Is Psychodynamic Therapy Unscientific?

The following information is a summary of Jonathan Shedler's "The Efficacy of Psychodynamic Psychotherapy": https://www.apa.org/pubs/journals/releases/amp-65-2-98.pdf

Shedler writes that modern approaches to psychodynamic therapy omit many of Freud's "more outlandish and inaccessible speculations".

He opens with a discussion of the unique aspects of psychodynamic psychotherapy:

1. A focus on emotion and its expression in words, particularly conflicting, troubling, or threatening feelings, which the patient may not initially recognize. Intellectual insight is not the same as emotional insight, which resonates on a deeper level and is more likely to lead to durable behavior change.

2. An exploration of attempts to avoid distressing thoughts or feelings: People resist knowing things which are hard to know. Whether it be missing sessions, arriving late, changing the subject, or talking about matters in an overly detailed and intellectual (but devoid of emotion) way. Associated jargon are "resistance" and "defense mechanisms".

3. Identifying recurring themes and patterns: Sometimes patients are caught in distressing and self-defeating patterns, know this, but cannot escape them...such as the person who repeatedly finds herself drawn to emotionally unavailable partners, experiencing high chemistry at the beginning of the relationship, only to realize her partner does not meet her needs, yet again. Sometimes, people are unaware of their patterns until a therapist can point them out. The classical term for this concept is "the repetition compulsion" - that we may repeatedly encounter situations which are not entirely good for us because they are familiar, or perhaps because we are trying to master them in some way.

4. Discussion of past experiences and other factors which may have influenced development: Psychodynamic therapists tend not to explore the past simply for the sake of doing so, but rather to create linkages between how formative experiences may be affecting current functioning. While talking about the past does not change or erase it, it can change how it affects us.

5. A focus on relationships: Psychodynamic therapists assume that our personality and other aspects of our mental life form in the context of our attachment experiences. Psychological difficulties may occur in an interpersonal context - arising when our patterns interfere with our ability to meet emotional needs. The classical jargon for such concepts are "object relations" and "attachment".

6. A focus on the therapy relationship: The relationship with one's therapist can be deeply meaningful and emotionally-charged. Our attachment styles and other templates for seeing ourselves/others/and the world sometimes direct toward the therapist. A person who fears disapproval or abandonment may expect such treatment from the therapist. The repetition of interpersonal themes in the therapeutic relationship provides an opportunity to name and work through them, as they occur in the here-and-now. The therapeutic relationship can be a 'laboratory' for such change, likely translating to benefits in other and future relationships. The classical terms for these ideas are "transference" and "countertransference".

7. An exploration of fantasy life. Psychodynamic therapy does not actively structure sessions or follow a pre-existing curriculum. Instead, we encourage patients to speak freely about whatever may be on their minds. When people are able to speak freely, they cover a variety of relevant areas of mental life - such as wishes, fears, fantasies, and dreams - all of which provide rich information about the person's view of themselves and others.

The goals of psychodynamic therapy go beyond symptom reduction. Successful treatment not only eliminates troubling symptoms. It also results in some degree of personality change, and the blossoming of internal resources and capacities.

Shedler then summarizes a variety of meta-analyses which studied the effectiveness of therapy, in general, psychodynamic therapy, and medication.

The statistical metric "effect size" measures the difference between a group receiving a particular treatment vs. a group receiving no treatment. An effect size of 0.8 is considered large in psychological and medical research. 0.5 is considered moderate, and 0.2 is considered small.

For context, a pill with an effect size of 1 would make the average American male 5 inches taller, the average female 77 pounds lighter, or IQ 15 points higher, on average. An SAT study program with an effect size of 1 would increase the average score by 100 points.

The first major meta-analysis of the effectiveness of general psychotherapy (Smith, Glass, and Miller, 1980) yielded an effect size of 0.85. Lipsey and Wilson (1993) yielded 0.75 for general psychotherapy and 0.62 for CBT and behavior modification. Robinson, Berman, and Niemeyer (1990) indicated that psychotherapy for depression had an effect size of 0.73. These represent large effect sizes.

An analysis of the FDA databases reported in the New England Journal of Medicine suggest an overall effect size, for newer antidepressant medications, of 0.31 between 1987 and 2004. Moncrieff, Wessely, and Hardy (2004) reported an effect size of 0.17 for tricyclic antidepressants. Shedler mentioned in a presentation I attended in 2023 that the effect sizes for antidepressants likely reflect that the medications are very helpful to some, and unhelpful to others, averaging a lower overall effect.

How does psychodynamic psychotherapy compare?

A large and rigorous meta-analysis of psychodynamic therapy, published by the Cochrane Library, included 23 randomized controlled trials of 1,431 patients (Abbass, Hancock, Henderson, & Kisely, 2006). The studies compared patients with a range of common mental health disorders, who received short-term psychodynamic therapy with controls, and yielded an overall effect size of 0.97 for general symptom improvement. The effect size increased to 1.51 when the patients were assessed 9 months after treatment had ended. In addition to change in general symptoms, the meta-analysis reported an effect size of 0.81 for change in somatic symptoms, which increased to 2.21 at long-term follow-up; an effect size of 1.08 for change in anxiety ratings, which increased to 1.35 at follow-up; and an effect size of 0.59 for change in depressive symptoms, which increased to 0.98 at follow-up. The consistent trend toward larger effect sizes at follow-up suggests that the benefits of psychodynamic therapy continue to accrue after graduation.

These and additional findings are summarized in the following table, from Shedler's article:

In contrast, Shedler argues, the benefits of other empirically supported treatments, like CBT, may decay over time, perhaps because of their emphasis on symptom reduction.

He further argues that psychodynamic therapy may be especially useful for personality disorders, perhaps because these disorders are relational in nature, relate to attachment issues, and may involve difficulties in understanding one's own mind.

Shedler offers the following limitations: There are many more studies of CBT than psychodynamic therapy, which he feels reflects our current Zeitgeist that only cognitive and behavioral treatments are scientific - and that short-term, manualized approaches to therapy are best.

In conclusion, Shedler posits that psychodynamic therapy is 'pretty darn effective' and at least as effective as CBT. And much more effective than antidepressant medication alone, on average.