Violence and mental disorders—rightly or wrongly—appear to be irreversibly linked in the popular mind. Articles in this issue of the Journal shed light on two key questions about this relationship:
Sweden’s comprehensive national registers of hospital admissions and criminal convictions provide the data for Fazel and Grann’s exploration of the impact of severe mental illness on violent crime in this issue. Linking the two registers, the authors find that persons with psychoses are about four times more likely than the general population to have been convicted of a violent crime but that the psychotic group accounts for just 5% of such offenses. Age, gender, diagnosis, and type of criminal offense all affect the odds ratios for violent convictions and the percentage of crimes attributable to persons with psychoses. Of particular note, women with severe mental illnesses make a negligible contribution to the overall rate of violence.
These Swedish data confirm an evolving consensus on the relationship between serious mental illnesses and violence. Studies using a variety of methods have shown an elevated risk for violence among persons with mental disorders (1). However, the proportion of violence that they account for is relatively small, suggesting that the well-documented public perception of the mentally ill as dangerous persons is substantially exaggerated (2) and that the disproportionate attention given to their acts of violence by the media and by our elected representatives is unwarranted.
It is important to note that the results of Fazel and Grann regarding the percentage of violence attributable to persons with psychoses cannot be extrapolated directly to the American context. In countries such as Sweden, with low rates of violent crime, persons with serious mental illnesses are likely to account for a larger percentage of criminal violence than in countries such as the United States. Nevertheless, the finding of an 18% attributable risk of homicide for people with psychoses in that population, although unlikely to be replicated in the United States, underscores the importance of continuing research on factors that mediate the risk of violence in our patients and on the means of reducing that risk.
How, then, given the current state of our knowledge, ought we to respond to the possibility of violence among persons with mental disorders? Swanson and colleagues explore one approach in this issue. Drawing on a five-site survey of 1,011 outpatients of community mental health centers, they document the extent to which legal mandates and access to money and housing are used to leverage compliance with treatment among persons who report acts of violence or physical aggression. Legal mandates for treatment are significantly more common among persons with histories of more serious violence and among persons with any level of violence who also report poor medication compliance.
Taken at face value, the data of Swanson et al. suggest that when legal mandates are available, they are targeted at patients with elevated risks for violence. (As the authors note, the cross-sectional nature of the data makes it impossible to determine the causal relationship, i.e., to demonstrate that it was violence and not some other factor that led to the imposition of the mandates.) Highly publicized acts of violence by persons with mental disorders often evoke calls for expanded mandatory treatment; outpatient commitment, in particular, has been adopted by a number of states in response to such events. Indeed, an earlier study by some of these authors suggested that outpatient commitment, when paired with frequent clinical contacts, may reduce the subsequent risk of violence (3). From this perspective, the data presented here suggest a rational use of leverage for treatment.
But it remains an open question whether the full panoply of approaches to mandatory treatment—including those imposed by mental health courts, terms of probation and parole, and outpatient commitment orders—is effective in reducing the risk of violence. Among the variables likely to determine effectiveness in a given population are the extent to which violence is linked to psychiatric symptoms, the efficacy of treatment in reducing those symptoms, the availability of treatment, the degree of compliance with treatment (which may relate to how aggressively the mandates are enforced), and the degree to which positive effects carry over after the termination of the mandate. For example, substance abuse and delusional ideation, both frequently proposed as important determinants, have shown inconsistent strength as predictors of violent outcomes across studies, perhaps because of differences in the methods used and the populations studied (4–7). Given the complex interactions among these variables, claims that widespread use of mandatory outpatient treatment will significantly reduce the risk of violence, although very much worth investigating, are decidedly premature. Indeed, at this point, a stronger argument can be made for mandates as a means of improving the treatment of people with serious mental illnesses than as a mechanism for increasing public safety.
Another widely embraced approach to reducing violence by persons with mental disorders involves restriction of their access to firearms. Norris et al., in this issue, provide a comprehensive review of federal and state statutes directed to this end. Although the federal statute defines the minimum criteria to be applied in determining whether someone can purchase a gun, states can enact more restrictive laws. With most states having enacted legislation, there is substantial variation across jurisdictions; some states limit restrictions to persons who have been involuntarily committed for treatment of mental disorders or convicted of substance abuse-related offenses, whereas others appear to encompass a much broader range of persons who have sought treatment for mental disorders, including substance abuse.
These statutes pose a dilemma for advocates for persons with mental disorders, including psychiatrists and their national organizations. Many such persons and groups are probably appalled at the ready access to firearms that prevails in much of the United States and would favor greater restrictions for all people. Thus, it is difficult for them to oppose any law that makes it harder to acquire a gun. But given that only a tiny fraction of violence, including gun violence, is perpetrated by persons with mental disorders, efforts that center disproportionately on restricting their access reflect a deeply irrational public policy. Moreover, by once more linking mental disorders and violence in the public mind, these firearms laws reinforce the stigmatization of persons with mental disorders as inherently dangerous.
Compounding concern about the effects of these statutes, many of them call for the creation of data banks that accumulate information about persons who meet the criteria for exclusion (e.g., patients who have been involuntarily committed to a psychiatric facility) that can be accessed by one or another state agency before the sale of a firearm. The threat to the confidentiality of psychiatric treatment is evident, and in many cases, it continues indefinitely. Like clinical interventions, public policy initiatives should be subject to evaluations of their effectiveness and adverse consequences. Whether singling out persons with mental disorders, including substance abuse problems, for restrictions with regard to gun purchases is an effective means of protecting the public cries out for careful assessment.
The relationship between mental disorders and violence is complex. Among the variables that have been identified as increasing the risk of violence, in addition to psychotic symptoms and substance abuse, are socioeconomic status and even the neighborhoods in which persons with mental disorders reside. No single approach to reducing the risk is likely to be completely effective. And given the relatively modest contribution to the overall risk of violence by persons with mental disorders, the likelihood and magnitude of adverse effects from any intervention must be carefully considered before it is embodied in law.
Psychiatrists and organizations such as APA have an important role to play as the “honest brokers” in this process. When passions become inflamed by tragic acts of violence, we should be clear voices of factual information and advocates of reason. Real risks should be acknowledged and appropriate interventions endorsed, while distortions are exposed and recourse to discriminatory and stigmatizing policies is discouraged; a tall order perhaps in what is often a politically charged environment, but not a bad set of aspirations.