November 7, 1995

Panel Discussion: The Training Case: A Unique Psychoanalytic Situation

Moderator: Steven Roose, M.D.
Presenters: Robert Glick, M.D.
Deborah Cabaniss, M.D.
Discussants: Jeff Halpern, M.D.
Arnold Cooper, M.D.
Reporter: Stuart Taylor, M.D.


“There are three great things in the world: there is religion, there is science, and there is gossip.

Robert Frost, 1963

This panel represented a move from gossip to science. Dr. Robert Glick attempted to investigate “the cynical folklore that candidates find ways to drop training cases after graduation because of training burdens, and Dr. Deborah Cabaniss began her inquiry by considering “what candidates talk about all the time -- in the halls, in the library, in the cabs downtown: will my patient stay in treatment? I think my patient is going to leave; will I get credit?” Drs. Glick, Cabaniss, and Roose discussed these issues from a scientific perspective, presenting data on the phenomena. Drs. Halpern and Cooper endeavored to put the related folklore and faith in cultural contexts, with comments on subjects ranging from ethics to cults. In a large sense, the panel was about how, in Dr. Roose’s words, “it is important to subject what we believe we know to the most rigorous scrutiny.”

Dr. Glick’s Presentation

Dr. Glick’s study entitled “The Fate of Training Cases,” assumed that clinical immersion is an important value in itself. Pointing out that there is no data on cases after candidates’ graduation, Glick surveyed by anonymous questionnaire all graduates from the Center between 1983 and 1992. He asked about all the cases candidates had begun. Were they terminated? How long was the treatment? Was the case supervised after graduation? Was the case considered successful?

Of the 71 graduates in the study period, 70% returned the questionnaires, which Glick felt was a representative sample. The total number of cases yielded was 184. Of these, 35% (65) were terminated prior to graduation. The great majority of these pre-graduation-terminated cases (89%) were judged by the respondent to have been unsuccessful. Of the 119 cases that had continued past graduation, 48% were considered successful at the time of the study, and 72% had terminated by the time of the study. Of the 151 total cases that had terminated by the time of the study (pre- and post- graduation terminations), the cumulative success rate was 28%. This rate, Glick noted, was consistent with other available data. The length of treatment differed in the cases terminated prior to graduation (average length=2 years), compared to those terminated after graduation (average length=6 years).

Glick considered two other findings highly significant: 78% of cases were supervised after graduation. Furthermore, when the number of cases terminated was plotted against time, the rate was roughly linear, suggesting that there was no precipitous unloading of training cases after graduation.

Glick drew the following conclusions from his data. The total amount of clinical experience accumulated by candidates during training was impressive, since the 184 total cases provided an average of nine years of experience per candidate. Since 35% of control cases terminate prior to graduation, and these essentially represent unsuccessful cases, premature termination can be expected and can be learned from.

He further argued that since there is no evidence of a precipitous increase in terminations after graduation, graduation has no discernible effect on the timing of termination; these cases have a life of their own, independent of training status. Finally, since it is unlikely that a candidate will successfully complete an analysis prior to graduation, Glick stated that training casework will mainly focus on induction and the midphase, and therefore training does not end with graduation, which is only a marker. As training about termination issues frequently follows graduation, and most cases are supervised after graduation, Glick suggested that the choice of supervisor after graduation is important. Training ought to have as a goal the development of one strong mentor-relationship. Dr. Glick felt that this was the next direction for study: the factors that contribute to the choice of a post-graduate supervisor, and ways in which this experience can be supported and maximized.

Dr. Cabaniss’s Presentation

Dr. Cabaniss’s study started with an assumption that graduation is far more than a marker. Arguing that candidates’ need to get credit for their “control” cases was one of the most affect-laden and politically charged aspects of training, she sought to investigate what she thought were profound ways this factor influenced candidates’ conduct of these training analyses.

Cabaniss surveyed by anonymous questionnaire all current candidates (except first-year candidates, who did not have cases at the time of the study). Among her queries were the following items: Did the candidate feel the patient was suitable for analysis? Did the candidate feel that the need for credit affected the ability to make interpretations? What about setting fees, charging for missed sessions, establishing an appointment schedule? And what about presenting the case in supervision and writing it up? She also asked if the patient had brought up the candidate’s training status, and whether this issue was discussed in supervision. Finally, she asked about candidates’ understanding of what was required for “credit” on a case.

Of the 33 current candidates, 27 completed the questionnaire: Cabaniss felt this 81% yield provided a representative sample. The total number of cases studied was 56. In 48% of the cases, candidates reported that the need to get credit influenced their conduct of the analysis. Perhaps more importantly, according to Cabaniss, 60% of the candidates reported that at least one of their cases had been influenced in this way, which suggests that the phenomenon is not limited to a small group.

