The Historical Origins of the Mortality Gradient : Socioeconomic Inequalities in Adult Mortality over Two Centuries in Sweden

Mortality differentials by socioeconomic status (SES) are among the most pervasive facts of contemporary demography. However, while the mortality gradient by income, class and education is well-established for the period after 1970, evidence regarding the origins of the gradient is still scarce. The aim of this thesis is to explore the development of SES differences in all-cause and cause-specific adult and old age mortality over the last 200 years, exploiting unique longitudinal individual-level data for a regional population in the south of Sweden, as well as full-count decennialmicrocensuses for the whole country. It was confirmed throughout all the four studies that the socioeconomic gradient in mortality is a recent phenomenon starting after the second world war. This result was independent from the dimension of socioeconomic status used in the analysis. A similar late emergence of social differences in mortality dating back to not earlier than the 1950s is evident regardless of whether the analysis was based on social class or income. Even when examining more detailed occupations, more prestigious jobs such as architects, engineers, physicians, and lawyers were not associated with lower mortality before the second half of the twentieth century. Analyzing more specific groups of diseases showed that the advantages related to one’s higher social class or to one’s higher income appeared at approximately the same time and did so regardless of preventability. Interestingly, when looking at mortality from circulatory diseases for men both by social class and income during the nineteenth and first half of the twentieth century a reversed gradient emerged. Moreover, empirical models including both social class and income showed that they are both independently related to mortality and that the income gradient appeared at an earlier stage, around the 1950s, while when looking at social class it emerged a couple of decades later. When the relation between social class and mortality is broken down in more detailed occupations, results showed significant differences in mortality by occupation within the same social class. Eventually by analysing the different role of cohort and period effects on the relation between social class and mortality it emerged that cohort factors may have had a greater contribution to explaining mortality patterns by social class. Several mechanisms were considered as possible explanations for such patters, including early life factors, material resources and lifestyle. Taken together the results point toward the importance of lifestyle factors. Such mechanism is consistent with the reverse gradient in circulatory diseases before the 1950s and the turn-around that happened later. Indeed, higher social classes were more likely to be heavy consumers of alcohol, to smoke tobacco and to have a sedentary life. In more recent years, the opposite is true. Moreover, unhealthy behaviors were more common among men, which is a potential explanation for why the reverse gradient is not present for women. Overall, the studies presented here looking at long-term developments of the SES-mortality relation reveal that the impact of SES on survival chances has not always been the same but it rather depends on which coping mechanisms each SES group exploits to avoid risk factors at each point in time