An ecological age, period and cohort analysis of German breast cancer mortality from 1999 to 2008

Berkemeyer S, Lemke D, Hense HW

Abstract

Background: Longitudinal analysis investigates period (P) data, often in yearly intervals. Additional scales of time, such as, age (A) and birth cohort (C) also exist. Investigations usually account for age while cohort is rarely investigated. We conducted an ecological APC analysis for breast cancer mortality in women in Germany.Methods: We obtained nation-wide deaths due to breast cancer from Robert Koch Institute, Berlin and estimated the female population size for the period 1999 to 2008. Data were stratified into fourteen 5-years age-groups starting at 20-24 years (y), fourteen birth cohorts starting with 1914-18, and the two calendar periods 1999-2003 and 2004-2008. Data was analyzed using R statistical package. Annual mortality rates were calculated as the number of deaths due to breast cancer per 100,000 per year.Results: Breast cancer mortality rates increased consistently with age up to age-group 85+, only age 60-70 years had somewhat lower mortality compared to the adjoining age-groups. Period analyses showed declining secular trends from 1999 to 2008. The birth cohorts 1950 to 1960 showed the highest mortality rate while the lowest was observed for cohort 1984-88.Conclusions: While confirming reports of declining mortality based on age and period analysis, our study reveals that German birth cohorts 1950-1960 were unique in having highest mortality rate compared to other birth cohorts, adding another facet to breast cancer mortality analyses.