In the aftermath of Virginia Tech, a fair number of Monday-morning pundits have weighed in with opinions about how school officials, the police, and psychiatrists should have acted before tragedy hit. To clarify some of the issues about the intersection of law and psychiatry, MIWatch spoke with forensic psychiatrist Dr. Andrew P Levin.

Q: Thanks for talking to MIWatch. To set some groundwork, can you tell us what a forensic psychiatrist does?

A: Forensic psychiatrists are asked to draw on their scientific, psychiatric knowledge to answer a legal question. That legal question can fall into different realms. It can be the issue of criminal responsibility, or in lay terms, insanity. There can be issues like civil damages, the psychological damages that people have sustained, and that can be as simple as a car accident or as complex as sexual harassment. There can be issues that relate to professional responsibility or malpractice, and that has to do with standards of care. For psychiatrists that would be psychiatric care. Other physicians would provide opinions for the standards of other types of medical care. And the last area where forensic psychiatrists work is the care of the mentally ill in the prison system, correctional psychiatry, and as you’re well aware, the proportion of people now housed in prisons rather than in state mental institutions or other kinds of facilities, is growing all the time and this is a very serious issue.

Q: Does that mean they work in prisons?

A: Yes, a significant proportion of forensic psychiatrists work in the correctional system.

Q: Who decides whether a person is able to stand trial. Would that be a forensic psychiatrist?

A: Yes, now you're talking about the issue of competence. There's competence to perform a certain set of tasks. There's competence to stand trial, competence to enter into a contract, competence to make a will. Competence is an ability that is determined by a forensic psychiatrist or a psychologist, usually a person who has special training and experience. Most competency hearings are pretty straightforward and most people are found competent. The standards are relatively low in terms of the criteria.

Q: What are the standards?

A: They are based on a series of criteria from various legal cases the best known being a case called Dusky v. US. Basically they ask is the defendant aware of the charges against him and is he aware of the potential penalties? Does he know who the actors are in the courtroom –his attorneys, the prosecutors, the jury, the judge, witnesses? Does he know how to and is he able to assist his attorney? And more recently they have added, does he know what a plea bargain is? Does he understand a defense of insanity?

Q: What's the difference between someone who behaves this way and you say that’s his "personality" and someone who has a sudden break as an adolescent, and someone who is a sociopath?

A: People have enduring ways they see themselves and other people, what we would call their "personalities," and under stress there are characteristic ways those systems break down. The asocial man like Cho deteriorates in certain ways. There is a lot of scientific evidence to support the enduring quality of temperament that is identifiable early in life, the first couple of years, and is predictive of a lot of later behaviors and styles. The predictive quality of temperament was established two generations ago by Stella Chess.

Q: There's been a lot of talk about violence and mental illness.

A: It is important to recognize that although psychotic illnesses lead to a small increase in the risk of violence, the research shows that substance abuse and a prior history of violence are far and away the greatest predictors of future violence. The likely victims of someone who is psychotic is an intimate and family member. Random violence toward strangers, although it makes good headlines, is much rarer for the psychotic individual.

Q: What about stalking - is it a psychiatric illness?

A: Stalking is now a crime in all 50 states. It's a criminal definition. Stalking is defined as the imposition of unwelcome and fear-inducing communications and approaches.

The issue is: "who are the people who are most likely to stalk," and "who are the most likely victims?" By far and away, about 60-70 percent of stalking situations involve men who are ex-intimates of the woman whom they stalk. These typically are the men who call the ex-girlfriend, show up at her job, continue to call, show up unwanted at her home, and attempt to interfere with her personal life. A woman is mostly likely to be murdered by a man who is or was an intimate.

Q: Are there others?

A: A second type are people who are socially awkward and who misinterpret social signals. They tend to be men pursuing women. But not always. Sometimes they are women pursuing men. The young man who pursues the clerk at the corner grocery store because she smiles at him is a benign example. He develops unrealistic fantasies about her affection for him often stemming from his lack of social experience and inability to read social signals. He may try to show his affection with flowers, cards, etc., and usually the woman is mystified by his interest, or possibly flattered, but feels uncomfortable.

Then there is a third category of stalkers who have a condition called eroto-mania. Often they are women who believe they are in love with, and loved by, famous and powerful figures. Usually it doesn’t go any further than that but sometimes it can become quite ominous.

Then there are people who fall psychotically in love with their therapists, or their teachers or their bosses, or other people in their environment.

Q: Basically can we tell who will become violent?

A: Violence, like suicide, is a very low frequency event. Statistically it is very difficult to predict low frequency events. It is difficult to say with certainty that a certain behavior is going to occur. Instead, we now think of "risk assessment" and "risk management."

Q: So what are the risks?

A: Avariety of factors have been identified as increasing risk. For violence, of course, substance abuse is number one. Previous violence is number two. Anti-social personality is a risk, psychosis is a much smaller risk. Other considerations include access to means - weapons and peer groups.

Q: What do you do when you have someone who presents these risks?

A: When you have a series of risks, in treatment you try to remove or minimize them. Your job is to treat substance abuse, you try to ameliorate acute symptoms of illness, you try to develop better support systems, eliminate access to weapons. There are times when the risk seems imminent, and this is a judgment – a person has a gun or expresses anxiety about injuring others – although statistically it may not even be 50 percent, when they’re talking about life and death situations, then we’re willing to take away people's freedom, even when it's five percent.

So the field has really moved away from prediction.

Q: How would this manifest into a sensible social policy?

A: That's the million-dollar question and this kind of a case really brings that out. The standard, the bar, that exists now for taking away people's rights – and those rights are confidentiality and liberty, the bar for that is relatively high—the test is imminent danger to self or others and imminent is hard to define.

Currently we don't ask people who sell guns, or who rented him the car, or who sold the bullet – they don’t make an assessment of somebody, that's not part of their job. Somebody who sells you bullets doesn’t say "gee how are you feeling, are you mad at anybody, have you ever thought about killing anyone," but by statute we do ask them to check a mental health data base.

Q: Is there a slippery slope here?

A: This young man would probably not have met the criteria for AOT (assisted outpatient treatment). He had not been violent, not been hospitalized, This wasn't a young man who had been violent previously; stalking alone is not usually thought of as violence although it is a crime; it can be a predictor of violence but only a small number of stalkers are violent. The same is true for people with mental illness except where substance abuse is in the picture.

So there is the central rub, how to protect people from a serious but relatively rare event and weigh this against their rights to privacy and liberty.

Because the news focuses on high profile events, it's not a statistical approach and it leads the public to believe the mentally ill are all murderous and crazy people. People are prone to draw conclusions from single, serious events such as what we see in the media even though these events are poorly predictive of future events and risks.

Andrew P. Levin, M.D., is Assistant Clinical Professor of Psychiatry at Columbia University College of Physicians and Surgeons and Medical Director, Westchester Jewish Community Services where he specializes in the treatment of trauma. He consults and publishes widely, and has participated in joint international joint ventures training professionals in the identification and management of trauma.