Dear Colleagues,

It has been a quiet summer in many ways but a lot is going on at SFPRG. The first piece of exciting news is that Patsy Wood has analyzed data from therapies from our own SFPRG clinic from a project started several years ago by Linda Tetzlaff, Marshall Bush, and Lynn O'Connor. Patsy has presented the findings elsewhere in this newsletter, but the results are heartening on many levels. The first is that in a small number of thirty-seven cases, which had enough data to evaluate, there were significant improvements in multiple outcome measures. This is significant because clinics and trainees are notoriously unreliable in delivering effective psychotherapy, but it appears that this clinic with SFPRG interns supervised by SFPRG supervisors was able to afford significant symptomatic relief to a difficult population of clients/patients. Read On

The SFPRG Education Committee is sponsoring three terrific workshops in the fall - so put these in your schedule! On September 27th, we are offering a four hour workshop at the San Francisco JCC entitled, "Aggression, Adaption, Evolutionary Theory and Control Mastery Theory" which will be taught by Drs. Heather Clague and Helene Goldberg.

Our second fall workshop on October 11th is entitled "Law and Ethics from a Control Mastery Perspective" and is taught by Jules Bernstein at 9 Funston Avenue.

Finally, on October 18th at the San Francisco JCC we have our second annual Introduction to Control Mastery workshop taught by Drs. Steve Foreman and Steve Kanofsky and Dr. Jan Schreiber. This workshop is six hours in length.

In addition all our wonderful ongoing case conferences will be starting up in early September including Steve Foreman's Conference on Wed afternoon; Peter Schumaker's on Monday morning; Micheal Lowenstein's case conference in Lafayette; and Helene Goldberg's Case Conference in Berkeley.

For the past four years the SFPRG Clinic and Training Center interns, post docs and trainees have been dutifully collecting intake and in some cases, follow-up data from their clients and inputting this into a database. These intake and follow-up protocols were originally developed from the joint efforts of the late Linda Tetzloff, Marshall Bush and Lynn O'Connor back in the early years of the SFPRG Clinic. These measures allow us as a Training Clinic to track the potential effectiveness of Control Mastery therapy as practiced by our interns with their clients. Critical to the ongoing momentum of this project was the lead role that certain post docs and interns played in making sure that survey forms were readily available; these interns also continued to encourage their fellow interns to keep inputting the data. Zohar Itzhar-Nabarro was in this lead role for many years and even did some preliminary analyses of the data which she presented to her colleagues several years ago. Over the past year since Zohar has been gone, Tim Lewis has been in this role of trying to keep the momentum of this project going. As a result of their efforts and that of their fellow interns and post docs who continued to collect and input their clients' data, we currently have intake data for some 444 clients and follow-up data for 37 clients. This is an achievement we want to build upon. Read On

Formal support for the control mastery theory of psychotherapy process was originally derived from case studies (Fretter, Bucci, Broitman, Silberschatz, & Curtis, 1994; Silberschatz & Curtis, 1993; Silberschatz, Fretter, & Curtis, 1986) designed in accordance with the "events paradigm" (Rice & Greenberg, 1984). These labor intensive studies involve several steps. First, a reliable plan formulation must be developed for each case. Three or more control mastery oriented clinicians careful review early clinical material (e.g., transcripts of the first few sessions) and individually infer the client's plan. Their proposed plan items are combined and randomized into a master list of potential plan items, each of which are rated by all of the clinicians as to its relevance to their assessment of the client's plan. The extent of agreement between clinicians is calculated and items are excluded from the plan that are either redundant or consensually rated as having lesser relevance (Curtis, Silberschatz, Sampson, & Weiss, 1994). Before the resulting plan formulation can be applied to test hypotheses, the "event" of interest has to be identified and quantified. Control mastery studies have typically focused on plan compatible interventions or key tests. Events are assessed by reading through all of the remaining transcripts, identifying and extracting relevant passages, randomizing them, and presenting them to another team of raters who have been trained to judge how well each passage supports the client's plan formulation. In addition, the client's immediate response to the event must be assessed by extracting passages of patient speech before and after each event. These passages of patient speech are also randomized and presented to yet another group of judges who are trained to assess the client's level of response. The final step of the process is to correlate the quality of the event (e.g., passed versus failed test) with the amount of response (e.g., increased versus decreased insight). Studies following these procedures have convincingly shown that plan compatible interventions and/or passed key tests are associated with immediate improvements in experiencing, boldness, and insight (Fretter et al., 1994; Silberschatz & Curtis, 1993; Silberschatz et al., 1986) and that treatments with more plan-compatible interventions have better overall outcomes (Norville, Sampson, & Weiss, 1996; Silberschatz et al., 1986). Read On

