Mental Health

Deconstructing and Reconstructing the “Goldwater Rule”

Original article published, October 07, 2016 by Ronald W. Pies, MD in The Psychiatric Times:

Thank you, Ron, for your thorough reply that I largely agree with.

First though, let’s be clear, your article was on psychiatric diagnosis in INDIVIDUALS not on the validity of the existence of psychiatric diagnoses in the population. You stated, “On occasion psychiatrists are asked for an opinion about an individual who is in the light of public attention or who has disclosed information about himself/herself through public media.”

You are also falling into a logic trap by noting that SOME medical or neurological problems have no specific marker, i.e., migraine. This is because there are also many like stroke or brain tumor that DO have an objectively measurable parameter such as seeing infarction, hemorrhage, or mass on a brain scan. Unfortunately there is NO psychiatric condition that has an objectively measurable parameter in an INDIVIDUAL in spite of some persons in a POPULATION who fit diagnostic criteria showing various DIFFERENT and subtle brain abnormalities. Brain scans can be said to have near 100% validity in physically proving that the condition specified because scan results match physical findings on direct examination of the brain seen in surgery or autopsy. This is what is meant by “clearly definable and objectively measurable”.

The first link you provide below by Aboraya states:

“The authors encourage clinicians to use as many validity criteria as possible to improve the validity of their diagnosis.”

They do not state or prove anywhere that a psychiatric diagnosis in an INDIVIDUAL is the same as having objective proof as in neurology (I would place Alzheimer’s in neurology not psychiatry for this discussion).

The paper by Levy only discusses that pursuit eye movement findings are commonly found in POPULATIONS of persons who fit criteria for schizophrenia (and in the CLINICALLY UNAFFECTED relatives of these persons). As they show in Fig 5 and in other places in the article, findings are trends in these groups with a standard error but they nowhere state that these tests can be used clinically as proof of diagnosis in an individual. It’s not the same as seeing a mass in one’s head CT which is clearly a mass: there is no trend or standard error, nearly all INDIVIDUAL cases are either yes or no.

Coincidentally, I was also the lead author on a published study of eye-movements in schizophrenia:

Berger D, Nezu S, Iga T, Hosaka T, Nakamura S: INFORMATION PROCESSING EFFECT ON SACCADIC REACTION TIME IN SCHIZOPHRENIA, Neuropsychiatry, Neuropsychology & Behavioral Neurology, (Journal Name Changed to: Cognitive and

Behavioral Neurology) 3:2, 80-97;1990.

Full paper here:

We found a strong trend for differences between normals and those diagnosed with schizophrenia in the small group we studied, but there was overlap and there were outliers. Thus, it is just interesting research data, it is not useful clinically for individual diagnosis, nor does it prove the validity of schizophrenia.

Now, evaluating public figures who are more likely to have personality issues, ADHD, mild forms of mood disorder etc. and not schizophrenia is even more problematic because there is less robust evidence of a consistent biologic finding in populations of these persons compared to schizophrenia, especially in personality disorders.

So while we would agree on Goldwater, I think you need to take it to the next level of uncertainty when it comes to labeling individuals, examined or not. Psychiatric diagnoses are helpful in clinical treatment, but are unproven constructs and never more than concepts that guide us in what to do, and for mental health workers and institutions to bill medical insurance companies.

For more information about Douglas Berger Psychiatrist Tokyo visit the following websites: