The tragic killing of 30-year-old Jordan Neely in a New York City subway this month has once again thrust the challenging topic of mental health treatment to the forefront of public debate. A closer examination reveals that different states have sharply divergent standards for emergency civil commitment, often based on outdated medical understandings. The time is ripe for a serious discussion about how we should care for society’s most vulnerable and exposed populations during times of severe crisis while never losing sight of the exceptionally serious civil liberties implications of such committals.

Consider that many states, including New York, have laws narrowly tying emergency civil commitment to evidence of present danger and violence to self and others. Although these criteria are of course sensible, they arguably give insufficient attention to an equally important factor, namely, the more passive inability of those in crisis to meaningfully tend to their own basic survival needs.

Further, a significant number of states continue to disallow civil commitment when the cause of the danger the person poses to themselves or others arises out of chronic alcoholism, drug abuse, and other lasting substance abuse disorders, as opposed to “pure” mental illness.

Moving from the general to the specific, under most state laws, including New York’s anachronistically-titled “ Mental Hygiene Law ,” emergency service workers, including police and EMTs, as well as family members and others, can initiate emergency civil commitment to qualifying hospitals. The triggering event, in addition to a finding of mental illness, is that the person poses a “substantial risk of physical harm to himself . . . or others . . . .”

New York’s emergency civil commitment standard, in short, puts the focus on present “dangerousness.” The listed indicators of such dangerousness to self or others, in turn, include “attempts at suicide or serious bodily harm,” “threats,” as well as “homicidal or other violent behavior.”

But in a caring and empathetic society, should we not also expect our government to equally exercise its emergency parens patriae powers to help those posing similarly serious, but more passive, threats to themselves? Put differently, it is challenging to think of a persuasive public policy argument for not considering those in this small but growing vulnerable population as equally dangerous to themselves as a result of their inability to attend to their basic survival needs, including obtaining food, clothing, safety, necessary medical care, or shelter.

Perhaps an even bigger issue is that emergency civil commitment in New York is still statutorily limited to those suffering from a “mental illness.” That term conspicuously does not include chronic substance abuse disorders, including life-threatening chronic alcoholism and drug abuse.

Several states continue to follow New York’s approach, though many other states, for good reason, have begun to reject it. States ranging from Alabama , Arizona , and Arkansas to California , Idaho , Illinois , New Jersey , Maryland , Montana , Utah , and Wyoming also do not mention, or, more often, in fact, explicitly exclude, substance abuse disorders, regardless of severity, from their civil commitment laws. The laws’ legislative histories, often going back many decades, reveal the reason for this approach: state lawmakers once considered substance abuse a “volitional impairment,” whereas they considered mental illness a disability.

To be sure, the problem with considering even chronic substance abuse as something squarely within a person’s control ignores today’s understanding of addiction. In fact, the Rehabilitation Act, Americans with Disabilities Act, and Affordable Care Act all treat diagnosable drug addiction and alcoholism that substantially limit a major life activity as disabilities .

Even from a public policy perspective, it is not clear why legislators would want to draw sharp lines around those threatening serious physical harm to themselves and others due to untreated mental illness, as opposed to (or, as often is the case, in addition to) untreated chronic substance abuse disorders. If a meaningful distinction exists, it is difficult to tease it out based on the dire real-world consequences.

There is little question that our system's frequent failure to ensure appropriate treatment for those suffering from severe mental illness and chronic substance abuse disorders, both prior to and after the emergency circumstances arise, has spawned very public tragedies. Yet, in even more cases, the heartbreaking consequences are largely invisible to the broader society. Too many men and women in need freeze to death on city streets, die in childbirth, commit suicide, are felled by readily treatable illnesses, succumb to traffic accidents, and become victims of violent crimes, including murder.

Of course, and as the U.S. Supreme Court has repeatedly cautioned, civil commitment undeniably imposes a massive curtailment of a person’s individual liberty. Skeptics, particularly in the disability and civil rights advocacy community, for good reason, raise serious concerns about equal protection and due process.

We, as a society, must, therefore, ensure that the provable and urgent harms staved off because of emergency commitment significantly outweigh the losses to personal liberty suffered by the inpatient. In short, every reasonable effort must be taken to minimize harm to the patient.

To this end, we must guarantee safe and effective evidence-based care, appropriately promote the patient’s autonomy, and assure effective professional services in the least restrictive setting (and with the goal of release at the earliest reasonable opportunity). Additionally, any emergency civil commitment must always be accompanied by rigorous due process guarantees, including the right to be heard on whether the commitment elements were met, and continue to be met, by clear and convincing evidence.

In the late 1960s, policymakers, reacting to widespread reports of abuse, began shuttering state psychiatric institutions. In 1967, for example, then-Governor Ronald Reagan signed California’s Lanterman-Petris-Short Act , effectively ending the practice of emergency involuntary civil commitment. But available studies establish that during the intervening decades, the prevalence of mental illness and chronic substance abuse disorders have changed for the worse, even though more enlightened and effective treatment options have emerged.

In fact, even those who have a healthy concern about expanding government power and the attendant financial costs that come with it are likely to be more sympathetic when the goal is to devise a more compassionate, humane, and morally compelling way to potentially save the lives of those who find themselves perilously near life’s edge.

Nobody wants a return to bygone policies of detaining people just for being sick or for other untoward reasons. But it would be a grave mistake not to at least consider whether today’s inconsistent patchwork of state laws restricting what we can do to help those truly in crisis deserves a rethink.

T. Markus Funk (@TMarkusFunk1) is a former federal prosecutor who has taught law at Oxford University (where he received his PhD in law), Northwestern, the University of Chicago, and the University of Colorado, among others.