Background To explore time trends in deaths attributable to work in England and Wales, and identify priorities for prevention, we conducted a proportional analysis of mortality by occupation over a 22-year period.

Methods Analysis was based on deaths in men aged 20–74 years during 1979–1980 and 1982–2000 with a recorded occupation. Proportional mortality ratios, standardised for age and social class, were calculated for pre-specified combinations of occupation and cause of death, for which excess mortality could reasonably be attributed to work. Differences between observed and expected numbers of deaths by cause and occupation were expressed as annual excess death rates.

Results Mortality attributable to work declined substantially over the period of study, with total excess death rates of 733.2 per year during 1979–1990 and 471.7 per year during 1991–2000. The largest contributing hazards were chronic obstructive pulmonary disease and pneumoconiosis in coal miners, pleural cancer from asbestos, and motor vehicle accidents in lorry drivers. In contrast to most other hazards, there was no clear decline in excess mortality attributable to asbestos, or in deaths from sino-nasal cancer associated with exposure to wood dust.

Conclusions The overall decline in mortality attributable to work is likely to reflect reduced employment in more hazardous occupations, as well as improvements in working conditions. It is imperative to ensure that occupational exposures to asbestos and wood dust are now adequately controlled. Further research is needed on accidents involving lorries with the aim of developing more effective strategies for the prevention of injury.