Interpersonal Psychotherapy: History and Future

The story of interpersonal psychotherapy (IPT) began at Yale University in 1970, when Dr. Gerald Klerman was joined by Dr. Eugene Paykel from London to design a study to test the relative efficacy of a tricyclic antidepressant, both with and without psychotherapy, as maintenance treatment for ambulatory nonbipolar depression. The evidence for the efficacy of tricyclic antidepressants in reducing the acute symptoms of depression was strong, yet the main treatment for depression at the time was psychodynamic psychotherapy. The few studies testing psychotherapy focused on behavioral treatments and were limited in scope and sample size. A manual for cognitive-behavioral therapy (CBT) was under development by Dr. Aaron Beck. At the time, it was clear that many patients with acute depression relapsed after termination of tricyclic antidepressant treatment. It was unclear how long psychopharmacologic treatment should continue or whether psychotherapy had a role in the prevention of relapse. Some psychotherapists thought taking medication would make patients less interested in psychotherapy, whereas some psychopharmacologists felt psychotherapy would undo the positive effects of medication by having patients talk about upsetting experiences.

Dr. Klerman, then head of the Connecticut Mental Health Center and faculty at Yale School of Medicine, felt that a clinical trial of maintenance tricyclic antidepressants should, as much as possible, mimic clinical practice. Because many patients received both psychotherapy and medication, either together or in sequence, he felt psychotherapy should be included in the maintenance treatment trial, if for nothing more than a milieu effect. He was not convinced he would find a psychotherapy effect because no positive clinical trials of psychotherapy had been conducted. In fact, prior trials that examined patients receiving psychotherapy lacked the design necessary to draw any conclusion. However, Dr. Klerman was convinced that psychotherapy could be subjected to rigorous testing in a clinical trial.

In planning the trial, Dr. Klerman was first tasked with defining the type of psychotherapy and specifying the procedures to be used. Psychotherapists could then be trained, and the quality and stability of treatment could be tested. He felt that the psychotherapy tested in the trial should be suitable for a time-limited treatment of depression. This common-sense clinical approach, initially called “high contact,” was the basis for IPT. A time-limited psychotherapy, such as IPT or CBT, defined in a procedural manual used to train therapists and subjected to testing similar to medication testing, was a novel idea at the time and drew much skepticism from clinicians.

Three guiding principles governed this early work. First, randomized controlled clinical trials were important for testing and establishing the efficacy of all treatments, including psychotherapy. Second, a broad range of standardized assessments was needed for measuring outcomes, including quality of life and social functioning. Finally, treatment results needed to be replicated. Dr. Klerman did not develop training programs or attempt to disseminate IPT until findings were replicated outside the original group of enthusiastic developers.

The team tasked with developing IPT included Dr. Klerman and Dr. Paykel, both psychiatrists; Dr. Brigette Prusoff, a recent graduate from the Yale Department of Biostatistics; and me. I was, at the time, an inexperienced social worker who could work only 2 days a week because I had small children. The clinical trial tested amitriptyline, placebo, and no pill, with or without weekly psychotherapy, for 8 months among ambulatory patients with depression who had responded initially to medication. We set about designing the psychotherapy. Dr. Klerman was impressed with Dr. Beck’s progress in defining cognitive therapy. He gave me a typed document of approximately 100 pages from Dr. Beck describing the procedures for CBT. Dr. Klerman said a similar document would be needed for supportive psychotherapy, which was to be used in the new trial. He noted that “supportive psychotherapy” was a vague term that needed to be defined, as Beck had done for CBT. He then said to me, “Design the psychotherapy.” I was so surprised and pleased to be given this assignment—I had had little experience designing and documenting clinical procedures—that I set about writing scripts detailing the procedures and dialogue of the psychotherapy on days when I was not supposed to be in the office. It was a challenging and exciting assignment. It became an iterative group process: Dr. Klerman, or Gerry as we all called him, took the lead and reworked the scripts I wrote, and others contributed data (e.g., life events or cases) to illustrate a point.

Several contextual features, I later realized, defined our approach to designing the psychotherapy. We were working in a psychopharmacologic clinic where we routinely made a diagnosis using standardized assessments and followed the patient’s clinical course. Dr. Klerman and Dr. Paykel were developing standardized assessments and using control groups to assess the role of life events in the onset of depression and relapse. Dr. Klerman was both a psychopharmacologist and psychotherapist, and, as leader of the team, he held a broad view of depression that influenced the direction of our work. Although he thought depression was basically a biological illness, he was impressed with how social and interpersonal stress could exacerbate onset and relapse. As he said, “One of the great features of the brain is that it responds to its environment.”

Dr. Paykel, a London-trained psychiatrist, had a healthy skepticism about psychotherapy, an excellent knowledge of research design, and an open mind. Working together, we did not set out to define a new psychotherapy but to define what we thought were the important components of good clinical supportive practice for ambulatory patients with depression.

We were given many books and articles to read, but the ones that guided us the most were the works of Harry Stack Sullivan (1), with his focus on the current interpersonal context of a psychiatric illness. Sullivan stated that interpersonal behaviors of others form the most significant events that trigger emotions. Our progress was also influenced by the ideas of Adolf Meyer (2), who put great emphasis on the patient’s relationship to his or her environment, and by the writings of John Bowlby (3), who stated that individuals make strong affectional bonds and that the separation of these bonds, or the threat of separation, gives rise to emotional distress and depression. We were also influenced by the work on life events—both our own and others’—which showed, consistent with theoretical writings of Sullivan, Meyer, and Bowlby, that events that represented exits from the social field were associated with depression. (Weissman and colleagues [4] provide a description of the theoretical and empirical thinking underlying the development of IPT.)

