Research, Nonresearch, and CDC Policy 557

The Centers for Disease Control has a document–Policy 557–that lays out when Institutional Review Board involvement is necessary with regard to distinctions between patient care and the involvement of patients in research.

Here’s the basic policy:

CDC has an ethical and legal obligation to ensure that individuals are protected in all public health research activities it conducts. All CDC activities must be reviewed to determine whether they are research involving human participants. When an activity is classified as research involving human participants, CDC and its collaborators will comply with 45 CFR part 46 in assuring human research protections.

Policy 557 then distinguishes Public Health Research from Nonresearch. Here’s a bit of the description of Research (my emphasis):

The purpose of the activity is to develop or contribute to generalizable knowledge to improve public health practice;

intended benefits of the project can include study participants, but always extend beyond the study participants, usually to society;

and data collected exceed requirements for care of the study participants or extend beyond the scope of the activity.

Generalizable knowledge means new information that has relevance beyond the population or program from which it was collected, or information that is added to the scientific literature.

Knowledge that can be generalized is collected under systematic procedures that reduce bias, allowing the knowledge to be applied to populations and settings different from the ones from which it was collected.

Research intends to use patients to undertake systematic study that produces knowledge that does not depend on any specific patient and may not even provide a benefit specific to any given patient. That is, patients are used–agree to be used–to ascertain something that is not particular to any given patient.

Nonresearch, however, is different:

The purpose of the activity is to identify and control a health problem or improve a public health program or service;

intended benefits of the project are primarily or exclusively for the participants (or clients) or the participants’ community;

data collected are needed to assess or improve the program or service, the health of the participants or the participants’ community;

knowledge that is generated does not extend beyond the scope of the activity;

and project activities are not experimental.

We might pause here with regard to “experimental.” In the context of Policy 557, a research experiment is one in which the purpose is to obtain generalized knowledge. It may be distinguished from an effort to diagnose or identify an effective treatment that might involve trying one intervention and then assessing it for efficacy before continuing or trying an alternative intervention. Whether a given intervention is well established or involves a degree of uncertainty is not material. A well established intervention used to gain generalizable knowledge (such as serving as a control) is experimental; a highly uncertain intervention used for the purpose of discovering an efficacious treatment for a specific patient is not experimental.

Policy 557 then gives a number of examples of research and nonresearch in the context of CDC programs, such as surveillance, emergency response, and program evaluation. For physicians with front-line health care responsibilities, a patient-first approach contrasts substantially with a research-first approach, even though both approaches may use–and compete for–commonly used vocabulary. To diagnose easily may be described as research to identify the cause of a disease. To prescribe may easily be described as an experiment to see if a given intervention provides the benefit that many people are reported to receive. But to diagnose or to prescribe are directed at the benefit of the patient–that is the purpose–even if what is learned may be used by doctors treating other patients. The knowledge gained is not generalized outside the population of patients and doctors treating the same or comparable conditions.

In a sense, doctors working together to share experiences and insights and methods transmit the craft of medicine rather than an generalized knowledge of medicine. It is the same with learning a sport, such as hockey. One may construct a theory of hockey game play, but hockey is learned especially by deliberate practice–taught by others who play and who observe play–without generalized knowledge. It is a craft, or as it was called in the middle ages, mystery. It’s still that way, now, in the teaching of magic. One does not reveal one’s tricks, but rather puts others in the position to develop their own tricks. In any case, to learn to perform magic tricks from a generalized knowledge of such tricks would take a very different turn of mind, just as learning hockey from reading books would fail to connect awareness and physical and mental performance–skating technique, muscle fatigue, the pain after blocking a shot, elation after having scored a goal.

The idea that research will provide generalized knowledge from which discovery will be made and new products developed is a seductive one, and there are instances in which such a thing must indeed happen. But in the great sweep of technology change across history, much of what we have discovered has come because we share and develop the craft of what we know–we participate in the mystery of it, and from that observation and participation comes insight. The problem is not so much insight as finding standing for others to act on that insight. One may realize that it matters that doctors wash their hands before delivering babies, but that realization must take a form that doctors will act upon.

Generalized knowledge–a theory that has been validated in parts, a general law or principle that authorities on the matter have announced, a systematic study that generates quantitative data that shows significant if not profound results–has carried greater weight in inducing practice change in science than announcements of insight. But not all change comes about because there is a change in science. Indeed, in both craft and medicine, science often has trailed practice change by decades. We know something works long before we know how or why.

In these other areas of medical or technology change, the sequence may take the form of insight, trial, advantage or desired function, and envy or emulation by others. Thus, all manner of new craft and industrial processes operated under trade secret protections until someone figured out how to replicate the process or someone defected with the requisite knowledge–and hence the early role of patents offered by competing city-states to induce defections of craft with the promise of some period of exclusive control in the new territory.

Let’s return to Policy 557 again. The crucial distinguishing characteristic of research and nonresearch is the purpose. That is, one must rely on intention, on statements of intention, to separate research and nonresearch. It is, then, to a great degree a matter of the habits of expression one uses to set out what is being done. A doctor may treat patients and come to realize that a given combination of results from a lab panel suggests a particular intervention will have good (or adverse) results. That much is craft, mystery, nonresearch, so long as the purpose of the activity is to treat patients. One does not have to propose or test a theory regarding why the combination marks a likely intervention–that may of course be a valuable addition to generalized knowledge, and to pursue it involves research.

