Abstract
In 2003 the Columbia Center for Psychoanalytic Training and Research began a prospective study of graduates designed to both describe and understand their professional trajectory. The study has two components: a quantitative component based on an anonymous comprehensive questionnaire given analysts yearly starting with their graduation, and a qualitative component comprising analyst interviews beginning at the end of the first postgraduate year and repeated every two years. Analysis of the first six years of the qualitative study shows that analysts will talk openly about their practice and careers and that when they do, practical issues are a dominant concern. Analysts both immersed and not immersed in four-times-weekly analytic cases experiment with adapting skills developed in training to treat cases in analysis seen less frequently. Analysts without four-times-weekly case immersion are engaged in analytic careers, participate as faculty at the institute, and report a high degree of career satisfaction. The major findings of this study compel changes in psychoanalytic training programs. The field would do well to address actual clinical practice experience in institute curricula and training programs, thus making analytic training more relevant.
For the past decade our group at Columbia has been studying the career paths of analysts after graduation from training. The starting point for this work was a concern that there might not be a sufficient flow of new training analysts, which is dependent on analytic immersion, to sustain our training program. Our first study (Cherry et al. 2004) documented the clinical immersion of analysts for the first nineteen years after graduation. We demonstrated that while little four-times-weekly analysis was being conducted by most of our graduates, there was a subset that maintained the immersion necessary to meet APsaA’s training analyst requirements. These results prompted our interest in the career paths that develop after graduation, why analysts choose one path over another, and whether they are happy with their choices.
We decided that the best study design to answer these questions was to follow graduates prospectively from the point of graduation. In 2003 the Columbia Center for Psychoanalytic Training and Research began to collect data for a prospective study of graduates designed to describe and understand their professional trajectories. The study has two components: a quantitative component based on an anonymous comprehensive questionnaire given analysts yearly starting with their graduation, and a qualitative component comprising analyst interviews beginning at the end of the first postgraduate year and repeated every two years. Thus, we chose a mixed-method study design. This approach has recently been reviewed by Tillman, Clemence, and Stevens (2011) and cited as a methodology specifically useful for psychoanalytic research. In addition, we had had positive experience with this approach in a study of candidates’ experience of supervision (Cabaniss, Glick, and Roose 2001). We were hoping that the interviews would add depth to our understanding of the motivation for career choices; while we were confident the quantitative data would tell us what was happening, we believed the qualitative data would give us a deeper and more complex view of why. Here we will describe the systematic process used to analyze the interview data and will report initial qualitative data from the first five years of interviews.
Quantitative Findings
We have previously reported on the first five years of the quantitative data (Cherry, Wininger, and Roose 2009). Twenty-six of 29 Columbia Center graduates from the years 2003 through 2007 (90%) participated in the study (New York State Psychiatric Institute IRB #5726R). The cohort was 58% female (15/26), and the mean length of analytic training (including leaves) was 5.4 years (SD 1.4). Ten of 26 graduates (38%) terminated their training analysis before graduation; the mean length of these analyses was 6.5 years (SD 1.7). The mean length of training analyses terminated in the first postgraduate year was 6.9 years (SD 1.9). Nearly all graduates reported a positive experience in training and high career satisfaction. The most important finding was that within five years of graduation two subgroups of analysts emerged: half who continue to sustain immersion in four-times-weekly analysis by starting new cases as their control cases terminate, and half who do not. Surprisingly, the groups were more similar than different; they did not differ in gender, professional degree, family income responsibility, whether they had a salaried job, or in the amount of analytic case immersion completed during training. Both groups reported positive feelings about their training experience and the future of psychoanalysis, and were active teachers at the institute. While it had long been believed that graduates who do not conduct analyses after training were demoralized and/or less skilled, our data suggest otherwise. Analysts in the group who do not maintain immersion reported conducting as much psychotherapy as the other group, felt very positive about their training, were optimistic about the future of psychoanalysis, and were engaged in the field. They felt a strong connection to the Columbia institute and have taught there and in other training programs, including psychiatric residencies.
We looked in our prospective data for characteristics that might herald the emergence of these two distinct groups. At graduation, we had asked analysts to report, using a four-point Likert scale (4 = most likely), whether they thought they were likely to pursue training analyst (TA) appointment and whether they were likely to make being a psychoanalyst a primary career identity. Those who were more immersed in analytic practice in the first three postgraduate years were more likely to state they wanted to become a TA (immersed analysts: Mean 4.0, nonimmersed analysts: Mean 2.9, p < .006) and make being a psychoanalyst a primary career identity (immersed Mean 4.0, nonimmersed 2.4, p < .001). These differences in anticipated career path at graduation were the only significant difference between those who did and those who did not continue with immersion in four-times-weekly cases.
Method
Interviews
All graduates from the Columbia Center for Psychoanalytic Training and Research since 2003 were invited to participate in a longitudinal study of careers after analytic training. At five years, 26 of 29 graduates (90%) had completed confidential baseline questionnaires focusing on four domains: experience in analytic training, current private practice, postgraduate activities, and career goals. These members of the longitudinal study cohort were invited to join the interview component of the project at the end of their first postgraduate year. Participants signed a written consent form to join the interview study; the form outlined procedures for audiotaping and transcribing the interviews (New York State Psychiatric Institute IRB #5726R).