In preparing the first draft of the psychotherapy (“high contact”) manual, we decided to begin by defining the dose and frequency of the treatment and the diagnostic process that became the first phase of IPT. This phase included a diagnostic evaluation; a psychiatric history; patient education about depression, its symptoms, and treatment alternatives; an interpersonal inventory of important people currently in the patient’s life; the sick role allowing the patient to receive help; a discussion linking symptom onset to interpersonal situations; and the identification of problem areas associated with the onset of the depressive episode. The group’s work in life events allowed us to identify and define the key problem areas: grief and complicated bereavement after a death; role disputes and conflicts with a significant other in renegotiations, dissolution, or impasse; role transitions and changes in life status (e.g., divorce, moving, retirement); and interpersonal deficits, such as lack of social skills, boredom, loneliness, or paucity of attachments.

The basic assumption was the onset and recurrence of a depressive episode were related to the patient’s social and interpersonal relationships at the time. We met weekly to develop the manual, reviewing cases and creating scripts from actual practice to describe how to carry out the treatment procedures and in what sequence. We were concerned that the procedures be specific enough so that therapist training would be consistent.

The Depression Research Unit, where the project was housed, was a cozy, somewhat dilapidated, wooden-frame house converted into offices two blocks from the new Connecticut Health Center and Yale Medical School. Our work, testing psychotherapy and conducting clinical trials of medication, was not highly regarded in academic circles. Even though Dr. Klerman was head of the Mental Health Center, he could not get space for the project on campus. The Yale Drug Dependency Unit headed by Dr. Herb Kleber soon moved into the other side of the building to work on understanding and developing treatment for the emerging drug epidemic in New Haven and many major U.S. cities. Dr. Klerman, Dr. Kleber, and I frequently collaborated. We were much later joined by Dr. Bruce Rounsaville, then a young psychiatrist, to carry out a treatment study we developed of patients undergoing methadone maintenance, which included IPT.

The final piece of work to accomplish before starting the maintenance study was to develop the target psychotherapy outcome measure. Dr. Klerman said medication should help the patient sleep and eat better, but it would probably not help the patient get along better with his or her spouse—that is where he felt psychotherapy would likely have its effect. The next task was to design the social functioning measures. Dr. Paykel and I found that the social functioning measures designed for newly ambulatory patients with schizophrenia as part of the National Institute of Mental Health’s multisite studies of antipsychotic medication included tasks that were too simple, such as dressing and grooming. Other measures were designed for college students and focused on sex and dating. Our patients were primarily middle-aged, married women who had children and were from large extended families. Their important areas of functioning were not captured by any of the existing social functioning measures. Borrowing from several preexisting scales, we developed questions on functioning in relation to marriage, children, family, and extended family. Dr. Klerman then advised us to test the scale’s validity. Could we discriminate cases from controls or acutely ill from recovered patients? The measure we developed was titled the Social Adjustment Scale, which eventually became a self-report (SAS-SR). It was first published in a book on women with depression in 1974, when interest in women’s health, accelerated by the women’s rights movement, was beginning to emerge (5, 6).

The maintenance study results, also first published in 1974, showed that drugs prevented relapse and that psychotherapy improved social functioning (7). The 1-year follow-up on the patients after maintenance treatment became my Ph.D. dissertation in epidemiology at Yale (8).

When the first maintenance study showed the efficacy of “high contact,” we began to more fully describe the treatment and termed it interpersonal psychotherapy. We designed a trial of acute depression treatment that tested drugs and IPT alone and in combination (9). This was my first large National Institute of Mental Health (NIMH) grant. The positive results of the acute study, and particularly the findings that the combination of drugs and psychotherapy was the most efficacious treatment for depression, led the NIMH Treatment of Depression Collaborative Research Program to test drugs, CBT, and IPT for treatment of patients with acute depression (10).

In 1984, after the efficacy of IPT had been demonstrated outside our research group, we published the first IPT manual (11). Dr. Rounsaville and Dr. Eve Chevron, a psychologist, made key contributions to the IPT treatment training program for the NIMH collaborative study and joined us in preparing the manual for publication. This publication was followed by numerous modifications to and studies of IPT, including examinations of IPT for adolescents, the elderly, pregnant and postpartum women, women after miscarriage, medical patients, and as part of maintenance therapy for recurrent depression and bipolar disorder, eating disorders, anxiety disorders, posttraumatic stress disorder, and depression in developing countries (4).

The evidence for the efficacy of IPT is strongest for depression, is strong for some adaptations, is tentative for others, is negative for the treatment of drug abuse, and remains untested for some new adaptations (12). The IPT manual has been translated into 10 languages. Adaptations have included group, conjoint, and telephone IPT. Dr. Klerman could not have anticipated the great interest in IPT that has ensued nor the recommendation of IPT in several official guidelines in the United States and elsewhere as the first-line treatment for depression. In February 2019, IPT and CBT were recommended by the U.S. Preventive Services Task Force for treatment of depression during pregnancy (13, 14). Dr. Klerman died on April 3, 1992, before these developments occurred. Dr. John Markowitz, one of Dr. Klerman’s last trainees, has joined me in updating the manual over the years. An updated version of the manual was last published in 2018 (4). We have always kept Dr. Klerman as an author because his seminal contributions continue to form the core of IPT, unchanged since its inception. Many other specialized manuals of IPT have appeared using the fundamental principles our team established with special adaptations and case material for specific age groups or disorders. Descriptions of the major adaptations can be found on the Web site of the International Society of Interpersonal Psychotherapy (isIPT) (www.interpersonalpsychotherapy.org).