A nonresearch activity can develop or contribute to generalizable knowledge after the project is undertaken even though generating this knowledge was not part of the original purpose. In this case, because the purpose was not to develop or contribute to generalizable knowledge, the project is not classified as research at the outset.

Policy 557 works its way toward the observation that a research activity can overlay a nonresearch activity. What matters, however, is whether those involved in the nonresearch activity then take their direction from those involved in systematic study to produce generalizable knowledge or whether they continue to pursue their craft with their rules of thumb, their insights, and their shared understanding of things that work or do not work. Even if they supply their own explanations for their understanding, their work is not research, even though their methods might be described by others as involving research. It is their purpose that controls, not the methods at hand.

However, if subsequent analysis of identifiable private information is undertaken to develop or  contribute to generalizable knowledge, the analysis constitutes human research that now requires further consideration under 45 CFR part 46.

And we are then back around to the requirements imposed on the protection of patients brought about when one moves from providing care to using care as a basis for study and creation of generalizable knowledge that in turn depends not merely on de-identifying personal information but rather on abstracting specific data to construct a general statement or rule or account.

It’s interesting, then, that how people set their intentions, their purposes, affects significantly the safeguards that need to be put in place to protect “human subjects.” If their purpose is to assist those “human subjects,” then there is less need for the intervention of a “review board” (more humans) to establish protocols and oversight. One’s own commitment to assist is often sufficient. But when the purpose is to create generalizable knowledge, then other humans must intervene–ones that do not share that purpose and instead focus on the safety and dignity of the people involved.

Our university work is dominated by the idea that we must create or seek “new knowledge,” that research must come first, before discovery and before public benefit. But in the deep past it was never that way, apparently, and even in the recent past–perhaps any time before 1945, there was a much expectation of technology change from craft as there was from research. Indeed, Vannevar Bush’s great endeavor to engage industry, academics, and gadgeteers in development of new technology that incorporated both new (or unexplored) science with industrial methods and capabilities represents an important change in the idea of where research might be placed in a broader pattern of technology change.

Bush argued that there were advantages for his approach to development if there were new areas of science continually being opened up and mapped–as science. These areas could then be exploited along with anything else at hand to create stuff that otherwise could hardly be imagined, or imagined only as magic or science fiction. People got the idea then that research had to come before development, and development had to develop what the research had invented. Just twistedly wrong-headed, but that’s what happens when policy administrators get hold of ideas too complicated for them.

Here with Policy 557 we see the potential for the dominance of research vocabulary. That the use of research terms–systematic study, new knowledge, experimentation, even public benefit–may imply that the purpose of an activity is research, that the goal is some generalized contribution to knowledge. That’s the attitude baked into administrative policy statements and wordsmithed accounts of university missions.

That’s also what’s baked into accounts of university patent policy and technology transfer. Just as research exists to extract from individuals generalizable knowledge to serve a purpose greater than (or other than) the benefit of those involved, technology transfer appropriates research as merely the means to extract patentable inventions and other proprietary assets to be used not to advance scholarship or science (that generalizable knowledge) but to stimulate economic activity, to generate money, to create commercial–generalizable–products, and that in doing this activity, providing a generalizable benefit to society.

It’s all of roughly the same character–exploit something not for its own purpose, but impose on it some other purpose that is claimed to be more important. To help patients is nothing compared to research that uses patients to develop a general theory of disease, and developing a general theory of disease is nothing compared to extracting something proprietary from that effort that can increase the profits of a company or at least make policy makers feel good about the potential for companies to increase their profits. Anything done directly is taken to be silly idealism, patsies snookered into doing menial things like helping people, rather than doing generalizable things like making new knowledge or making new knowledge profitable.

We need means to resist such exploits. One way is to be firm with our intentions. We can discover new things by the diligent practice of craft, by entering the mystery, by the direct apprehension of the Thing, by working for the benefit of a patient, or a student, or the imaginative delight of another human being. That stuff is not patsy menial–it is right there with the best things we can choose to do. To displace the purpose in doing these things for some other thing–status, money, influencing economic statistics–is perhaps the menial thing, the activity in denial that we live together and make life what it is. Martin Buber’s I-Thou word is every bit as needed as the more readily trotted out I-It.

It may be then, that what we need is a diligent, firm displacement of research and technology transfer vocabularies from the activities of the university–that the engagement of person with person, whether in instruction, mentoring, craft, collecting, exploring, or inquiry, is every bit as much the highest calling and most productive activity as derivative or meta practices that exploit the existence of such engagement but treat it as only a necessary means to get at something more important, something that in the end takes up way too many administrative resources and way too much administrative imagination and way too much administrative fear of failure.

Commit to purpose, refuse to use volunteer research or commercialization vocabulary, refuse to cooperate, even, with anything that diminishes the practice of craft, participation in the mystery, the direct engagement whether with people or nature. There is room for the other things–research and technology transfer–but not at the expense of displacing the primary thing, sucking its life out of it for some other purpose, and pinching it out of existence.

The premise then, is that direct engagement, this Nonresearch–something Policy 557 has no name for, this I-Thou word that is so powerful it would never fit in an administrative policy discussion–is a great generator of benefits, of discovery, of technological change. We need not engage others or any thing in the world merely to seek benefits or discovery or change–these things come about because we stay on the purpose of engagement, of craft, of deliberate practice. From activities motivated by such a purpose, others may then conduct research or seek financial advantage. Never let them make you serve them, however, in your engagement, your Nonresearch.

This entry was posted in Policy, Sponsored Research and tagged , , . Bookmark the permalink.