In the aftermath of the massacre in Newtown, Connecticut, America is having a long-overdue national conversation about guns, mental health, and avoidable violence. The slaughter of elementary schoolchildren has a way of clarifying things.

Much of the conversation thus far has focused, quite rightly, on the guns. President Barack Obama has presented a package of valuable measures that closely match the recommendations of the nation’s leading gun violence experts. If enacted into public policy and American law, the president’s proposals, especially closing loopholes on background checks, have the potential to save many lives.

But there’s a certain irony in the timing of this national conversation, and the way in which we are conducting it. High-profile mass shootings are relatively rare, resulting on average in a few dozen deaths a year; ordinary, day-in, day-out gun crimes, on the other hand, wipe out more than 10,000 lives a year. Mass shootings are also quite difficult to prevent. I hope the sheer horror of Newtown catalyzes passage of a strong assault weapons ban, but the number of lives we will save through such a measure is likely to be modest. That’s because the overwhelming majority of gun-related murders and injuries, whether by angry spouses or by professional criminals, aren’t committed with assault weapons (defined, loosely, as semiautomatic firearms with military-style characteristics). Rifles and shotguns of all types, assault versions and not, together account for less than 12 percent of murders for which the type of gun is recorded; handguns account for the rest. The flow of ordinary crimes committed with ordinary weapons doesn’t command the same attention, but it accounts for many more gun deaths.

A similar irony applies in the area of mental health. The Newtown massacre has elicited serious calls to do something to stop the violently prone mentally ill. Yet the fact is that very little of the violence perpetrated by people with mental illnesses takes the form of mass shootings. Much more common are cases in which deranged individuals commit assaults, muggings, and robberies, or cases in which individuals with mental illnesses commit crimes connected with the misuse of alcohol or illicit drugs. In many cases, the primary victims are not strangers, but the perpetrator’s own loved ones and friends.

Just like those ordinary crimes committed with ordinary weapons, these individual acts of violence seldom make headlines. And of all the mentally ill likely to commit violence, mass shooters may be the hardest to identify in advance—in part because of the unusual, often idiosyncratic nature of such atrocities.

To be sure, in rare cases—often the cases that make headlines—mental health systems have failed to stop specific individuals who were already known to be dangerous. Seung-Hui Cho, the twenty-three-year-old who killed thirty-two people at Virginia Tech, had previously been found by a judge to be “an imminent danger to himself as a result of mental illness” and forced into outpatient mental health care. But things are usually far murkier. After an atrocity happens, one can often spot some ominous warning sign, some missed opportunity to intervene. Jared Loughner displayed strange and scary tendencies before he shot Gabrielle Giffords and eighteen others. James Holmes saw three mental health professionals before he shot dozens of Aurora theatergoers. Such red flags generally appear brighter in retrospect, the dots easier to connect, than they do in real time. One rarely sees the equally red flags waved by tens of thousands of others who have struggled with mental illness or who have sought help for personal challenges who never subsequently hurt anyone. There is little in the biography of Adam Lanza that should have led law enforcement authorities to specifically identify him as a threat to public safety before the events in Newtown.

In a New York Times op-ed last December by Richard A. Friedman, Columbia University clinical psychiatrist Michael Stone noted that “most of these killers are young men who are not floridly psychotic. They tend to be paranoid loners who hold a grudge and are full of rage.” They are damaged, but not disabled, by whatever impels them to hurt others. Indeed, they remain sufficiently lucid to plan and execute horrific crimes. Except perhaps to the people closest to them, shooters’ signs of possible violence are rarely sufficiently obvious or precise to provide the basis for effective clinical or law enforcement intervention.

That doesn’t mean we should abandon the effort to identify and stop potential mass killers. Rather, as with gun policy, we need to be realistic about where the bigger problems lie, and what is likely to be accomplished. We must use the political energy of this moment to take actions that can reduce far more prevalent kinds of violence perpetrated by the mentally ill.

