Geographical variation in prostate cancer testing explained by socio-economic position in Stockholm

Abstract Background Prostate cancer screening using prostate-specific antigen (PSA) testing is controversial but remains prevalent in many countries. Despite its prevalence there is little information in Sweden or globally on the spatial heterogeneity in PSA testing. This study aims to describe the spatio-temporal variation in PSA testing in the Stockholm region specifically at the municipality and small area level. Additionally, it aims to quantify the extent to which socio-economic position (SEP) indicators such as education, income, and country of birth explain the observed variation in PSA testing. Methods A retrospective population-based register study was conducted, encompassing men aged 40 years and older residing in the Stockholm region from 2007 to 2016. Utilizing spatial smoothing techniques, we calculated the age-standardised proportion of men undergoing PSA testing prior to any prostate cancer diagnosis within each area for every calendar year. Furthermore, we used Bayesian logistic regression models with spatial random effects to estimate the individual-level association between PSA testing and socio-economic position. The proportion of variance in PSA testing explained by various measures of SEP was also estimated at both the municipality and small area levels. Results Significant geographical disparities in PSA testing were observed across small areas within Stockholm. The proportion of variance explained in PSA testing at the small area level was highest for income (32.6%), followed by education (9.7%) and country of birth (7.4%). Incorporating all SEP measures led to a higher proportion of explained variance at the small area level (41.6%) compared to the municipality level (24.8%). Conclusions PSA testing was significantly associated with various socio-economic indicators and some of the spatial variation could be explained by SEP. It would be useful to ascertain whether similar patterns are observed in other regions or countries. Key messages • The findings suggested widespread and consistent opportunistic prostate cancer screening practices within the Stockholm region, raising concerns regarding potential overdiagnosis and overtreatment. • The socio-economic differences in prostate cancer screening could partially explain the differences in cancer incidence and mortality.