The interview is an open-ended semistructured interview lasting approximately forty minutes. The initial interview, or baseline interview, conducted one year after graduation, covers three distinct periods: before analytic training, during training, and the year after graduation. The interviewer asks the analyst to tell how he or she initially became interested in psychoanalysis and psychoanalytic training. Participants are asked to comment on each aspect of training—classes, cases, supervisors, the training analysis—as well as the overall experience. They are encouraged to reflect on how training and earlier experience shaped their feelings about psychoanalysis as a treatment, a professional field, and a career path. Finally, they are asked about the year since their graduation, including questions about their clinical experience with psychoanalysis, ongoing learning, first experiences as a faculty member, and goals and plans for the future.
Follow-up interviews are conducted every other year for eleven years after graduation. The follow-up interviews ask about the analyst’s professional evolution, experience conducting analysis, and plans for the clinical and academic future since the last interview. While the interview covers all of these domains, it encourages interviewees to elaborate in areas that seem most meaningful at the moment and not dwell on less engaging items. They are asked to consider what has influenced their choices. The interviews are conducted by two of the authors (SC and JM), both graduate analysts from Columbia. A transcription service generates printed text copies of the digital recordings.
After six years of conducting interviews (2004–2010), we had 31 baseline interviews, 24 first follow-up interviews (three years after graduation), and 13 second follow-up interviews (five years after graduation) and began analyzing the qualitative data. By the end of this period, 2 of 31 participants (6%) had “dropped out” of the interview study due to relocation or career changes.
Data Coding and Grounded Theory
Names of colleagues, courses, and other identifying specifics were expunged by our research assistant, after which the interview transcripts were uploaded into the N-Vivo 8 software package. We formed a research team consisting of both interviewers (SC and JM) and an analytic candidate at Columbia (LH) to analyze the interviews. Team members read the same three baseline and three follow-up interviews. We then identified and agreed on emergent themes using the principles of grounded theory (Glaser and Strauss 1967; Glaser 1992). Grounded theory involves an iterative process whereby we read interview transcripts as a group and identify categories and themes as they emerge. This is a process of repetition and integration of themes that group together. The process enabled us to make a list of categories and themes, which we put together for use as an initial codebook (Glaser and Strauss 1967). At times our analytic listening skills made it difficult to choose what level of depth and inference to use in selecting emergent themes. Deciding to stay relatively close to the surface, we identified themes evident in the manifest content rather than elements that the interviews were not designed to elicit and that have long been difficult to agree on, elements such as defenses and other process-oriented variables (Vaughan et al. 1997). The list of themes included broad categories such as supervision, control cases, or converting new cases, which helped us subdivide and organize the text. We also identified subthemes that rather than simply reflecting categories prompted by the interviewer’s questions, emerged unprompted in the narratives. These emergent themes and subthemes include items such as feasibility of starting new cases, effectiveness of analysis as a treatment, conviction that analysis works, and affect regarding cases. These organizational categories and emergent themes and subthemes were organized into a formal codebook. Each item was assigned a code number.
For over a year, the research team read the same interviews, first independently and then meeting to discuss the presence of categories and themes, formulate new codes and revise existing ones, and update the codebook based on consensus. This was a time-intensive process. Independent reading of a single interview required on average an hour of one’s time. Meeting to review an interview could sometimes take up to three hours. Eventually we noted that when we compared our independent readings, we made few additions or changes. Once we agreed on the codebook’s organization, we independently coded three additional interviews and began to assess our interrater agreement. We continued to modify the codebook based on areas of agreement and disagreement, streamlining the codebook to delete codes that were unreliable across the coders, and ultimately generated a final version. (This is the iterative process involved in grounded theory.) Finally, we decided to have each coder write a brief narrative summary after reading each interview. This summary notes the respondent’s attitude toward psychoanalysis, his or her experience in training, view of him- or herself as an analyst, and whether he or she has a strategy for beginning new psychoanalytic cases. These narrative summaries were not included in our measures of interrater agreement, as noted below.
Interrater Agreement
Before proceeding to independent coding of the interviews, we established agreement among the three coders by having them independently code three baseline and three follow-up interviews. This created a problem in that standard measures of chance-corrected agreement (kappa) rely on two-rater formulas. As a result, we estimated agreement using a generalized form for kappa known as Krippendorf’s alpha (Hayes and Krippendorf 2007). Our estimates revealed substantial agreement (in the .61–.8 range) for four interviews, whereas two interviews were estimated to be in the moderate range (.41–.6) (Landis and Koch 1977).
We then discussed each instance of discrepant coding and came to an agreement about whether to use a code. The consensus was generally easy to reach and indicated that omissions and commissions were generally due to a factor of level of detail in coding or to lack of clarity in defining the exact meaning of the code. We continued to annotate the codebook to clarify when to select a given code and any areas of confusion. We made some minor codebook adjustments and added ten additional codes to allow greater coding accuracy, ending up with a total of 85 codes for the baseline interviews and 46 for the follow-up interviews.
Given that interviewees were allowed to elaborate in divergent areas depending on which items in the interview were salient at the time, this level of concordance was determined to be acceptable; we felt we had standardized our independent coding of the interviews sufficiently to proceed with having only one of us code each interview. Moving on using a single coder for each interview, we entered the codes into N-Vivo 8. We checked for “rater drift” midway, by having all three raters code the same interview again. At this juncture, our interrater agreement remained robust (Krippendorf’s alpha = .66).