The SFPRG Board, under the new leadership of Steve Foreman, is forming a Fundraising Committee. Our goal is to broaden SFPRG's donor base and achieve full financial sustainability to ensure our future. While we have earned a reputation of excellence in the psychotherapeutic community both locally and internationally, membership fees alone cannot support our important work and the ambitions of our organization. Many SFPRG members and friends of SFPRG have valuable experience and/or ideas on how to promote the marvelous work we do. Anyone interested in joining this committee should contact Karen Hubble at (415) 921-7106 or kchubble@comcast.net

I've started publishing videos on YouTube that will in a small way increase the visibility of control mastery. They will be commentaries on the the HBO series "In Treatment." The video that's posted now is at http://www.youtube.com/watch?v=15B7vCpSoz8

Secondly, this is an example of research that our group has been conducting but which has not been seen by many of us. There is a fair amount of research currently being conducted through SFPRG that has not been publicized and is under-recognized. George Silberschatz and Marshall Bush are currently involved in research projects. Lynn O'Connor has produced a steady stream of research findings, published over the last few years. There are many dissertations currently underway at the Wright Institute supervised by Control Mastery researchers. I have asked the chair of the Research Committee, John Curtis, to catalogue the recent and current research, so that we are aware of what our colleagues are achieving and so we can plan subsequent research in a more systematic, accessible way. Research has been a hodge-podge, partly because of the lack of communication and awareness of what we all have been doing. I'm hoping by cataloguing our work, it will make it easier for people who are interested to participate in research and increase our productivity as a group.

There are many other projects that we have launched in the last month. Karen Hubble is putting together a fundraising committee, which she chairs. The committee's first task will be to improve public relations with the community. We need to revamp our brochure, revitalize our Web presence, and explore other ways to publicize our group. In addition to being a prerequisite for fundraising, publicizing our group will also help expand awareness of our teaching program and our clinic.

We have expanded our number of directors on the Board from 12 to 15. We are currently interviewing and nominating a slate of directors to get the Board to full capacity, which I hope will be ready to be voted on by our August meeting. Other changes on the Board include the election of Molly Sullivan as Secretary. Congratulations to Molly and congratulations to the Board which has not had a secretary for quite a while. Our budget for 2008-2009 is almost ready for passage. I am still requesting help from accountants or bookkeepers who are friends or relatives of SFPRG, who can advise us on budget matters and assist Peter Schumacher who has done a terrific job as Treasurer and head of the Budget Committee.

The Clinic is doing well as many of the current interns are coming to the end of their training program. Jessica Broitman is doing a wonderful job running the Clinic as is Carol Drucker, its new Training Director. A new cadre of excellent interns is ready to go for their start date in September. The Clinic is setting up a waiting list so the new interns will have enough clients when they begin their rotations. Clinic staff is revamping paperwork and reevaluating research protocols in the Clinic. Please remember to tell colleagues and refer low fee clients to our clinic.

The Education Committee, under the energetic chairmanship of Patsy Woods, is preparing a full educational program for the fall. We will be offering an exciting fall conference on Aggression, led by Heather Clague and Helene Goldberg. Because the Introductory Course on Control Mastery Theory has been so popular and sells out every year, we will be offering an additional Introductory Course in the Fall led by Steve Kanofsky, Jan Schreiber, and myself. George Silberschatz and Jan Schreiber will continue to offer the original Introductory Course in the spring. There will also be a full contingent of case conferences, research conferences, the Post-Graduate Psychotherapy Training Program, and the Ethics Conference. The Education Committee welcomes more help from the membership planning conferences and teaching.

This year's Membership Drive is starting again at the end of the summer. I encourage all of you who are members to sign up again and those of you who are not to join. The challenge of our San Francisco Psychotherapy Research Group and Clinic is to remain vibrant and do meaningful work, even when our founders, Joe Weiss and Hal Sampson, are no longer present to teach, inspire and lead. Our obligation is to teach the theory, test it, expand it, and improve it. If Joe and Hal were active today, SFPRG would still have the same challenge.

