For many years it has been politically correct to claim that people with schizophrenia or other forms of mental illness are no more dangerous than the general population. Even after the recent homicides in Arizona attributed to Jared Lee Loughner, an obviously mentally ill young man, the Bazelon Center for Mental Health Law stated: “Studies show that having a mental illness, in itself, does not increase one’s propensity to commit serious violence.” Likewise, Mental Health America, previously known as the National Mental Health Association, publicly claimed that “people with mental health conditions are no more likely to be violent than the rest of the population.” Advocates for the mentally ill seem to think that if they repeat this often enough, it will become true.
The facts are otherwise, as shown by several studies over the past two decades. It’s true that most individuals with mental illnesses, including all those taking the medication needed to control their symptoms, are not more dangerous than the general population. However, among the 4 million individuals with serious mental illnesses, primarily schizophrenia and bipolar disorder, a small number are clearly more dangerous — and, as happened to Jared Loughner, they become headlines when their delusional thinking and auditory hallucinations lead them to commit some heinous act.
Since 1990, eight major studies have been carried out in the United States assessing violent acts by individuals with serious mental illnesses. The 1998 MacArthur Foundation–funded study, which is usually cited by mental-health advocates as proving that these individuals are not more violent, actually proves the opposite. The study followed 951 patients who were discharged from psychiatric hospitals. Although many patients had refused to participate in the MacArthur study, thus removing some of the most paranoid members from the sample, in a one-year period the 951 individuals committed a total of 608 acts of serious violence (physical injury, threat or assault with a weapon, or sexual assault). The violent acts included six homicides. Overall, 18 percent of the patients who were not substance abusers, and 31 percent of those who were, committed an act of serious violence. These rates were 25 percent higher than the rates among other residents of the high-crime neighborhood in which the study was carried out.
An alternative means of measuring violence is to question those who have a mentally ill family member. In 1992 the National Alliance for the Mentally Ill asked more than 1,400 of its members whether their mentally ill family member had physically injured anyone or threatened to do so in the previous year. Eleven percent responded that their family member had harmed someone, and an additional 19 percent said their family member had threatened to do so.
Given such studies, it should not be surprising to find that approximately 10 percent of all homicides in this country are committed by individuals with serious mental illnesses, a figure that was reported in two small studies in California and New York and a recent large study in Indiana.
Can we predict which mentally ill people are most likely to become violent? The two strongest predictors of violence in all individuals, mentally ill or not, are a history of violence and the abuse of alcohol or drugs. For those who are mentally ill, however, there are additional predictors. The strongest is a failure to take their medication. Virtually every high-profile violent act committed by a mentally ill person was done by someone not taking medication. Another factor that increases the risk of violence is the presence of certain kinds of psychiatric symptoms, especially paranoid delusions or a belief that someone is trying to control your mind. Jared Loughner had both of these risk factors, according to those who knew him.
Given the prevalence of such episodes of violence, what should we do? The real tragedy is that they continue to occur even though serious mental diseases are treatable in most cases. We have a variety of effective anti-psychotic medications that control a patient’s symptoms and let him live a reasonably normal life, even if they don’t cure him. If Jared Loughner had been properly treated, he would probably now be finishing community college and able to hold a job, rather than facing the probability of life in prison. But medications are, of course, effective only if the person takes them.