Fortunately—and unlike the situation with guns—Congress doesn’t need to pass major legislation to deal with mental illness-related violence. We might begin by doing a better job implementing programs we’ve already put in place. In Illinois, a 2011 report indicated that the state police should have reported an estimated 120,000 mental health records to the FBI National Instant Criminal Background Check System, which is used to keep guns out of the hands of the dangerously mentally ill. Only about 5,000 of these records had been provided. Illinois law enforcement also struggles to effectively retrieve weapons when people who bought guns legally in the past are found to have serious mental health problems. Boring nuts-and-bolts issues, funding challenges, and bureaucratic wrangling thwart such efforts. Obama has proposed federal money to support such state and local efforts. We should make sure that happens.

As for more sweeping mental health efforts, much of what’s needed is already on the books, in the form of the Affordable Care Act. The trick will be making sure that the new law gets translated into regulations and on-the-ground policies that actually address the problems we face.

Millions of Americans suffer from some form of severe mental illness, or SMI. It’s important to remember that the vast majority of these men and women have never committed a violent crime and never will commit one. (Indeed, the mentally ill are often victims of violent crime, a social problem that has not received sufficient attention.) The most careful studies in several wealthy democracies suggest that the severely mentally ill account for perhaps 5 percent of violent crimes.

Despite these rather low numbers, violence perpetrated by a subset of disturbed individuals is a genuine public safety concern. These are certainly familiar issues for law enforcement. As one influential report notes, “Officers spend more time managing incidents related to severely mentally-ill persons than they do responding to traffic accidents, burglaries, or assaults.”

Criminal offending is especially prevalent among young men with emergent symptoms. An Australian study explored crimes committed by individuals with schizophrenia. Among those convicted of violent offenses, 64 percent had their first conviction before their first contact with mental health services. This pattern arises in many other populations. Serious psychiatric disorders often emerge or strike during adolescence and young adulthood, precisely the point in the life cycle when people have the most tenuous contact with social services or medical care—and precisely the point when people face the highest general risks of becoming either perpetrators or victims of violent crimes.

“Dual diagnosis,” the combination of SMI and substance abuse disorders, appears to be an especially strong risk factor for violent offending, with alcohol use playing a particularly prominent role. A much larger fraction of violent crimes is associated with alcohol disorders than is associated with severe mental illness. Indeed, crime rates among individuals with SMI who do not have co-occurring substance use disorders appear surprisingly similar to those of their peers of the same age and gender, living in the same communities, who don’t suffer from mental illness.

The Australian study underscores the strength of this relationship. Among male schizophrenia patients, 29.7 percent of those with known substance abuse problems were convicted of a violent offense, compared with only 7.4 percent of their counterparts without substance abuse problems. Studies here in the U.S. suggest that roughly two-thirds of jail inmates with severe mental illnesses are dual diagnosed. Dual-diagnosed individuals are also far more likely than others to re-offend after they are released from jail or prison.

Mental health services are not cure-alls. Fisher and colleagues examined arrest records of almost 14,000 individuals who had received services from the Massachusetts Department of Mental Health during 1991 and 1992. Between 1991 and 2000, 28 percent of this group was arrested for committing a crime. More than half the young men ages eighteen to twenty-five were arrested over the same period. Especially concerning were the 13.6 percent of the entire sample arrested for “serious violence against persons.”

Harvard mental health expert Richard Frank notes that many of the most dangerous young men have contact, albeit sporadic, with mental health service systems. Yet they are rarely properly assessed for drug or alcohol disorders or referred to the most effective interventions designed for dual-diagnosed individuals. “These people are touching the system,” Frank says. “They’re one step away from what you need to do.… We just don’t manage to make the connection.”

Many disturbed and violent individuals are especially easy to spot for one simple reason: they’ve been violent before. They are already locked up, or they are under some other form of criminal justice supervision. In Chicago, the massive jail complex at Twenty-sixth and California may be the largest de facto psychiatric facility in the Midwest. The sheriff’s office estimates that about 22 percent of the jail population has been diagnosed with some form of mental illness. On any given day, that’s about 2,000 people. Research spearheaded by Northwestern University’s Linda Teplin suggests that there may be an even higher prevalence of psychiatric disorders among juvenile detainees.

