Geographic disparities of breast cancer incidence in Portugal at the district level: A spatial age-period-cohort analysis, 1998-2011. (October 2021)

• Record Type: Journal Article
• Authors: Froelicher, Joseph H.; Forjaz, Gonçalo; Rosenberg, Philip S.; Chernyavskiy, Pavel
• Is Part Of: Cancer epidemiology. Volume 74(2021)
• Publication Date: 2021-10
• Subjects: Age-period-cohort -- Geographic disparities -- Breast cancer incidence -- Portugal; Cancer -- Epidemiology -- Periodicals; Cancer -- Prevention -- Periodicals; Cancer -- Diagnosis -- Periodicals; Carcinogenesis -- Periodicals; 616.994005
• DOI: 10.1016/j.canep.2021.102009
• Languages: English
• ISSNs: 1877-7821

Abstract

Graphical abstract: Highlights: We show district-level disparities in relative risk and age-adjusted trends in breast cancer incidence for Portugal. Densely populated coastal districts (Lisboa, Setúbal, Porto) had the highest incidence rates. There is a gradient in age-adjusted trends over time, with steeper trends in the sparsely-populated north of Portugal. Geographic disparities were especially prevalent among heavily-screened and late-onset age groups. Abstract: Background: Breast cancer is the most common malignancy in women world-wide and the most common cause of cancer deaths, which can often be managed with early diagnosis and subsequent treatment. Here, we focus on geographic disparities in incidence within Portugal for three age groups of women (30−49; 50−69; 70−84 years). Methods: Age-period-cohort (APC) models are widely used in cancer surveillance, and these models have recently been extended to allow spatially-varying effects. We apply novel spatial APC models to estimate relative risk and age-adjusted temporal trends at the district level for the 20 districts in Portugal. Our model allows us to report on country-wide trends, but also to investigate geographic disparities between districts and trends within districts. Results: Age-adjusted breast cancer incidence was increasing over 1998–2011 for all three age groups and in every district in Portugal. However, we detect spatially-structured between-district heterogeneity in relative risk and age-adjusted trends (Net Drifts) for each of the three age groups, which is most pronounced in the highly-screened (50−69yo) and late-onset (70−84yo) groups of women. Conclusions: We present evidence of disparities in breast cancer incidence at a more granular geographic level than previously reported. Some disparities may be due to latent risk factors, which cannot be accounted for by age, birth year, and geographic location alone. Impact: Our study motivates resuming data collection for breast cancer incidence at the district level in Portugal, as well as the study of exogenous risk factors.