Results
We are reporting our initial analysis of the first six years of interviews, which included graduates from 2003 through 2009. All but two of the analysts enrolled in the quantitative study agreed to participate in the interview component (31 of 33, or 94%). As noted in Table 1, we have excellent retention of the respondents (80–100%). The interviews included in this analysis at the six-year point include 31 graduate analysts who have been interviewed at least once, 24 analysts interviewed twice, and 13 interviewed at least three times (See Table 1).
Participation and retention of qualitative study respondents in the interview study
In order to select areas of focus, we first tabulated how often each code was selected in the interviews as a group. We then focused on codes appearing in at least 50% of the interviews.
Baseline Interviews
In the first interview, analysts are asked to reflect on why they undertook analytic training, their experience in training, and their experience during the first year after graduation. With regard to the pre-training years, the most often cited dimension of how they became interested in pursuing training (noted by 25 of 31 analysts interviewed) was the influence of a mentor, teacher, or person other than one’s therapist or analyst. Seventeen talked about a long-standing interest in psychoanalytic ideas and/or being psychodynamically minded as a factor in why they pursued training. Personal treatment experience (16/31) and wanting to improve their clinical skills (15/31) were other frequently cited motivations for pursuing analytic training.
Follow-up Interviews
At each subsequent interview, analysts are asked about their experience in practice and overall career in the two years since the last interview, allowing us to track their responses over time. The overall frequency of themes discussed did not vary from the first to the third interview, a space of four years. At each juncture, at least two-thirds of the analysts interviewed discussed the following themes:
Overall experience of training, which is generally positive
Thoughts about ongoing supervision, regardless of whether they are still in supervision
Other postgraduate activities, most often teaching at the institute or elsewhere
Experiences with trying to convert psychotherapy patients into psychoanalysis, both successful and unsuccessful
Practical concerns about whether four-times-weekly analysis is feasible for themselves and their patients
Their level of interest in being certified or becoming a training analyst
All of these topics reflect specific questions asked in the interview, except for comments about the practical aspects of starting new analytic cases, which came without prompting.
Individual Analysts: Graduates from 2003–2005
We took a detailed look at the thirteen analysts who had been interviewed at least three times. We chose this subgroup because they had the longest career trajectory in the study. For the remainder of the results presented, we refer to these thirteen analysts. We first read the narrative summaries written by the coder after reading the interview transcription. We reached consensus on the presence of themes. Themes from the narrative summaries included the following:
They identify themselves as analysts.
They were satisfied with the training program.
They have an ongoing sense of connection to the analytic community, largely through their role as teachers, whether at the institute or in other training programs.
They are satisfied with their careers regardless of whether they had ongoing four-times-weekly analytic cases.
Bitterness and dissatisfaction were rare.
While these themes dominated the reviewers’ summaries, there were rich descriptions of many aspects of the postgraduate experience coded in the full text. The most frequently cited topics emerged as themes in the narratives of these thirteen analysts are noted below:
Clinical immersion in a range of psychodynamic treatments
Discussions of how they are grappling with the realities of present-day analytic practice
Experimentation with techniques learned in training and developing their own views about how best to help their patients
Reflection on their ongoing postgraduate education and whether to attain further credentials such as certification from APsaA and appointment as a training and supervising analyst
Reactions to the transition to joining the institute faculty
We took these coded subsections of text for these thirteen analysts and read through them again as a group (SC, LH, and JM), reaching consensus on the presence of themes. In essence, we did another round of grounded theory analysis on coded interview subsections from these analysts. The topics most often discussed naturally grouped into two areas: (1) clinical practice and conducting more analysis; (2) postgraduate education and further credentialing.
Clinical Practice
The analysts talked at length and in depth about the challenge of continuing to find cases for analysis, as well as their current efforts to adapt their new skills to practicing psychodynamic treatments ranging from one to four times a week. The interviews contained rich descriptions of patients in both therapy and analysis. In reading these descriptions we were especially interested in expressions of affect related to clinical work.
Feelings about their patients
Segments of text were coded for the presence of affect and then further subdivided into positive and negative valences. The affects associated with their caseloads of both therapy and analytic patients include a wide range of positive and negative feelings. Only one of the thirteen analysts did not express affect when discussing cases. In general, descriptions of control cases and new analytic cases generated more affect than did therapy cases —which may reflect the fact that we encouraged more discussion about their experience doing analysis. Positive and negative valence was about equal and often coexisted when describing a patient or clinical moment. Analysts most often talked about transitional moments in treatments, and these descriptions generated the most affect.
Examples of affect around transitions with patients
Getting a new case going: “Picking up my first case at the end of the first year was so exciting and scary at the same time. I felt very charged up.” Case abruptly ending: “She quit in a stormy borderline, very primitive, intense, angry, you know, upsetting way—and, that was so tough. . . . I really questioned how much my need for a training case was affecting my judgment.” Patients not doing well: “She had spent a huge amount that year on drugs alone, she looked awful, and I felt terrible.” Changing the frame: “We just agreed that she should come once a week . . . and even that has been a very erratic experience, but it feels like for me it is better. . . . it has been helpful to me in taking the pressure off and I feel able to say things to her that I wasn’t able to say before.”
