A 1993 survey of residents and family physicians found that many felt uncomfortable about the process of death certification. More than half the 131 practising family physicians who completed the survey acknowledged a need for more training in this area.6
The results of our study confirmed that there was a need and a desire for more training in the process of death certification.
The effect of experience on physicians’ ability to complete death certificates accurately is controversial. Both Maudsley and Williams6
and Messite and Stellman11
reported no difference in accuracy in completing mock certificates based on case scenarios between physicians who had completed many certificates and those who had not. Two more recent studies found that more experienced physicians completed death certificates more accurately.12,13
In our study, physicians with more than 10 years’ experience with certification of death scored substantially better on the pretest. However, more recent graduates who attended our workshop had greater improvement on their posttest scores compared with more experienced physicians; 3 months later, there was no significant difference in scores of physicians experienced and not experienced in completing death certificates.
Results of our pretest confirmed that physicians did not know enough about how to apply the Coroners Act. Misconceptions included the belief that death certificates could not be amended once completed, and there was uncertainty around definitions of natural and non-natural deaths. Participants frequently made errors when completing mock certificates based on case scenarios. Use of mechanisms (eg, cardiac arrest) instead of specific underlying causes of death was a frequent error, as was choosing conditions that did not have specific diagnoses.
Many investigators have documented the frequency and types of errors made by physicians when completing death certificates, but very few studies have evaluated strategies to enhance the knowledge and skills required for completing death certificates accurately. A randomized controlled trial of first-year medical students evaluated the effects of an educational video on students’ ability to certify death. The intervention group showed a small but significant improvement over the control group in overall knowledge and skills.8
The effect of such an intervention so early in medical training on eventual skill in practice is unclear.
Weeramanthri and Beresford9
sent written educational materials on certifying death as part of a questionnaire to house officers at a teaching hospital in Australia and compared rates of errors on death certificates 1 month before and 1 month after the intervention. Although the rate of errors dropped, the change was not statistically significant. Written educational materials and guidelines have been shown to be inadequate for bringing about sustained changes in physicians’ practice, and this might have contributed to the negative results of this study.14
Our workshop for family physicians was an expanded version of a 90-minute workshop for residents. In addition to receiving written guidelines, participants had the opportunity to interact with opinion leaders in the area of death certification and to practise completing death certificates. We believe this has advantages over use of educational materials alone. Results of our evaluation of both residents and family physicians showed that their knowledge and skills were greatly enhanced by the workshop.1