Search
Program Calendar
Browse By Day
Search Tips
Personal Schedule
Sign In
Since the 1990s, identity categories such as ethnicity have been widely used to explore and understand patterns of health and disease. However, historical and localised perspectives show that the ways categories are constructed and used have varied considerably over time and across different locations. This variation is partly due to differences in data availability. Epidemiologists rely heavily on access to data, and Sweden (along with the other Nordic countries) is notable as an epidemiological goldmine. The availability of extensive, digitised population records and health data collections, combined with the ability to link these to individuals via the personal identity number (introduced in 1947), has laid the foundation for a scientific success story. Swedish population registers appear to be natural resources primed for exploitation, thereby contributing to what Bauer (2014) describes as a distinctive Scandinavian environment for epidemiological research.
The metaphor of the goldmine suggests a scientist extracting undisputed facts, rather than engaging in epistemological and practical work. Nevertheless, the methods and concepts chosen by epidemiologists have consequences not only for the classifications and those classified but also for the knowledge derived from those categories. The Swedish case is particularly intriguing because Swedish population statistics do not include information about ethnicity. The 1945 census was the last to record ethnicity in Sweden, and the 1990 law on data security effectively made ethnic registration illegal. Epidemiologists working with Swedish data are, in essence, forced to attempt to classify the unclassifiable when exploring potential links between ethnicity and disease. Focusing on the development of epidemiological knowledge based on Swedish data from 1950 to 2020, this paper examines the scientific practices and categories involved in these processes.