In addition to affects grouping around transitional themes, we also noted that analysts are emotionally affected by their patients, particularly when they leave. For example, one analyst states: “I still have case loss anxieties, though I now have seven people in analysis. . . . I don’t always know if it is the patients or myself, maybe I can analyze it in myself instead of being in a panic over it.” Another analyst reflects on a patient who left: “It was a disappointing and painful experience—to see that he could break things off in that dramatic kind of way after all that time.”
A majority of these analysts made positive general comments about their experience practicing analysis. While two of the thirteen made superficial and vague positive comments, nine made more detailed positive statements about practicing analysis. Only two expressed more negative experiences of analysis, though these were noted in reference to specific patients.
Positive experiences conducting analyses
“There’s enough variety and one is different than the next.” “I like it, it wears well.” “It’s amazing—a totally different kind of process you are in.” “It’s a totally different feeling of working with somebody.” “I find it challenging and very interesting. And one thing I have found this last year, which has been nice, and new for me, is a kind of comfort with what I don’t know that is actually very relaxing, sort of, it just makes me more curious about what I don’t know as opposed to anxious about what I don’t know.” “I guess relaxing isn’t the right word, it’s just the fluidity of it, the immersion in it is a more comfortable place for me than constantly having to readjust to the next person coming in the door.” “I like thinking about the complexities and really getting to know everything that’s going on with somebody.” “Watching all the ins and outs of how their behavior changes and how they express different views about it is very satisfying when that happens.”
Mixed and negative experiences conducting analysis
“It is difficult—an arduous process and at other times exhilarating seeing the kind of changes that accrued. . . . There are times when I get frustrated with the pace or dealing with tough character pathology. One patient is very dug into her masochism, it’s frustrating, but I comfort myself and calm myself down by saying what else am I supposed to do to help her?” “I think if I saw people improving more, or more able to really discuss their transference it might be more gratifying for me right now . . . though I still enjoy it of course.”
Getting new cases started
Almost all of the analysts interviewed state that they would like to do more analysis than they are currently doing (11/13). Four of the responses reflected anxiety about doing more analysis.
Anxiety about getting to do more analysis
“I haven’t recommended it to anybody for quite a while. . . . it feels like gee, I hope there are others to follow.” “At least if I had one or two cases in analysis I would be happy.” “I’m not sure if five years down the road I might feel like I don’t want to have not done any adult cases.” “I wish I’d been bolder trying to get patients to think about analysis—sometimes I am comfortable and just feel that this is not a huge problem for me. . . . The point where I find it most troubling is that I feel like it would be good for me to have more analytic experience.”
Others are more comfortable with their amount of case immersion. One analyst who has some immersion states: “I like having the mix; I don’t think I’d want to just do psychoanalysis.” Another with no cases states: “I’m just as happy with it as without it—I’d be happier with it, but I’m not beleaguered.”
Converting cases
Transitioning therapy patients to four-times-weekly analysis is central to starting new analyses. All those interviewed are asked about whether they are thinking about converting patients from psychotherapy to analysis. Responses break down thematically around analyst-centered and patient-centered concerns. Nine analysts spoke about their own motivation to convert cases, and five discussed the patient’s suitability, motivation, and resistance. Analyst-centered themes around motivation include wishing to employ analysis to address patients’ problems and general belief in the treatment’s effectiveness.
Analyst-centered themes around starting new analyses
Conviction: “After the patient’s mother’s death it was clear to me that she has to be in analysis.” “It makes sense with a lot of people I see that it’s a valuable treatment.” Conflict in the analyst: “Whose benefit is it for?” “In training I was more motivated and more willing to lower my fee—there was a time limit.” “I definitely want to be doing analysis but haven’t figured out yet how much of my practice it’s really going to be. I want there to be some of it, definitely.” Conflict in both patient and analyst: “There are two people where there is huge resistance—and I don’t know no matter how expert I was if I would be able to get past their resistances. These people are making significant compromises in their life . . . and it’s a shame. On the other hand, they’re people who are doing very well in therapy. I just think they could do better.” Therapy vs. analysis: “This person has been in analysis before and doesn’t really want to do it again. . . . my own feeling is that they are doing very in-depth work in this format so I can’t totally feel like it should be changed.”
The patient’s motivation and suitability also comes up regularly. “At some point they are either going to be interested or not.” Or: “If the climate was different and others told them you should really be in analysis, it’s a really great thing, and then they might think differently.” In addition to motivation, some discuss the clinical challenge of specific cases and their appropriateness for analysis given their significant pathology or, in other cases, their lack of psychological mindedness.
Practical concerns
Practical matters related to seeing patients in analysis were a common theme. Looking at these references as a group, we subdivided comments about the feasibility of psychoanalysis into those that referred to the challenge of practicing analysis from the patient’s vantage point (i.e., whether they are suitable and whether analysis is feasible) and those considering feasibility from the analyst’s perspective. Of the thirteen analysts we reviewed in detail, four referred to feasibility issues from both the patient’s and the analyst’s perspective, two mentioned only the patient’s perspective, and six only the analysts’.
Practical concerns and resistances from the patient’s perspective
“It’s hard to get people to come four times a week.” “He had lots of resistances to it and concerns about money.” “The tension is between having to make a living and wanting to do something right for these people, who are already committing to two or three times a week, but that would require opening up more time and reducing my fee.” “I am having a hard time finding people who are able and willing to come four times a week. I would consider seeing them for a reduced fee, but I’m seeing patients who come twice a week who could benefit, could afford it, but don’t want to come four times a week and they’ll say it’s the money. But I’m not going to collude with their resistance by lowering the fee; when I know the fee’s not the issue, I’d rather interpret the resistance. Whether it’s just resistance or practicality, I’m having a hard time getting people to come even three times per week. I’ve had several twice a week but the greater frequency is just not happening for me.”
