Search tips
Search criteria 


Logo of jrsocmedLink to Publisher's site
J R Soc Med. 2008 January; 101(1): 19–21.
PMCID: PMC2235920

Kenneth Barlow: the first documented case of murder by insulin

On the morning of 4 May 1957, just after 2 am, Dr David Price, a forensic pathologist, was called to the home of Elizabeth and Kenneth Barlow in a residential suburb of Bradford, Yorkshire. The story was that Kenneth Barlow had discovered his wife unconscious in the bath at about 11.20 pm the previous night and called his own doctor, who diagnosed her as dead. Kenneth, a 38-year-old state registered nurse, was unemployed. He had married Elizabeth 11 months earlier and was, to all outward appearances, living happily with her and his 10-year-old son by his first wife.

According to Kenneth, Elizabeth had had tea at about 5 pm on the day of her death. Shortly afterwards she announced that she was tired, and went to bed. When Kenneth came to bed at about 9.30 pm he found that Elizabeth had vomited on the bed. Together they changed the sheets. She put on some pyjamas but took them off because she said she felt too warm and decided to take a bath. Kenneth lay on the bed and went to sleep at about 9.45 pm to the sound of the bath running. When he woke up at around 11.20 pm Elizabeth had not returned to bed. When he went into the bathroom he found her submerged beneath the water. He tried to lift her out but did not have the strength to do so. Nevertheless, he held her head above the water until all the water had run out of the tub. He said he then tried artificial respiration by ‘pressing on her abdomen’ as he was unable to lift her from the bath. Only after this was unsuccessful did he run next door to his neighbours, who had a telephone, and ask them to call a doctor.

The family doctor arrived 10 minutes later and found Elizabeth in the empty bath in a position simulating natural sleep. He did not touch her beyond ascertaining that she was dead. With such an unexpected death he felt it necessary to call the police, who in turn called Dr Price, who was on-call as Home Office forensic pathologist. His main job was as consultant pathologist at the nearby Beckett Hospital in Barnsley.

Dr Price suspected from the beginning that this was not a natural death for two reasons. First, death from drowning in a domestic bath in a previously perfectly healthy 32-year-old woman is rare. Second, but even more telling, was the 110 mL (small cupful) of water that remained in the cavity where the crook of Elizabeth's arm abutted the side of the bath. This made Kenneth's story that he had tried to resuscitate her difficult to accept, and consequently his account of the events suspicious. Meanwhile the police had made a thorough search of the house and uncovered nothing very much except two vomit-stained pillow cases in the bathroom wash basin, a set of sweat-drenched pyjamas in the bedroom and a couple of used syringes in the kitchen. Because of the latter, the police searched the house for vials of insulin or other injectable medications but found none.

Dr Price began a post-mortem examination in the local mortuary at 5.45 am, only 3½ hours after he first saw Elizabeth's body and 6 hours after she had died. He noted that her pupils were widely dilated and that there was bloodstained froth in her nose, mouth and throat. Samples of her lungs, when examined under the microscope, were bulky, congested and wet. They also revealed fluid retention and small haemorrhages, confirming the initial diagnosis of death by drowning. Apart from this, he found no abnormalities, but did observe that Elizabeth was eight weeks pregnant. In addition, Dr Price took the precaution of collecting blood from a number of different sites in the body as well as some urine from the bladder to send to the poisons laboratory, just in case she had been poisoned. The samples were examined by Dr Alan Curry of the North-Eastern Forensic Science Laboratory, who went on to become one of the most distinguished directors of the UK's nationwide forensic science service.

Dr Curry found none of the common poisons or abortion-causing substances in any of the samples he examined. Nevertheless, Dr Price and his senior police colleagues remained convinced that Elizabeth had been rendered unconscious before she drowned, and they considered the possibility that she had been injected with insulin. This would explain her excessive sweating and dilated pupils before death.death.

Figure 1
Kenneth Barlow

Four days later, on 8 May, the decision was made to re-examine Elizabeth's body more thoroughly and under bright light. On this occasion, with the benefit of a magnifying glass, two hypodermic injection sites were identified in each buttock. Dr Price removed these, with their surrounding tissues, and stored them in a refrigerator until he could find a scientist with the expertise and facilities to undertake an insulin test. The methods available at the time were comparatively crude by today's standards and could only be performed by a handful of specialist laboratories. They relied on finding the dose that caused hypoglycaemic convulsions in mice, and comparing the sample with standardized samples containing known amounts of insulin. It was the method used at that time to measure the strength of pharmaceutical insulin extracted from animal pancreases before releasing it for use by patients.

Dr MR Gurd of the research laboratories of the Boots drug company, which was one of three British manufacturers of insulin at the time, undertook to perform the test. Dr Gurd did not, however, hold out much hope of success. He thought the technique might not be sensitive enough to detect the small amount of insulin that would be found in the tissues of someone who had received an insulin injection. However, Dr Gurd did find easily measurable quantities of insulin in extracts of the tissues taken from Elizabeth's buttocks. For comparison, Dr Price had removed tissues from other corpses, which Dr Gurd treated in the same way. As he expected, he found no trace of insulin in them. The findings from Elizabeth were thus clearly abnormal and therefore highly suspicious.

On 5 July, two months after Elizabeth's death, Dr Gurd reported that he had been able to recover from three separate samples of Elizabeth's buttocks a total of about 84 units of insulin, enough to keep two insulin-dependent diabetic patients going for a whole day. Elizabeth did not have diabetes, nor had she been prescribed insulin.

On the basis of this evidence, Kenneth Barlow was confronted by the police on 26 July, where-upon he admitted injecting Elizabeth, but not with insulin. He said he had, with her permission and collaboration, injected her with ergometrine, a drug used legitimately in obstetrics at the conclusion of a delivery, and by lay people—when they could get hold of it—to try to induce an abortion, and which was clearly illegal. He was unaware that this possibility had already been considered—and ruled out—by the toxicological examination conducted by Dr Curry on Elizabeth's body immediately after her death and, subsequently, by examination of the needles and syringes found in the kitchen.

Between the time that Kenneth Barlow was charged with murdering Elizabeth and his case coming to court, a new, much more sensitive and precise method for measuring insulin in body fluids had become available. This involved measuring the uptake of radioactive glucose by muscle tissue (rat diaphragm) incubated at body temperature for several hours in a flask containing a sample of the patient's blood serum diluted in a special type of saline. The rate of glucose uptake by the muscle is directly proportional to the amount of insulin present in the incubating fluid. The only exponent of this technique in Britain at the time was Dr Peter Wright of Guy's Hospital, London, with whom I was collaborating in the investigation of non-diabetic patients with hypoglycaemia and neurological symptoms.

Dr Curry asked Dr Wright to examine the extracts made from Elizabeth's buttocks. His results confirmed those of Dr Gurd. To ensure that the methods he was using measured insulin and not just something like it, Dr Wright used four different techniques to improve the specificity. One of these techniques used a guinea pig anti-insulin antiserum. The effect of this was to neutralize any insulin present in the extract by binding to it and making it unavailable to the rat diaphragm. In Elizabeth Barlow's case it abolished the insulin effect produced by the untreated extracts. This use of insulin antibodies foreshadowed their use in the immunoassay first employed in the Archerd case,1 and indeed in all subsequent cases of murder by insulin that have come to trial. No longer would it be possible to sustain the myth that insulin is the perfect weapon because it cannot be detected after death.

Among the other substances tested for by Dr Curry were phenolic preservatives used in commercial insulin preparations. None were found, but he explained this by their rapid disappearance from injection sites—unlike insulin, which remains there acting as a depot for up to 24 hours, depending on the type of insulin used. He also measured the glucose content of the blood Dr Price had removed from Elizabeth's heart, as it would surely be low if she had died of an overdose of insulin. Imagine his surprise, therefore, when instead of finding a low blood glucose concentration, he found it to be abnormally high. To someone less expert than Dr Curry this might have been the end of the matter. He knew, however, that scientists had previously observed a rise in the concentration of glucose in blood that was collected from the right side of the heart of people who had died a violent death, but that it was usually low in blood collected from leg veins or from the left side of the heart. With the assistance of various forensic pathologists who collected blood from victims of violent deaths for him to analyse, Dr Curry confirmed the earlier discovery that blood glucose concentrations in blood from the right side of the heart do not reflect that in the rest of the body after death. As a result the investigating team were not put off pursuing an otherwise very strong case by what, on the face of it, seemed an insuperable obstacle.

With Dr Gurd's results to go on there seemed so little doubt that Elizabeth had been injected with insulin that the prosecution proceeded to trial even before Dr Wright's results became available. The prosecution recognized that some people deliberately inject themselves with insulin for a variety of reasons, including suicide, but as it is difficult for anyone to inject themselves in their buttocks, it was obvious to them that someone else had done it and that that person was Kenneth.

Before the trial it emerged that earlier in the year Kenneth had told a fellow employee at the hospital where he was working about an accident his wife had suffered the previous September. On that occasion, he claimed, he had found her collapsed in a hot bath and rescued her by removing her from it. He had also, it transpired, boasted to fellow workers over a period of two years that it would be easy to kill someone with insulin since it was undetectable in the body after death. There appeared to be no motive for killing his wife apart from the rather tenuous one that he did not want her to have a child.

Kenneth vehemently denied the charge of murder but at his trial at Leeds Assizes in December 1957 could not explain the insulin found in Elizabeth's body apart from suggesting that she had administered it herself. The absence of insulin vials and of any insulin in the two syringes that had been found, together with the improbability of self-administration into the sites where the insulin was found, all militated against this possibility.

The jury found Kenneth Barlow guilty, on the evidence, and Mr Justice Diplock sentenced him to life imprisonment on 13 December 1957. He was released from prison 26 years later, in 1984, still maintaining his innocence.

Although usually described as the first case of murder by insulin to lead to conviction, Elizabeth Barlow did not in fact die from an insulin overdose, although it played a crucial role in her death. The amount injected into her was sufficient to render her unconscious, and whether she would or would not have died had Kenneth left her long enough will never be known. It is probable that he had expected her to die more quickly than she did and so he made the decision to drown her—which he did. Had he left her in her bed, she might well have been dead in the morning, or at least have suffered irreversible brain damage, and all of the insulin he had injected would have been absorbed into her bloodstream and destroyed. There would have been no ‘smoking gun’ for the pathologists and toxicologists to find, and there would have been no case against him.



Competing interests None declared

Funding None

Ethical approval Not applicable

Guarantor VM

Contributorship VM and CR collaborated in writing Insulin Murders

Acknowledgements None

This is the first in a series of articles adapted from the book Insulin Murders, by Vincent Marks and Caroline Richmond (ISBN 13: 978-1-85315-760-8). The book is available from the RSM Press website at


1. Marks V, Richmond C. William Archerd, USA: a serial matrimonialist killer. In: Insulin Murders. London: RSM Press, 2007, 9-18

Articles from Journal of the Royal Society of Medicine are provided here courtesy of Royal Society of Medicine Press