Inequalities at the Brussels Graveyard. Residential and Socio-Economic Mortality Differences in the Belgian Capital, 1910

Patrick Deboosere, VUB
Tina Van Rossem, VUB
Isabelle Devos, University of Ghent

At the turn of the twentieth century, life expectancy in Brussels amounted to only 41 years, while the national average was approximately 10 years higher. In other large cities in Belgium, the level was also 5 to 7 years higher than in Brussels. In previous research, we suggested that the most important determinants of the Brussels health penalty were unhygienic working conditions in cottage work and small-scale companies, together with the high levels of crowding and the absence of pure drinking water in many neighbourhoods. As these factors suggest large mortality variations within the city of Brussels, an individual level analysis is required. In this article, we examine whether residence and social class influenced the age at death of the Brussels population. We use individual death certificates for the year 1910 with data on age at death, sex, residence and profession, combined with data on sanitary conditions from the Brussels’ city reports and the special census of habitations conducted in 1910. We perform a linear regression analysis with age at death as dependent variable. Some independent variables such as social class (HISCO classification based on the profession of the deceased, their spouse or parents) and sex are measured on the individual level, while other variables such as the level of crowding, and cadastral class of the inhabited houses are based on street level. To visualize the variation in health conditions and mortality within the city, the results of our regression analysis are complemented by maps. As a result, our study will deliver important insights into the excess mortality of the Belgian capital at the turn of the twentieth century and into the historical debate on social health inequalities.

1) Objective

At the turn of the twentieth century, life expectancy in Brussels was remarkably low. Around 1910, it amounted to 41 years, when the national average was approximately 10 years higher. The differences with large cities such as Liège, Antwerp and Ghent fluctuated between 5 and 7 years (Devos and Van Rossem, 2015). Only a small part of the differences can be attributed to artificial effects, i.e. to the large number of unregistered people dying in Brussels, mainly in hospitals (Van Rossem, Deboosere and Devos, forthcoming (a)). In previous research, we posited that the Brussels health penalty was mainly related to the unhygienic working circumstances in cottage work and in small factories, and the high levels of crowding and the absence of pure water in many neighbourhoods (Van Rossem, Deboosere and Devos, forthcoming (b)). Because these determinants suggest large mortality differences within the city, an individual level analysis is required.

First, we examine how residence influenced age at death, as environmental circumstances such as housing conditions, access to pure water, sewage systems, air quality etcetera can strongly affect survival chances (e.g. Kesztenbaum and Rosenthal, 2017; Thornton and Olson, 2011). Second, we examine the link between social class and age at death. Health can also be connected to people’s socio-economic position through financial resources, living and working conditions, lifestyle, social networks, etcetera (e.g. Bengtsson and van Poppel, 2011; Razzell and Spence, 2006; Van Rossem, Deboosere and Devos, 2017). As such, this research will permit to link up with the historical debate on social health inequalities, of which the evidence is currently mixed (Bengtsson and van Poppel, 2011).

2) Data

The main sources for our study are the individual death certificates for the year 1910, delivering data on the age at death, sex, place of death and usual residence for each deceased in Brussels. We only take into account people who had their usual residence in Brussels. Social class is proxied here by the profession of the deceased, or the profession of their spouse, father or mother. We complement these data with residence-specific information on health conditions and general demographic measures from the Brussels’ city reports and the special census of the habitations in Brussels conducted in 1910.

3) Method

In order to determine the impact of residence and socio-economic status on mortality, we perform a linear regression analysis. The dependent variable is age at death. Unfortunately, as data on the age- and sex-specific composition of the Brussels’ population by street or division was not preserved in a coherent way, we cannot calculate death rates or life expectancies. As independent variables for residence, we include the average level of crowding and the cadastral class of the inhabited houses per street. To study the impact of socio-economic status, we include occupation using the HISCLASS scheme (Historical International Social Class Scheme) based on the HISCO classification (Historical International Standard Classification of Occupations). Finally, we present the health variations within Brussels via maps, hereby visualising the healthiest and unhealthiest areas of the Belgian capital. Moreover, these maps will also demonstrate whether the living areas of the poor and wealthy were clearly separated from each other or not.

4) References

Bengtsson, T., van Poppel, F., 2011. Socioeconomic inequalities in death from past to present: An introduction. Explor. Econ. Hist. 48, 343–356.

Devos, I., Van Rossem, T., 2015. Urban health penalties. Estimates of life expectancies in Belgian cities, 1846-1910. J. Belg. Hist. 45, 74-109.

Kesztenbaum, L., Rosenthal, J.L., 2017. Sewers’ diffusion and the decline of mortality: The case of Paris, 1880–1914. J. Urban Econ. 98, 174–186.

Razzell, P., Spence, C., 2006. The Hazards of Wealth: Adult Mortality in Pre-Twentieth-Century England. Soc. Hist. Med. 19, 381–405.

Thornton, P., Olson, S., 2011. Mortality in late nineteenth-century Montreal: Geographic pathways of contagion. Popul. Stud.. 65, 157–181.

Van Rossem, T., Deboosere, P., Devos, I., 2017. Death at work? Mortality and industrial employment in Belgian cities at the turn of the twentieth century. Explor. Econ. Hist.

Van Rossem, T., Deboosere, P., Devos, I., forthcoming (a). The de jure/de facto enigma. The impact of unregistered attendees and absentees on urban death rates in early twentieth century Belgium.

Van Rossem, T., Deboosere, P., Devos, I., forthcoming (b). The Brussels graveyard. Disease-specific mortality of men in the Belgian capital at the beginning of the twentieth century in a comparative perspective.

Presented in Session 1058: History