Rapid Fire Australian Epidemiology Association ASM 2018

Explaining geographical variation in bowel cancer screening participation in Victoria (#141)

James A Chamberlain 1 , Julie K Bassett 1 , Geoffrey W Stuart 1 , Kate H Broun 1 , Dallas R English 1 2 , Roger L Milne 1 2
  1. Cancer Council Victoria, Melbourne, Victoria, Australia
  2. Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Parkville, Victoria, Australia

Background: Bowel cancer screening is a highly effective public health initiative, although National Bowel Cancer Screening Program (NBCSP) screening participation was only 41.2% in Victoria in 2015-16.

Aim: To assess factors associated with geographic variation in NBCSP % screening participation in Victoria, Australia.

Methods: We analysed publicly available bowel cancer screening data, Australian Bureau of Statistics 2011 census data and Victorian Population Health Survey data, using Statistical Area 2 (SA2) as the unit of analysis. The dependent variable was NBCSP participation in 2015-2016.  Using linear regression and dominance analysis, we investigated a range of potential area-based determinants of screening, including remoteness, socio-economic status, education, income, community amenity, proportion of people speaking English well, country of birth, ancestry, religion and language.

Results: We produced maps of Victoria showing the variation in NBCSP participation. A multivariable model explained 77% of the variance in screening participation. Screening rates are strongly associated with cultural variables (country of birth, ancestry, religion, whether both parents were born overseas, language), community amenity, socio-economic status, remoteness, income and education. There was a non-linear inverted U-shaped relationship between screening and income.

Conclusions:  People living in regional areas, with English background, high community amenity and medium income are more likely to participate in screening. People living in cities, culturally diverse areas, areas with low community amenity, and areas with low or high income, are less likely to participate in screening.