Psychotherapy over the Last Four Decades

Psychological treatments are the other arm of therapy for depression and many psychiatric disorders. The literature is full of trials indicating that they are in some contexts more effective and better at preventing relapse than pharmaceutical interventions. There are no physical side effects, but some of the best-validated treatments depend on therapists with training. Unlike drug therapies, there are no multibillion-dollar suppliers of these treatments with the motivation to market them and make them easily obtainable . . . And yet such therapies have the potential to make a substantive difference to new categories of patients.

—“Therapy Deficit: Studies to Enhance Psychological Treatments Are Scandalously Under-supported” (2012 editorial, Nature)¹

The dawn of evidence-based psychotherapy began over 40 years ago. Twenty-five years ago when the Harvard Review of Psychiatry was launched, psychotherapy clinical trials were still in adolescence. The quotation from Nature sums up the situation today.

The major catalyst for psychotherapy efficacy research was the psychiatric pharmacologic revolution. Rating scales to test efficacy regarding a range of psychiatric symptoms, along with diagnostic interviews to better define patient populations, were developed to test the new psychotropic drugs. Patient deinstitutionalization and the increase in ambulatory psychiatric care, fueled by the community health center movement, required measures of community functioning that took into account more than mere symptoms. Assessments that measured family, work, and interpersonal relationships were developed and became one target for psychotherapy outcome studies. While medications were being widely tested, psychotherapy was, in practice, the most common treatment for major depression and other psychiatric disorders. The conventional wisdom among both psychopharmacologists and psychotherapists was that one could not test the efficacy of psychotherapy. The effects were believed to be the product of each individual therapist’s own skills. There was disagreement as to the value of drugs or psychotherapy. Some felt that psychotherapy would undo the good effects of medication, and others felt that medication would divert the patient from dealing with the personality problems that produced their symptoms. Eysenck’s view of psychotherapy² was popular. Based on 7000 cases from studies of psychoanalysis and other forms of psychotherapy, he concluded that there was no evidence for their efficacy. In the late 1960s, not one psychotherapy clinical trial that used a randomized design and standardized assessments was published. This is no surprise; it was only in 1962, with the passage of the Kefauver-Harris amendment, that the Food and Drug Administration (FDA) required substantial efficacy data for the approval of new medications.

It was in this context that Aaron Beck and Gerald Klerman—psychiatrists and also friends—independently began writing manuals with scripts in what were later called cognitive-behavioral therapy (CBT) and interpersonal psychotherapy (IPT), respectively. These manuals were used to train therapists and to carry out the first randomized psychotherapy clinical trials. By 1979, the trials had flourished, and 17 clinical trials were identified testing the efficacy of behavioral, cognitive, group, marital, or interpersonal psychotherapy alone, in comparison to, and in combination with, pharmacotherapy in homogenous samples of depressed outpatients.³

In the mid 1980s, a large, multisite clinical trial was launched by the National Institute of Mental Health to test the efficacy of CBT, IPT, imipramine, and placebo for treating depression.⁴ This study included two years’ preliminary work to establish and test both manual-guided training and the procedures for determining fidelity in the clinical trial. The methods developed for that trial were the precursors of the methods used today for training and for ensuring fidelity in psychotherapy trials.⁵ With the growth of psychotherapy testing, London and Klerman⁶ suggested that an FDA for psychotherapy be established to review psychotherapies and determine that the evidence from clinical trials was sufficient to justify clinical use. This did not happen.

What’s the current status of psychotherapy research? Since the first randomized trials of psychotherapies in the 1970s, hundreds of trials have been conducted. Psychotherapies for depression have been examined most extensively, with about 500 randomized trials.⁷ These trials have shown that different types of psychotherapy are effective in the treatment of depression, including CBT, IPT, behavioral activation therapy, problem-solving therapy, nondirective counseling, and psychodynamic therapy.^7,8 A large group of comparative outcome studies have shown that these therapies all seem to be equally effective or about equally effective. Psychotherapies are also about as equally effective as pharmacotherapies, and combined treatment is more effective than either psychotherapy or pharmacotherapy alone.⁷ In the longer term the effects of psychotherapy are stronger than those of pharmacotherapy, with high relapse rates after stopping with antidepressant medication and lower rates for psychotherapies. These therapies have also been found to be effective in specific target groups, including older adults, women with postpartum depression, minority groups, and patients with comorbid general medical disorders. They may be somewhat less effective in chronic depression, comorbid alcohol abuse, and subthreshold depression, although the effects are still significant⁷ (Cuijpers et al., 2016). They can be delivered in various treatment formats—individual, group, guided self-help, or Internet based—with no or only minimal differences between them.⁷ Considerable evidence indicates that psychotherapies can also be delivered effectively in low- and middle-income countries.⁹

