Over at Slate, Dahlia Lithwick has written a helpful primer on “excited delirium,” the non-medical diagnosis that is often used to clear police officers for in-custody deaths.

“Excited delirium” is the name given to a condition in which a person, either as a result of mental illness or protracted use of stimulants such as cocaine or methamphetamines, becomes extremely violent; hyperaggressive; and is often found naked, agitated, incoherent, feverish, and displaying extraordinary strength. The phenomenon is reported most often in police encounters, requiring, on average, four officers to restrain the suspect. In approximately 10 percent of cases, according to the literature, the person with excited delirium may die suddenly. The heart or breathing simply stops. So when someone dies in that agitated state and no other cause of death is found, the medical finding is that excited delirium was the cause. It accounts for approximately 250 deaths in the United States each year, with one expert speculating that about 800 cases occur each year nationwide.

The obvious problem is this: What do we make of a syndrome that seems to occur almost unerringly when a police officer is choking, hog-tying, or stunning with a Taser someone with a mental illness or drug addiction? And why do many experts dispute that the diagnosis even exists? While excited delirium is used to explain a significant number of deaths occurring in police custody, the term has not been recognized as a genuine mental health condition by the American Medical Association, the American Psychological Association, or the World Health Organization. Excited delirium—which sounds, to the naked ear, something like “crazy-craziness”—is not found in the current version of the Diagnostic and Statistical Manual of Mental Disorders, either. Yet medical examiners and police departments keep claiming it as the cause of death of people in custody. In 2014, the International Association of Chiefs of Police issued a white paper that tried to bridge the gap, concluding, “Despite what it is called or whether it has been formally recognized, it is a real clinical concern for both law enforcement and the medical communities.” But is it a real medical phenomenon? Is it a convenient way to blame the victim, as civil liberties and prison reformers claim? Or is this a genuine syndrome that occurs largely in fights between the mentally ill and the cops? Now more than ever, when suspicious deaths in police custody are making headlines, should we consider excited delirium an illness or a cover for police abuse?

The spread of the diagnosis among medical examiners is, conveniently enough, largely due to the tireless work of stun gun manufacturer Taser International.

In 2013, Amnesty International, the only organization that has compiled data on this issue, claimed there were 552 incidents since 2001 where a Taser was used on a victim who then died, yet autopsies cited Taser use as a cause or contributing cause of death in only 60 of those cases.. Excited delirium diagnoses have also protected Taser from liability in many use-of-force suits against the company. In a 2007 interview with NPR, Taser International spokesman Steve Tuttle said that each year his company “sends hundreds of pamphlets to medical examiners explaining how to detect excited delirium. Taser also holds seminars across the country, which hundreds of law-enforcement officials attend.” . . .

A piece in Mother Jones details the extent to which ER doctors and medical examiners get most of their information about excited delirium from Taser. . . .

At a Canadian public inquiry set up in 2008 to study the appropriateness of allowing cops to use Tasers, Mike Webster, a police psychologist, went further. He blamed Taser International for “brainwashing” cops and testified that “police and medical examiners are using the term [excited delirium] as a convenient excuse for what could be excessive use of force or inappropriate control techniques during an arrest.” He went on to add that members of the law enforcement community “have created a virtual world replete with avatars that wander about with the potential to manifest a horrific condition characterized by profuse sweating, superhuman strength, and a penchant for smashing glass that appeals to well-meaning but psychologically unsophisticated police personnel.”

Lithwick also points out that determining that a death was caused by excited delirium isn’t a diagnosis so much as a best guess due to an inability to find an alternate cause of death.

Appropriately enough, this all evokes an National Institute of Justice study from several years ago on the safety of Tasers. The study was widely reported to have concluded that Tasers are safe, and that the “non-lethal” description often attached to them is accurate. But note this line from the study:

The panel determined that there is no conclusive medical evidence in the current body of research literature that indicates a high risk of serious injury or death to humans from the direct or indirect cardiovascular or metabolic effects of short-term CED exposure in healthy, non-stressed, non-intoxicated persons.

The problem here ought to be obvious. The whole point of using a stun gun is to subdue someone who presents an immediate threat. And in fact, we should hope that most people on whom stun guns are used are under a significant amount of stress — if the police are regularly using stun guns on people who aren’t experiencing a high degree of stress, then we have a whole different problem. We can also assume that a fairly high percentage of people police feel compelled to shoot with a stun gun will also be intoxicated. To say that a device is perfectly safe when test on a population on whom the device is extremely unlikely to be used in the real world tells us nothing at all.

As for the health of those hit by stun guns, there’s a concept in tort law called the “eggshell skull rule,” or “you take your victim as you find him.” The gist of the rule is that if your victim has a medical condition that causes him to suffer unusual harm due to your actions, you’re still liable for the additional injury, even though you couldn’t have known about the medical condition. If we were to apply it here, we could say that if a police tactic is likely to kill someone with a heart condition, mental illness or some other medical problem, we can’t simply dismiss those cases because we have studies showing that those same tactics do minimal damage to healthy people.

Even these risks could possibly be justified if stun guns were used as a substitute for lethal force, as they were once intended. Police officers rarely need to use lethal force and are supposed to use it only when a suspect presents a threat to the officer or others. If that were how stun guns were used, we’d be talking about limited use on a very small percentage of the population, only in extreme circumstances, and only to save lives. But stun guns today aren’t used as a substitute for lethal force. They’re used to force compliance. As Lauren Kirchner recently wrote in the Pacific Standard, police today “taser the homeless to get them to leave an area, they taser the mentally ill who aren’t understanding their instructions, and they even once tasered a six-year-old who was having a tantrum in kindergarten.” (And that’s far from the only example of police Tasering a child.) Taser International itself recently told the Miami New Times that its product alone is used by police more than 900 times per day.

With those kinds of numbers, we’re now talking about a significantly higher risk of Tasering someone with a condition that could result in death. And those people are being put at risk of death not to save lives but because some police officers get impatient, angry or annoyed.

All of which is why the debate over the medical legitimacy of “excited delirium” is really beside the point. Let’s assume for a moment that it’s a real condition. If so, it’s a real condition that only manifests and kills when the police are subduing or Tasing someone. Its only symptoms seem to be (a) the presence of police officers and (b) the absence of any other clear cause of death. The fact that hundreds of people have died with these symptoms suggests that a not insignificant portion of the population is susceptible to death if they’re subjected to high levels of stress while under restraint or hit with a stun gun. That alone should guide use-of-force policies when it comes to things like Tasing people in handcuffs, the use of restraint chairs, and the amount of weight officers put on suspects while they’re cuffed and on the ground.

Giving these deaths a name and a diagnosis doesn’t remove our obligation to do what we can to prevent them.