Attack of the Killer Nurses: A look at a curious phenomenon - nurses who kill their patients

National Review, May 28, 2001 by Theordore Dalrymple

In August 1975, three patients suffered cardiopulmonary arrests within 15 minutes of each other at the Veterans Administration Hospital in Ann Arbor, Michigan. An astute anesthesiologist noticed that all three patients looked as if they had been administered a muscle-paralyzing agent-which, as it turned out after much investigation, they had.

Rarely can the commencement of a tradition be placed so precisely in time-the tradition in question being that of nurses murdering their patients wholesale. Of course, it is quite possible that an underground tradition of this nature existed undetected even before the events at Ann Arbor, but from that time forth there have been epidemics, both large and small, of unexpected death at regular intervals wherever murderous nurses have taken employment. On March 26 of this year, for example, the nurse Kristen Gilbert, somewhat unimaginatively dubbed the "Angel of Death," was sentenced to life imprisonment for having fatally injected three of her patients with a stimulant to mimic heart attacks. And in November 1999, a nurse named Orville Lynn Majors was sentenced to 360 years' imprisonment for having killed six of his patients by injection of potassium chloride.

There have been outbreaks of nursing murder in many different hospitals in the last quarter of a century. They are often surprisingly difficult to detect. Indeed, many may have gone completely undetected. During the 1980s, The New England Journal of Medicine and The Journal of the American Medical Association published several articles about epidemics of unexpected cardiopulmonary arrest in hospital intensive-care units, usually coming to the cryptic conclusion that patients were 47.5 times as likely to have had a cardiopulmonary arrest while Nurse 14 (who could not be named for obvious legal reasons) was on duty as when any other nurse was on duty.

One of the most notorious outbreaks occurred in Toronto in 1980 and '81, in the cardiology ward of the famous Hospital for Sick Children. The death rate suddenly increased by four times in Ward 4A. High levels of digoxin were found in the blood of four of the deceased children, and it was estimated that, during the epidemic of sudden and unexpected death, the children in Ward 4A were 64.5 times as likely to die when Nurse A (who was a woman named Susan Nelles) was on duty as when any other nurse was on duty. She was arrested but released 45 days later for lack of evidence. The police, however, sought no other suspect.

In fact, Nurse A may have been innocent, and the deaths may have been no murder. The high digoxin level found after the deaths of the children may have been an artifact of the methods used to detect the drug. In those days, the rubber in syringes and other medical apparatus was manufactured using a potentially toxic substance that could have killed the children, either directly or by allergic reaction to it. To this day, therefore, the deaths at the Hospital for Sick Children have not been satisfactorily explained, and these doubts (which are similar to those that accompany several other such epidemics in hospitals) have been used to suggest that the very concept of mass murder by nurses is a reversion to the medieval witch mania or to the Salem trials of the 1690s. American courts have ruled that epidemiological evidence-that an excess of deaths in a hospital ward when a particular nurse was on duty-is not sufficient to convict, even when there is also evidence that a fatal substance was administered.

But there have been so many cases of undoubted murder of their patients by nurses (and hospital orderlies) that popular hysteria or mass psychosis in the public and the media is not a plausible explanation of the phenomenon. The question, then, is why should this pattern of behavior have emerged in the last quarter of the 20th century?

Do the nurses who behave like this have a deep-seated personality trait or a motive in common? It is natural to assume that they must have characteristics that distinguish them from others of their profession (after all, mass murder by nurses is still very uncommon) that would enable them to be identified in advance, and therefore tragedy to be averted.

This assumption leads to official responses that are like those of generals, who are always inclined to fight the current war as if it were the last war. For example, not long ago I was approached by a senior nurse in my hospital who was worried about the conduct of one of our nurses. She seemed in several ways remarkably like Beverly Allitt, a nurse in Mrs. Thatcher's home town of Grantham, who was given 13 life sentences for murdering three children, attempting to murder three others, and causing grievous bodily harm to six more.

In retrospect, Allitt was clearly a young woman with peculiar traits. Among them were both bulimia nervosa (binge eating followed by vomiting) and Munchausen's syndrome: the habit of presenting oneself to doctors and hospitals with dramatic symptoms of illness that are entirely spurious (it is characteristic of our age that repeatedly lying about illness should be considered itself an illness).