Joe and Hal's achievement, among many, was to attract a great number of able clinicians, smart researchers, and nice people to the theory and to the group. Most of these people are still around, as are a great number of younger people who have come to the group more recently. I invite all of you to study with us, to teach, to write, to supervise, and to participate in any way you find meaningful. Join the membership, join a committee, and join the Board. This is a very exciting time for the group and a wonderful opportunity to make a difference.

To analyze this data, I included only those participants where we had both intake and follow-up data so we could track progress in therapy over time; this turned out to be a relatively small sample of only 37 clients. So using SPSS, a statistics program, this is what the data that could be analyzed demonstrated: The 37 participants in this sample of clients were treated by our SFPRG Clinic interns between 2004 and 2008. Outcome was measured by examining changes in reported depression using the Beck Depression Inventory (BDI), in anxiety using the Beck Anxiety Inventory (BAI) and in three subscales on the Lambert Therapy Outcome OQ-45 Questionnaire (OQ-45), a common instrument used in therapy outcome research to track progress in treatment. These three subscales included a measure of subjective discomfort, a measure of interpersonal relations and a measure of social role performance. To evaluate outcome on these five measures, the mean difference scores from intake to follow-up were analyzed using a t test for dependent samples to determine if statistically significant change had taken place over time. In addition to these five outcome measures, a number of predictor variables were also analyzed to see if they correlated with the outcome variables. First, there was a 74 item Parental Characteristic Questionnaire (37 questions for each parent) that looked at both strengths and potential weaknesses of participants' parents. There was also an 18 item subscale that looked at self identified personality characteristics in clients measured by the Lambert Outcome Extension Questionnaire. Unfortunately the number of participants in our database where we had both parental characteristic data and outcome data and personality and outcome data was too small to analyze.

The demographics of the clients in this sample were quite diverse. In terms of reported gender, 62% were female and 38% were male. The ages of clients ranged from 18 to 77 years old with a mean of 37 (SD=1.4). In terms of ethnicity, 65% of the clients identified as European American, 8% as African American, 14% as Asian American and 13% identified as biracial. Overall it was an educated sample with 3% having had some high school, 5% with a high school degree, 19% with some college, 41% with a college degree, 10% with some graduate school and 23% with a graduate degree. In terms of sexual orientation, 81% identified as heterosexual, 10% as bisexual, 5% as "questioning," and 3% as lesbian. In terms of relationship status, 27% identified as single, 22% as divorced, 17% as a domestic partner, 11% as married, 5% as separated and 18% as in a monogamous relationship. As for having had prior therapy, 22% had never had therapy before while 78% had had some kind of prior treatment. In terms of prior or current experience of being on psychiatric medications, 41% of the sample had been on or were currently on psychiatric medications and 59% had never been on medication.

Overall there were mixed results in the outcome measures in this study in terms of demonstrating progress over time in therapy. Also the time from intake to follow-up varied from a minimum of three months to up to four years in some cases: the average duration across the sample from intake to follow-up was approximately six months. Also the majority of these therapies were ongoing treatments so follow-up did not constitute termination data but ongoing progress in therapy.

To measure progress in therapy, the differences in the mean scores at intake and follow-up were computed in the following areas: depression, anxiety, subjective discomfort, interpersonal relations and social role performance. Of these, three outcome measures showed a significant reduction in symptoms over time in therapy. Depression levels as measured by the Beck Depression Inventory dropped from a mean of 23.6 at intake to 13.6 at follow-up: t (21) = 4.883, p= .000. While this is a significant drop in scores, the mean scores at follow-up remained in the mild to moderately depressed range (according to the BDI manual, scores of 10 to 18 are considered in the mild to moderately depressed range). There was also considerable improvement on two subscales on the Lambert Outcome OQ-45 Scale. In the first subscale, the Subjective Distress scale, mean scores dropped from 45.7 at intake to 34.5 at follow-up: t(34) = 4.355, p = .000. This scale measures a mixture of anxiety and depression and overall subjective stress levels. According to the OQ-45 Manual, a drop of at least 8 points signifies a reliable change in therapy on this subscale and the cutoff for experiencing less distress and more modal well being is a cut off score of 36 or less; this change in subjective distress scores met both these criteria. There was also a marked improvement on the interpersonal relations subscale of the OQ-45 which examines the degree of distress experienced in key relationships including friendships, family, family life and intimate relationships: t(34)= 2.26, p=.03. In this subscale, the mean scores dropped from 19.5 to 16.6 over the course of therapy. Here the change was less dramatic and according to the manual, a drop of at least 8 points is necessary to show reliable improvement in this area. The cut off score in the manual for experiencing greater comfort with one's interpersonal relations is a score of 15 or lower and this mean fell just above that. Nevertheless this drop in scores was statistically significant.

