How much of the deprivation gap in cancer survival can be explained by variation in stage at diagnosis: An example from breast cancer in the East of England


Session type:

Gary Abel1, Mark Rutherford2, Sally Hinchliffe2, Georgios Lyratzopoulos1, Paul Lambert2,3, David Greenberg4,1
1University of Cambridge, Cambridge, UK, 2University of Leicester, Leicester, UK, 3Karolinska Institutet, Stockholm, Sweden, 4National Cancer Registration Service, Public Health England, Eastern Office, Cambridge, UK


Socioeconomic differences in cancer patient survival exist in many countries and across cancer sites. Reasons for these differences are multifactorial and include inequalities in stage at diagnosis. Previous work have shown substantial inequalities in stage at diagnosis exist for a number of cancer sites1. Here we present an example estimating the number of deaths in women with breast cancer that could be avoided within five years from diagnosis if it were possible to eliminate socioeconomic differences in stage at diagnosis2. Further examples may be discussed.


We analysed data on East of England women with breast cancer (2006-2010). We estimated survival for different stage-age-deprivation strata using both the observed and a hypothetical stage distribution (assuming all women acquired the stage distribution of the most affluent women).


Data were analysed on 20,738 women with complete stage information (92%). Affluent women were less likely to be diagnosed in advanced stage. Relative survival decreased with increasing level of deprivation. Eliminating differences in stage at diagnosis could be expected to nearly eliminate differences in relative survival for women in deprivation groups 3 and 4, but would only approximately halve the difference in relative survival for women in the most deprived group (5). This means, for a typical cohort of women diagnosed in a calendar year with breast cancer, eliminating deprivation differences in stage at diagnosis would prevent approximately 40 deaths in the East of England from occurring within five years from diagnosis. Using appropriate weighting we estimated the respective number of avoidable deaths for the whole of England to be 450.


The findings suggest that policies aimed at reducing inequalities in stage at diagnosis between women with breast cancer are important to reduce inequalities in breast cancer survival.