Many dual-diagnosis inmates are what you might call frequent flyers. They are detained and re-detained because they shoplift, yell at passers-by, urinate in the street, or in other ways become a public nuisance. Some receive mental health services while in jail; some pass through in therapeutic isolation. They are then returned to the street, where their mental health and substance use disorders often go (or remain) untreated. Over time, some portion of this group turns violent. This dual-diagnosis group is rearrested much more frequently and quickly than are other released inmates, including those suffering from SMI without co-occurring substance use disorders.

We can’t ethically perform randomized trials that compare a treatment group that is provided mental health services to a control group that is not. We can, however, compare innovative approaches to usual modes of care. Evidence-based programs appear especially valuable when they provide transitional services into or out of secure institutions.

Because the social costs of crime are so high, even highly imperfect services—which is to say, most of the services offered to this population—tend to wildly pass any reasonable cost-benefit test. Anirban Basu, David Paltiel, and I published an economic analysis of substance abuse treatment services that included (but were not restricted to) individuals with co-occurring disorders. We found that the economic benefits of reducing the incidence of armed robbery alone—which was only committed by a small subgroup of treatment patients—were sufficient to offset the entire cost of the intervention.

One promising approach is known as Forensic Assertive Community Treatment (FACT). This intervention engages severely mentally ill patients through the delivery of psychiatric and addiction treatment, transportation, financial services, and vocational support. FACT augments treatment models for severe mental illness used outside the criminal justice context with the legal leverage provided by judicial monitoring.

A randomized trial performed by Chapel Hill mental health services researcher Karen Cusack and colleagues indicated that mentally ill individuals supervised with a high-quality FACT program were less likely to commit crimes. Over a two-year period, the FACT group demonstrated a 41 percent reduction in the number of jail days and a 19 percent lower conviction rate than was observed among their counterparts receiving usual care.

Why do so many people at risk—many of them young low-income men—fail to receive appropriate mental health services? The most important single reason is this: most are categorically ineligible for Medicaid. These men are not custodial parents. They are not veterans. They have not (yet) been diagnosed with federally recognized disabilities. Many get into trouble because they have serious drug or alcohol disorders. Since 1996, substance use disorders are no longer qualifying conditions for federal disability programs.

People often assume that an unattached adult might qualify for Medicaid if he is sufficiently poor. In fact, only a handful of states provide such coverage, even for individuals with no income at all. The homeless man who is not deemed physically or mentally disabled but who does suffer from chronic alcohol or crack dependence is typically ineligible for Medicaid. He might need substance abuse treatment or mental health care (not to mention care for any number of other health needs). If he’s locked up, the correctional service is required to provide these services. If he’s out in the community, he’s both more vulnerable and more bereft of help. He’s reliant on a patchwork of safety-net services, public hospitals, clinics, and emergency departments that’s financially stressed and disorganized under the best of circumstances, and that often allows vulnerable and occasionally dangerous people to fall through the cracks.

This will begin to change in 2014. That’s when the ACA will start providing subsidies that will eventually reach thirty-three million Americans without health insurance. An estimated sixteen million will eventually be covered by expanded Medicaid to low-income Americans with incomes below 138 percent of the federal poverty line. That number will include the hundreds of thousands of mentally ill men cycling in and out of places like Chicago’s Cook County Jail and sleeping on grates in cities from Washington, D.C., to Seattle. For the first time, nearly all of these individuals (undocumented immigrants are the big exception) will gain access to regular health care. Moreover, if the law is properly implemented, these same individuals will gain access to mental health services that can reduce their propensity to commit violent acts.

But that’s a big “if.” In theory, both within Medicaid and outside it, the ACA expands the range of mental health and substance abuse services. Insurance plans within the new state insurance exchanges must cover mental health and substance abuse services—and do so under conditions of “parity” so that copayments and other details match plans’ coverage of physical health conditions. In the fine print, evidence-based approaches including screening, brief intervention, and referral to treatment for alcohol disorders are supported in emergency departments and other medical settings.