Much research has investigated psychological treatments of anxiety disorders, posttraumatic stress disorder, and obsessive-compulsive disorder. Most of those studies have examined CBT, although other types of therapy may also be effective.¹⁰ Several dozen trials have shown that generalized anxiety disorder, panic disorder, social anxiety disorder, obsessive-compulsive disorder, and posttraumatic stress disorder can be treated effectively with CBT, especially compared to wait-list control groups, although the effects are considerably smaller when other types of control groups are used.^11–13 Although psychotherapies for depression and anxiety disorders have been examined most extensively, therapies for several other mental disorders have also been tested in randomized trials. A recent meta-analysis of more than 30 trials showed that psychotherapies for borderline personality disorder, such as dialectic behavior therapy and CBT, have small but significant effects.¹⁴ The same is true for psychotic disorders, where small but significant effects have been found in meta-analyses of several dozen trials examining psychotherapies.^15,16 In one meta-analysis it was found that in psychotic disorders, CBT was significantly more efficacious than other therapies in reducing positive symptoms, whereas social-skills training was significantly more efficacious in reducing negative symptoms. CBT was also found to be significantly more efficacious for positive symptoms when compared directly to supportive counseling.¹⁶ A growing number of studies have also examined the effects of psychotherapies for eating disorders. Although no effects have been found in the small number of trials focusing on anorexia nervosa, significant effects for CBT and IPT have been found for bulimia nervosa and binge-eating disorders, especially on recovery and reduction of symptoms.¹⁷ Psychological treatments of bipolar disorder show that these therapies have small but significant effects on depression, and result in considerably lower rates of relapse.¹⁸

It can be concluded that in a period of 40 years, an extensive body of knowledge has been created on many psychotherapies for all kinds of mental disorders, as well as for physical health problems, in all age groups. This research has shown that psychotherapies are effective in treating many mental disorders, with the consequence that these therapies are included in guidelines as first-line treatments. In recent years it has become clear that the effects of psychotherapies have been overestimated because many negative studies may have not been published, affecting the outcomes of meta-analyses, and because especially older trials do not meet current criteria for randomization, assessment, analyses, and outcome reporting. This situation is similar for studies of pharmacotherapy.^19,20 Despite this overestimation, the effects of psychotherapies for most mental disorders are still considerable and clinically relevant.

There are new developments in psychotherapy. An Institute of Medicine study defined a framework for establishing evidence-based standards.²¹ It was concluded that psychosocial interventions are found effective in research settings but are not used routinely in clinical practice; the evidence base needs to be strengthened. A common terminology of the psychotherapy elements should be identified, and accountability measures be developed that would allow for monitoring and improving quality for purchasers and providers. The efficacy of an element approach would need to be tested.

A meta-analysis of over 90,000 participants found more than a threefold higher patient preference for psychotherapy over pharmacotherapy.²² Despite patient preferences and the availability of a number of evidence-based psychotherapies, all psychotherapy use, regardless of evidence, has declined in the United States, whereas use of psychotropic drugs has increased.²³ Barriers for evidence-based psychotherapy include insurance coverage and limited competency-based instruction in the major programs training providers, psychiatrists, psychologists, and social workers.²⁴

Against this background in the United States, there is a strange paradox. The World Health Organization, through its mental health Gap Action Program (mhGAP), has sought to increase access to evidence-based psychotherapy, especially in low- and middle-income countries, and to persons exposed to adversity from natural and unnatural disasters. Manuals for problem management and for group IPT have been made available free of charge through the organization’s website (at http://apps.who.int/iris/bitstream/10665/206417/1/WHO_MSD_MER_16.2_eng.pdf and http://apps.who.int/iris/bitstream/10665/250219/1/WHO-MSD-MER-16.4-eng.pdf, respectively).

Given the current state of evidence, what can one suggest for the future? In the United States, the funding emphasis is on clinical trials aimed at understanding the biological target of specific psychotherapies. Few such studies are ongoing. Without the stick of accreditation, training in evidence-based psychotherapies among the major trainers of psychotherapists remains scant. Training continues to be unregulated and ad hoc.

Currently, a number of evidence-based psychotherapies are available, which is a good development. Like antidepressant medication, they may be similar in overall efficacy and better than placebo. As with medications, however, some patients do better on one rather than another form of psychotherapy. It is therefore good to have different choices even if, overall, they may appear the same. As with medication, we do not really know which ones do best with which patients, how to deal with treatment-resistant patients, or how to prevent relapse, which is high even after successful treatment. Promising new areas of research are focused on methods to examine which patients with which biological or clinical characteristics respond better to one treatment than to another, and to examine whether each treatment affects different clusters of symptoms.²⁵

Computerized psychotherapies, whether clinician guided or not, are in their early stages and need testing in the United States. Several countries, including Australia, Holland, and Sweden, are ahead in testing and using electronic psychotherapy. The various psychotherapies will need testing to determine how they can be used in a sequence or for triage, and for which persons at which stage of symptoms. Finally, the lesson learned from the World Health Organization that the psychotherapies can be simplified sufficiently to be used by health workers should be heeded, tested, and made available for persons with early symptoms. In the past 40 years, the industry of psychotherapy has grown and flourished. The question of whether there should be an FDA for psychotherapy seems even more important.