In contrast to these three outcomes measures, two other outcome measures showed no statistically significant change over time. The first of these measures was anxiety symptoms as measured by the Beck Anxiety Inventory and the second was Social Role Performance as measured by the Lambert OQ-45 Social Role Performance subscale. In the Beck Anxiety scores which measure classic fight or flight anxiety symptoms, the mean intake score was 17.9; this score dropped to a mean of 14.2 at follow-up (t (32) = 1.535, p=.135). According to the BAI manual, a score below 21 points is considered low anxiety so while there was not a significant reduction in scores over time, both mean intake and follow-up scores fell solidly in the low anxiety range with only a slight reduction of anxiety at follow-up. Similarly the OQ-45 Social Performance subscale which measures the client's reported distress and sense of adequacy dealing with tasks related to employment, family roles and leisure activities showed no substantial change in therapy (t(34)=.234, p=.816). Scores on this measure are often correlated with overall measures of life satisfaction with lower scores suggesting higher levels of life satisfaction and vice versa. These scores changed very little in this sample: the mean score dropped very slightly from 13.3 at intake to 12.9 at follow-up. Compared to the normative data provided by the OQ-45 manual, these means were slightly higher than the mean for a non-clinical community sample which was 9.81 (SD= 3.91) but lower than the mean for a sample at a university outpatient clinic which was 14.24 (SD=5.62).

In summary, clients receiving therapy in this sample at our SFPRG Clinic and Training Center experienced significant relief over time in therapy. Most dramatic relief was found in clients' reported depression, subjective distress and in the difficulties they reported in their key relationships. In the areas of anxiety symptoms and social performance related to life satisfaction, reported scores at intake and follow-up were close to or in the normal range for these measures so less dramatic change took place. Nevertheless this is a good initial effort to document the positive changes that can occur over time in Control Mastery therapy as practiced by our interns and post docs at our SFPRG Clinic and Training Center. Given this relatively small sample, it is the hope of the Clinic Committee in collaboration with the Research Committee to build on these efforts so we can continue to document the effectiveness of Control Mastery therapy with larger samples and more varied intake and outcome measures.

Carrying on in the tradition of single case research, my colleagues and I have been pursuing an alternative way to determine whether and when control mastery prescribed therapy processes are associated with successful outcomes (Pole, Ablon, & O'Connor, 2008; Pole, Ablon, O'Connor, & Weiss, 2002). Our approach combines prototype methodology pioneered by J. Stuart Ablon and the late Enrico Jones (Ablon & Jones, 1998, 2002) with statistical procedures developed by John Gottman (Gottman & Ringland, 1981) and originally applied to psychotherapy by Dr. Jones (Jones, Ghannam, Nigg, & Dyer, 1993). In this method, international experts of particular schools of therapy are identified and asked to describe an "ideal" therapy session from the perspective of their theoretical orientation. Their description is given in the form of ratings of 100 Psychotherapy Process Q-set (PQS; Jones, Hall, & Parke, 1991) items. The PQS includes items describing a broad range of therapist behavior (e.g., therapist is empathic), client behavior (e.g., patient is demanding), and the therapist-client interaction (e.g., silences occur during the hour). Experts rate each item in terms of how characteristic (or uncharacteristic) it would be of their hypothetical ideal session. A prototype is created by combining the ratings of experts who share the same theoretical orientation. Ablon and Jones used this method to create prototypes of ideal psychodynamic therapy (PDT), cognitive behavioral therapy (CBT), and interpersonal therapy (IPT). They were able to measure the extent to which archived PDT, CBT, and IPT therapies conformed to these prototypes by training raters to use the PQS to independently assess the actual process that occurred in the archived sessions. Each session was assessed by 2 to 3 raters in a process that takes approximately two hours per session. After establishing that there is adequate agreement between raters, their ratings are combined into a composite score. The degree of similarity between the actual therapy process and both the target prototype (e.g., the CBT prototype in cognitive behavior therapy) and non-target prototypes (e.g., the PDT prototype in cognitive behavior therapy) can be calculated by correlating the PQS prototype scores with the composite PQS ratings of each session. Using this method, Ablon and Jones found many interesting results including: (a) specific forms of therapy can contain more ingredients belonging to the non-target therapy than ingredients belonging to target therapy (e.g., the interpersonal therapies of the NIMH Treatment of Depression Collaborative Research Program study were found to adhere more closely to the CBT prototype than to the IPT prototype); (b) greater adherence to the target prototype can predict better outcomes (e.g., greater adherence to the psychodynamic prototype predicted better outcomes in an archival psychodynamic therapy dataset); and (c) sometimes the presence of processes belonging to a non-target therapy can be more predictive of outcome than adherence to the target therapy prototype (e.g., in one archival cognitive behavior therapy dataset, better outcomes were more consistently associated with the level of prototypical psychodynamic process than the level of prototypical cognitive behavior therapy process) (Ablon & Jones, 1998, 2002).