The factors most commonly affected by the need for credit were: the setting of fees -- 48%; writing up the case -- 43%; and reluctance to interpret issues such as negative transference and the analysand’s wish to leave treatment -- 39%. In contrast to the influence on writing about their work, in only 13% of cases did candidates say the need for credit affected the way they presented the case in supervision. Cabaniss speculated that the more intimate setting of supervision might make candidates more comfortable discussing their work than would the prospect of describing their work on paper for an uncertain audience.

Candidates reported that in 79% of cases, the patients brought up the issue of the candidate’s training status. But they reported that the issue was discussed in supervision in only 61% of cases. Ideally, Cabaniss said, the issue should be discussed in supervision in 100% of cases, and it should certainly be discussed in the 79% of cases in which the patient brought it up. She felt these discrepancies were highly significant.
The data also showed that 98% of candidates knew that establishment of an analytic process was a requirement for case-credit. Although 50% of them knew there was a time requirement to receive credit, only 19% knew that the time required was 18 months. No candidates reported an awareness that other aspects of their academic performance could be taken into consideration in deciding on credit for a case.

The major finding of her study, Cabaniss observed, was that the need for credit affected the conduct of “control” analyses by candidates, a phenomenon that was reported by a majority (60%) of candidates and affected a significant minority of cases (41%). She suggested that this probably underestimates the problem. In essence, Cabaniss feels that candidates’ agenda to keep their patients in treatment causes the candidate-analyst to avoid certain conflicts with patients, thus compromising their ability to analyze and to develop analytic skills.

Does this phenomenon represent countertransference? In the broadest context, this conceptualization may be correct, but Cabaniss preferred to cast the issue in terms of “the inevitable impact of reality” on the training situation. She explained: First, since a majority of candidates experienced this dilemma, it should be considered a shared experience rather than a manifestation of intrapsychic conflict within the analyst. Second, acknowledging the reality of the situation should allow candidates to discuss the situation more openly.

Finally, Dr. Cabaniss pointed out that her data suggesting under discussion of the training situation in supervision concurs with Mayer’s (1972) idea that patient, analyst, supervisor, and institute may collude in not thinking of the training case as different, in order to think of it as a “real” analysis. Dr. Cabaniss’s major point was that data such as she obtained should perhaps both indicate and inform future discussions and research. Possible areas to be explored are ways to discuss and teach about issues such as the need for credit and the apparent differences between “control” cases and post-graduate analytic cases. Also worthy of a more fundamental examination are the criteria needed for case-credit and the clinical-experience criteria for graduation, since these do not appear to be based on any empirically derived data

Dr. Halpern’s Presentation

Dr. Jeff Halpern departed from systematically derived data and used more “traditional” methods -- introspection and his experience with a case that quit treatment -- to enrich the discussion. As he sees it, the problem is more complicated than just keeping the case for credit. He hypothesized that idealization, as a function of training, leads to two distinct analytic processes for both candidate and supervisor. The candidate wishes to learn what is specific to analysis, is reading early analytic texts, is identifying with his/her teachers, and wants to graduate. The supervisor wants to teach what is unique to analysis and identifies with his or her former supervisors. Both consensually create an ideal of a technically correct analysis, “the exemplary analysis,” that may even be anachronistic to them both. However, certain limitations combine to create an “unruly analysis” -- a more covert process that may include non-analytic interventions. These limitations may be the candidates’ inexperience, the patient’s pathology, disjunction between the pace of the candidate’s educational progress and the patient’s psychic progress, and real life personal issues. Halpern described a training analysis in which he sensed the patient’s intuition of the analyst’s need for the analysis to be successful so that the analyst could successfully complete his training. The patient was pseudo-compliant with treatment, often making devaluing comments about treatment. Halpern felt he could not discuss the reality of the situation: that he did need the patient and was eager to pursue the “exemplary” analysis by consistently interpreting the negative transference. The patient left treatment. Halpern believes the treatment failed because of his attempt to do an “exemplary” analysis and his difficulty in discussing with his supervisor the issue of his need for the patient.

“Doggedly pursuing the negative transference” (Cabaniss phrase), Halpern suggested, can also be a countertransference, and analysts don’t always teach the way they practice. Halpern credited Cabaniss’s study for bringing into focus issues that had previously been blurry. Although there are similarities between the candidate’s need for credit and post-graduate analysts’ real-world needs from their patients, he argued, the candidates’ situation differs because of the central need to learn what analysis is. The consensual idealization of analysis combined with the need for credit, can produce distinctly different “exemplary” and “unruly” analysis. These pressures are greatest at a time when the candidate has the least experience in becoming an analyst. Halpern’s conviction -- like Dr. Cabaniss’s -- is that acknowledging the reality of the situation, rather than calling it countertransference, encourages candidates to think about it more analytically.

