Saturday, October 29, 2011

What I Learned Part II

Psychiatry residents on the APA listserv were surveyed about their experiences with social media. 9.7% of residents had gotten friend requests from patients. The remaining residents were asked what they would do if they got such requests. 85% of residents said they would automatically ignore them. 15% said they would discuss the request with the patient, then decline it. Less than 3% of residents received any training about proper social media use in residency. Half admitted googling their patients.

One poster presented the results of a one year study of four major media outlets: the New York Times, the LA Times, USA Today and Fox News. The outlets were culled for articles related to mental illness and violence. The stories were scored according to how many contained one of the follow true facts about mental illness and violence:
1. The majority of mentally ill are never violent.
2. People with mental illnesses are more likely to be victims of violence than perpetrators
3. People with mental illnesses are more likely to be violent against themselves than against others
4. When violent, they are more likely to victimize family than strangers.
Fewer than five percent of the articles contained any of these facts.

The state of New York has successfully used electronic monitoring and regular clinician feedback to reduce the use of polypharmacy in the state prison system.

All state prison systems were surveyed regarding their policies regarding pornography. Of 43 responding states, all banned material that represented a risk to institutional safety and security. There was high variability regarding allowed visual or verbal depictions of sexual acts. There is no published data to support any policy link between prison safety and the presence of pornography.

Crisis intervention training for police is thought to be helpful to reduce incidents of violent outcomes when responding to acutely mentally ill people, but the content of the training varies between program and lacks consistency. There also needs to be more outcome studies to learn if these programs do actually divert mentally ill people out of the criminal justice system.

The PCL-R is coming under pressure as a risk prediction tool. There was a great pro-con debate about this presented by a panel of Broadmoor Hospital psychiatrists. Although interrater reliability is 0.8, there is still significant variance in scores and this could be enough to cause inconsistency when using cutoff scores to determine readiness for release from their severe personality disorder program. It stood out as a risk prediction tool in the 1990's because it was the only one of its kind, but newer tools are being developed with better ROC-AUC scores.

Lastly, multiple personality disorder is coming back. In spite of recent books such as Sybil Exposed, Creating Hysteria and I'm Eve, which document the role the therapist played in the creation or course of the disorder, one presentation today still featured a criminal defendant thought to suffer from multiple personality disorder. This presentation would have been much more effective had there been another expert presenting the potential pitfalls of examining criminal defendants for this disorder. The psychiatrist in this presentation fell into many of those pits: he presented a videotape of the interview in which he asked probing and leading questions (admitting at one point he had to "dig" for symptoms for 20 minutes before the defendant reported any!), and occasionally referring to the defendant as a "patient" rather than as a defendant---clear evidence of bias. The redeeming feature of the presentation was an overview of case law regarding competency and insanity and MPD.

In 2006 there were 4000 civilly committed sex offenders in the US. In this panel they took maps of several large cities in New York, overlaid a map of schools and school bus stations, then overlaid a map of available housing. Finally, the last overlay was a map of legal exclusion zones---boundaries of areas that were legally "off-limits" to convicted sex offenders. This illustrated, at least in Buffalo, New York, that there was literally no place for a sex offender to legally live within city limits. Then they overlaid a map of reported home addresses of sex offenders: 90% were living in restricted areas. There is increasing evidence to suggest that sex offender registration and living restrictions may increase recidivism.

So that's the second conference day. I also enjoyed the computer committee's presentation, which was a two hour geek-fest off all things tech and forensic. It's something you just have to witness to appreciate, sorry. Lots of cloud-based software for document management and report-writing. Not relevant to most Shrink Rap readers but fun for me.