Recently, we (Pole et al., 2008) have applied this method to the study of a long term control mastery case. Nine experienced control mastery therapists used the Psychotherapy Process Q-set to rate a hypothetical ideal session from their perspective. Though control mastery therapy (CMT) shares some commonalities with psychodynamic (PDT) and cognitive behavioral (CBT) therapies, it is interesting to note that the resulting CMT prototype was distinct from both the PDT and CBT prototypes. For example, among the most characteristic items in the CMT prototype were: (a) therapist accurately perceives the therapy process, (b) therapist focuses on guilt, (c) therapist is reassuring, and (d) patient tests the limits of the therapy relationship. The most characteristic psychodynamic items were: (a) patient's dreams or fantasies are discussed, (b) therapist is neutral, (c) therapist interprets defenses, and (d) therapist interprets transference. Finally, the most characteristic cognitive behavioral therapy items were: (a) there is discussion of homework, (b) discussion focuses on cognitive themes, (c) patient's treatment goals are discussed, and (d) therapist encourages patient to try new ways of behaving with others. It is important to note at this juncture that each prototype actually includes all 100 PQS items but in a different rank order depending upon how much emphasis it was given by the experts. For example, readers familiar with control mastery theory may be surprised, given the CMT focus on pathogenic beliefs, that the item "discussion focuses on cognitive themes" was mentioned as among the most characteristic of CBT but not CMT. In fact, the CMT experts saw this item as characteristic of their hypothetical ideal session but, whereas the CBT experts ranked it as the second most characteristic item, the CMT experts ranked it as the 23rd most characteristic item. This is presumably because in CMT pathogenic beliefs do not have to be discussed directly but may rather be treated through corrective emotional experiences (see Pole et al., 2008 for a complete list of item rankings).

We applied the CMT prototype to a highly successful two year control mastery therapy that had been previously described in the literature both in terms of its qualitative/clinical features (Bloomberg-Fretter, 2005; Fretter, 1995) and in terms of its quantitative Psychotherapy Process Q-set (PQS) profile (Jones et al., 1993; Pole & Jones, 1998). This treatment was tested not only for how closely it conformed to the control mastery theory (CMT) prototype but also for how closely it conformed to the psychodynamic (PDT) and cognitive behavior therapy (CBT) prototypes. As expected, we found that the treatment was more similar to the CMT prototype than the PDT prototype. However, to our surprise, we also found that the treatment was more similar to the CBT prototype than the CMT prototype! This puzzling result led us to decompose the prototype scores into items reflecting therapist behavior, client behavior, and the therapist client-interaction in order to better understand the source of the unexpected result. Consistent with our original expectations, we found that the therapist's behavior conformed more closely to the CMT prototype than either the CBT or PDT prototypes. The client's behavior conformed most closely to the CBT prototype and the interaction between client and therapist was equally consistent with prototypical CMT and prototypical CBT. These results show that though the therapist was clearly behaving in accord with the teachings of control mastery theory, the overall process was strongly influenced by CBT elements in both the client's behavior and the client-therapist interaction.

