Efficacy of dynamic psychotherapy

unique flowerThe following post is an adaptation of an argument I presented on Sacramento Street Psychiatry, my blog on the Psychology Today website.  As usual, I welcome your comments.

Western medicine’s great strides are largely due to understanding etiology (the biological basis of disease), defining a nosology (a system of categorizing diseases), and testing treatments aimed at these nosological entities, not at individual patients. Take 100 healthy volunteers, swab their throats with Streptococcus, and perhaps 88 will soon develop strep throat. Both our knowledge of bacterial infections (etiology) as well as repeated empirical observation of similar cases leads us to conclude that Streptococcus causes a recognizable condition called strep throat (nosology). Once patients are diagnosed with strep throat — once their conditions become exemplars of this disease category — experiments can be done to show which treatments relieve the condition. Western medicine is the accretion of such knowledge.

Hypotheses about disease categories, and about treatments aimed at these categories, can be tested using randomized controlled trials (RCTs), our most powerful statistical method to assess the effect of independent variables. As in the rest of medicine, evidence supporting the efficacy of psychopharmacology, as well as manualized psychotherapies such as CBT, depends on sorting patients into nosological categories such as “major depression,” applying different treatments to comparison groups, and finding statistically significant group mean differences.  In psychology such a research approach is called nomothetic; the goal is to identify general laws of behavior.

However, another kind of knowledge is important too. Why didn’t the other 12 subjects get strep throat? Is it the same reason for all 12, or is the answer different for each of them? Looking at what makes people unique, as opposed to members of a category, is called idiographic research in psychology. This is the nature of psychodynamic theory and treatment, and why it resists the usual RCT approach to research. Patients who present for such treatment rarely fit neatly into a category such as “depressed.” They vaguely say their lives aren’t working well for them, or that their relationships are unsatisfying in a particular way. They lack meaning and purpose in life.  They get a “funny feeling” when dealing with competition. Their boss triggers authority issues.  They can’t trust their spouse’s fidelity. And on and on.  Such complaints are not exemplars of a nosological category. We may not know what causes schizophrenia or bipolar disorder — we have no etiological understanding of any psychiatric disorder, one reason they are called “disorders” and not “diseases” — but at least these labels reflect a coherent nosology.  Not so with the presenting complaints of most psychotherapy patients.

Psychodynamic therapists and psychoanalysts find little of value in the nomothetic approach. DSM-IV and similar nosology sheds no light on the particular patient in the office, with his unique history, dreams, fears, hopes, etc. The psychoanalytic/dynamic perspective is to understand the uniqueness of that specific patient, and to promote unique helpful changes that may have no relevance to any other patient seen in the practice.

This is not to discount the importance of the nomothetic approach where it applies. If a patient’s condition is exemplary of a nosological category, it should be treated that way.  Doing so allows us to use powerful research tools to separate bias and wishful thinking from real treatment effects.  If a patient presents with major depression, bipolar disorder, or schizophrenia, nomothetic research can and should guide treatment. In such cases, psychodynamic therapy must stand or fall on the same RCT basis as other treatments.  The evidence base for manualized psychotherapies such as CBT, IPT, and a few others is stronger than for dynamic psychotherapy. If someone is seeking relief of major depression, pure and simple, I am happy to refer them to a CBT therapist, and have done so on a number of occasions.  It would be nice to be able to claim strong evidence for the efficacy of prescription antidepressants as well, but unfortunately this is less clear.

CBT and other manualized therapies for specific conditions are much easier to study than dynamic therapy for ill-defined complaints. So it’s really no surprise there are more such studies.  Idiographic research methods, e.g., pre and post measures in single-case designs, have been used to study dynamic psychotherapy, both whether it works and how. But nomothetic researchers consider this “weak science”: There are no control groups — no groups at all, actually.