There is evidence that these analysts are making personal decisions to make four-times-weekly analysis an option in their practice.
Practical concerns from the analyst’s perspective
“I wish I could work fewer hours and not work four late nights . . . but I am not really able to do this if I want to keep doing as much analysis as I am doing.” “There’s only so many hours in a week that I can lower my fees for.” “Money and time—I think if I saw an adult who needed an analysis and could come at random times in the day I would think about it more.” “I’ve set my schedule up to revolve completely around my children—that’s the main constraint.” “I’m not willing to see people at this time in my life for a very low fee.” “I am worried about having a family and being able to support them because you can’t make much money doing this and that’s a real concern.” “There are people interested but I don’t have time to offer them.” “My first case pays a reasonable fee but lower compared to my full fee, my second case pays close to a full fee now, my third case pays a low fee. So I feel blessed.” “I have been fortunate in seeing psychoanalytic patients from my training who have been able to pay a full fee for psychoanalysis, so that hasn’t been a problem of saying I have to cut my fees.” The same analyst also notes: “But I have patients in the wings who could do psychoanalysis who I would have to drop their fees. And so there’s a tension there. At this stage of my life—my wife isn’t working, the kids are at home . . . this tension of what do I do—I have worry about that domain while at the same time seeing patients who could be in psychoanalysis.” Sacrifices for analysis: “I have a completely different tolerance level for the people who are in analysis—what times I will give them and what fees. . . .” “It’s been difficult to put in the hours to do analysis; that has been a real problem and a big financial strain because it is a decision to make less money.” “It causes some conflict at home that I’m donating my time for a lot of things. And we have children to put through school. . . . But I love my job.”
Others seem to adapt to these realities—whether it’s to refer patients to a candidate, make do with fewer than four sessions, or use the couch twice a week. As one analyst says: “One of the places I’ve arrived at in terms of three or four times a week is that I think you can get a lot of good analytic work done doing three times a week. As I saw on the other patient, four times a week and nothing really gets done. So the four-times-a-week requirement seems arbitrary in a way. I understand the history . . . but the realities of doing the work are very difficult.”
These barriers from the analyst’s vantage point are expressed as real conflicts, but often with a practical rather than intrapsychic basis. Patients are often described as having intrapsychic resistance to analysis, separate from external realities. Intrapsychic resistance to analysis within the analyst is generally not spontaneously mentioned.
Strategy
Nine of the thirteen analysts refer to strategies they employ to deepen treatments and convert cases to analysis.
Strategies used to transition patients to analysis
The gradual approach: “Somehow I feel more comfortable building up some kind of connection with the person and then suggesting maybe intensifying the treatment in some way—so there are a couple of relatively new patients who I can imagine bringing it up with further down the road.” Four years later this same analyst says: “We’ve started talking about it a little bit more, just in terms of her fears about it and why I think it would be a good thing for her, but I’ve been very gentle.” “I haven’t gone directly from once a week to four—which I think I’m just sheepish about. I usually put people in twice a week and see how they do for a while.” Direct approach: “I recommended analysis to him at the start of treatment with me. . . . It was one of the first times I saw a therapy patient and didn’t wait a long time to recommend analysis.” “I just think of all the rule-outs that we discussed in class—I am not sure whether I go through them systematically in my mind. It’s almost like if a person really has issues they need to work on and they don’t have rule-outs—just propose analysis!”
Some analysts just appear to have no one currently appropriate. One immersed analyst stated he has people “on the bench” waiting to start. And for some, despite their efforts, many do not take. One analyst says, “I have two or three twice-weekly patients, who would do very well in analysis, and I have discussed it—but for a variety of reasons it hasn’t happened.” Two years later the same analyst says, “I’ve discovered that to the extent that this comes up it’s usually something I am bringing. It’s not them spontaneously bringing it up. It’s not something I find clinically helpful to bring up a lot because it ends up working against you— . . . you can sort of overplay it.” Another analyst at three years out has “mentioned it to the occasional few—but my practice is pretty stable, so it’s not like there are that many new people—the people to whom I raise it over time are utterly uninterested in coming that many times per week.”
Finally, one analyst seems to make do with what is realistic: “I’ve tried suggesting analysis—and they all complain about the frequency—that they don’t want to come more than twice a week. So sometimes I do put them on the couch twice a week. I don’t consider it analysis—but in many ways it is analysis, it’s analytic process. . . . another one . . . did not tolerate therapy face to face and it was easier to do therapy on the couch.”
Technique
This theme of flexibility regarding the frame emerged as the most often discussed aspect of analytic technique. No direct questions were asked referring to analytic technique, so this emerged as an unprompted theme in interviews with ten of the thirteen. Many of the references to flexibility focused on frequency of sessions, and most often touched on conducting analysis three times weekly.
