Balint group participants and enthusiasts (myself included) could not be more emphatic about the value of the process, its beneficial impact on group members and the need that practicing physicians (and other health care professionals) have to process the emotional impact of their work. However, at the same time that this chorus of support exists, there is an equal but opposite sentiment about the effects of Balint groups due to the paucity of measured evidence in support of these benefits. It’s not that there is zero evidence in support of Balint groups. It seems that the more quantitative efforts to measure Balint group benefits have been equivocal - with no clear result. How could this be? Why is there such a gap?
At the same time that I’m asking myself some of these questions, I became aware of a recent publication in Patient Education and Counseling (June 2015): “Research on Balint groups: A literature review” by Van Roy, Vanheule and Inslegers - all from Ghent University in Belgium. What a valuable effort! They summarize and organize published Balint related papers by the type of participants, the type of evidence or the instruments used, topics raised and the results or findings. In the words of the authors, “Research on B(alint) G(roup)s proves to be diverse, scarce and often methodologically weak. However, indications of the value of BG work were found. Therefore, further research is strongly indicated.”
As I reviewed this paper, one observation that puzzled me is the wide range of different measuring instruments that have been used. My belief has always been that Balint group participation teaches empathy, so using the Jefferson Empathy Scale makes sense - but these results are unconvincing. I also buy into the value of Balint group participation in preventing burnout, so using the Maslach Burnout Inventory makes sense - but these results are also unconvincing. These are two of the most commonly used measures, and then I see all the many other measuring instruments reported in the literature.
Is it possible that in our uncertainty of what to measure or in
our differing ideas of what to measure we are unclear about our target?
I finally read John Salinsky and Paul Sackin’s book: What are You Feeling Doctor: Identifying and Avoiding Defensive Patterns in the Consultation. This book represents an incredible commitment to the Balint process and to each other. A group of physicians agree to not only get together regularly to Balint their challenging cases, but they also agree to put themselves under each other’s gentle, but still analytic microscope to try to make conscious their unconscious processes in responding to their patients. Interestingly, their target for study is neither empathy nor burnout! Their target - appropriately enough given Michael Balint’s analytic orientation - is their own defensive patterns. And their idea is that if we become more conscious of how we unconsciously distance patients, we can choose responses and reactions that are less distancing while still taking care of our own needs.
So I ask myself: Are our defense mechanisms a reasonable target for our measuring efforts? While Salinsky and Sackin clearly make the case that defensive patterns intervene and interfere in the doctor patient relationship, their discovery is not the result of a balint group discussion. It required an extensive seminar like strategy. While their entire project is a most valuable and revealing contribution to our understanding of conscious and unconscious processes, I’m not sure that it gets us closer to identifying a target outcome to measure. In the back of my mind, I continue to consider the prototype Realist methodology question: What is it that is working, for whom and under what circumstances?
One of my several opportunities to lead Balint groups is with third year medical students from University of South Florida’s (USF) Select program. Their last two years of med school are conducted in Allentown, PA at the Lehigh Valley Health Network’s campuses. Because I am a member of their faculty, I had the opportunity to participate in a three day Emotional Intelligence (E.I.) Immersion program. While E.I. is not new to me, it was good to participate with colleagues who are part of these medical students’ training. And, it gave me another opportunity to think about Balint in yet another different framework.
As one can see from the above diagram, this is a 2 X 2 grid that considers both recognition and regulation of self-awareness and social awareness. As you can also see from the examples in each quadrant, empathy is part of social awareness. I suspect most people would place defense mechanisms in self-awareness, and it makes sense to me to place doctor patient relationships in self management and relationship management. Cases that are offered to Balint groups may be the result of inadequate self management or less than ideal social management. Using this model, my target behavior would be neither empathy nor burnout; it would be one of the components of self awareness!
Finally, a third model has emerged on my Balint radar. I have the advantage of having conversations from time to time with my department chair, Will Miller. He was our keynote speaker when the ABS sponsored the 2011 International Balint Congress in Philadelphia. And he and I typically share ideas about doctor patient relationships.
Will shared this variation of the Johari Window that he worked on for another project. If you are not familiar with this model, it was described by Joe Luft and Harry Ingham in 1955. In it, Joe and Harry juxtapose what is known and unknown about ourselves and about others to yield a 2 X 2 grid with four distinct quadrants. In our residency, we introduce it as a way to encourage residents to solicit feedback which can make them aware of blind spots. It is also an opportunity for residents to explore and experiment revealing aspects of their secret or private selves (like fears or medical mistakes) to reduce sources of shame or embarrassment. The perspective this diagram adds to the Balint discussion is that the disclosure or offering of a case to the group is in fact revealing something that is secret. In turn, the group’s explorations of that case serves as possible feedback to the presenter which may contribute to reducing one’s blind spots. What is known to self but unknown to others gets revealed by a presenter; what is unknown to self but known to others gets exposed by the group. Now with this perspective, what would be a more direct target behavior to measure?
Michael Balint suggested that participation in his seminars led to a small but perceptible change in physicians. A colleague, Clive Brock, says that participation in Balint groups makes good doctors into better doctors. I wonder if in consideration of these three very different models, we (who are interested in the Balint evidence gap) might give more thought to a remedy for this evidence gap. What specifically are the changes balint group participants experience and what are the conditions that support and allow these changes? And how do we measure them??? I’ll attempt to answer, or at least address some of these questions in my next post. In the meantime, I invite your reactions / reflections / and especially any new ideas stimulated by these thoughts.