We were also able to determine the extent to which conforming to each prototype predicted symptom improvement in this case using a sophisticated statistical approach called bivariate time series analysis (Gottman & Ringland, 1981). Though the details of the analysis are too complex to describe here, suffice it to say that this method has two big advantages over simple correlations. First, it directly assesses and corrects for dependencies that are likely to exist in data obtained from the same subjects over time. Correlations assume that all data points within a given variable are independent from each other. This assumption is likely to be unrealistic in single case studies. Second, it can provide evidence of causality by testing whether adherence to a given prototype tends to precede (rather than follow) changes in symptoms. Correlations can typically only show associations but are ambiguous about whether the therapy process influenced the symptoms, the symptoms influenced the therapy process, or both. Using the bivariate time series analysis, we found that the full CMT and CBT prototypes both predicted symptom change but the full PDT prototype did not. When the full prototypes were decomposed, both the therapist (e.g., therapist accurately perceives the therapy process) and client (e.g., patient tests the limits of the therapy relationship) components of the CMT prototype predicted change and the therapist-client interaction items (e.g., patient's aspirations and ambitions are discussed) showed a trend towards predicting change. Recall that the client's behavior was not found to be particularly consistent with prototypical CMT. Yet, our results showed (as others have before) (Ablon & Jones, 1998, 2002) that low frequency therapy processes can have potent clinical impact. Conversely, even though the client's behavior was highly adherent to the CBT prototype, only the CBT therapist-client interaction items (e.g., there is discussion of homework) significantly predicted change. The fact that the client behavior component of the CBT prototype did not predict change reminds us that the most conspicuous therapy processes may not be the most effective. Finally, though the overall PDT prototype did not predict change, the client behavior (e.g., patient achieves a new understanding or insight) and client-therapist interaction items (e.g., patient's dreams or fantasies are discussed) did predict change. This last result is actually consistent with earlier work on this case showing that whereas the therapist's use of psychodynamic techniques did not influence change, the client's use of psychodynamic processes (e.g., free association) did influence change (Pole & Jones, 1998). In sum, the time series analyses indicated that the client's symptom change was significantly driven by the therapist's adherence to prototypical CMT behaviors, the client's adherence to prototypical CMT and PDT behaviors, and therapist-client interaction consistent with prototypical CBT and prototypical PDT.

These results must be qualified in several ways. First, as readers of the full article (Pole et al., 2008) and its predecessors (Jones et al., 1993; Pole & Jones, 1998) will discover, this study involved estimating a large number of data points representing the client's symptom status at various points in the treatment. This is not a flaw inherent in the methodology but a limitation in this particular dataset. The time series analysis requires a relatively large number of observations (e.g., greater than 50) but only fourteen symptom scores (assessed every sixteen sessions) were available. Thus, we interpolated missing data points between the observed data points to derive the needed number. It is unknown how this estimation influenced our results. A second limitation, inherent to single case research is that the results cannot be assumed to generalize to other cases. For example, our findings do not suggest that all or even most CMT treatments contain significant CBT elements. We merely show that a CMT treatment conducted by an experienced CMT practitioner can contain significant CBT elements and that these CBT elements can predict symptom change. A third and potentially important limitation is that the prototype approach seemingly ignores case-specific elements that are central to much of the thinking of control mastery theory (Fretter et al., 1994; Silberschatz et al., 1986; Weiss, 1993). It is possible that some of the unexpected results were due to failures of the CMT prototype to adequately capture key aspects of CMT process that would vary from case to case. This possibility should be tested in future research by comparing our "generic" CMT prototype with one constructed for a particular case (Pole et al., 2002) or by comparing the predictive power of the generic CMT prototype versus conventional CMT process measures derived using the events paradigm (e.g., plan compatibility of interventions).

Despite these limitations, this study has made several important contributions. First, it replicates and extends previous single case studies supporting CMT (Fretter et al., 1994; Jones et al., 1993; Messer, Tishby, & Spillman, 1992; Norville et al., 1996; Pole et al., 2002; Silberschatz & Curtis, 1993; Silberschatz et al., 1986). Replications are always important in science but they are especially so in a science built on single case research (Pole, 2001). Moreover, the prototype method increases the likelihood of further replication and extension of research on control mastery theory because it: (a) is much less time consuming than the events paradigm method, (b) does not require that any of the researchers have expertise in control mastery theory, (c) assesses psychotherapy process in terms that can be directly compared to other theoretical orientations (e.g., PDT, CBT, and IPT), and (d) can easily be applied to datasets involving multiple (rather than single) cases (Ablon & Jones, 1998, 2002). Consequently, this research offers the opportunity to advance the field by making research on control mastery theory more widely accessible and ultimately more likely to be accepted as evidence-based practice (APA Presidential Task Force on Evidence-Based Practice, 2006). It is my hope that this study inspires others to take this opportunity.

