Adjusting proportional mortality ratios for the influence of extraneous causes of death.

Proportional mortality ratios (PMRs) are frequently used where denominators of the population at risk are not available or are known to be seriously biased. The interpretation of proportional measures requires care and can be complicated by the fact that a PMR of a cause of interest may be affected by disproportionate mortality from one or more other causes of death. This problem is examined using occupational mortality data for England and Wales. A method of allowing for the influence of other causes of death on a PMR of interest is proposed and it is argued that this approach is a useful aid to the general interpretation of PMRs.