The bottom line is that dynamic psychotherapy has different goals than CBT or medication.  It doesn’t aim to treat a nosological category such as major depression.  Since it isn’t based on a nomothetic treatment model, RCTs are the wrong assessment tools to use.  Idiographic research methods may be statistically weaker than their nomothetic counterparts, but they are the best that this domain of inquiry allows.  (Seligman argues that naturalistic surveys have their place too.)  Dynamic psychotherapy is based on a rich theoretical foundation that has been scrutinized and refined for the past century. But ultimately it comes down to the individual and the unique mix of discomforting feelings and troubling thoughts that led him or her to reach out for help.

4 comments to Efficacy of dynamic psychotherapy

  • Jeff G.

    Hi Dr. Reidbord,

    Because your argument here is very similar to the one you presented in our exchange on your psychologytoday.com post, I’ll let my counterargument mostly stand and also refer readers to our exchange that you already referenced.

    In addition, I want to add some perhaps newer perspectives on our debate.


    Many people, all unique individuals, come to see doctors and health professionals (mental health or others) with all sorts of symptoms, life situations, stories, hopes, fears, etc. These people may recognize their symptoms as part of a recognized disorder or disease and some may not. It is up to licensed, expert clinicians to do some diagnostic interviewing/assessment to determine if a patient can be diagnosed with a particular disease category. Just like people presenting to a primary care physician due uncomfortable symptoms and pains go to a doctor because they themselves are not experts in diagnosing, people presenting to a mental health specialist for therapy typically are not expert diagnosticians for mental health disorders and are unlikely to present with their own words a case for meeting a diagnostic category.

    An example from a non-mental health situation may be helpful to illustrate. Someone may go see their primary care doctor due to what they think is a worsening of their frequent indigestion, but differential diagnostics through a more thorough exam may reveal that the person may have developed symptoms of a cardiovascular disease. While the patient may have benefited from some treatment for indigestion, not exploring beyond the presenting issue may have led to serious complications with heart disease. Similarly, I think it is important to actively assess for mental health disorders for people who present to therapy, as symptoms that meet criteria for particular disorders may be present and there may be evidence-supported treatments available to better address what a patient may be experiencing.

    Your post seems to imply (though I am not sure) that you don’t do further assessment of people seeking support from you, which I believe is a disservice.

    Assessing a patient/client for a psychiatric disorder that fits into a system of nosology doesn’t discount or ignore people’s individuality, deny their uniqueness, or turn them into a category. It does allow a clinician to note that a person has symptoms common to a particular well studied disorder, a disorder for which there may be effective treatments, and allow for more efficient, targeted, and effective treatments to be planned and carried out. When you treat someone for arthritis, it is recognized that the treatment has to work for a unique patient in their unique life and can be adjusted to do so as things like cost, comorbidities, fear of medications, family history of arthritis, access to continued care, side effects and other issues may be relevant individual factors, but there is no question that 1) there are effective treatments available for treating arthritis and 2)arthritis is the proper target for treatment, not the person.

    I think conceiving of what we call mental health disorders as part of a category and not intractably embedded in individuals is a really respectful way of approaching mental health, as it does not assume that there is anything off about someone or their life or how they want to live their life, just that they have symptoms of a disorder that can be treated, usually regardless of etiology, and it is up to them to determine if there is anything else about their experience that they find problematic and want to address.

    I suppose if someone who meets criteria for major depression who also has issues with authority only wants to address their issues with authority, that person and their clinician can choose to do that. My hunch though is that many people present with many derivative life circumstances directly affected by the symptoms of what is considered a mental health disorder. (ex. Someone with depression who has issues with authority may have those issues with authority caused or compounded by the worthlessness they feel from depression and could benefit from treating depression categorically.) Ignoring or disregarding treatments that could address these other symptoms I think is also a disservice to clients/patients.


    “The evidence base for manualized psychotherapies such as CBT, IPT, and a few others is stronger than for dynamic psychotherapy. If someone is seeking relief of major depression, pure and simple, I am happy to refer them to a CBT therapist, and have done so on a number of occasions.”

    Dr. Reidbord, you make it sound like dynamic therapy for nonpathological symptoms is somehow more important/meaningful/complex than the “pure and simple” relief of symptoms of depression, and that relieving symptoms of a disease category is not usually particularly useful or important to the people who may present to you. For depression, those symptoms may include insomnia, lack of pleasure from previously enjoyed activities/relationships, extreme feelings of worthlessness, weight changes, suicidal ideation/planning and others. All of these symptoms may have a large effect on someone’s life, happiness, health, livelihood, etc., and addressing them is neither a simple task nor the benefit of relieving these symptoms a consolation reward. I can’t speak for your patients, but if they have symptoms that are causing serious distress/impairment, do you not think it is important to address those and as quick as possible?


    I just want to be clear about your argument. I believe you are saying:
    1) if someone meets criteria for a mental health disorder, using evidence-based treatments should be the logical starting place for planning treatment (whether psychotherapeutic, psychopharmacologic, etc.).
    2) However, most people who come to therapy do not meet criteria for a mental health disorder.
    3) Psychodynamic therapy is useful and should be used to support the majority of people who do not meet criteria.
    4) People who do meet criteria for a psychiatric disorder who presents to a dynamic therapist or anyone else who is not trained in an evidence-supported treatment for that disorder should be referred to someone well trained in a therapy modality (psychotherapy, medication, etc.) shown effective for treating that particular disorder.

    I believe we mostly agree with components 1 and 4. However, it is unfortunate that may people who present to mental health professional of all kinds do not receive evidence-based treatments when indicated due to therapists’ disinterest/disbelief in effectiveness of evidence based treatments, lack of training in evidence -based treatments, unwillingness to refer patients, preference for a particular modality regardless of its effectiveness for a particular patient, and I think sadly some therapists use modalities not shown to be effective as a way to keep patients in treatment longer.

    For component 2 of your argument, there must be some trustworthy psychiatric epidemiology data that can answer whether the majority of people who seek outpatient mental health therapy do not meet criteria for a mental health disorder. You would probably know more than me about population level outpatient data and of course know more about the patients you see in your practice. My feeling is that people who are willing to seek outpatient treatment tend to have severe or distressing enough symptoms to often meet criteria, but it may not be the case.

    It is component 3 that I question the most strongly. I suppose I am still confused about what psychodynamic therapy can do for people who would be considered having “good mental health” that other interventions may also be able to do often more cheaply and more quickly. That is, I have no way to assess if other ways of addressing “unsatisfactory relationships,” feeling “lack of meaning and purpose,” problems with “competition,” “authority issues,” or other issues can be as helpful or more helpful than dynamic therapy. I have briefly talked about this before, but these issues, if people truly are presenting with just these issues, don’t seem like they are or should be the sole domain of health professionals or therapists. Other ways of intervening from other approaches, like coaching, assertiveness training, pastoral counseling, etc. may be as or more helpful than psychodynamic therapy perhaps with less risk, cost, and time invested (we just don’t know). I don’t disagree that many people feel that psychodynamic therapy has been helpful for addressing these issues, we still don’t have a way to assess whether psychodynamic therapy may be worth the cost and time vs. other approaches, as quality efficacy studies have not really been done. I also still maintain that they can be done, but until they are, claims about the benefits of psychodynamic therapy I still think should be limited.

    • “I think it is important to actively assess for mental health disorders for people who present to therapy”

      Jeff, I’m happy to continue this discussion with you, but you’ve got to give me more credit than that. You don’t think I actively assess for mental health disorders? Even though I plainly said I did, and gave examples? And where would you get the idea that I expect patients to come pre-diagnosed?

