Monday, January 17, 2011

PREDICTING BEHAVIOR AND PREVENTING TRAGEDY

Most of the readers of this blog know by now that I am a psychiatrist (an M.D. as opposed to a PhD psychologist). I specialize in emergency psychiatry (though I have worked in almost every area of the field, both inpatient and outpatient; and also have expertise in personality disorders and psychopharmacology). In my daily job, I am one of those doctors who is asked to predict behavior that might be a "danger to self" or a "danger to others" and then act to hospitalize the individual, either voluntarily or involuntarily.

The latter case is the more difficult situation, obviously, because if a person does not want to go into the hospital for treatment or, does not think they have a problem (many think I am the one with the problem); then I have the societally-imposed responsibility to determine the the dangerousness of the situation and force them to be hospitalized, if necessary.

Laws about this vary from state to state, of course; and during my 35 year career I have watched as the pendulum has swung against rather indiscriminate involuntary hospitalization of the mentally ill; all the way to overly protecting the rights of the the mentally ill at the expense of their families and the public in general.

Sally Satel reminds us that
"...according to mental-health law, we cannot restrict another’s freedom without first knowing whether he is poised to harm others or himself due to mental illness. Acting weird or lapsing into psychosis, while frightening for everyone and tragic for the ill person, is not, in itself, a crime. Nor does being psychotic necessarily mean one can be treated involuntarily."

This is important to keep in mind. Not everyone who is psychotic is dangerous. And, not everyone who "acts weird" has a mental illness.

I like to think I am good at my job. I have seen a lot of delusional and psychotic individuals over the course of the years. I have seen violent, homicidal and aggressively suicidal people; as well as angry and hostile non-psychotic individuals. I have seen them intoxicated and out of control. I have been knocked unconscious; hit, slapped; had knives and guns pulled on me in the course of my experience. You name it, I have probably seen it dozens of times.

If I do my job correctly, then those individual who are a damger to themselves or to others are temporarily neutralized until they can get help for their mental illness. When it is involuntary, doctors ususally have an initial period of 72 hours; after which a case must be proven in court to keep them longer--one or two 14-day extensions may be granted by the court; or eventually in severe cases, a conservatorship that strips the individual of the ability to make decisions and may result in him or her being placed in a locked facilty. The latter situation is actually rather rare, and even then, conservatorship is not a permanent situation, but one that is fluid and can be nullified if evidence is brought to the court that the person in question is no longer dangerous.

I have had patients say to me things like, "You can lock me up but sooner or later I will get out and do what I want to do"-- e.g., kill themselves or someone else. Sadly, they are correct. I have watched depressed individuals get out of the hospital after a court refuses to keep them hospitalized, and then they promptly go and kill themselves; or, a psychotic person with a delusion be discharged because everyone thought the delusion was under control; but they go and kill someone because it is really still there and they just learned to be really good at hiding it from the doctors. Both of these kinds of tragedies have happened multiple times over the course of my career.

Fortunately, most patients do get better and the suicidality or homicidality recedes.

Nevertheless, one big problem with my job of predicting dangerous behavior is that the consequences of what might have been are unrecordable. The thing about prevention is that...it prevents really bad things from happening. By the very act of preventing a bad thing from happening, you can never then prove that it might have occurred in the first place.

In fact, the only cases that ever make the news are the ones where the court freed the individual from the mental hospital and they do something terrible; or the one's where the doctor did not have a crystal ball and did not foresee correctly a person's behavior.

I've never liked being held responsible for someone else's behavior. It's damn hard to predict what someone is going to do in the next couple of minutes, let alone days or weeks from the time you saw them.

But this is one of the roles that society gives psychiatrists, and, as a libertarian, I have always tried to do the best job I can while respecting an individual's personal liberty as much as possible. It is particularly difficult when the patient refuses to acknowledge that something is wrong with their thinking processes; or when there has not been any specific behaviors that could be deemed dangerous--only hints and suspicions that such danger lurks beneath the surface.

Sometimes you just have to act on "soft" intelligence in making these life or death decisions.

It is a great responsibillity and one I take very seriously. I know that if I intervene I can often change the course of a person's life for the good--at least in the short term. But I also know that sometimes I simply make them angry and hostile toward mental health interventions and drive them away from getting the help they need after they eventually get out of the hospital.