The devil is in the details. Right now there are major fights under way over how precisely the law will be implemented, and which specific services Medicaid will cover. One fight that’s gotten a fair amount of attention is in states such as Texas and Georgia, where recalcitrant Republican governors so far have refused to allow their Medicaid rolls to expand, even though the federal government will pick up nearly 100 percent of the cost. Almost every health policy expert believes that these governors will eventually relent to pressure from their constituents and from local medical facilities of all kinds who stand to lose billions from the governors’ intransigence. The longer they wait, however, the longer poor and possibly violent mentally ill people walking the streets will be denied treatments that could help them and that might protect surrounding communities from preventable crimes.

Still, health reform represents a genuine sea change in this world of services. Here in Cook County, Sheriff Tom Dart notes that most inmates will be eligible for Medicaid after 2014: “The Medicaid waiver and the ACA have the potential to be game changers, ending the [federal, state, and city] cuts that have made the Cook County Jail the provider of first resort in the city of Chicago and Cook County.”
Other battles must still be fought. States don’t always cover specific promising services such as FACT. It’s especially important for Medicaid to cover evidence-based interventions for young men with cooccurring mental health and substance abuse disorders who pose the greatest public safety concern.

We must also reverse recent punishing cuts to our mental health system. Between 2009 and 2011, states experienced a cumulative shortfall of $432 billion. Mental health agencies in almost every state have cut expenditures, even as recession increases demand for community mental health services, crisis services, and emergency department services.

The ACA helps to address some of this problem, since outpatient mental health services for “non-Medicaid low income consumers” have faced particularly deep cuts. Yet the problems go deeper, including the continued decline in the number of state psychiatric beds. As state hospitals continue to contract in the face of fiscal difficulties, psychiatric emergency rooms, nursing homes, and acute care facilities face growing burdens, seeking to serve the sickest segments of the mentally ill population, who would once have received institutional care.

Finally, one encounters an obscure but important rule that few people have even heard of. Since Medicaid’s beginning, the program has maintained something called the “institution for mental diseases” exclusion, or IMD. Roughly speaking, Medicaid does not cover inpatient care for working-age adults at psychiatric facilities with more than sixteen beds.

This policy may once have made sense. Federal policymakers didn’t want to assume the full costs of extensive state and local mental hospital systems. Advocates for the mentally ill did not want to create further incentives to warehouse people. Today, though, the IMD exclusion hinders low-income patients’ access to inpatient psychiatric care, including specialized residential care for individuals with substance use disorders. Ron Honberg, the national director for policy and legal affairs for the National Alliance on the Mentally Ill, notes that some people require the help, monitoring, and supervision of long-term inpatient care. Family caregiver-activists such as Ilene Flannery Wells emphasize that when the mental health system steps back from providing this care, this leaves vulnerable, potentially dangerous people out on the street until (in all too many cases) a tragedy occurs, or until the correctional system steps in to fill the gap.

It’s a strange thing. Newtown was an atypical crime, committed by an atypical offender, using a murder weapon that I hope will be outlawed but that remains pretty atypical for gun homicides. Even though we may not be able to stop an event like Newtown from happening again, it seems to be moving public policy more than the routine smaller scale tragedies that we could more easily prevent. Newtown has provided a genuine occasion for Americans to think seriously about gun policy, and to consider the very real challenges to our mental health system. We should make the most of this moment.

It’s naive to believe that we could specifically identify someone such as Adam Lanza before he goes on a rampage, but improved policies could still prevent an unknown, maybe unknowable number of violent deaths. No one policy will dramatically reduce homicides, and the politics and administration of effective mental health policy are both daunting. But making these policies work would provide a fitting memorial to the victims of needless violence across America. While we may not be able to entirely solve the tragedies that occur at the intersection of mental illness and gun violence, surely we can do better than we’re doing now.

Harold Pollack

Harold Pollack is the Helen Ross Professor at the School of Social Service Administration at the University of Chicago.