      “I think conceiving of what we call mental health disorders as part of a category and not intractably embedded in individuals is a really respectful way of approaching mental health…”

      Why would you equate categorical diagnosis with not being intractably embedded? Schizophrenia is a pretty intractably embedded diagnosis. Adjustment Disorder With Depressed Mood isn’t. Dynamic issues can be intractable or not. Perhaps your point is that patients can see diagnoses as “other”: It’s not me, it’s my panic disorder (or whatever). But this cuts both ways. “Chemical imbalance” hand-waving may reassure some patients they’re not to blame for feeling sad or anxious. But it also helps others to avoid responsibility for their own behavior, not to mention healing. And it’s a fiction anyway. On balance, I don’t believe it’s any more stigmatizing to see a dynamic therapist for relationship or self-image issues than to see a psychiatrist or other mental health professional for a diagnosed condition. On the contrary.

      “I suppose if someone who meets criteria for major depression who also has issues with authority only wants to address their issues with authority, that person and their clinician can choose to do that. My hunch though is that many people present with many derivative life circumstances directly affected by the symptoms of what is considered a mental health disorder.”

      Your hunch is mistaken my friend, and you’re repeatedly overlooking (or ignoring) my point. I’m a psychiatrist with over 20 years of clinical experience. I know all about medical and psychiatric diagnosis. I diagnose new patients every week in my private practice, and treat them based on my diagnostic assessment. I have no beef with psychiatric diagnosis in general. My point is that most patients who seek dynamic psychotherapy have no diagnosis. They do not have major depression and issues with authority, they just have issues with authority — and some frustration, maybe self-doubt, and a wisp of shame here and there. In case there’s any doubt, I never “ignore” major depression if it is present. It just isn’t, most of the time.

      “Dr. Reidbord, you make it sound like dynamic therapy for nonpathological symptoms is somehow more important/meaningful/complex than the “pure and simple” relief of symptoms of depression, and that relieving symptoms of a disease category is not usually particularly useful or important to the people who may present to you.”

      You’re conflating two issues. I feel dynamic therapy is more meaningful (in a sense) and often more complex than symptomatic relief of depression. But it’s not more important. Relieving a psychiatric disorder such as major depression is extremely useful and important. Of course, there are many extremely useful and important pursuits we choose not to undertake ourselves. I’m glad there are oncologists out there, but also that I’m not one myself. Likewise, I’m glad there are CBT therapists (and psychopharmacologists) who specialize in treating major depression, but that isn’t my chosen area. I prefer to help patients with conflicts, anxieties, self-defeating patterns of behavior, negative self-image, etc. I do find dynamic therapy more intellectually stimulating than medication management (and what I’ve read, but not practiced, regarding CBT), so I suppose I telegraphed my bias by calling depression without dynamic issues “pure and simple.”

      Finally, regarding your “clarifying arguments,” yes, I am endorsing #1, 2, and 4. However, I never said dynamic therapy should “support the majority of people who do not meet [diagnostic] criteria.” Most people who do not have a psychiatric diagnosis need no treatment at all. Some would benefit by emotional or social support, which can (and should) be had without paying a psychotherapist. A subset who suffer from troubling thoughts, feelings, or recurrent maladaptive behaviors can (and do) benefit by dynamic psychotherapy. The reasons why dynamic therapy is indicated in some cases and not others, and why it can help where coaching, assertiveness training, and pastoral counseling can’t, would make for a very long narrative — e.g., a book. In any case, we are back to the beginning, where you seek nomothetic “proof” for the efficacy of an idiographic intervention. Can’t help you there.

  • Jeff G.

    In regard to psychiatric assessment, I was just unclear from your wording. I meant no disrespect and can see how what I wrote can imply that you don’t actively assess. Your original third paragraph seemed vague to me, in that it felt like someone comes to a consult with the issues you mentioned, and that was that, but I know that is not the case.

    I see categorical diagnosis as recognizing a cluster of symptoms that often co-occur and can hopefully be treated as such. I agree with you that in this juncture of science, mental health disorders are just syndromes for which we do not have a good sense of etiology (as now being illustrated with debates over definitions of mental disorders for the DSM5). I do think it can be useful to diagnose symptoms as part of a syndrome, but not diagnose people.