Comments about flexibility in analytic technique
Session frequency: “I don’t think this frequency business is important—I recently had to ask myself was I primarily insisting on four times for my own purposes? That’s lousy. . . . I’m not going to get into that power struggle over three or four times per week.” “One of the places I’ve arrived at is that you can get a lot of good analytic work done at three times per week. . . . One of the three-times-a-week patients had a process that really developed and a lot of work done in the transference—I don’t think there’s something magical about four times.” The couch: “I don’t feel every patient needs to lie down to experience analysis.” “[The patient] sits up for a while, goes back to lying down, then freaks out again and we try to figure out what’s going on.” “If they want to use the couch, I let them, no matter what frequency.”
References to one’s theoretical orientation or technical approach other than the frame were far fewer (only three of thirteen) than comments about frequency and the couch. One analyst comments on how much he has come to value working with the “momentary give and take” in the “here and now” rather than just learning something repressed from the past. Another analyst comments that she is increasingly using maneuvers that are intersubjective: “I would say some pan out and others really backfire.” Finally, someone else reports using supportive techniques interspersed with more exploratory work. Even with the theoretical comments referring to technique, analysts seem to be noting how their experience differs from a more stereotyped classical approach—whether to session frequency, use of the couch, or working with newer or blended techniques.
Postgraduate Education, Professional Development, and Career Path
Work/life balance
Eleven of the thirteen made spontaneous comments about work/life balance, stating that their personal life trajectory affects their career path and practice. Whether it is having time to find a life partner, be available to children, or tend to unforeseen circumstances, all are given primacy as the scaffolding to which analytic careers must adapt. Two analysts noted that their high analytic immersion has come with real personal sacrifices.
The training analysis
Our quantitative study revealed that about two-fifths of the analysts in our study were still in their training analysis during the early postgraduate period. Comments made about the training analysis experience were generally positive, though not detailed in nature. Five analysts noted without being prompted that their training analysis was a source of conviction about the effectiveness of analysis as a treatment. Comments such as “nothing proves treatment works better than this” capture that spirit. Using personal experience as a benchmark also comes through in the following comment: “I do think of my own experience in the transference with my analyst—and we [the analyst and her own patient] don’t get to that level, though he [the patient] could.” Another analyst mentions referring back to “interpretations by my analyst.”
Supervision
All were asked about their experience in supervision during training and had largely positive responses, though stories of bad match and some supervisory changes were well represented. Comments about supervision after graduation were far more nuanced; they come from those who have a stable ongoing attachment to their supervisor (four analysts), from those who are moving away from reliance on their supervisor (ten), and from those who are considering returning after a break (seven).
Comments about postgraduate supervision
Ongoing attachments: “I could go the rest of my life to supervision.” “I’ve been with her eight years and have no desire to change it. . . . I just find it so helpful.” “He knows my weaknesses.” “It helps me with insecurity about my professional identity. . . . it’s reassuring to know you are working like an analyst.” Separation themes: “I started feeling like there wasn’t much to say in supervision.” “Time to fly on my own.” “It was good to work unsupervised . . . but I had a consultation again with her which was great to know she was there.” “I feel like I’m hanging by a thread sometimes.” “It’s hard not to have someone to run things by constantly.” Returning in the future: “I’ve been getting a tune-up on a tough case.” “My plan is to go back this year; it just made sense developmentally for me.”
Certification and training analyst appointment
Echoing the comments on immersion, a range of approaches was evident regarding these credentials. Two analysts had these goals in mind from the beginning and were actively pursuing them. Others show ambivalence and cite a sense of limitation again, analyst-centered more often than not.
Reasons analysts cite for not pursuing certification and TA appointment
Other priorities: “Dating is really [at the] top of my list now, and I would put analytic stuff on hold for that.” “I’ve been delaying [doing write-ups] as I have little extra free time.” Practical limitations: “It feels financially undoable.” “There are people interested in analysis and I don’t have time to give them.” “I should be emphasizing making a living, taking care of my family.” “I’d be interested once my kids are older.” Self interest: “If I started taking process notes and thinking more of using her for my certification I would feel very awkward . . . it’s a personal feeling with this specific patient. . . . I feel it would almost be corrupting the treatment.” “This guy really would do better in four-times-weekly, so I don’t think it’s just for my own selfish interests.”
The theme of idealization/devaluation of the TA role came up overtly with two analysts. One, who is actively pursuing TA appointment, says “it was kind of the ultimate pie in the sky—it seemed ludicrous.” The other, in the process of choosing whether to defer thinking about appointment, says he is “sure there are certainly benefits to the little badge of becoming a TSA.”
The institute
Connection to the institute is strong among these analysts, with many describing activities they currently are engaged in, such as teaching or committee work (seven) or mentioning people they enjoy and with whom they feel connected (five). However, two referred to the lack of connection they experience and nine complained about institute politics. Of note, the four analysts who had no analytic cases did not express feelings about institute politics, whereas all but one of the nine with cases did. Some of the comments reflect disappointment over not getting referrals, not feeling included, or not being thanked for volunteering time. There were complaints about how the administration handles political issues, and some expressed dismay at “rigidity” they experience in the administration. Throughout these comments, as in the discussions about whether to stay in supervision, ran a developmental theme. There is the feeling of entering the adult world and becoming disillusioned with how the grown-ups behave.
Comments reflecting disillusionment
“I was naive [to think] that this analytic community would all be well-analyzed people.” “Things happen and you expect more from them, you know, analysts.” “I don’t like to get involved in things where I may have to, you know, watch the grownups fighting.” “I wish it were more of a happy family.”
