AMA Journal of Ethics^®

Physicians have become accustomed to and appreciate the value provided by the availability of organ donor registries, cancer registries, and other disease registries. However, applying the registry concept in the mental health and corrections arenas raises questions about its ultimate effectiveness. The rationale for such registries is that they protect the public in several ways: by restricting the actions of those considered dangerous (e.g., by limiting the access to guns of people in mental health treatment or where sex offenders can live and work) and by making the government or the public aware of the presence of those considered dangerous (as in registries of sex offenders or mandated reporting of threats of harm). But these registries increase the stigmatization of those listed on them, and, because of their focus on those with mental illness, they distort and magnify the role of mental illness in violence.

Gun Restrictions

In the wake of the December 2012 Newtown shootings of children and teachers by a young man who allegedly had Asperger syndrome, the Connecticut legislature enacted a series of bills to restrict firearms and size of ammunition magazines and to provide mental health services to schools [1].

The legislature also added gun restrictions for people who voluntarily admit themselves to mental hospitals. The new law restricts gun possession for any person who voluntarily admits himself or herself to a psychiatric hospital on or after October 1, 2013, for reasons other than drug or alcohol addiction. Anyone who has been so admitted will not be able to receive a permit or eligibility certificate or to possess a firearm for 6 months thereafter [1]. This means that the Connecticut Department of Mental Health and Addiction Services will now have to report all voluntary admissions to public and private psychiatric facilities (including names and other identifying data) to the Department of Emergency Services and Public Protection.

The legislature, however, left a significant gap in the reporting statute. Patients who were involuntarily committed by a court to a psychiatric hospital within the past 5 years were restricted from acquiring guns [1] (lengthening the previous restriction, which only restricted gun access within 1 year of court-ordered commitment [2]), but there are no restrictions on gun ownership for the mentally ill who have been involuntarily hospitalized by a single physician using a physician’s emergency certificate. This certification permits detention in a psychiatric facility for up to 15 days without a full court hearing for those who are deemed mentally ill and dangerous or gravely disabled [3]. This is the most commonly used mechanism for involuntary confinement, and those so hospitalized face no restrictions on the ability to purchase or own a gun.

The underlying assumptions must be that those who voluntarily seek treatment in mental hospitals are more likely to be dangerous than those who do not and, since those with alcohol or drug problems were excluded from the statute, that the latter group are less likely to be dangerous than people with mental illness. Both assumptions are questionable. It is inappropriate to select voluntary patients as the target group for the legislation. A significant number of people with mental illness are admitted because they are “gravely disabled” rather than being dangerous. Furthermore, the exception for those in alcohol or drug treatment is wrongheaded, since this group has a much higher violence rate than those with mental illness alone [4].

Another problem with this section of the new law is that it means hospitals will have to inform potential voluntary patients of these restrictions before they are admitted, which may deter people from seeking treatment. Thus, the law may not only discourage voluntary admissions, it will also increase the stigmatization of mental illness and will unfairly affect patients who are not dangerous. Overall, people with mental illness have been shown to commit only about 4 percent of the total violence in this country [5].

Unlike bans on semiautomatic weapons, mental health laws get bipartisan support. Gun advocacy groups have conceded to gun regulation that applies to people with mental illness. In response to Newtown, the National Rifle Association (NRA) argued that mental illness, not guns, was the root cause of the shooting spree. Restricting possession of weapons only by those with mental illness is not a solution to the larger problem of gun violence, and broad statutes aimed at people with mental illness seeking treatment make the situation worse, rather than better.

Threats of Harm

The headlines from recent mass shootings have focused on the mental illnesses of the perpetrators and have precipitated some reactive laws that legislators hope will decrease the likelihood of recurrences. These laws are not likely to be effective, given our lack of ability to predict which people with mental illness will become violent and when they will have violent exacerbations. Following the Tarasoff rulings of the mid-1970s, most states enacted legislation that requires mental health professionals to report to police or take steps (e.g., hospitalization) to prevent harm when they have concluded that a patient represents a danger to an identifiable victim or specific class of victims [6].

New York, also following the Newtown shooting, passed a law that requires mental health professionals to report to authorities when a patient is “likely to engage in conduct that would result in serious harm to self or others” so that that person’s access to firearms can be restricted [7].

This new statute is vague and will be difficult to apply. How “likely” does an event have to be before a report is required? Many patients express suicidal feelings or ideation during treatment or express anger towards others. Generally patients expect their comments to mental health professionals to be confidential. Confidentiality is not absolute in these treatment relationships due to other statutes that specify reporting of child abuse, elder abuse, and credible threats of harm to specific people. New York’s law introduces yet another warning that will have to be made in advance to patients. It will further erode the patient-physician relationship and will be more difficult to explain. The pressure to report will be high: many professionals will be concerned that not doing so will result in liability. So, once again, the provision could cause people to avoid treatment.

