Intended for healthcare professionals

Education And Debate

Monitoring mortality rates in general practice after Shipman

BMJ 2003; 326 doi: (Published 01 February 2003) Cite this as: BMJ 2003;326:274
  1. Richard Baker (, professora,
  2. David R Jones, professor of medical statisticsb,
  3. Peter Goldblatt, chief medical statisticianc
  1. a Clinical Governance and Research Development Unit, Department of General Practice and Primary Care, University of Leicester, Leicester General Hospital, Leicester LE5 4PW
  2. b Department of Epidemiology and Public Health, University of Leicester
  3. c Office for National Statistics, London SW1V 2QQ
  1. Correspondence to: R Baker
  • Accepted 4 October 2002

Harold Shipman's murderous career led to demands that steps be taken to prevent any recurrence, but devising an acceptable and workable method of monitoring mortality rates in individual general practices is not a simple matter

Letters p 280

Soon after the publication of the review of Harold Shipman's clinical practice,1 one of us (RB) went to a meeting for families of possible victims of Shipman. Each of the 100 people present was facing the possibility that at least one member of their family had been murdered by their general practitioner. They wanted the review explained, and to ask questions about how the health service had failed to detect Shipman's murders. One person asked, “How will I be able to trust a doctor again?” Whatever the answer given at the time, the only adequate response must be a collective one from the medical profession and its regulators together. One such response, recommended in the review of Shipman's clinical practice, would be the development of a system to monitor the mortality of general practitioners' patients.

Our aim in this paper is to stimulate debate about monitoring mortality in general practice; an appropriately designed system, as well as detecting illegal behaviour, might help general practitioners to plan improved methods of care.

Summary points

Analysis of excess numbers of deaths among Harold Shipman's patients reached a figure similar to the total determined by the inquiry

Monitoring mortality rates among general practitioners' patients would help maintain public trust

Such a system could detect high mortality at an early stage by using valid local comparative data and information about non-fatal outcomes

Procedures for investigating abnormal patterns need to be agreed

A monitoring system could also provide practices with information to help select clinical policies to reduce mortality

Findings of the review

Firstly, drawing on the review and updated information reported to …

View Full Text

Log in

Log in through your institution


* For online subscription