Views & Reviews Review of the week

Lethal practice

BMJ 2007; 335 doi: (Published 19 July 2007) Cite this as: BMJ 2007;335:159
  1. Wendy Moore, freelance writer and author, London
  1. wendymoore{at}

    A new book chronicles insulin's 50 year history as a murder weapon—including stories of doctors engaged in foul play. Wendy Moore reports

    If one fact emerges with abundant clarity from Insulin Murders it is that coauthor Vincent Marks would make the ideal dinner party guest. As a world authority on criminal use of insulin, Marks could spin startling stories of bigamous murderers, serial killers, and bungled miscarriages of justice sufficient to last until well after the last wafer thin mint has been eaten and the port bottle emptied.

    Documenting 50 years of legal cases that have implicated insulin as a murder weapon, Marks and his coauthor, medical journalist Caroline Richmond, have produced a compelling account that is at least as thrilling as any best selling crime novel, with many twists and loose ends to keep the reader guessing.

    Beginning in England in 1957 with the first murder proved to have involved insulin, though technically death was caused by drowning, the book details 14 of the most controversial trials in which insulin has played a determining role. In many—including the famous conviction and later acquittal of Claus von Bulow, wrongly accused of murdering his heiress wife by insulin injection, and the case of British nurse Beverly Allitt, convicted of killing four children in her care—Marks testified as an expert witness.

    Just as the book provides a galloping account of ingenious attempts by fortune hunting husbands and money grabbing wives to pull off the perfect murder, it also charts the development of the laboratory tests that in many instances have foiled their efforts—and in other cases quite possibly led to wrongful convictions or erroneous acquittals. In this parallel journey, doctors and nurses are as often the villains as the heroes, just as likely to be clinically efficient killers as they are to be the medical sleuths who unmask them.

    Kenneth Barlow, the first proved insulin murderer, convicted of killing his wife by injecting her with insulin then leaving her to drown in a bath, was an unemployed nurse. It was an astute forensic pathologist who cast doubt on his wife's seemingly natural death, having discovered a tiny puddle of water in the crook of her arm, discrediting Barlow's story that he had tried to resuscitate her. Subsequent tests with insulin antibodies on tissue taken from the suspected injection sites not only clinched Barlow's conviction but for the first time destroyed the myth that insulin was the route to the perfect—undetected—murder.

    Colin Bouwer, who was found guilty in 2001 of murdering his wife, probably by a succession of prescription drugs including insulin, was a professor of psychiatry at a New Zealand medical school. It was through a trail of emails he had sent to medical experts on hypoglycaemia around the world, purporting to be a forensic psychiatrist investigating a possible insulin death, that his involvement was first suspected.

    But if Insulin Murders is a roller coaster ride to equal any forensic detective drama on television, it is also a cautionary tale of medical mishap, misdiagnosis, and misinterpretation. True life, unfortunately, is rarely as straightforward as fiction. So as the laboratory tests to measure insulin and its criminal misuse have become ever more sophisticated, so evidence of their potential flaws has grown.

    Himself a pioneer of the immunoassay test used to measure insulin in blood—first developed in 1960 and still the lynchpin of criminal insulin investigations—Marks skilfully recounts the progress and the pitfalls. One difficulty is that tests that are both adequate and vital in diagnosing and treating patients on a hospital ward are not always foolproof in determining cause of death, proving murder, or fingering the possible culprit. Postmortem tests for insulin in brain tissue, which helped convict serial wife killer William Archerd, for example, are now discredited. Postmortem urine tests for insulin and C peptide, instrumental in jailing nurse Maria Whiston, are similarly doubted—although both these cases also featured overwhelming circumstantial evidence. But, as conflicting opinions from medical experts in various trials make clear, many results are prone to misinterpretation, inaccuracy, and mix up.

    Most fascinating is the recent case of Deborah Winzar, a nurse convicted in 2000 of murdering her husband by insulin injection, on the basis of a controversial immunoassay test and despite evidence of vomiting—a circumstance unheard of in insulin induced hypoglycaemia. Courageously casting doubt on the test he himself spearheaded, and indeed on the interpretation of his own colleagues who provided the critical result, Marks suggests that Winzar was wrongfully convicted.

    Tellingly, with the benefit of 50 years' research on insulin measurement, Marks concludes that none of the available tests are sufficiently accurate on their own to provide a safe conviction of murder, unless backed by mass spectrometry.

    Since Marks is probably unavailable for dinner party guest turns, his and Richmond's powerful and enlightening book makes a gripping substitute.


    • Insulin Murders: True Life Cases

    • Vincent Marks, Caroline Richmond

    • Royal Society of Medicine Press, £12.95, pp 189

      ISBN 978 1 85315 760 8

    • Rating: ****

    View Abstract

    Sign in

    Log in through your institution