Sometimes people tell me that they are having feelings of wanting to kill themselves; and they tell me the details of how they'd go about it. Occasionally, some patients will confide in me their desire to kill a specific person and ask for my help to stop them from doing this bad thing. When there is honesty and self-awareness, it makes my job a bit easier. But most of the time I have to act on "soft intelligence"

What do I mean by that? Well, most of the time I have to take many little things into consideration; for example:

-How honest do I think a person is being with me in regard to their feelings and intent?
-How honest have they been with me and/or others in the past?
-Do they have a history of ACTING on suicidal or homicidal feelings in the past? (one of the most significant predictors of future behavior is past behavior)
-Is the individual attempting to achieve a particular goal by saying they are suicidal or homicidal--even though they are not; e.g, (1) do they just desire to be be hospitalized and have no real intent to harm themselves or others; or( 2) do they NOT WANT hospitalization because they have real intent to harm themselves or others?
-Do they have the means to actually hurt themselves (e.g., if they have a plan to shoot themselves or someone else--do they actually have a gun at home? or access to a gun?)
-How seriously do their family and friends take the possibility of their suicidality or homicidality?
-How impulsive are they now? How impulsive have they been in the past?
-Are they psychotic, paranoid, or delusional?
-If they are on medication, have they been taking it regularly?
-What kinds of stressors are there in their life right now?

The above list does not include all the considerations and factors that come into play in my assessment, but they are some of the primary ones. A particular person might not meet ALL the criteria above; and each case is different, depending on the situation. I see many people who are what we call "chronically suicidal", meaning that they express suicidal ideation all the time. When is it approriate to hospitalize them, and when is it not?

I also see many intoxicated individuals who, when they are intoxicated with either drugs and/or alcohol are what we refer to as "drunkicidal". In most of these types of cases, when the person returns to normal (i.e., non-intoxicated) they are neither suicidal nor homicidal; and often they can't even remember what they said or did while under the influence.

The underlying reality of all these evaluation and careful consideration about what to do is actually quite painful; and it is this: if a person really wants to kill themselves or someone else, they will succeed eventually. In fact, many if not most of the people who do commit suicide and certainly most of those who commit homicide ARE NOT MENTALLY ILL in the formal sense of the world; that is, they do not suffer from a psychiatric illness. Most never come to the attention of a mental health professional.

But, for those who do come to our attention and who do have a mental illness; then an intervention--even an involuntary hospitalization -- can save lives.

And because people's lives are at stake, I try to err on the side of caution in most cases. The potential consequences are very high, and earlier in my career, I was a bit overwhelmed by the responsibility. I will admit that in the last 30 years, I have made mistakes. Sometimes I hospitalized someone against their will who really had no intention of hurting themselves or someone else. My judgement was incorrect about their imminent homicidality or suicidality. The worse consequence of that error was that the patient hated me and refused to ever see me or any other mental health professionals again.

OK, I can live with that.

But it has also gone the other way. I remember when I didn't hospitalize a patient of mine, because she didn't have the past history of trying to kill herself; had never been suicidal before; and was seemingly very cooperative in therapy. In addition, she didn't even appear to be particularly depressed or psychotic any of the times I saw her; and she was always willing to contract with me for safety. I was an intern at the time and somewhat inexperienced and uncertain what to do, but finally decided to believe the person when she told me that she didn't really have any "plan" to kill herself--and I really couldn't see that she had any real intent.

But to my horror, that person not only killed herself the next day, but she killed another patient who was in treatment with her. I was devastated at the time, and almost left psychiatry because of it.

If I had acted--if I had only known what was in her mind-- two lives could have been saved. Perhaps I could have only delayed what happened if I had acted when I had the chance. I don't know. I will never know. And neither I or anyone else will ever know what is in a person's mind or heart unless they are willing to tell us honestly.

In the end, a person--even a psychiatrist!--can only be responsible for their own actions--not for the actions of someone else.

Since that time I have studied all that there is to know about predicting suicide. I have tried to hone that "instinct" that makes a person aware that someone is not telling the truth. I think I am pretty good at it, and I have come to trust my instinct in these situations. Now, if I have any doubt; if I find myself not quite convinced that a person will be safe, then I go with the doubt.

Because I am completely sure only of this: If I have the opportunity to do something to save a life, then I must do it.

I cannot know the future and I cannot know how someone will behave in the future. But I can use all available information I have at that one point in time; and my own professional skills and experience to the best of my ability to assess each situation individually. I can balance and weigh the hard and soft evidence--and sometimes it is only the soft evidence that I have to work with; balance and weigh the risks and benefits; and make the best decision possible at that moment--for the individual and for the society that makes it my duty and responsibilty to decide.

The way I figure it, if I did not have doubts; if I did not constantly struggle with the decision to hospitalize involuntarily, then I would not be doing my job correctly.

No comments: