Associations between deprivation and colorectal cancer incidence and mortality in Scotland


Session type:

Peter Steele1, David Brewster2, Annie Anderson1, Robert Steele1
1University of Dundee, Dundee, UK, 2ISD, Edinburgh, UK


While there are recognised positive correlations between increasing levels of deprivation and both cancer incidence and mortality, there are variations between cancer sites and different countries. In colorectal cancer (CRC), an association between deprivation and incidence has not been a recognised phenomenon in the past1. However, recent evidence indicates that the prevalence of CRC lifestyle risk factors are significantly higher in deprived compared to less deprived communities2.


Using Scottish Cancer Registry data, European age standardised incidence and mortality rates per 100,000 person-years at risk (EASRs) were analysed by deprivation quintile for the combined period of 2004-2008 for incidence and between 2005 and 2009 for disease-specific mortality. The total number of registrations for all colorectal malignant neoplasms was 18069 of which 12032 were colonic and 6037 were rectal. Deprivation was estimated using the Scottish Index of Multiple Deprivation (2006).


For colorectal cancer, increasing deprivation was associated with increasing incidence (48.4-54.4 p<0.008).   However, when analysed by site within the colon, this correlation only held for rectal cancer (17.1-19.7, p<0.001), and not for colon cancer (34.6-34.6, p<0.36). When further analysed by gender, there was no effect of deprivation in females in either rectal or colon cancer, but in males there was an association between increasing incidence and deprivation in both rectal and colon cancer. Disease specific mortality increased with deprivation for both rectum (8.6-10.1, p<0.0003) and colon (11.5-13.5 p<0.0014) and this was observed in both men and women. 


An association between increasing incidence and deprivation in colorectal cancer is not widely recognised.  This is a recent phenomenon as it was not evident in the period 1986-19951, and appears to be confined to males and to be more pronounced in rectal cancer. It is hypothesised that this may be due to increasing differences in known lifestyle risk factors for CRC across the deprivation gradient.