Monday, July 18, 2011

Writing about Patients (and research subjects)

Last week I read a fascinating book by Charles Bosk - "What Would You Do? Juggling Bioethics and Ethnography."

Bosk's first book - "Forgive and Remember: managing medical failure" - a brilliant field study of surgical training, made him famous when the first edition was published in 1979. I can't evaluate his skills as an ethnographic field worker, but he's a terrific writer.

In a chapter on "Irony, Ethnography, and Informed Consent," Bosk reports the intense distress a genetic counselor experienced on seeing how he'd written about her and reflects on the nature of informed consent in ethnographic research. The ethnographer is not doing an experiment for which the risks and benefits can (and should) be described clearly to those who are being asked to participate. But he describes how ethnography characteristically seeks to "debunk" (his word) conventional social constructions, and he comments that ethnographers often write with irony. (That's part of what makes his writing so engaging.) The chapter reflects on whether the risk of reactions like that of the genetic counselor can be conveyed to folks in the environment the ethnographer hopes to study.

This is a tough question. How many potential "subjects" would agree to the study if the would-be-researcher came to the group and said: "I would like to observe you to understand how you function. I will write up my observations and may well debunk your own views of yourselves and what you are doing. I will write with irony, and you may feel misunderstood and condescended to."

Not many. I'm a great fan of ethnographic literature and have learned a lot from it. But if an ethnographer wanted to study me in action, I'd (a) agree if I had "must approve" rights to review any and all proposed public presentations that could possibly be connected to me or else (b) I would run for my life.

Over the years I've often used experience with patients in writings and lectures. I've applied this rule of thumb to the way I developed the material: The patient's closest friends shouldn't recognize who the story is about, and the patient should feel respected by the way he is talked or written about.

The most interesting experiences I've had in writing about patients involve communicating with other physicians. The group I practiced with used an electronic medical record. If I wanted to write to another physician involved in the patient's care, I often composed the note collaboratively with my patient. In one situation, after a lot of nudging and nagging on my part, a patient of mine finally acknowledged that alcohol was a big problem for him. He agreed that his primary care physician ought to be told. We sat in front of the monitor together. He balked when I wrote "X now recognizes that he suffers from alcoholism." We negotiated and agreed to substitute "X now recognizes that alcohol is a problem for him." The discussion itself was part of the work that goes into the first of the twelve AA steps - acknowledging the problem itself.

I don't know how contemporary ethnographers handle informed consent. Quite apart from what an IRB might say, I would not feel justified in writing about people in any way that could be connected to them without their "sign-off" on what I'd written. That was the agreement for the field work that went into the two health policy books I've written with colleagues. And in the grant proposals for funding the studies we stated that we thought of those we observed and spoke with in the field as "colleagues," not "subjects."

Bosk sees the issue of informed consent in utilitarian terms. Respect for those who will be observed and written about calls for robust informed consent, with full explication of the potential risks of the ethnographic study. But the potential value of the study itself pushes against informing about risks in a way that is likely to squelch the study. I may have misunderstood him, but it looked to me as if he came down on different sides of this equation in different chapters.

Apart from federal regulations about human subjects research, I vote for an open process of informed consent. Discussing risks may indeed squelch promising projects, but my optimistic view is that transparency will, over time, enrich the kind of research Bosk exemplifies so well. Negotiating with those one is observing (or, for clinicians, treating) can itself be a rich source of insight. And for ethnographers, that process may lead to less determination to be "debunkers," and to a writing style that depends less on irony about the subjects to engage the reader. A writer as good as Bosk doesn't require a put-down mode to make the story he tells vivid and alive.

No comments: