A prohibition against psychiatrists discussing the mental health of public figures — a rule that has become especially controversial, and sometimes flouted, since the inauguration of President Trump — is “premised on dubious scientific assumptions,” researchers concluded in an analysis scheduled for publication in a psychology journal.

The American Psychiatric Association (APA) defends its “Goldwater rule” by arguing that an in-person psychiatric examination is the gold standard for diagnosing mental illness and psychological traits — given that there are no blood tests or brain scans for psychiatric disorders. In fact, however, numerous studies suggest that the interview-based exam can be misleading, psychologist Scott Lilienfeld of Emory University and colleagues argue in the paper, which will appear in an upcoming issue of Perspectives on Psychological Science.

Patients lie or hide facts, they often have poor self-insight, and psychiatrists err, the authors write. In contrast, the accounts of people who know the individual, plus his or her public behavior, writing, speech — and, yes, tweets — can provide more accurate insights into a public figure’s mind, they contend.


The flaws of psychiatric exams and the usefulness of other data make the Goldwater rule “scientifically indefensible,” said Dr. Leonard Glass, a psychiatrist at McLean Hospital and Harvard Medical School who was not involved in the new analysis and is a critic of the rule. “I sort of believed” that interviews offer the clearest window into someone’s mind, he said. “But I believe that a whole lot less now that I’ve read this paper,” which he called “thoughtful, thorough, and a major contribution.”

But a past president of the APA, Dr. Paul Appelbaum of Columbia University, called the paper “unpersuasive.” Despite the very real shortcomings of in-person psychiatric evaluations, he said, “without them you are very limited in what you can say about someone,” particularly the person’s innermost thoughts and feelings.

The Goldwater rule has its roots in a 1964 magazine survey asking psychiatrists their opinions of the mental health of Sen. Barry Goldwater, that year’s Republican presidential nominee. The majority of those responding said Goldwater was psychologically unfit to be president (they called him “emotionally unstable,” “grossly psychotic,” “paranoid,” “delusional,” and “a dangerous lunatic”). In 1968, Goldwater won a libel suit against the magazine, and in 1973 the APA adopted the rule barring its members from sharing their views of the mental health of public figures.

Trump’s behavior as president, including claims that President Obama had him wiretapped and that his inauguration crowd was the largest ever, has triggered a flood of commentary by, for the most part, psychologists, speculating that his narcissism may be extreme enough to qualify as a personality disorder or that he has other dangerous mental traits, including impulsivity. Psychiatrists have been mostly, though not entirely, absent from the discussion, owing largely to the Goldwater rule.

During the 2016 campaign, the then-president of the APA, Dr. Maria Oquendo, justified the rule by arguing that making diagnoses-at-a-distance “would be irresponsible.” (Oquendo said she is traveling and unable to comment on the upcoming paper.)

The American Psychological Association has no Goldwater rule per se, but its code of ethics says that “opinions of psychological characteristics” must be based on a psychological exam.

The argument in the new paper has two parts: that a large body of scientific research shows such exams can mislead the examiner, and that inferences about a public figure’s psychological makeup and mental health based on his or her public behavior and the observations of close associates can be valid. Together, the data undermine the claim that psychiatric exams provide the highest-quality information on someone’s mental state.

The flaws of in-person exams are no surprise to experts. “People can snow you,” said Dr. Sally Satel, a psychiatrist and resident scholar at the conservative-leaning American Enterprise Institute. “The in-person interview can be highly misleading, especially with people who are superficially charming”: They’re good at “impression management,” have no self-insight, lie, or otherwise deceive the psychiatrist, intentionally or not. Satel supports the Goldwater rule, however, as a guard against politicizing psychology and psychiatry.

“Psychiatrists are taught that if a patient says he has three drinks a day, double that,” said Glass, the McLean psychiatrist. “The psychiatric interview is flawed.”

His and other psychiatrists’ clinical experience is backed up by studies going back to the 1970s in which patients were examined by two psychiatrists back-to-back. The rate of agreement was so low (often around 20 percent) that “the unreliability of psychiatric diagnosis has been and still is a major problem in psychiatry,” researchers concluded. Later studies reported similar discrepancies, especially for personality disorders, finding that examination-based conclusions by a psychiatrist often disagree with assessments in a structured research setting.

