IRB Case:Public health practice vs. research
Is a new context coming? Who benefits?
While public health research and practice may sometimes be easily distinguishable, the case presented here brings to bear some of the most challenging questions faced by Institutional Review Boards (IRBs) when a single researcher proposes to undertake the two activities in a single course of a study. In truth, I think the members of the IRB in this case did an admirable job of parsing apart the comingled concepts. I certainly have no delusions about my own ability to have thought any more clearly than did they. Still, in light of my brief training in public health, I’ve been asked to respond to the case with a brief commentary of my own and the first part of this commentary will encapsulate just that response. Perhaps more interestingly though, in the second part of this commentary, I hope to move beyond the facts of the case to offer a broader set of concerns about the future of this subtle distinction in the emerging field of global health.
The ruling IRB and I are in agreement in much of our thinking about the distinction of research and practice in this case. I found their use of the National Bioethics Advisory Commission definitions to be prudent since the locus of benefits seems to me the most compelling, albeit imperfect means of differentiating the two activities. In light of the definitions provided, it seems clear that the stated goal of the study, which was “to establish a population-based surveillance system for persons newly identified with this infection in one town in Connecticut in order to develop individual and population-based prevention strategies for use across the state,” represented research. In fact, there appear to me to be at least two ways in which the locus of benefit lies outside the study population. The first is geographical scope - the goal is to generate knowledge applicable to the whole state as a result of findings generated from a small population of infected people in one town. The second is infection status – the study is to be conducted on infected persons for the purposes of the development of prevention strategies (presumably for non-infected persons).
It was only in our interpretation of the Objectives that my opinion and that of the IRB’s diverged. While I agreed with their ruling that Objectives #1, #3, and #4 fall clearly within the bounds of the NABC’s definition of research, I disagreed with their deeming of Objective #2 as practice. It is on this point that I would respectfully push for closer consideration, as I would argue that Objective #2, like the rest of the researcher’s proposed Objectives, also fulfills the necessary ‘locus of benefits’ condition to be considered research.
The researcher has stated his second objective to be
“Assess the stage of illness and the need for medical and personal preventive services (e.g., health insurance status, access to primary and specialty care, need for mental health treatment etc.)”
The IRB’s ruling that this activity was ‘practice’ rested on their understanding of the “theoretical benefits” accruing “to those local participants with the disease.” I, on the other hand, am at a loss as to what benefits would be accruing to the local participants with the disease. There seems to me to be no medical treatment or financial compensation provided, nor is the researcher even proposing to advocate on the study participants’ behalf for such services. In the absence of any such follow-up to the assessment, I would suggest that this objective exemplifies research. The primary benefit of such an assessment might reasonably include a heightened understanding of the disease demographic, the standard progression of the illness and the corresponding courses of treatment a future patient might reasonably expect to undergo. This information would be of great value to hospital administrators hoping to conduct research mapping, insurance companies looking to accurately assess financial risk, researchers interested in cost-effectiveness studies or people in the future who are newly-diagnosed and want to more fully understand their illness. In all cases, those benefitting most from the assessment lie outside the study population of currently diseased patients.
The confusion over Objective #2 seems to arise from differing interpretations of the term surveillance. The IRB ruled that because the data to meet the assessment goals were to have come from a surveillance system, Objective #2 represented public health practice. I, however, would argue that surveillance is little more than a means of informing future decision making, meaning the locus of benefits almost always lies exclusively outside the study population.
Surveillance is, in its purest form, an act of research. Standing alone, surveillance does precisely nothing for the population who takes part in it. There are no real benefits accrued from data, or even the knowledge data creates. And yet it has quite successfully made its way into the popular notion of public health practice. Why? I would suggest its inclusion to be the result of a modern conception of ‘good’ pubic health defined by a continuous loop between planning, doing, studying and acting. This same loop is sometimes also captured in three phases: surveillance, implementation and evaluation. As in all loops, the distinctions between the phases are poorly defined, and barring a clear chronological demarcation, we have trouble drawing the proverbial ‘lines in the sand.’ Here is where the issue of who benefits from the activity becomes delicate. Depending on how long each phase lasts, the results of surveillance could be used as ‘real-time’ feedback to inform decision making - or not. Even so, had surveillance been described in Objective #2 as part of a loop such as those cited above, I would have considered it practice rather than research. But absent any mention of an ‘intervention’ informed by the surveillance, I would suggest that the activity remains an exercise in research.
I will now move into the second part of my response, expanding the scope of my commentary to include a note on how these already bleary distinctions relate to the emerging field of global health. Many have suggested that global health is simply a new name for classic public health driven solely by misplaced donor interest and overhyped student enthusiasm. Most however have settled upon a conception of global health as something of a novel breed of public health that has newly broadened horizons, methods and challenges. I fall into this latter category. I raise the temporal relationship between public health and global health here only because I believe their relationship to have important effects on the difficult distinctions I have already discussed here between practice and research.
For the most part, the leadership of global health knows that it cannot simply carry over the substantive ‘lessons’ of 20th century public health into the new contexts of 21st century global health. Where public health was locally bounded, global health is (as its name implies) just the opposite. What worked in sanitation for New York City in 1910, even if it worked in Philadelphia in 1915, will decidedly not work for Monrovia in 2010. Variations in culture and context are simply too great to support the extrapolation of past results. But the leadership of global health can and has carried over the research ethos of public health. In truth, this is largely because the personnel at the top ranks of 21st century global health are the same personnel who worked at the front-lines of 20th century public health. Regardless, the result of these two temporal effects is a cadre of global health leaders who 1) will readily admit that they do not have ready-made answers to the new challenges they are confronting; 2) feel ethically obliged to do something rather than nothing; and 3) know enough based on their collective historical consciousness to rely on research to guide their actions.
And so, the only way forward in global health necessitates that the line between research and practice be blurred still further. We must constantly seek answers to the questions of “what works?” To solely seek answers through research is both impossible (we won’t know what works until we try) and potentially ethically questionable (even if we could, would we stand by and do nothing until we felt our solutions perfect?). As a result, the importance of integrating our search for answers (research) with a trial-and-error approach (practice) is underscored. In fact, the future of global health suggests that the sequential nature to the two activities that I suggested earlier will likely not last much longer. The loop may soon become obsolete, as practice and research increasingly happen simultaneously. If nothing else, this intertwined and self-reinforcing model of research and practice is quickly becoming the model of our work here at Yale’s Global Health Leadership Institute.
Understanding that the relationship between practice and research is changing to fit the needs of global health, I fear that future IRB cases – and in particular those cases that constitute the most innovative and potentially meaningful work in global health - will continue to present ever greater challenges in distinguishing the two activities.