Samantha Copeland, Thinking about guidelines:
I have been interested for a while in how we justify the move from a single case to conducting research on other patients as research participants. For instance, there have been cases where unexpected (positive) results suggest that a novel approach to treating difficult patients may be found: such as the case in Toronto, Canada where electronic stimulation of a patient’s brain had an effect on his memory that suggested a new method for treating Alzheimer’s might be available; or in Bergen, Norway, when a patient with chronic fatigue unexpectedly recovered from her symptoms while undergoing treatment for Hodgkin’s disease. When a decision is made to start research, researchers must justify why they think the observations made of this particular patient could be repeated in others.
I believe this justification process is most simply expressed as an analogy: one argues that this particular patient is like some other patients in the right ways, and so we can reason that the same effects can be caused by the same treatment approaches. Therefore, it is both correct and ethical for research to attempt to recreate the same results by doing the same things to research participants as was done in this one case with this one patient.
Such analogies are redrawn when guidelines are made and applied to individual patients. The results of research, in particular when these results are obtained by randomized controlled trial, are known to apply to a specific group of participants who are selected for and thus share specified characteristics. When those results are then applied to a patient in the clinical context by a clinician deciding about the best treatment options, that clinician is also making a judgment about whether her patient is sufficiently like the participants in that research for the results to be the same.
But whereas research participants are selected for a trial because they share a specific set of characteristics, patients in the clinic—whether the original case that inspired a new direction of research, or the case to which the results of research are meant to apply—are not selected in that way. And some kinds of patients are particularly difficult to fit into narrow categories. For instance, the broad category of chronic fatigue syndrome (CFS) is argued to have several sub-categories, and so we should not use this disease category to judge whether a treatment will apply to a particular patient who fits into the category. Another group that makes it difficult to judge whether research results will apply in a single case are patients who suffer from co- or even multi-morbidity—two or more conditions, sometimes chronic ones, at the same time. How do we decide whether the results observed in a single case, or even the results observed in a trial done with a group who share a set of characteristics, will apply in these more complicated clinical situations?
Drawing analogies is often done without explicit recognition that we are doing it. Assuming that one situation or person will be like another is often a result of bias or assumptions, rather than reasoning. Cases at the fringe, such as research conducted because of an unexpected observation, or the difficulties of applying guidelines to patients who suffer from multi-morbidity, call attention to the kinds of analogies we do and must draw. How often are the analogies themselves getting lost in the translation from a single case, to groups of research participants, and back again to the individual patient? It seems to me that this is a problem at the root of many of the questions our workshop participants and colleagues have raised in this discussion about guidelines already.
Finally, I am interested in the discussions the group has been having about the role of guidelines in decision-making at the clinical level. There are a lot of questions (philosophically important ones) about how the rule model of guidelines and the expertise model of clinical decision-making intersect. These waters are additionally muddied by the political aspects of creating rules and professional standards for medicine that individual practitioners must interpret and follow. My current research focusses on the place of strategy and complexity reasoning in situations where both uncertainty and high expectations can be found (what I call high-stakes inferences). Healthcare guidelines and their use present just such a situation.