Individual-level high definition spatial distribution of mammography uptake in Geneva (Switzerland). (13th November 2019)

• Record Type: Journal Article
• Authors: Sandoval, J L; Himsl, R; Joost, S; Guessous, I
• Is Part Of: European journal of public health. Volume 29(2019)Supplement 4
• Publication Date: 2019-11-13
• Subjects: Epidemiology -- Europe -- Periodicals; Public health -- Europe -- Periodicals; 362.109405
• DOI: 10.1093/eurpub/ckz187.136
• Languages: English
• ISSNs: 1101-1262

Abstract

Abstract: Background: The local social and physical environment influences health outcomes and individual behaviour. It remains unknown if the environmental influence is independent of individual socioeconomic status (SES). We studied the spatial distribution of mammography screening uptake in the Swiss urban population of Geneva and determined its independence from SES. Methods: We used individual-level geo-referenced data (n = 5 002) from the participants in the Bus Santé study, a population-based cross-sectional study ongoing in Geneva since 1992. Data from 1992 to 2014 have been included in this study. Local indicators of spatial association (LISA) were calculated and used to assess the spatial dependence of mammography uptake (defined as having had a mammography in the past). We report unadjusted spatial clusters; adjusted for neighbourhood income and individual educational attainment; and for demographic variables (age and Swiss nationality). Furthermore, we evaluated the association between the distance to the nearest screening centre and the identified spatial clusters. Results: Mammography uptake was not randomly distributed, with spatial clusters overlapping with those of SES. Spatial clusters were reduced to 56.2% of their initial size (n = 1, 033) after adjustment for SES variables. Adjustment for age and nationality further reduced the proportion of individuals exhibiting spatially-dependent behaviour (to 36.5% of the initial size). Distance to the nearest screening centre was not associated with the spatial distribution of mammography uptake. Conclusions: High definition spatial distribution studies using analysis of individual data allow identifying spatial clusters which are not based on pre-determined administrative units (e.g. postal code) but rather on individual behaviour. Persistent spatial clusters after adjustment for SES and demographic confounders suggest additional area-level determinants influencing the spatial distribution of mammography uptake. Key messages: High definition spatial studies can help uncover new patterns of health outcomes distribution independent of SES and based on individual behaviour, rather than administrative units. While SES explains a great proportion of the spatial distribution of mammography uptake, other still elusive factors may be contributing to the observed spatial inequalities in mammography uptake.