Correlations with the Amount of Analysis Practiced
To look for associations between emergent themes and degree of immersion we focused on the thirteen analysts surveyed one, three, and five years after graduation. We were able to group the analysts in three subgroups according to their current degree of immersion. Of the thirteen, four had no ongoing four-times-weekly cases, three averaged one or two, and six averaged three or more. Reading through the narratives and coded text, we were unable to ascertain any clear thematic differences correlated with level of immersion. Themes related to clinical practice and career path appear scattered and unrelated to the amount of immersion in four-times-weekly analysis.
Discussion
Recent graduates are willing to be interviewed and are forthcoming about their career and practice experience. Their interviews bring to life the postgraduate experience, revealing that graduates have strong emotional engagement with their analytic cases and much affect around how these early analytic treatments unfold after graduation. Whether their patients stay and get better really matters to them. They are living out the trajectory of their early cases and are emotionally invested and engaged.
These interviews support findings from the quantitative study that the postgraduate experience is largely similar for those who do and do not maintain immersion during the immediate postgraduate period. However, these data add perspective to the quantitative data in a number of important ways. First, the interviews reveal that the postgraduate period is a time of transition and further adult development for graduate analysts. Second, they show that analysts’ decision-making about conducting more analysis is dominated by real-life practical issues about the feasibility of four-times-weekly practice. Finally, they reveal a much more active experimentation with session frequency than was apparent from the quantitative data.
Analysts’ focus on the practical aspects of the frame can be viewed from a variety of perspectives. One view is that graduates are in a developmental phase involving experimentation with the frame they were taught in training. Another view is that graduates are using practical factors, such as time and money, defensively, possibly as an internal resistance to doing more analysis. Finally, it is possible that these analysts are grappling with truly challenging external realities and being both flexible and creative in putting their large investment in analytic training to use. Real-life obstacles may be a central challenge to integrate with their training experience during the early postgraduate period.
A Developmental Period
There is a rich literature referring to the early postgraduate period as a time of professional development, as analysts complete their training analysis and sort through identifications with their supervisors, teachers, and theorists (Craige 2002; Fogel and Glick 1991; Gabbard and Ogden 2009; Rosenbloom 1992; Smith 2001; Skolnikoff 2000; Spurling 2003). Our interviews corroborate many of these themes in both language and content. Developmental themes were prominent in graduate analysts’ reflections about whether to seek further supervision and in their thoughts about how much involvement they seek at the institute. Analysts’ experimentation with new techniques and different treatment frames reflects a process of working out their relationship to classical models and trying alternatives on for size. Later in the study, we should be able to delineate specific developmental trajectories for the early career period for the complete sample of graduates studied. We can then clarify whether there are specific developmental trajectories that apply to those who do and those who do not maintain immersion.
Feasibility of Analytic Practice
A striking dimension of the interview data is the extent to which feasibility and issues of a practical nature dominate the discussion about starting new analyses and sustaining immersion. Practice time, fees, and balance between work and family were the most frequent emergent themes when analysts were questioned in an open-ended way about their analytic practice. On the one hand, this is not surprising given the decades-long decline in the popularity of psychoanalysis and in insurance reimbursement for it. This finding is consistent with our main finding from the quantitative data: that the path toward TA status and practicing psychoanalysis as a primary career identity correlates with starting new cases and sustaining immersion. Perhaps upon graduation the desire to become a TA generates a different cost-benefit equation right from the start. Just as candidates accept low fees and work additional hours to meet training needs, aspiring TAs are similarly motivated to sustain the immersion needed to qualify.
Beyond Practical Issues
Despite evidence that current practice realities may be the most pressing factor affecting the analytic practice patterns of recently graduated analysts, it is unsatisfying to accept this without deeper reflection. Why aren’t deeper explanations of motivation revealed in these lengthy, self-reflective interviews? Our graduates are certainly taught to consider their counterresistance to doing deep work and whether the countertransference-transference relationship is somehow getting in the way of confronting a patient’s resistance. Resistances to analytic practice among recent graduates are described in the literature (Parsons 2006; Ehrlich 2004). While these interviews reflect a strong awareness that patients use practical issues as a resistance to deepening involvement in treatment, very little is said about the analyst’s own reluctance to do deep work or to confront their own focus on practical reality. Are we just hearing analysts’ defensive resistance in these interviews?
Some authors go beyond the concept of defensive resistance and view graduates who avoid full frequency cases as having larger conflicts about being an analyst. Wille (2008) attributes analysts’ low case frequency to their having an ambivalent “internal psychoanalytic object.” He argues that these analysts have additional manifestations of ambivalence, including disappointment with training, hostility toward analysis, and reluctance to identify oneself as an analyst. We do not see this in our study. Our graduates are not consciously ambivalent about psychoanalysis, they engage with the institute actively and with pleasure, and speak positively about training and their own analysis.