Can a public figure’s public behavior and speech, which experts who ignore the Goldwater rule rely on, provide a better window into the mind? “Personality traits are pretty stable over time,” Lilienfeld said. Moreover, markers of, say, pathological narcissism — grandiose self-importance, entitlement, exploitativeness, and lack of empathy — can be seen in someone’s public behavior, especially in the case of a public figure whose behavior and speech has been documented in hundreds of hours of TV interviews, speeches, offhand remarks, and tweets, plus the accounts of people who have lived or worked closely with him. This evidence can also be more accurate than information from formal examinations, the new paper argues.

That is especially true for antisocial and narcissistic personality disorders, the authors argue. One study found that accounts of a person’s behavior from those close to the individual were more accurate than the person’s own (the basis of psychiatric exams); another found that reports from “informants” did a better job at capturing an individual’s personality traits and predicting future behavior than self-reports did.

Columbia’s Appelbaum, who supports the Goldwater rule, agrees that “information from the outside can be extremely helpful in getting a more complete [psychological] picture of a person.” But direct evaluation, he argues, allows a psychiatrist or psychologist “to do things that observations of public behavior and third-party accounts do not,” such as getting a glimpse “of what’s going on inside someone’s head.”

For the Goldwater rule, the question is whether those glimpses are so crucial that they override other arguments against it, including that it robs the public of expert opinion on important questions.

“Is it better to have an in-person interview? Probably,” Lilienfeld said. “Is it necessary to get deep, accurate insights into someone? Probably not. Justifying the Goldwater rule on that basis just does not have scientific support.”

Leave a Comment

Please enter your name.
Please enter a comment.

  • From the Stalin regime in the Soviet Union to the Mao era in China to the Pol Pot tyranny in Cambodia, mental health professionals have been used to control political opponents. It would be wise to be careful…very careful…about giving a large amount of credence to those who are self-appointed to answer the call to the duty to warn when in fact they simply politically despise the one whom they claim is mentally challenged. It has the appearance that that is exactly what is going on here, and if so, it appears that they are the ones who are unhinged.

  • When the Goldwater episode happened 50 years ago, psychiatric nosology in the US was fraught with unreliable, psychoanalytically inspired, inference. Back then, the Goldwater rule was a good idea. Today, not so much. In the UK tradition, information from knowledgeable informants always was factored in to diagnostic assessments, and sometimes the data from those sources was decisive. There is enough information in the public domain about Trump’s cognitive difficulties, his personality, his projective style, and his insecurity to allow opinions that a full psychiatric evaluation is warranted. Provisional diagnoses include early dementia, narcissistic personality disorder with pathological lying, and possibly an attentional disorder.

    • Had it been me vice Trump our problem would be how to divide what would be left of N Korea after a Pre-emptive (Nuclear) strike post ICBM test. I am amazed at how intellectuals pander to an Animal with Nuclear Weapons? Time for change in America.

  • Diagnosing public figures may be hazardous for the profession, I’m afraid. And most people, which I’m sure includes psychiatrists, have political bias which may be hard to exclude. Having said that, it would be nice to have a relatively reliable idea of a candidates mental issues before voting. But I can’t see a reliable, unbiased way to do that. Of course, anyone with an open mind can tell that certain politicians have some kind of mental problem, even if they can’t quote the DSM 5. Disclaimer: In this field, I’m just a layman.

  • If a personal interview is not the best way to diagnose a patient, then psychiatrists should avoid them, wherever possible! If Facebook and Twitter entries can do a better job in much less time (and, therefore, expense), personal interviews can be reserved for those unfortunates (like myself) who have no Facebook or Twitter presence. I shudder to think how psychiatrists would diagnose my online comments elsewhere. But I am not the President of the United States, so I have nothing to worry about. Pres. trump is the obvious intended target of this implied change in the “Goldwater Rule, and probably the only one.

  • Neither does Lilienfeld have any basis of scientific evidence to support his point. Neither sides do. And thus goes listening to him.

  • I would concur about difficulties with in-person evaluation. But that same problem is compounded by assessment at a distance or by press release. Consider Harvey Weinstein, could you argue that a precise or accurate assessment of his personality could be made upon his public persona before his recent outing? The value of an expert opinion is degraded by acting outside of professional bounds and standards. I have no problem with physicians expressing views, as long as it is clear that they are personal rather than professional.

  • This is clearly an important study, which goes to validation of personality diagnoses by behavioral observation. Caution in public statements is always important, but is offset by a duty to warn.

Sign up for our Daily Recap newsletter

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy