Identifying specific non-attending groups in breast cancer screening - population-based registry study of participation and socio-demography

Background: A population-based breast cancer screening programme was implemented in the Central Denmark Region in 2008-09. The objective of this registry-based study was to examine the association between socio-demographic characteristics and screening participation and to examine whether the group of non-participants can be regarded as a homogeneous group of women. Method: Participation status was obtained from a regional database for all women invited to the first screening round in the Central Denmark Region in 2008-2009 (n=149,234). Participation data was linked to registries containing socio-demographic information. Distance to screening site was calculated using ArcGIS. Participation was divided into 'participants' and 'non-participants', and non-participants were further stratified into 'active non-participants' and 'passive non-participants' based on whether the woman called and cancelled her participation or was a 'no-show'. Results: The screening participation rate was 78.9%. In multivariate analyses, non-participation was associated with older age, immigrant status, low OECD-adjusted household income, high and low level education compared with middle level education, unemployment, being unmarried, distance to screening site >20 km, being a tenant and no access to a vehicle. Active and passive non-participants comprised two distinct groups with different socio-demographic characteristics, with passive non-participants being more socially deprived compared with active non-participants. Conclusion: Non-participation was associated with low social status e. g. low income, unemployment, no access to vehicle and status as tenant. Non-participants were also more likely than participants to be older, single, and of non-Danish origin. Compared to active non-participants, passive non-participants were characterized by e. g. lower income and lower educational level. Different interventions might be warranted to increase participation in the two non-participant groups.