Correspondence concerning this article should be directed to Dr. Pole by e-mail: npole@email.smith.edu.

References for this article follow.

References

Ablon, J. S., & Jones, E. E. (1998). How expert clinicians' prototypes of an ideal treatment correlate with outcome in psychodynamic and cognitive-behavioral therapy. Psychotherapy Research, 8, 71-83.

Ablon, J. S., & Jones, E. E. (2002). Validity of controlled clinical trials of psychotherapy: Findings from the NIMH Treatment of Depression Collaborative Research Program. American Journal of Psychiatry, 159, 775-783.

APA Presidential Task Force on Evidence-Based Practice. (2006). Evidence-based practice in psychology. American Psychologist, 61, 271-285.

Bloomberg-Fretter, P. (2005). Clinical use of the plan formulation in long-term psychotherapy. In G. Silberschatz (Ed.), Transformative Relationships (pp. 93-109). New York: Brunner-Routledge.

Curtis, J. T., Silberschatz, G., Sampson, H., & Weiss, J. (1994). The plan formulation method. Psychotherapy Research, 4, 197-207.

Fretter, P. B. (1995). A control-mastery case formulation of a successful treatment for major depression. In session: Psychotherapy in practice, 1, 3-17.

Fretter, P. B., Bucci, W., Broitman, J., Silberschatz, G., & Curtis, J. T. (1994). How the patient's plan relates to the concept of transference. Psychotherapy Research, 4, 58-72.

Gottman, J. M., & Ringland, J. (1981). The analysis of dominance and bidirectionality in social development. Child Development, 52, 393-412.

Jones, E. E., Ghannam, J., Nigg, J. T., & Dyer, J. F. P. (1993). A paradigm for single-case research: The time series study of a long-term psychotherapy for depression. Journal of Consulting and Clinical Psychology, 61, 381-394.

Jones, E. E., Hall, S., & Parke, L. A. (1991). The process of change: The Berkeley Psychotherapy Research Group. In L. Beutler & M. Crago (Eds.), Psychotherapy research: An international review of programmatic studies (pp. 98-107). Washington, D.C.: American Psychological Association.

Messer, S. B., Tishby, O., & Spillman, A. (1992). Taking context seriously in psychotherapy research: Relating therapist interventions to patient progress in brief psychodynamic therapy. Journal of Consulting and Clinical Psychology, 60, 678-688.

Norville, R., Sampson, H., & Weiss, J. (1996). Accurate interpretations and brief psychotherapy outcome. Psychotherapy Research, 6, 16-29.

Pole, N. (2001). Making the case for single case research. Psychologist Psychoanalyst, 10(5), 19-21.

Pole, N., Ablon, J. S., & O'Connor, L. E. (2008). Using psychodynamic, cognitive behavioral, and control mastery prototypes to predict change: A new look at an old paradigm for long-term single case research. Journal of Counseling Psychology, 55, 221-232.

Pole, N., Ablon, J. S., O'Connor, L. E., & Weiss, J. (2002). Ideal control mastery technique correlates with change in a single case. Psychotherapy: Theory, Research, Practice, Training, 39, 88-96.

Pole, N., & Jones, E. E. (1998). The talking cure revisited: Content analyses of a two-year psychodynamic psychotherapy. Psychotherapy Research, 8, 171-189.

Rice, L. N., & Greenberg, L. S. (1984). The new research paradigm. In L. N. Rice & L. S. Greenberg (Eds.), Patterns of change (pp. 7-25). New York: Guilford.

Silberschatz, G., & Curtis, J. T. (1993). Measuring the therapist's impact on the patient's therapeutic progress. Journal of Consulting and Clinical Psychology, 61, 403-411.

Silberschatz, G., Fretter, P. B., & Curtis, J. T. (1986). How do interpretations influence the process of psychotherapy? . Journal of Consulting and Clinical Psychology, 54, 646-652.

Weiss, J. (1993). How Psychotherapy Works. New York: Guilford Press.