In Spain, the study of socioeconomic differences in mortality has been limited by the fact that death certificates often do not include complete information on occupation.

In this study, we chose those geographic areas with the highest quality information on occupation of the deceased in order to study socioeconomic differences in mortality from various causes of death.

We used information from the death certificates ofmales who died between 30 and 64 years of age in eight Spanish provinces to compare mortality from the leading causes of death in professionals and managers (group I) and in manual laborers (group II) in 1980-82 and 1988-90.

In each period the standardized mortality ratios (SMRs) were higher in group II, except for ischaemic heart disease during the first period, and cancer of the colon and rectum in both, although in the latter case the differences were not statistically significant.

The relation between socioeconomic level and mortality for ischaemic heart disease was reversed, a phenomenon similar to that which took place in the 1960s and l970s in the developed countries.

These socioeconomic differences in mortality are probably underestimates since the use of the economically active population excludes those persons belonging to population groups with higher mortality.

The magnitude of this effect is greater in the second period, therefore the observed increase in the difference is also an underestimate.