Crooked Timber has a link to a story in the Guardian by a psychologist who repeats an experiment from the 70s in which a psychologist, David Rosenhan, and 8 friends, faked their way into mental hospitals. Rosenhan's volunteers told doctors they were hearing a voice that said "thud" - they added some physical ambience (not showering or shaving for a week), but nothing else - it was not an elaborate deception - and once admitted, said they were no longer hearing the voice... Rosenhan was diagnosed as paranoid schizophrenic, locked up for 52 days - then relesaed, just like that. The writer, Lauren Slater, repeats the experiment, exactly - going to hospitals in a shabby condition, saying she hears a voice that says "thud" - and is, herself, diagnosed - usually with some form of psychotic depression, and invariably proscribed with anti-psychotic and anti-depressive drugs.

It's a fascinating story, both of them. The differences between 1972 and now are revealing - Slater writes that in the 70s, pathology drove diagnosis - now, medication drives diagnosis - she suggests that both were/are fads more than good science. On the other hand, she notes that there are changes - she says the doctors who treated her were all very nice - they treated her with respect and compassion. Rosenham and his associates, in the 70s, reported a far harsher climate - they were institutionalized - abused (or saw others being abused), and generally had it very rough.

So give it a read. It raises questions - about the changes in the psychiatric community (some for the better, some - just changes), and probably about medicine and science over all. The answers to those questions may vary - Rosenham seemed to suggest that, since doctors could not recognize someone who was faking psychotic symptoms (though some of the patients apparently did spot him), all their diagnoses were flawed - this, even to the extent that their diagnostic systems are flawed, driven by fashion, or the desire to say something, to appear to be in control, does not appear fully justified. The fact that someone can fake symptoms doesn't mean that the symptoms aren't real when they're real... though weeding out false positives is crucial. And Slater's article does illustrate one very good rule - even if the science is approximate and pronse to errors, it pays to treat people well - the ultimate scandal is probably not that diagnosis isn't 100% accurate - it's that mental patients were treated like shit in the 70s. Treating them with respect mitigates many of the pernicious effects of misdiagnosis.