Tuesday, August 21, 2007

Fourteen Research Topics In Humanistic Therapy

Entry for August 20, 2007:

Offered by participants at “Doing Research on Humanistic Psychotherapy and Training: Practical Strategies for Practice-Based Research”, Workshop given at Humanistic Psychotherapy Conference, San Francisco, California, August 2007. Used by permission.

Topics identified by focusing on areas of curiosity. For an earlier list of research topics/curiosities, see Blog entry for 22 May 2007. My comments and suggestions are given in italics after each idea.

1. Take data from therapeutic alliance ratings and use them to work backwards to help consumers pick a therapist.
Robert’s commentary: Nice idea. Alliance ratings often show ceiling effects, so it would be important to use an instrument with good variability at the top end of the scale. Also, it might be a good idea to use ratings from multiple clients, i.e., to identify therapists who were generally seen by their clients as participating in high positive alliances with their clients.

2. Students in doctoral [i.e., postgraduate in the UK] programs in clinical psychology and counselling (psychology): What do they learn and what do they use? How much of what they do is from what they learned vs. what they brought to training. Try comparing first vs. third year students.
Robert: This is why pre-post designs are such a good idea for training research, because of the importance of prior training/skill levels. A cohort design (that is, comparing first vs. third year students) can be useful, but a more powerful design would be to track one or more groups of students longitudinally over training.

3. How do I measure clients’ intensified sense of their lives, existential presence, engagement, commitment?
Robert: This sounds like a good outcome variable for existential therapy research. I would start by collecting different descriptions of the desired state; I would also want to think about doing a qualitative study of this state to provide additional descriptions. Then, I would put these descriptions together into a draft questionnaire, get feedback from friends and experts, then test it out on 100 or so people. I would use the results of identify and get rid of bad items and to develop subscales.

4. Effects of shifts in client language: Is there a relationship between client therapy outcome and changes in client use of language (used to describe experiential states) from (for example) “I’m angry” to “It’s angry in here”.
Robert: This sounds like a really interesting topic. One way of getting starting and making this more manageable would be as part of a larger pre-post outcome study on the effects of focusing-oriented or other experiential/humanistic therapies. In such a study, one can then identify small subsets of high and low outcome clients in order to study their discourse practices for describing their experiential states.

5. [From the same person:] The effects of therapist use of language as in “You’re sensing”; “Something in you”; and “That place”. The hypothesized effects include (1) Changes in client language [see Idea 4]; (2) change from start to end within particular sessions, in terms of client ease, comfort, anxiety, and feeling less overwhelmed; (3) overall therapy outcome.
Robert: Ideas 4 & 5 are related and could be combined within the same larger research project. Client mood ratings could be collected at the beginning and end of sessions, using, for example, parts of Stiles’ Session Evaluation Questionnaire (SES), or building on the list you've given. It would be important that outcome measures include something that measures emotion regulation (Jeanne Watson’s group has developed one), or something that ties in to the change pointed to by the changes in client language.

6. Client nonverbal indicators of change: Does client change correlate with changes in client gestures, body movement, and especially gestures that differ from verbal content?
Robert: This is similar to idea 4, except that the variable is nonverbal (gestures) instead of verbal (discourse markers). Again, I would start with outcome and work backwards via high and low outcome clients.

7. How do clients tell their stories in therapy and how do their content and structure change over time? How can this inform the therapy process? How do changes in client stories relate to therapy outcome and process?
Robert: There is one study I’m familiar with from Edna Foa’s research group on this topic in the treatment of trauma (Amir, N., Stafford, J., Freshman, M. S. Foa, E. B. (1998). Relationship between trauma narratives and trauma pathology. Journal of Traumatic Stress, 11, 385-392.). This topic is also similar to ideas 4 and 6 in the interest in relating client discourse to therapy outcome. Because discourse analysis is difficult and time-consuming its generally a good idea to focus on high or low outcome clients (the working backwards strategy). In this case, a rather different line of research on the same topic could be done as task analytic research on client narrative retelling work. I have a graduate student at the University of Toledo, Emily Beighner, working on the latter angle.

8. Self-discrepancy as an outcome measure: What is the relationship of self-discrepancy ratings to anxiety and depression? What implications does this have for understanding the process of change in therapy?
Robert: A self-discrepancy measure compares different ways of rating oneself, typically, how I see myself with how I would like to be (real - ideal discrepancy) and takes decreases in discrepancy as a sign of positive change. This idea sounds like a pretty straight forward construct validation study. It would be interesting, however, to add another person-centred outcome variable, such as the Strathclyde Inventory, which is supposed to measure psychological flexibility vs. rigidity.

9. (A) Temperament and focusing in therapy: Do persons with certain temperaments, such as intuitive sensing, thinking, feeling enjoy Focusing more vs. less? (B) What client markers and guidelines do I use in order assist my own perceptual decision process re: C temperament in relation to whether I use Focusing or not?
Robert: A and B suggest fairly different lines of research. In any case, there several angles to this, ranging from client predictor research using something like the Myers-Briggs Type Indicator scale to predict outcome, to task analysis of Focusing events in therapy.

10. What are the key humanistic principles that support and make more effective other traditional interventions with survivors of sexual or physical emotion abuse?
Robert: This seems a bit tricky as a topic. Could it be that you’re asking about studying the humanistic elements of psychodynamic and cognitive-behavior therapy? One approach to that would be to adapt Jones and Ablon’s method of creating prototypes of different types of therapy (they did psychodynamic and CBT) in order to develop a profile for humanistic therapy. That would be a very interesting thing to do!

11. In the UK, what are the effects of recent government funding policies on counselling training?
Robert: This is policy research, but timely. You’d probably want to start with a survey of course directors, then see if you could back that up with some kind of quantitative trends, for example, declining applications.

12. How does change really happen from the client’s point of view? Including all the pieces, what is change all about? For example, much change is nonconscious, such as clients acting different in everyday life. How do we get at this?
Robert: Asking clients is a good starting point, as in the Change Inteview, but as noted may not yield a comprehensive account of changes, because nonconscious changes are likely to be left out. Asking other people has been proposed in the past as a way to get around this. This could include the therapist but significant others could also be interviewed. Another possibility is using in-session process (e.g., client experiencing level or repeating interpersonal themes) as an outcome measure. These possibilities are tricky and time-consuming, which is why they are not used much. The problem with indirect measures and measures of nonconscious changes is that they either correlate highly with direct/conscious measures, in which case they are redundant and not worth doing, or they don’t correlate with direct/conscious measures, in which case their validity is suspect. That means you really have to believe this is important to want to do this kind of research!

13. Are there tools or attitudes that clients develop to handle future crises by acting as a fully functioning person? What are the long term effects of developing such tools or attitudes?
Robert: The Strathclyde Inventory, developed by Freire, is an attempt to develop a measure of the fully functioning person. Another possibility would be to look at measures of coping style, of which there are many, for example, Carver’s COPE (http://www.psy.miami.edu/faculty/ccarver/sclCOPEf.html). Your research question, then, comes down to assessing the long-term predictive validity of this sort of measure, tested by following a group of people over time while giving them one or more measures of general functioning (problem distress, life satisfaction) and your measure of full-functioningness.

14. Client and therapist experiences of futility about therapeutic process.
Robert: This topic seems to me to be closely related to the research topics of therapeutic difficulties and alliance ruptures, an interesting, clinically-relevant topic about which there is a small scientific literature. For alliance ruptures, see the work of Safran and Muran; for therapeutic difficulties, see studies by Davis et al. (e.g., Davis, J.D., Elliott, R., Davis, M.L., Binns, M., Francis, V.M., Kelman, J., & Schroeder, T. (1987). Development of a taxonomy of therapist difficulties: Initial report. British Journal of Medical Psychology, 60, 109-119.) Measures of each of these exist, such as the Working Alliance Inventory 12-R and the Therapeutic Difficulties Scale.

As always, I really enjoyed hearing about all these different research ideas, and hope that at least some of them can be developed further, possibly along some of the lines suggested. Good researching!

No comments: