Intended for healthcare professionals


Medical manslaughter

BMJ 2013; 347 doi: (Published 18 September 2013) Cite this as: BMJ 2013;347:f5609
  1. Sarah E McDowell, knowledge transfer senior research associate1,
  2. Robin E Ferner, director2
  1. 1University Hospitals Birmingham NHS Foundation Trust, Edgbaston, Birmingham, UK
  2. 2West Midlands Centre for Adverse Drug Reactions, City Hospital, Birmingham B18 7QH, UK
  1. r.e.ferner{at}

More prosecutions won’t ease the problems for lawyers, doctors, or patients

“It would be most fatal to the efficiency of the medical profession if no one could administer medicine without a halter round his neck.” Baron Pollock (1859)1

A doctor who makes a “bad enough” medical error to cause the death of a patient can be prosecuted for criminal negligence manslaughter,2 3 but prosecutions were rare until the 1990s. When we surveyed cases previously, the numbers of doctors charged had increased during the period 1990-2005, although few were convicted.4 Those who were convicted were rarely imprisoned, and if they were, sentences were generally short. Negligent acts, however reckless, that have non-fatal consequences, are not crimes in English law. In France, by contrast, claimants can invoke criminal proceedings for involuntary harm short of death, and not necessarily as a consequence of recklessness.5 Some have suggested that a similar approach be adopted in England.6

Three new factors will alter the relationship between medicine and criminal law in England and Wales. Firstly, the Court of Appeal’s ruling that a sentence of two years’ imprisonment imposed on a surgeon who pleaded guilty to manslaughter was “not manifestly excessive.”7 Secondly, revisions to the law on corporate manslaughter.8 Thirdly, recommendations of the Francis and Berwick reports that reckless or wilful misconduct causing harm short of death should be prosecuted.9 10

The Appeal Court decision recognised that before the implementation of the Criminal Justice Act 2003, “medical gross negligence cases were regarded as requiring levels of punishment at a more modest level than those appropriate to different forms of the offence of manslaughter by gross negligence.”7 However, the new sentencing tariff should, in the court’s view, reflect “the fatal consequences of a criminal act” even when applied to medical manslaughter.

Where practitioners wilfully neglect their patients or are reckless, they deserve punishment.11 However, manslaughter charges can also follow unintentional medical errors of the sort that in one circumstance cause no harm but in a similar circumstance prove fatal.12 For example, giving 10 times too much intravenous digoxin is much more dangerous than giving 10 times too much benzylpenicillin. Focusing on the person who makes a fatal error can harm future patients because it distracts attention from flawed systems that permit errors, many of them non-fatal, to occur.

The Francis report recommends that “where serious harm or death has resulted to a patient as a result of a breach of the fundamental standards, criminal liability should follow.”9 This is, as Berwick’s report implies, precisely the opposite of what is needed for a safer NHS. Prosecution of some doctors is unlikely to encourage others to take greater care. Humans inevitably learn by trial and error, with the result that new tasks lead to more errors than familiar ones.13 Familiar tasks, by contrast, are carried out without much conscious thought, so are susceptible to distraction. The corollary is that punishing more doctors for making errors is unlikely to improve patient safety. Furthermore, broadening the scope of the criminal law threatens to put retributive punishment before the openness needed for systemic improvement,14 because health professionals may hide errors for fear of prosecution.

The number of doctors tried for gross negligence manslaughter, and the number of convictions, seems to have fallen recently. Although the Crown Prosecution Service recognises the unusual status of “medical manslaughter,”15 it still does not keep records of the number of doctors who are investigated, tried, or convicted (personal communication, 2013; Freedom of Information Act request 3831).16 Using media reports and data provided by the General Medical Council, we identified eight trials in England and Wales, and three convictions, in the seven years 2006-12, compared with 23 trials and eight convictions in the preceding seven years.

The fall in prosecutions for medical manslaughter seems unlikely to be a consequence of fewer cases of medical error. Although reporting rates may not reflect error rates,17 the number of reports to the NHS Reporting and Learning System increased from 659 038 in 2006 to 1 258 860 in 2011.18 The fall in prosecutions may reflect the difficulties with the current law on criminal negligence manslaughter. Lawyers complain that the test used to decide is circular: “it is a crime if the jury think it ought to be a crime.”19 The test of recklessness or wilful neglect, pursuing the doctor “who does not care rather than the doctor who tries to care and fails,”12 would help.

If a jury accepts that a negligent act was “bad enough” to be criminal, the prosecution must still prove beyond reasonable doubt that a patient has died as a consequence of the negligence. This may be hard—after all, patients can die even with expert care. There is a further barrier to conviction. Most avoidable harm in medicine is a consequence of the acts and omissions of several people, working in an imperfect clinical world. It is therefore difficult to choose one responsible person to charge with manslaughter if negligence causes death. The current moves towards clinical teams and the “pass the parcel” of emergency department, admissions unit, and specialty ward magnify both the opportunities for error and the diffuseness of responsibility. The Corporate Manslaughter and Corporate Homicide Act 2007, which might have clarified the circumstances in which a hospital trust was criminally responsible for a death,20 seems to have done the reverse.21

We certainly need safer healthcare. Berwick is right to advise that “Recourse to criminal sanctions should be extremely rare, and should function primarily as a deterrent to wilful or reckless neglect or mistreatment.”10 Neither an increase in prosecutions for manslaughter, nor longer sentences for those convicted, will ease the problems for lawyers, doctors, or patients.


Cite this as: BMJ 2013;347:f5609


  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and declare the following interests: REF has received payment for medicolegal work and SEMcD has none.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract