Is “Mental Illness” A Disease?

Well, I do admit, people seeking help for emotional disorders are certainly ill-at-ease, dis-eased. But that is not what we mean.

We are talking about the Official Disease Model, the DSM 5. In Scientific American, Peter Kinderman makes a case for abandoning the whole DSM 5 model.

You will want to read his argument for yourself:

I offer you a few concepts of my viewpoint.


Traditionally, diagnosis and treatment are linked. The proper diagnosis tells the physician what treatment to pursue.

Is that true in the psychotherapy profession?

Of course not. Regardless of the diagnosis, all treatment is essentially the same. We engage in supportive, heart-to-heart talks with a troubled person. That is, while the “evidence based treatment” argument says that we should follow different treatments for different conditions, in reality, the evidence that such targeting makes any difference if quite weak. There may be some exceptions, but frankly, most therapists are helping people struggle with dilemmas and emotional pain, not so much with particular diagnoses.

You see, again with a few exceptions, treatment programs that compare different approaches, such as cognitive therapy versus psychoanalytic therapy, the outcomes are equal.

That is called “The Dodo Bird verdict.” It comes from Lewis Carroll, in Alice in Wonderland.” The animals get wet and run in different directions for different amounts of time, so as to dry off. Then they decide they have run a race, and want to know who won.

The Dodo Bird replies, “Everyone has won, and all must have prizes.”

Saul Rosenzweig first raised this idea in 1938. Since that time, most of the evidence supports the Dodo Bird verdict.

CBT practitioners claim that for anxiety, it is the preferred treatment, but again, a recent long-term analysis showed that, compared with insight-oriented therapy, the outcomes were the same. Admittedly, the short-term advantage goes to CBT, but we aren’t so interested in that as we are in long-term prognosis. (Also, see this writeup.)

Now, you ask, “Where is Johnson going with all the blah-blah-blah?”

Simple: If there is little or no difference between different treatment strategies, if they tend to all get roughly the same outcome, then what good is diagnosis?

Now for a technical point: Kinderman points out that reliability of the DSM 5 is very poor, almost non-existent. That is very important.

Reliability is the measure to track the same factor accurately. If five diagnosticians all arrive at the same diagnosis for the same patient, then diagnosis is reliable

It may or may not be valid. Validity is whether the measure actually tracks what it is supposed to track. Now for technical reasons, the validity cannot be greater than the square of the reliability. Perhaps the reliability is .80, which is a good, high number. The validity cannot exceed .64, or in other words, the measure cannot explain more than 64% of the variability. There would be 36% of the variability that is unexplained.

If the reliability of the DSM is low, then there is virtually no validity. There simply cannot be any validity. (A good deal of why the DSM 5 is unreliable is that there are too many diagnoses, and many of them are new.)

Some diagnoses like obsessive compulsive disorder have a relatively clear and direct path between diagnosis and treatment. That is also the argument about anxiety generally. There I am not as sure. Yes, exposure and relabeling does help. But does that justify all of the separate anxiety diagnoses? 

Schizophrenia, like Bipolar, seems to be something we can reliably diagnose. There is a recent study showing that CBT without medication is a helpful treatment for schizophrenia.

Similarly, depression can be diagnosed as a specific condition that needs such general approaches as raising activity, raising mastery and pleasure experiences, and developing a hopeful image of the future.

Embracing diagnosis is necessary for the people who believe there are specific treatment, but embracing the Dodo Bird verdict is useful for clinicians who believe that the general factors are mostly operant in psychotherapy. A trusting relationship, emotional support and release, finding new ways of seeing the problem are examples of the general factors.

Which do you think is right? These are value driven positions, and we know that humans will always cherry pick evidence to support their point of view.

I prefer to believe in the Dodo Bird, considering him as a wonderful source of wisdom. Others prefer to believe in specific treatments, so they must throw out the Dodo Bird’s evidence.

Those people are in a crisis currently. The DSM 5 has created that crisis. No reliability equals no validity. So there cannot be any specific treatments without reliable diagnoses.

Why are we engaging in applying diagnosis with no validity? We may as well say that our patients are possessed of demons.

Your opinions? Has DMS 5 gone too far? Are the diagnostic categories meaningful? Discuss! Please leave your opinions below. I will read them and post a follow-up.

By |2016-11-26T15:53:46+00:00December 3rd, 2014|Psychotherapy|14 Comments

About the Author:


  1. Tina December 4, 2014 at 4:31 am - Reply

    The overall categories in any DSM may be helpful in a shorthand sort of communication between providers, but inconsistent use of labels is often counterproductive. I have used a strength based model instead to inform my choice of interventions and my communication with clients and families. This serves to personalize rather than objective anyone I work with. That is how I would want to be treated as well.

  2. Kevin Geraghty December 4, 2014 at 4:47 am - Reply

    We created this problem for ourselves when we asked insurance companies to pay for counseling and psychotherapy. They legitimately ask for proof of an illness and a treatmen

  3. Kevin Geraghty December 4, 2014 at 4:58 am - Reply

    We created this problem for ourselves when we asked insurance companies to pay for counseling and psychotherapy. They legitimately ask for proof of an illness and a treatment plan. We have spent decades inventing the DSM in order to get paid, not because it promotes health. We can’t have it both ways.

  4. Carol Daker LCSW December 4, 2014 at 6:48 am - Reply

    I only use the DSM diagnoses so that my clients can access insurance reimbursement .I always try to use the most benign diagnosis in order to avoid over labeling someone with a condition that might lead anyone to see them as somehow internally impaired. I design my treatment plan for each client based on their presenting issues with a CBT base.

  5. Stephanie Robertson December 4, 2014 at 9:03 am - Reply

    While I appreciate the Dodo Bird verdict, and I’m familiar with literature indicating that only about 5-15% of psychotherapy outcomes are attributed to specific techniques, you seem to undermine your own argument against the DSM-5. Instead of painting the diagnostic manual with a broad brush of “no reliability and no validity,” (then undermining your own argument by identifying diagnoses with validity and reliability like schizophrenia and OCD), it might be more helpful to discuss the DSM-5 in historical context. Of course, as the profession adjusts to new diagnostic criteria and new diagnoses, the reliability and validity of diagnosis is going to fluctuate, but that doesn’t mean that we should abandon it altogether.

    In my view, the DSM-5 offers a framework by which we are able to gain some insight into individual behavior, and to guide our treatment planning. I know that exposure and response prevention is the most helpful treatment for OCD, and I understand OCD because of the insight provided by the DSM. A “heart-to-heart” is unlikely to help in this situation. By the same token, when I work with a child with Autism Spectrum Disorder, I know that behavioral and social skills training interventions are critical because the DSM allows me to understand Autism. Again, the therapeutic relationship is not enough. While there are some disorders for which treatment is similar (anxiety spectrum disorders or depressive disorders), I still utilize information gleaned from the DSM in treatment. Even with very similar diagnoses, I use the information to guide treatment. For example, panic disorder interventions are different from social anxiety interventions, even though they’re both anxiety disorders.

    The DSM-5 isn’t perfect, but in the words of John Norcross, “Science is not a set of answers. Science is a series of processes and steps by which we arrive closer and closer to elusive answers.” Blind adherence to the DSM-5 for manualized/prescriptive treatment of specific disorders wouldn’t be helpful because so many of our struggles are complicated, interconnected constellations of symptoms nestled within the context of our social world. However, the DSM-5 allows us to develop hypotheses and to work with our clients to move closer the “elusive answers” of becoming “well.”

  6. John Gallagher December 4, 2014 at 9:37 am - Reply


    It is a remarkable thing how often you present very interesting and meaningful information. Thank you for your valuable service. I’ve been a fan of yours since I attended a workshop of yours twenty years ago or so and it is always a pleasure when I have an opportunity to reconnect.

    Your article, “Mental illness. A Disease?” strikes close to what have been for me a fundamental guiding principle. That is, with few exceptions what I routinely treat in outpatient psychotherapy I conceptualize as unhappiness rather than as any of the 340+ various diagnostic conditions in the DSM-IV.. I have not yet tallied up the DSM-V inventory. I have presented a program several times which I call “The Etiology, Diagnosis and Treatment of Unhappiness” in which I make that case for the centrality of unhappiness and lay out what I regard as meaningful answers to five important questions:

    1) What does it take to be happy? Answer: A minimum dose of three distinct emotional needs as follows.

    A – Meaningful Work (for Self-respect),
    B – Affectionate Relationships (to internalize confidence that one is lovable) and
    C – Fun/Recreation/Pleasure/Enjoyment/Playfulness (for internalizing confidence that the world is a lovely place).

    2) What are the most common varieties of unhappiness? Answer: Bad moods which cluster predominantly in three different domains:

    A – Disappointment (disappointment, depression, sadness, the “blues”, resignation, grief, boredom etc.),
    B – Anxiety (anxiety, fear, worry, feelings of inadequacy, jealousy, self-consciousness, insecurity, apprehension etc.), and
    C – Anger (anger, rage, contempt, defensiveness, judgmental or critical attitudes, irritation, annoyance, impatience, etc.).

    3) What are the relative costs of these three ways of experiencing unhappiness? Answer: Generally speaking Anxiety is worse than Disappointment and Anger is worse than Anxiety. Making that case takes some time but I end up arguing that anger is always a relatively pathological reaction or at least a non-optimum reaction. Whenever someone can respond to a situation and do what needs to be done without getting angry, that person is more competent in that situation than someone who does get angry. This is very evident in how parents discipline children.

    4) What is the optimum psychological adaptation for getting through life with as little pain and suffering as possible? Answer: Strive to develop the capacity to react to disappointments graciously. If one does that then there is no need to live in anxiety and fear about potential disappointments and one is not frequently driven to anger over either a disappointment or a worry.

    5) What principles can one live by to cultivate this optimum psychological adaptation?

    A – Never express anger.
    B – If one becomes anger one should usually forego any attempts at conflict resolution and focus on getting over being angry. When a person gets angry that person usually looses about 20 IQ points. Most of us can’t afford to loose 20 IQ points especially if there is a significant conflict to be addressed.
    C – After having gotten over being angry, determine whether it was an anxiety or a disappointment that led to the anger. It will always be one or the other.
    D – Consider talking about the disappointment or the anxiety if the anger occurs in the context of an ongoing relationship but make sure you are no longer contaminated with anger and are congruently grounded in the disappointment or anger before you try to discuss it.
    E – Don’t blame others (i.e., go to anger) in response to your anxieties or insecurities.
    F – Don’t worry about things over which you have no control (reference Reinhold Niebuhr’s Serenity Prayer).
    G – Cultivate the cardinal virtues of gratitude and humility. With a reasonable degree of humility one is much less likely to become embroiled in pointless power struggles or patterns of judgmental contempt. With an appropriate degree of gratitude most middle class Americans, as citizens in what is arguably the wealthiest nation on the face of the earth, living at the dawn of the twenty-first century are living under better circumstances than have ever previously existed in the entire history of homo sapiens on this planet.

    I have not written extensively on any of these topic, Lynn, but if you go to my website

    and click on the links for Patient Information and then on Patient Handouts

    you’ll find links to a couple of documents that are relevant to my thinking on these matters. The Anger Essay is where I outline my argument that anger is a pathology. I often get push-back that “Righteous anger” is a good thing. I remain convinced that to act with unbounded compassion is better.

    The second piece is entitled “Principles of Psychological Well-being” in which I talk about the fundamental emotional needs in more detail than I have here.

    Thanks again for all you do. I admire the impact of your contributions to the profession and the sphere of influence you have achieved.

    John Gallagher

  7. Zak Zaklad December 4, 2014 at 10:03 am - Reply

    re the dodo bird theory – in the 1950’s thru the 1979’s Carl Rogers published research that supported – somewhat – the dodo bird. he found that therapy does make a difference and outcomes are different, but what matters is the quality of the therapeutic alliance, not the psychotherapeutic technique. and he created his well-known structure for how the therapist can build a strong alliance. i met Rogers briefly in 1976, and was amazed at his power and his humility. i said to myself then, “i want to be like him”. i’m still working on it.

