Crime

The Killer Nurse

She didn’t brag. She didn’t leave clues. She killed her patients, then went home to play computer games.

AFP_HH30L
Elizabeth Wettlaufer, a nurse accused in the murder of eight elderly patients, leaves the courthouse in Woodstock, Ontario, on Oct. 25.

Geoff Robins/AFP/Getty Images

It may be the most Canadian police interrogation ever captured on video. A middle-aged woman in sensible slacks sits quietly with her hands on her thighs and a pink purse on the table beside her. She’s waiting to confess. “Sorry about that,” a detective says when he finally joins her. “Too many people moving and shaking around here, and you can’t really keep track of who’s doing what … ”

“That’s OK,” she says.

The two go on to make some small talk—they chat about the price of Toronto Blue Jays playoff tickets, among other things—and then Bethe Wettlaufer begins to tell her story. “It seems so stupid now,” she says, referring to the eight murders and six attempted murders she committed between 2007 and 2016 while caring for elderly patients at a series of nursing homes. She then clambers to her feet and tries to clear her head. “Sorry, this is pretty major,” she explains. “I’ve only ever had parking tickets.”

Wettlaufer goes on to describe an urge to kill that came on during her nursing shifts like a “red surge” of pressure in her chest, a compulsion that went away only when she’d injected someone with an overdose of insulin. “I thought this was something God, or whoever, wanted me to do,” she tells the detective. Later she adds, “I’m so embarrassed.”

Want to listen to this article out loud? Hear it on Slate Voice.

Wettlaufer, who pleaded guilty to her crimes last month and was sentenced to life in prison, is said to be among Canada’s most prolific serial killers. Yet this dowdy cat lady from Ontario—dubbed the “shadow of death” by the judge who sentenced her—seems about as far away as humanly possible from the classic image of a murderous predator. Wettlaufer did not seduce her victims, nor did she torture them so she could linger over their demise. Instead she’d tell them, “The doctor wants you to have a vitamin shot.” Then, after administering a dose of slow-acting poison, she’d go home to play computer games.

Another sort of serial killer—the more cinematic type—might have left some cryptic clues for the authorities. But Wettlaufer was neither coy nor even particularly careful about her criminal conduct. During the nine-year murder spree, she confessed to her pastor, to a co-worker, to a friend from a support group, and then, finally, to the police. (“Maybe they didn’t believe me,” she says to the detective.) Another sort of serial killer might have preened in jailhouse interviews, haughty and self-serving to the end. Wettlaufer, by contrast, talked about her violent acts like she was an addict in recovery. “I had some anger issues that I’ve had to deal with,” she told a friend.

But while Wettlaufer may not fit the ubermensch-y mold of those who hunt and kill for sport, she is fairly emblematic of another well-established (but much less widely covered) type of killer: the homicidal nurse.

The archetype of the killer caretaker, slaughtering those she’s been entrusted with nurturing back to health, dates back at least a century. In the 1910s, Connecticut nursing-home proprietor Amy Archer-Gilligan was found guilty of poisoning several of her residents, and suspected of having killed a few dozen more, perhaps to hide the fact she’d been stealing their money. Her case would later serve as inspiration for the play and film Arsenic and Old Lace.

The idea that homicidal nurses might compose a special class of criminals, however, has been very slow to form. That’s in part because killer nurses aren’t often caught or put in prison. As scholars of the type have pointed out, hospital-based serial killers don’t need to be that brilliant or conniving to get away with murder. First, they tend to work alone, in close contact with potential victims who may be very old, very young, or otherwise unable to defend themselves. (When serial killers strike in health care settings, about 70 percent of their victims are elderly; in other locales that proportion is more like 5 percent.) Second, they have access to a broad array of deadly drugs and other killing tools. Third, they commit their violence in a milieu where patients die of natural causes all the time and where frequent cleaning means evidence of a crime may quickly be destroyed. Fourth, even when they become suspects, rules of patient privacy can make investigations difficult.

The initial evidence in these cases tends to be statistical and correlative—an unexplained blip in the rate of patient deaths that happens to be clustered on certain days or nursing shifts. (No one noticed such a blip on Wettlaufer’s watch.) Over a six-week stretch in the summer of 1975, for example, the mortality rate jumped at the VA hospital in Ann Arbor, Michigan. An epidemiological investigation of this pattern, later published in the New England Journal of Medicine under the headline “An Epidemic of Mysterious Cardiopulmonary Arrests,” used the location and timing of the deaths to narrow down a list of suspects. Eventually, a pair of female nurses on the evening shift were charged and then found guilty of poisoning five patients.

The legal case against a killer nurse is often weak and based on little more than circumstantial evidence. The nurses in Ann Arbor had their verdicts overturned a year later due to alleged prosecutorial misconduct, and amidst claims that the two Filipina immigrants had been unfairly targeted by a racist investigation. (One important eyewitness, who was hypnotized to enhance his memory, reported that the nurses “had two strikes against them” in his mind because “they ain’t American.”) A 1981 article in the journal Nursing called their trial sexist, too, noting that certain other (male) suspects had been let off the hook, among them doctors at the hospital, a psychiatric patient, and a mysterious “man in a green [scrub]suit.” The case against them, the article said, was not unlike eight other “witch hunts” in which nurses had been overzealously accused of heinous crimes.

The possibility that some nurses might indeed be killers, and that their prosecutions could at times be justified, was first made by Beatrice Crofts Yorker, who in 1988 published an article in the American Journal of Nursing called “Nurses Accused of Murder.” Yorker, who was then an assistant professor of nursing at Georgia State University, summarized nine criminal cases dating back to 1975, including the one in Ann Arbor and three others in which defendant-nurses had confessed. She acknowledged the “‘witch hunt’ atmosphere” that surrounded many of these trials but also took the charges seriously and tried to understand why a nurse might choose to kill. Some were putting patients into cardiac distress because they relished the excitement of an attempted resuscitation, she wrote. Others seemed to think of what they did as a form of euthanasia—more compassionate than evil.

That same year, police began investigating a pair of nurse’s aides at a nursing home in Walker, Michigan, who had conspired to smother five elderly women in their care. Contemporary reports asserted that the killings were meant to build intimacy between the women, who at the time were romantically involved. (According to true-crime writer Lowell Cauffiel, the murders were “an unholy cement in a union of lust and death.”) Like Wettlaufer, the Walker nurses wouldn’t have been discovered if one of them hadn’t confessed. Also like Wettlaufer, one of the nurses was diagnosed with borderline personality disorder and was said to kill because it “relieved tension.” (Wettlaufer would describe “a release of pressure” after committing each murder.)

Soon criminologists started to take killer nurses seriously. In 1995, British forensic chemist Alexander Forrest reviewed about 40 examples of the type and suggested that one or two new cases might be seen each year in the United States. He proposed calling these murders “CASKs,” for carer-associated serial killings, and noted that “the numbers of patients involved are not trivial.” Sure enough, just a few years later, British doctor Harold Shipman was found guilty of murdering 15 of his patients with opiate injections. Investigators guessed he’d killed at least 200 others.

Finally, in 2006, Yorker and Kenneth Kizer published a landmark review of CASKs. They found that 90 health care serial killers had been prosecuted since 1970, of whom 54 had been convicted. These murderers were associated with 2,113 suspicious patient deaths. That death toll suggests health care serial killers take at least 60 lives per year, though the actual count is likely to be much higher given the obstacles to detection and punishment. (By comparison, experts claim serial killers of all types commit at most 150 murders per year in the U.S.)

Women made up about half of all the murderers in Yorker and Kizer’s data set—a much higher proportion than you’d find among “standard” serial killers. Nurses comprised 86 percent and doctors 12 percent of CASKs, which is about what one would expect given these professions’ relative numbers in the workforce. The most common mode of killing was injection, mostly with opiates or insulin. (Suffocation came in second.) And while Yorker and Kizer noted a diverse set of motives, they did observe that very few health care killers have criminal records. The most predictive factor, they said, was a history of falsifying their credentials.

A 2007 book by Katherine Ramsland, Inside the Minds of Healthcare Serial Killers: Why They Kill, tried to give more guidance, laying out 22 “red flags” for hospital administrators. Wettlaufer displays about half of them, including disciplinary problems (she’d been fired from one nursing home for misusing medication), a history of mental instability and depression, an apparent personality disorder, and a track record of moving from one institution to another. It’s easy to see that some of these red flags overlap, and subsequent scholarship from British researchers Elizabeth Yardley and David Wilson suggests certain items on the list—“hangs around during investigations of deaths,” for example—could well be irrelevant. According to the British researchers, even the red flag Yorker and Kizer identified as a leading indicator—“lied about personal information”—isn’t all that useful.

Among the items on Ramsland’s list that may well have some predictive value: “had a substance abuse problem.” Harold Shipman was addicted to heroin. Amy Archer-Gilligan was said to be an addict, too. So was Kimberly Clark Saenz, a nurse at a Texas dialysis center who was seen drawing bleach into syringes and injecting the substance into patients’ IV lines during a four-week killing spree in 2008. And Wettlaufer has a long history of alcoholism and drug abuse. (Among the people to whom she confessed was a friend from Alcoholics Anonymous.)

In the interrogation video, Wettlaufer tells the detective she’d steal doses of the opiate Dilaudid from her patients and use it to calm down. “I was always putting this pressure on myself to be a really good nurse and do everything perfectly,” she says. After taking half a pill, “that pressure was gone.”

The way she talks about Dilaudid—as a way of easing pressure—mirrors how she talks about her murders. The two compulsions seem to blend together in her mind, as if the drugs and the killings were both nasty habits she hoped to kick. Wettlaufer went to rehab twice, first in the fall of 2014, several weeks after killing 75-year-old Arpad Horvath; she didn’t kill again until the fall of 2015, by which point she says she was using drugs again. She returned to rehab in 2016 after having been assigned to work with diabetic children. “I was afraid that I might get that feeling of wanting to give them insulin overdoses,” she tells the detective. Faced with her breaking point—“I panicked; those were just kids”—she quit her nursing job and checked into Toronto’s Centre for Addiction and Mental Health. That’s where she finally made the statements that led to her eventual discovery by the police. “I needed help with whatever this was,” she says. “I didn’t want this to keep going on.”

It makes sense that a health care serial killer might come to understand her criminal behavior as a health care problem—the sort of thing that could be treated, in-patient, by licensed psychiatrists and rehab nurses. But Wettlaufer is not the only serial killer whose murders have been medicalized. Just as criminologists have come to understand the prevalence and danger posed by serial killers in hospitals and nursing homes, so too have they begun to think about—and normalize, in a way—serial-killing as a clinical condition.

Forensic psychiatrist Helen Morrison claims to have invented this idea, which she calls “the addiction model of murder.” In truth, the theory goes back to the late 1980s and early 1990s, when the experiences of addicted gamblers were likened to those of multiple murderers. In more recent years, the addiction model has been overlaid with neuroscience buzzwords. Ramsland cites the involvement of dopamine and the “neural reward system” in the urge to kill. Neuroscientist James Fallon claims serial killers experience withdrawal between each murder stemming from “a buildup of hormones” in their amygdalas.

The image of the killer as a hapless addict—less Hannibal Lecter than Nurse Jackie—takes us one more step away from the classic genius-stalker serial murderer archetype. And it only makes someone like Wettlaufer more disturbing. During her confession, the detective asks about some words she has tattooed across her breast: Hopes & Dreams. “I have hopes for the future and dreams of the future,” she tells him. “Maybe somebody can study me and come up with answers and new medications so this doesn’t happen again. That’s my hopes and dreams.”

Watch the tape and you’ll see the “shadow of death” in all her terrifying banality. She’s not some clever sociopathic mastermind, an evil, charismatic rara avis. She’s just your basic homicidal nurse working through her issues.