Finally, Dr. Halpern raised ethical questions about the potential of trying to make the patient fit into an analytic frame, and of moving the treatment along too fast -- in order to satisfy the candidate-analyst’s need. In Dr. Halpern’s words, we want to engage the patient in a common endeavor, and at the same time to analyze that engagement.

Dr. Cooper’s Presentation

Dr. Arnold Cooper’s stimulating remarks, which ranged widely, served to set many of the issues in larger contexts. One of his themes was that, although these studies and discussions are essential beginnings in trying to understand psychoanalytic education, we must be willing to examine our most basic assumptions. For example, he noted that the tripartite model of education is an antique one, begun in Berlin nearly 75 years ago, and continuing without any empirically derived research to support its utility. Many tenets of analytic education are the result of ideological battles, sometimes representing compromise between the convictions of cultist believers and self-appointed realists. In particular, he called for thorough examination of the concepts of immersion and identity. Both are not well defined, he feels, and need to be thoroughly researched, rather than simply taken for granted. Even the use of the term “candidate”, originally describing Roman political aspirants, evokes some of the more cultist aspects of psychoanalytic identity formation. Cooper suggested that the personal analysis may be more important than the supervised cases in demonstrating the power of psychoanalysis. He also noted that one function of mentors is to eventually be overthrown -- in the interest of propagation of a field through the development of new ideas. Cooper specifically questioned Glick’s conclusion that the lack of precipitous increase in terminations, and candidates retaining supervisors after graduation demonstrate graduates’ commitment to being analysts. He suggested many other reasons that candidates may retain supervisors, and other ways in which graduates can obtain postgraduate “training” and maintain a connection the field besides having a supervisor.

On the other hand, Cabaniss’s finding that the candidates felt graduation and supervision affect their conduct of training cases seemed to Cooper common-sensically evident; he suggested that any person who has to report on what he or she is doing will do it differently than if he or she does not have to report on it. He also felt that all reality factors that influence the attitudes of the analysts and potentially influence interactions with the patient should be included in the category of countertransference. In today’s practice climate, the need to keep patients in treatment affects not only candidates, Cooper suggested, but all analysts. The crucial issues, he asserted (in accord with Dr. Halpern) is not the difference between candidate cases and graduate cases, but understanding the effects of education in leading candidates to make the kinds of interpretations that will help maintain patients in the treatment best suited for both the patients’ and the candidates’ interests. In Dr. Coopers’ view, these interests need not diverge; they may converge.

Discussion

The discussion section yielded many interesting comments; A sampling follows. Dr. Robert Michels argued methodologically that the lack of a control group in Dr. Cabaniss’s study diminished the usefulness of her conclusion that the training case is unique; he argued that extra-analytic issues are an inevitable part of conducting analysis post-graduation as well. Steven Roose pointed out that while technically the lack of a control group meant that the characteristics of candidates’ analyses could not be called specific, they nevertheless had been established by the study as characteristics. He also noted that one attribute of empirical research is that each study allows for further definition of issues. Drs. Lane and Levitan, both speaking from the perspective of the progression committee, pointed out the progression decisions, including “credit” for cases, are complicated and include many variables, many difficult to define. Dr. Roger MacKinnon expressed the view that the entire discussion was a happy commentary on changes in psychoanalysis: not long ago, such remarks as were being made would have provoked heresy trials. He felt that, since people spend a lot of time and money on this training, there must be something of value in it, and we need to study what is valuable in order to be able to continue to offer a useful experience. Several people expressed the opinion that the candidate’s situation was analogous to postgraduate realities.

Dr. Roose concluded by pointing out that the pedagogic questions regarding how and how well teaching and learning are occurring are not strictly separable, at least at this stage, from issues of what analysis and the analytic process are and how do we assess outcomes. In other words, if we are going to study how well we teach something, we have to know what it is we are trying to teach.

Reporter’s Commentary

Parsing the theme of this panel a bit differently reveals that it was also about the “fate of training” and perhaps, in turn, a prelude to discussions about the fate of psychoanalysis. Many comments from the audience anxiously referred to “the current practice climate.” Does psychoanalysis have some work to do in defining itself and demonstrating its relevance to our culture? One way to put the question would be: “What do we offer that neither science nor religion offers?” Studies like these two will help us articulate our position. But as we move forward, we may find that establishing an empirical/scientific basis for our field and reifying our own faith in our endeavor are insufficient. We want to show our culture that we can heal wounded and ill psyches. Both science and religion exist -- in fact, they coexist -- because they have convinced the culture that they are valuable. In short, they have “spread the word.” We must do the same.