  8. Lee Ann December 4, 2014 at 11:52 am - Reply

    The diagnostic labels are beneficial only for billing insurance purposes. I threw away the last DSM in 1999 because I rid my Self of viewing people with a label on their forehead. I learned to relate to the soul of the other from my own soul. This is what people are starving for. They want/need to be accepted as they are, where they are by someone and they need to be able to accept themselves as they are/where they are. I respect and use the methods of Dr. Lynn Johnson and Eckhart Tolle in maintaining my presence in the moment, embracing NOW and doing what, if anything makes sense. Talking from a deep feeling soulful place to a person who is able to listen from a deep feeling soulful place is healing. Conversely chronically complaining from an egotistical narcissistic place to a person who is in an equally egotistical/analytically biased place creates suffering and does not heal anything. I love my Self therefore, I do not label my Self or others. I love life and I understand people need to become conscious of who they really are: consciousness.

  9. Linda Redding December 4, 2014 at 1:01 pm - Reply

    I fully agree that the diagnoses are not reliable. However, health insurance only pays for the treatment of illnesses; once you remove a medical model diagnosis, you have lost a major source of funding for psychotherapy–> back to “the good old days” when few people could afford therapy.

  10. Derek V. Roemer December 4, 2014 at 1:10 pm - Reply

    I’m guessing that 5 or 6 general diagnostic categories would be helpful in planning therapeutic strategy, and talking about our patients to each other, with enough reliability to make sense and have some validity. Beyond that diagnosis serves the needs of the 3rd party payers and bean counters, as far as I can see.

  11. Jean brugger December 4, 2014 at 6:25 pm - Reply

    Recently retired I worked first with people with severe chronic physical illness and later with people with severe and persistent mental illnesses. They were mostly low income due to the illnesses, but some had been rich and powerful.once. Some had folk wisdom, some book learning and a few were genuses,much more intelligent than myself. My pattern with both groups was to get ongoing medical treatment ASAP and to establish a relationship. As most of my employment fell under case management there was the advantage of social exchange, ie service linkage and seeing the person in their living setting. This relates to DSM5 as those classifications are for billing and research money, even for people with one dx. However, most people I had were on at least three arises. In these cases a blending of approaches appeared to work best. Children were not my speciality, but from what I know the DSM5 has the aim of reducing expanding the “normal” range and reducing over dx.

  12. Nick Gallo December 5, 2014 at 2:27 pm - Reply

    Dr. J.

    Enjoyed your blog.

    While # 5 was being developed much was written about the problems with the publication. Some of the most powerful statements were written by specialists who were on some of the committees . The criticisms ranged from “no scientific basis for many of the diagnostic catagories” to “it is just a money maker for the American Psychiatric Association. Few would listen or take up the challange to not use # 5.

    If folks are interested in undertanding the basis of the writings, I would suggest they read “Lies, Damned Lies and Medical Sciense” by David Freedman. He’s been saying these things for years.

    Nick Gallo

  13. Trudi Strasberg, MA, CCC December 7, 2014 at 8:27 am - Reply

    Thank you for bringing Dr. Kinderman’s article to my attention Dr. Johnson. I was quite interested in his viewpoint that focusing less on diagnosis–especially if diagnosis might not be reliable/valid–might be more empathic and empowering to clients.
    Trudi Strasberg, MA, Canadian Certified Counsellor (& CA LPC but currently living/practising out-of-state)
    Psychotherapist, Career Counsellor, Life Coach
    Toronto, Canada
    I attended one of your Positive Psychology seminars in 2011 (thanks!)

  14. Jean Eva Thumm December 13, 2014 at 1:48 pm - Reply

    Dr. J, Thank you for your many interesting articles on your blog. I do agree with you that the RELATIONSHIP between therapist and patient is of prime importance. Certainly listening so that the person feels understood and then reframing the situation s/he is facing are keys in treatment. I also believe that certain approaches such as EMDR, Clinical Hypnosis, Autogenic Training, and Brain Spotting can be helpful after the relationship has progressed to a good stage of trust. I was one of the Collaborating Investigators in the DSM-5 field trials and found the work interesting. However, nothing can substitute for a heartfelt, caring relationship.

Leave A Comment