On the other hand, many analysts feel that the definition of analysis as treatment four times weekly is outdated or superficial; today it is common to experience rich analytic process treatments of lower frequency. Some analysts favor using more process-based criteria to determine whether a patient is in analysis. We continue the four-times-weekly definition in this study because it is the way our graduates are taught and remains an absolute criterion for receiving credit for training cases at Columbia. Our findings suggest that after training, analysts adopt a more process-oriented view and embrace the idea (and often the reality) of analytic treatments with fewer sessions weekly. With this approach, they are content to foster analytic process more broadly with their patients. In this way, they feel clinically engaged in psychoanalysis, continue to identify as psychoanalysts, and stay active in the institute. Those who aspire to become TAs must adopt the four-times-weekly criterion to qualify, but those not pursuing that path may foster analytic process without strict adherence to that frame. Our data show that at least half of our graduates embrace this postgraduate path with real career satisfaction and that even those who aspire to become TAs are amenable to analytic treatments at lower frequency.
Limitations of the Study
One limitation of this study is that it is conducted at a single institute. Thus, in designing our study we were especially cautious that analysts not be identifiable. To achieve this, our study design includes keeping the questionnaire data, which are anonymous, uncoupled to the interviews on an individual level. In consequence, the quantitative and qualitative data sets can be considered and compared only by group. This protects privacy, especially in the quantitative study, but limits our data analysis. An additional limitation of a one-institute study is that we do not know if our findings apply to other institutes. While some data suggest that immersion is similar in other settings, both nationally and at other local institutes, the specific climate at other locations has not been studied and could differ from Columbia’s experience (Brauer, Brauer, and Falk 2006; Kaplan, Wininger, and Cherry 2007; Pretzky, Aizaga, and Cherry 2006). In addition, our small sample size makes it hard to look for differences within our population and between graduation-year cohorts. This should improve over time as we incorporate a full ten years of graduates.
Another potential limitation is that the interviews are conducted by analysts whom the respondents know. This may facilitate ease and candor if there is rapport based on friendly contact at the institute. On the other hand, familiarity could be inhibiting if the interviewee perceives undesirable affiliations and biases. While we plan to inquire about this specifically in later interviews to clarify how this potential bias operates for each study participant, this will remain a possible limitation to our approach.
One possible explanation for the lack of deeper self-reflection in these interviews is that their structure was not conducive to elucidating this dimension. Possibly the questions were not pointed enough in that direction. Or possibly respondents were wary about self-revelation on these more personal topics. We plan to adapt the later interviews with an eye toward more directly asking analysts to reflect on their possible internal resistances and on factors other than practical matters that influence their practice decisions.
Finally, many analysts believe that the training analysis experience is central to postgraduate analytic practice. Our interviews so far do not contain a deep level of introspective, candid discussion about analysts’ experience of their training analysis and how that may have affected their motivation to practice analysis. While we did prompt interviewees to discuss their training analysis, responses were generally brief and few reflected on whether their experience has affected their desire to practice similar treatments. One reason for this absence could be that most graduates have either just recently ended their analysis, are in the process of ending, or were still in their training analysis at the time of the interview. We hypothesize that this proximity could feel “too close” to reflect on comfortably during an interview and likely prompted analysts to draw a personal boundary. We are considering adding direct questions about the impact of training analysis on the motivation to start new cases, but in later interviews, when analysts are further along in the post-termination period.
Conclusion and Future Directions
The major findings of this study inspire modifications in psychoanalytic training programs. Not only will analysts discuss the details of their practice and career path, but when they do, practical issues are a dominant concern. Analysts are experimenting with adapting skills developed in training to treatments at lower frequency regardless of their level of immersion. In addition, immersion does not predict engagement in analytic careers, participation on the institute faculty, or career satisfaction. Thus, the four-times-weekly criterion now seem largely out of step with current practice patterns and the postgraduate career paths of analysts.
Institute curricula could be updated to better prepare candidates for actual postgraduate clinical practice. Topics might include how to conduct analysis at a lower frequency, when in the course of such treatments to consider integrating psychotherapy techniques, and the analyst’s countertransference to making such choices. These topics go beyond simply interpreting a patient’s resistance to full analytic frequency. At Columbia we are beginning to explore some of these topics in a psychotherapy course for senior candidates. As a field, we would do well to evolve toward reflecting current clinical practice in our training programs.
Our plan is to continue the interviews along with the survey data collection and to revise the later interviews in an attempt to encourage greater self-reflection. Over time, with larger numbers, we may be able to observe any divergent experiences between analysts who do and do not continue with case immersion. Eventually we will have longitudinal descriptions of each individual analyst to compare with others. We encourage others to embark on similar research in their local psychoanalytic communities.
Footnotes
Acknowledgements
The authors thank Susan Tross and Robert Waldinger for methodology consultation, Lionel Wininger and Jenna Baldachin for research assistance, Joel Sneed for statistics consultation, and Nicole Yoskowicz for software consultation. This research was supported in part by grants from the International Psychoanalytical Association and the American Psychoanalytic Association.
Sabrina Cherry, Associate Clinical Professor of Psychiatry, Columbia University; Chair, Admissions Service, and Training and Supervising Analyst, Columbia University Center for Psychoanalytic Training and Research. Juliette Meyer, Lecturer in Psychiatry, Columbia University; faculty, Columbia University Center for Psychoanalytic Training and Research. Luke Hadge, candidate, Columbia University Center for Psychoanalytic Training and Research. Madeleine Terry, doctoral student and graduate assistant, Department of Clinical Psychology, City University of New York. Steven P. Roose, Professor of Clinical Psychiatry, College of Physicians and Surgeons, Columbia University; Director, Neuro-Psychiatry Research Clinics, New York State Psychiatric Institute; faculty, Columbia University Center for Psychoanalytic Training and Research.