Chapter 8
Auditing cause of death determinations

Options for additional oversight in determining cause of death

Two doctors

8.4We considered whether two doctors, rather than one, should be required to certify the cause of death. While the second doctor could, in theory, be a useful check on the accuracy of the first, this option would likely create further delay, would divert limited medical resources and may not provide many gains in the accuracy of the cause of death. Funeral directors already tell us of their frustration in obtaining the MCCD from the certifying doctor so that they can begin funeral preparations. That frustration is likely to increase if two doctors are required to certify, particularly in rural areas. We also doubt whether the second doctor would be able to provide meaningful oversight of the first doctor if he or she is not previously familiar with the deceased’s medical history.

Review committeesTop

8.5As we described earlier, some hospitals conduct their own systematic reviews of MCCDs. Those reviews have been successful at detecting errors in both cause of death determinations and referrals of deaths to the coroner. They have a range of mechanisms for using the lessons learned from the reviews to upskill certifying doctors, some more effective than others.

8.6There are two limitations of such systems. First, the feedback loop is limited to lessons learned from experiences within the hospital itself. There is no capacity to learn from the experiences of other hospitals nationally.

8.7Second, there is no equivalent system for the review of deaths in the community. In theory, it would be possible to require all deaths in the community to be reviewed by similar committees made up of general practitioners. However, that is likely to require large resources of time and money.

Improving the medical referee systemTop

8.8In Chapter 3 we described the problems with the current system of medical referees. We have examined various ways in which that system could be amended to provide greater checks on the accuracy of post-death documentation. One of the best features of the current system is that it is local. This means that the medical referee can respond quickly after a death. It also means that the medical referee is more likely to know the local doctors and can follow up easily with the doctor if an error is detected. However, it may also mean that the medical referee is less objective in his or her assessments.

8.9If medical referees were to continue to be employed by local crematoria, there should be formal systems of training and support implemented from a centralised and independent body. The aim of that training and support would be to standardise levels of practice in reviewing cremation documentation. Also, there is an increasing trend for medical notes to be stored electronically. If that trend enabled medical referees to compare the cause of death determination to the medical notes of the deceased person, significant gains in the safeguards provided by this system could be made.

8.10However, we do not consider that improving the medical referee system in this way would provide adequate safeguards because three significant problems remain:


8.11We consider that all of these options have significant weaknesses and would fail to deliver a robust system of scrutiny and safeguards over the quality and accuracy of cause of death determinations. Instead, we propose that a national system of random audits of cause of death determinations should be introduced as we describe in more detail below.