Tuesday, January 16, 2007


I'm writing to comment on an opinion piece published recently in the New York Times entitled "Mentally Ill, Behind Bars". I felt it was necessary to post about this because it was another story where people in the lay public would accept it as knee-jerk truth when in fact it seems to be repeating another correctional myth.

Mr. Harcourt, a professor of law and criminology, begins his piece by presenting a very familiar graph. The graph shows the number of people hospitalized in psychiatric facilities versus the number of people incarcerated in jails and prisons. The number of patients drops as the number of inmates rises over the years. Well and good, we've all seen that before. What concerns me is the fact that a full professor at an American university does not appear to appreciate the difference between correlation and causation.

Of course the number of psychiatric inpatients have fallen over the last fifty years. Prior to that we had no real treatments for mental illness. Thorazine wasn't invented until 1954. Tricyclic antidepressants were invented in the 1950's. Lithium was still experimental in the early 1970's. The majority of advances in psychopharmacology have taken place in the last thirty years, coincident with the rise in incarceration. (And I emphasize the term 'coincident'.) In addition to advances in pharmacology we also have more non-pharmacologic interventions. We have programs that act as therapeutic alternatives to hospitalization. We have partial hospitalization programs and assertive community outreach or mobile treatment programs. We have forensic pretrail diversion and alternative sentencing programs for forensic patients.

What Mr. Harcourt fails to mention is that there has been no increase in the relative number of people with mental illnesses in correctional systems. The absolute numbers have increased, but not the proportions. There is an increase in absolute numbers of mentally ill prisoners because the prison population as a whole has increased.

But the following quote is where he really takes a leap. More than one, actually:

But the graph poses a number of troubling questions: Why did we diagnose deviance in such radically different ways over the course of the 20th century? Do we need to be imprisoning at such high rates, or were we right, 50 years ago, to hospitalize instead? Why were so many women hospitalized? Why have they been replaced by young black men? Have both prisons and mental hospitals included large numbers of unnecessarily incarcerated individuals?
Wow, so many conclusions, so little data! There's nothing in the graph whatsoever that would suggest that there has been a change in how psychiatrists make diagnoses. And it's odd to see a psychiatric inpatient referred to as an "unnecessarily incarcerated individual". They aren't incarcerated at all. The demographics are different in hospitals versus prisons because they are different populations.

One could also draw a correlation between the increase in incarceration and the rise in, say, prescriptions for antibiotics or the number of people travelling in commercial airliners. But you'd never hear anyone suggest that we are only locking people up because the airlines are overbooked. But the biggest thing that bothers me about the transinstitutionalization theory is that it's based on the premise that all psychiatric inpatients are potential criminals who would be in jail but for their hospitalization. This just flies in the face of reality. The majority of psychiatric patients don't become involved with the law, even when they are ill. But Harcourt is a law professor not a clinician.

It is important to study the causes of incarceration and institutionalization; nothing in this commentary is intended to dismiss the topic or make light of it. My concern is that the use of conclusory statements, in the absence of data, leaves the article as nothing more than a political statement. We need more than that.