Published in:
01-10-2010 | Editorial
It does take two to tango! On the need for theory in research on the social determinants of health
Author:
Matthias Richter
Published in:
International Journal of Public Health
|
Issue 5/2010
Login to get access
Excerpt
Research on social determinants of health has witnessed vivid changes and improvements over the last years. These developments concern, for example, new possibilities of data analysis (such as dynamic or multi level modelling), the generation of new data (cohort or cross-national studies) and even an increase in the political awareness of the issue in some countries. The report of WHO’s Commission on Social Determinants of Health is a new milestone in documenting many of these improvements (CSDH
2008). Yet, one needs to state a strong imbalance in research on social inequalities in health. Almost all of these developments refer to data-driven empirical research. The further development of theory, however, is lacking way behind other advancements in the field (Lahelma
2006). In our research we sometimes even refer to the same theoretical approaches that have first been proposed over 30 years ago, for example in the Black report. Of course, even in this field there have been improvements and progression, for example in linking psychobiological processes and aggregate deprivation to social inequalities in health (Marmot and Wilkinson
2006). However, many of these approaches are still largely driven by empirical data. Without any doubt, social epidemiology as well as public health in general are disciplines that strongly focus on empirical research. However, this cannot mean that the gap between empirical and theoretical work continues to increase. While it is not possible to look at the reasons of this development in detail here (due to necessary length-restrictions of an editorial), it is very obvious that the status quo in research on social determinants of health needs a change to a stronger accentuation of explanatory approaches. The full potential of a theory-driven analysis is hardly utilised (Potvin
2009). …