Background Age period cohort (APC) analyses separate the impact of age effects linked to the process of ageing, period effects; external factors experienced at a particular time, and cohort effects, an interaction of age and period where a birth group experience external factors at a specific age. Mortality in Scotland has increased in recent years while life expectancy has reduced, believed to be largely due to austerity. This study aims to identify the cohorts with the highest mortality rates and interpret what this might mean in the current context.

Methods Scottish mortality 1980-2020 data by sex, age and year published by ONS were first graphed to assess age period and cohort patterns. Median Polish APC methods were used to remove the effects of age and period. Cohort residuals for males and females were logged and modelled using linear regression with residuals as dependent and birth cohort as independent variables. Single year and 5- year cohorts were modelled for both full dataset and a truncated 1980-2010 dataset in order to remove the impact of austerity and other recent.

Results Among males [females], with 1976 as the reference year, after removing period effects, those born before 1973 [1973] and after 1979 [1990] had a significantly lower risk of mortality. These effects were also observed when 5 year birth cohorts were used; Cohorts 1976-1980 had significantly greater rates than birth cohorts prior to 1971-1975 [1971-1975] and after 1981-1985 [1986-1990] for males [females]. As an example, the rate ratio (and CI) for males for 1951-1955 cohort was 0.74 (0.70, 0.79) compared with reference cohort 1976-1980, and for 1966-1970 cohort was 0.91 (0.86, 0.96) dropping to 0.63 (0.59, 0.68) for cohort 1996-2000. Cohort effects for females were smaller than males, e.g. rate ratio for 1951-1955 cohort was 0.88 (0.83, 0.92) compared with 1976-1980. When the truncated 1980-2010 dataset was modelled for males and females similar outcomes were observed.

Conclusion The male cohort born in 1971-1985 but particularly those born in 1973-1979 and the female cohort born in 1971-1990 have highest mortality rates after adjustment for period and age. Effect sizes are particularly large for males. This is likely to become more concerning as this cohort approaches their mid-50s and mortality rates due to age increase exponentially, in combination with period effects due to austerity measures, the pandemic and current cost of living crisis. The sum total of these factors is likely to affect future mortality and morbidity trends.