In the aftermath of the horrible tragedy at Marjorie Stoneman Douglas High School, there has been an understandable desire to do something. Much of the debate fueled by that desire has been focused on gun control. And the anger, fear, and raw politics on both sides of the divisive issue have largely overshadowed other important questions that deserve serious discussion.

Among those is the tricky issue of mental-health reform. In fact, it has become a bizarrely automatic call and response that when Democrats say “gun control,” Republicans respond with “mental-health reform.” And, clearly mental-health reform is needed. According to some estimates, at least 60 percent of mass shooters are mentally ill. Many would probably consider this a vast understatement. Indeed, it’s hard to think that anyone who commits such a heinous crime could not be mentally ill.

Mental illness is also at the heart of many other social problems as well. Researchers estimate that between 25 and 45 percent of the homeless have mental-health issues, depending on how one defines such issues. Roughly 16 percent of those in prison have been diagnosed with a mental illness. And America’s substance-abuse epidemic has undeniably been made worse by mentally ill people who self-medicate.

But we should be every bit as cautious in giving the government more power to deal with mental-health issues as we are in giving it more power to regulate gun ownership.

Recently, for example, President Trump spoke to the nation’s governors about the need to reopen mental institutions, suggesting a return to the era of widespread involuntary commitment. Many social scientists agree that the mass deinstitutionalization that took place in the 1970s and 80s didn’t turn out as well as planned. The country once had more than 500,000 inpatient psychiatric beds available to treat the mentally ill; today, that number has dropped below 40,000. Many of those who were released during deinstitutionalization were ill-equipped for life on their own, and alternative treatment methods were frequently unavailable. At the same time, legal changes made it increasingly difficult to commit someone against their will, which is problematic since those with the most serious mental illnesses are often the least likely to seek treatment on their own. It is a tragic Catch-22, and it is certainly possible that we have moved too far in a direction that puts both the mentally ill and the wider society at risk.

Yet, we should understand why deinstitutionalization happened in the first place, and why it enjoyed broad bipartisan support. Many psychiatric facilities at the time weren’t much more than warehouses, offering little real treatment. Others were literal houses of horrors, utilizing shock treatment and other discredited approaches, while confining patients in squalor and neglect. The abuses and concerns that sparked deinstitutionalization were real and justified.

We should also be aware of how frequently diagnoses of mental illness has been abused in the past. We are all familiar with stories about how authoritarian regimes have called dissent evidence of mental illness. But there was a time in this country when a husband could have a wife committed to a mental institution for being disobedient or too independent. Likewise, parents often sought the commitment of stubborn or recalcitrant children. More recently, a since-repealed Obama administration rule, designed to limit access to guns, suggested that Social Security recipients who relied on others to manage their finances were mentally incompetent. And many on the right continue to brand transgender Americans as mentally ill.

One would think that conservatives who are understandably concerned about the potential slippery slope of gun control would not recognize a similar slickness when it comes to mental illness.

We should be careful about suggesting that even those who really are mentally ill are, as a matter of course, also dangerous. The vast majority of those suffering from mental illness pose no danger either to themselves or others. As the National Institutes of Medicine concludes, while there is “a link between mental illnesses and violence, the contribution of people with mental illnesses to overall rates of violence is small, and further, the magnitude of the relationship is greatly exaggerated in the minds of the general population.” According to a recent study, only 5 percent of all gun-related killings in America could be linked to perpetrators’ symptoms of mental illness. So while mental illness may be in part responsible for the deadly sprees we hear about in the news, the mentally ill are not to blame for most gun deaths in America.

Rash moves on mental health might even be counterproductive, dissuading the mentally ill from seeking treatment at all. This is particularly true of proposals to create registries or databases of the mentally ill.

We should also be wary of simply throwing more money at the problem. Unlike many physical illnesses, some mental-health conditions are hard to diagnose and even harder to declare “cured.” That’s a ripe opening for abuse. Moreover, roughly 25 percent of mental-health care is provided through Medicaid, a program that is rapidly consuming unsustainable portions of both state and federal budgets. It may well be that states need to shift funds from, say, providing nursing-home care for the middle class to treating mental illness, but that raises its own set of difficulties.

None of this is to say that we can’t or shouldn’t do more when it comes to mental health. There may well be steps on mental health that we can take to reduce the chances of another mass shooting. But with gun policy and mental-health policy alike, we should be careful to balance the desire to “do something” with the need to protect the rights of all law-abiding Americans, whether they’re mentally ill or not.