    I suppose I disagree with an inherent goal of psychotherapy that seeks personality change (even for what are considered personality disorders), as I think that there are many ways people can live fully, happily, and successfully, and that the world should relish in all sorts of personalities, beliefs, thoughts, feelings, behaviors and feel great for having them, where autonomy is maximized and coercion/oppression minimized. I think when folks feel like they don’t feel control over thoughts/feelings/behaviors that keep them from feeling like themselves or cause distress, then supporting people in addressing those syndromes or symptoms so that they can live more like themselves is great.

    Looking at your examples of people who present for dynamic therapy, I think that often people seek and clinicians attempt to change people, when perhaps their sense of problem/concern is more related to social/economic/environmental power dynamics and expectations. Perhaps it is “right” for someone to experience “issues with authority” if they face systemic oppression, or have trouble with relationships if they face societal/familial/economic pressure to be in one that looks a certain way. In this regard, recognizing people’s feelings are legit as they are and making some changes to the world around us is more appropriate. I think more succinctly, and this is just opinion, I prefer conceiving individuals as inherently well and address symptoms/concerns that make them feel not so.

    I agree that people do have responsibility for their behavior and their healing. But healing to me implies that there is something in a way “other,” like a pathogen or wound, that is not part of someone that needs to be treated or healed, and healing is the restoration or return to someone as they were before wounding. Healing to me does not necessarily mean changing. As for the hypothetical person with the panic disorder, panic attacks can of course be severely detrimental to someone’s life, and supporting someone in reducing them helps them live their life the way they would without being afraid of unexpected attacks and the terrible feeling experiencing panic attacks come with. To me, “it’s not me, it’s my panic attacks,” seems like it is the opposite of othering, it is incorporating the inevitability of panic attacks and panic disorder as part of who someone is.

    More personally, I have a relative diagnosed with schizoaffective disorder who for some time has been tremendously successful and happy with limited management, I know he would argue that he is not intertwined with his diagnosis or the terrifying symptoms he experienced. They have played a big role in his life, and he is “changed” by his experience in the same way people grow/change with all experiences, but his personality and sense of self are not linked with his history of psychoses.

    I don’t mean to overlook or ignore your points. As you have said, psychodynamic therapy has goals unrelated to relieving symptoms of psychological syndromes, and if people understand and agree with those goals and are ok with going through the experience with without the research support, then I am in no position to mount a complaint. I guess our difference in this regard is more macro, in that I don’t see why an often expensive, variable-length intervention from a health professional is a preferable option among many for dealing with the issues people present with when other ways of addressing these issues are possible (with equally limited evidence). As you have said that discussing indications and contraindications would be book length, we don’t have to get into it here (though if you could recommend a book I’d appreciate it ;-)). Again, if everyone is clear about what they are getting into, then great, dynamic/analytic therapy it is.

    So I guess in finale, we mostly agree with the arguments you presented. Evidence-based treatments should be first line treatments for people who have symptoms of a mental disorder, and should be referred to clinicians skilled in in those treatments/modalities. As I’m not a clinician, if you say most people present without symptoms that don’t meet criteria, then I can’t disagree with you. If you say that many people with many life issues could benefit from other supports (not dynamic therapy) I of course agree with you.

    I still maintain that dynamic therapies can be tested to assess if they are effective for engendering certain outcomes through experimental design (even treating mental health disorders), and I think it would be great if dynamic therapies for mental health disorders can be shown to be effective, but I think you would still say that dynamic therapy is intended for treatment goals that are too individual to be assessed in a such a way.

    I think I’m done with my comments. As always, thank you for engaging and allowing for discussion with your readers.

  • […] practice setting, might respond, OK, I see your point, but some psychiatrists really do provide comprehensive, thoughtful care to their patients.  To which I would say, yes, but they are truly in the minority.  My own career […]

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>




This site uses Akismet to reduce spam. Learn how your comment data is processed.