Health of the Public in the 21st Century. The Future of the Public’s Health in the 21st Century. 2003 Washington, DC National Academy Press
7. Weed DL, McKeown RE. Science and social responsibility in public health. Environ Health Perspect. 2003;111:1804–1808
8. American College of Epidemiology. . Ethics Guidelines. Annals Epidemiol. 2000;10:487–497
9. Greenland S. Transparency and Disclosure, Neutrality and Balance: Share Values or Just Shared Words? J Epidemiol Community Health. 2012 DOI: 10.1136/jech-2011–200459
10. Rose G. Sick individuals and sick populations. Int J Epidemiol. 1985;14:32–38
11. MacIntyre AC After Virtue: A Study in Moral Theory.. 19842nd ed Notre Dame, Ind University of Notre Dame Press
12. McKeown RE, Learner RMCoughlin S, Beauchamp T, Weed D eds. Ethics in public health practice. In: Ethics and Epidemiology. 20092nd ed New York Oxford University Press:147–181
© 2012 Lippincott Williams & Wilkins, Inc.