In the 3 months after New York’s mental health professionals were required to report potentially dangerous mentally ill patients to the state database that screens gun-permit holders and applicants, approximately 6,000 names were submitted to the state—but fewer than a dozen resulted in any action to remove weapons [8]. The major problem is overidentification. Reporting will always include many who do not pose a significant risk and will have a chilling effect on help seeking by people concerned about state intrusions.

Sex Offender Registration

Sex offenders are subject to stigmatizing restrictions that don’t protect would-be victims. Indeed, the management of sex offenders has been controversial for more than 75 years. From the 1940s to the 1970s it was hoped that mental health treatment would provide some solution, so offenders were transferred to treatment facilities in lieu of long-term prison sentences [9]. Even when longer sentences were imposed in the 1980s, offenders still had to be released after serving them. Some, of course, reoffended. In 1990 new legislation proposed indeterminate mandatory civil commitments to mental health facilities for sex offenders after they had served their entire criminal sentences, to protect the public [10]. About 20 states have enacted these “sexual predator statutes” since 1990 [11]. These were challenged but upheld by the Supreme Court in 1997 [12].

Then registration for released sex offenders was proposed. The federal government provided incentives for states to adopt them by tying federal funding for law enforcement to the enactment of these statutes. These statutes were named for victims, usually children. The Jacob Wetterling Crimes Against Children and Sexually Violent Offender Registration Act of 1994 required states to establish registries for sex offenders for at least 10 years after their release from prison [13]. In 2006, the Adam Walsh Child Protection and Safety Act expanded the crimes covered by the registries, made failure to register a federal crime, and mandated reporting to a national database [14].

The number of registered sex offenders in the U.S. in 2012 was 747,408 [15]. Keeping track of this data is difficult. California acknowledged losing track of an estimated 33,000 offenders [16]. Many also require community notification, which was in response to the 1996 Megan’s Law, later expanded to mandate that states place information on web sites available to the general public [17].

Many states have enacted more stringent statutes requiring lifelong registration and adding restrictions concerning where offenders may live [18]. Child molesters are barred from living within specified distances (usually 1,000 to 2,500 feet) from schools, child care centers, and playgrounds. There are, however a number of studies that suggest that this restriction is not particularly effective. Zandbergen, Levinson, and Hart [19] compared residential locations for a group of sex offenders who reoffended over a 2-year period with those of offenders who did not offend again. Recidivists were not more likely to live within the buffer zone around schools or day care facilities. Nonrecidivist sex offenders were, in fact, significantly more likely to live within a 2,500-foot radius around at least one school.

In a Minnesota study, the Department of Corrections examined the relationship between residential proximity to areas with high concentrations of children and sexual recidivism. The study found that residence restriction laws, had they been in place, would have been unlikely to deter any of the 224 registered offenders they studied [20]. The fact that 93 percent of sexually molested boys and 80 percent of sexually molested girls know their offenders before the attack [16] highlights the importance of personal relationships rather than residential proximity.

Several authors have argued that forcing offenders to live at a distance from family, friends, and resources can result in social isolation and make reintegrating more difficult [21]. The location restrictions tend to push offenders toward more nonurban areas where jobs may be harder to find and prices for housing can be higher [22]. The public’s easy access to registries also makes it harder for released offenders to find jobs. Furthermore, studies have tended to indicate that a high percentage of offenders do not obey location restrictions and that these laws are not strictly enforced [23].

A particularly severe housing restriction statute in Georgia was overturned by the Georgia Supreme Court in 2007 [24]. Georgia’s restrictions include a prohibition on loitering where children congregate, including school bus stops. The plaintiff, who purchased a home and established a business in locations away from any schools and child care centers, was then asked to change his residence when new child care centers opened up near his home and business. The Georgia Supreme Court found the restriction around his residence, but not the one around his business, to be unconstitutional. The court’s sensible point was that a registered offender could possibly have to move repeatedly if he or she relocated to a less developed area that subsequently became built up with schools, day care centers, and the like.

Conclusion

Tracking of those with mental illness has become so heterogeneous that it is not easily classified. Registries range from those for criminals with mental disorders that may or may not relate to violent behavior to those for people seeking treatment for mental illness who would not be considered dangerous. In the legislature’s rush to “do something” following a tragic event, policy makers are less open to consultation and more prone to sweep with a broad brush where a more nuanced approach is necessary. While some people with severe mental illness do commit heinous crimes, there is no easy way to predict who or when or where the next such event will occur. Since the role played by those with mental illness in our national violence problem is small, they should not be scapegoated.