Originally published in Cognitive Therapy Today, the Beck Institute Newsletter in 2004.
I have recently been supervising a number of novice therapists and experienced therapists new to cognitive therapy.
At the beginning of (and throughout) supervision, I stress conceptualization, the cognitive model, and structure.
Also from the beginning I stress the importance of using the conceptualization to plan treatment within individual sessions. I ask therapists to keep two questions in mind during a session:
- “How can I help this patient feel better by the end of the session?”
- “How can I help this patient have a better week?”
These two questions have helped my supervisees deliver more effective treatment. One supervisee, Carol, had been a therapist for 22 years. For the most part, before our supervision began, she delivered supportive psychotherapy mixed with some psychodynamic and some problem-solving techniques. Her sessions were quite unstructured. She and patients discussed whatever was on the patient’s mind at the moment. Sometimes the topic involved a current difficulty; sometimes it was related to distressing childhood experiences. Patients tended to drift from one topic to another without closure and Carol followed their lead. When Carol occasionally made suggestion to help solve a problem, she rarely checked on the implementation of her advice at the next session and the problem was dropped unless the patient herself brought it up again.
Carol had been treating Cynthia, a 35-year-old divorced woman, for almost a year. Cynthia’s managed care company, noting that Cynthia had made no progress in that time, contacted us to supervise Carol. Carol was a willing supervisee. She truly wanted the best for her patients and realized that Cynthia was stuck in therapy. Cynthia was severely depressed and alcohol dependent. She had recently gone on disability from her job as a manager in a department store. I could hear from the initial audiotape that Carol sent me of her session with Cynthia that the two had developed a good therapeutic alliance. Therefore, my first goal (along with cognitively conceptualizing the patient) was to get Carol to actively do problem solving with Cynthia, teaching her the cognitive and behavioral skills she needed in the context of solving those problems. I role-played with Carol how to set agendas with Cynthia that contained current problems.
Three difficulties arose. First, when Carol asked Cynthia what problems she wanted to discuss, Cynthia started to describe the problem instead of just naming it. Because Carol did not interrupt and guide her to just name the problem at the beginning of the session, Cynthia did not bring up other crucial problems until the end of the session, if at all. Second, Cynthia put items on the agenda that were not problems she needed help with or was willing to work on. Often these agenda topics were about the way her ex-husband had treated her in the past or complaints about her children. Third, Carol did not add to the agenda herself. Cynthia, hoping to avoid discussions of her alcohol dependence, did not put her drinking on the agenda and neither did Carol.
To solve the first problem, I suggested that Carol model for Cynthia how to name a problem: “Can you tell me what problems you want my help in solving today? Can you just name the problem, for example, problem with drinking, problem with feeling lonely, problem with money?” When Cynthia again launched into a description of the problem, I role-played with Carol how to interrupt her, “Sorry to interrupt, but can you just tell me the name of the problem? Should we call this, “Problem with your son?”
After I recognized the existence of the second and third problems, I helped Carol conceptualize how it was that she thought Cynthia would get better. Through questioning, she was able to recognize that Cynthia most urgently needed to learn skills to deal with her urges to drink, to manage her negative emotions (the precursor to drinking), to solve problems around being alone in the house (which was the only time she drank), and to structure her day and feel productive. I also helped her evaluate Cynthia’s other difficulties and we agreed that Cynthia’s chronic problems with her grown children and with her finances, and the problem of returning to work, were of lesser immediate importance, unless these difficulties led her to feel so upset that they triggered urges to drink. Carol related this new treatment plan to Cynthia and elicited Cynthia’s agreement to give these four areas priority in treatment.
It was difficult for Carol initially to interrupt Cynthia and steer the discussion. I ascertained that Carol did not have any negative thoughts about imposing more structure; she simply did not recognize when she got off course. At first I gave Carol a written list of questions to review with Cynthia. When Carol had difficulty allotting sufficient time to go through the list, I gave her a list of questions for Cynthia to read aloud and answer. Data from these questions were crucial to help Cynthia plan the session. They included questions about frequency and circumstances of drinking, frequency and strength of urges, automatic thoughts related to urges and drinking, and use of coping behaviors. They also included questions about how Carol was spending her days and what had most upset her during the week, as well as problems Carol predicted might arise before their next session. I helped Carol figure out with Cynthia, given their limited amount of time together, what was most important to work on, i.e., what would help the patient feel better by the end of the session and what would help the patient have a better week.
Before we began supervision, Carol’s formula was: To get better, patients need to unburden themselves and receive support and encouragement from their therapist. Carol was beginning to learn a new formula: To get better, patients need to work on solving specific problems with their supportive and encouraging therapist, with a focus on what they can do (and how they can think differently) to have a better week. Setting a good agenda was the first step. Following the agenda and teaching skills in the context of solving problems was the second step. In a future column, I will address this step and further challenges in supervision with Carol.