Surnames and Survival: Mortality Differences across the Life-Course in Sweden

Elien Dalman , Stockholm University
Martin Hällsten, Stockholm University
Martin Kolk, Stockholm University

The socioeconomic gradient in mortality and health is widespread across societies today – existing regardless of societies’ level of public provision of healthcare. Previous research suggests that such a gradient in men’s adult mortality was non-existent or even reversed historically. This puts into question typical structural explanations of the health disadvantage of lower socioeconomic groups. Historical socioeconomic status is typically measured using occupations. However, historical occupations are often generic (laborer, farmer) and miss important social inequalities. We use the Swedish Death Index, covering all deaths in Sweden between 1815 and 2022, and a historical status measure based on surname types. Distinguished surnames reflect pre-industrial status groups and national minorities. Different potential mechanisms behind the mortality gradient likely weigh differently for surname types as for occupational classes. With this alternative measure and high-quality full population death records, we find a sizeable positive gradient in infant, childhood, and mid-life mortality for cohorts born 1906-1910. We find a gradient in adult mortality for women born 1866-1870 - while the gradient is reversed for men in this cohort, as in previous occupation-based studies. Geography explains some, but not all, of the gradient. These findings contribute to the literature in three ways: they are the first to use full-population data (no record linkage) to study historical mortality gradients, they are the first non-local estimate of the historical gradient in infant mortality, and they use a new status measure.

See extended abstract

 Presented in Session P5. Health, Mortality, and Ageing 1