Autumn Conference of the Section Sociology of Medicine and Health of the German Sociological Association 
03/04 October 2013 European University Institute, Florence, Italy 

The conference ›Bias in Health Data‹ was located in the wonderful city of Florence and covered several recent findings in the field of public health and methodology. Presentations focused on diverse challenges in measuring health outcomes and predictors such as the gender bias in measuring depression, applying differential item functioning (DIF) to identify measurement bias by group membership, estimating the correlation of education and health over time by referring to relative and absolute educational level, and survey methodological effects occurring when changing the mode of data collection. Studies dealing with self-rated health (SRH) as the most prominent health outcome refered to the intra-individual consistency in answering two different versions of SRH, the biases of SRH by a social event such as the 2011 earthquake in Japan, and differences in self-rated diseases versus diseases measured by blood tests. Consequences of biases for the results in health research have been shown by reconsidering the healthy migrant effect or the causal relationship of socioeconomic background and health. Further approaches criticized the postulated objectivity in facial analysis and resulting aesthetic reconstruction and presented biases that occur by varying acceptance of using new technologies. The open session contained several talks about explaining health decisions, which has a great impact on (bias in) register data.