Age, period, and cohort effects in the population-level incidence and mortality of cancer types associated with early onset increases in Australia

The incidence of early-onset cancers has increased in recent decades, with hypothesised inter-generational changes in overlapping environmental and behavioural risk factors implicated in changing trends. This descriptive-analytic study investigates the age, period, and cohort effects in population-level incidence and mortality of cancer types previously associated with early onset increases in the Australian population, compared to selected cancer types without early-onset increases. Age-period-cohort modelling examined whether there were differences in temporal trends in Australian cancer incidence (1982–2021) and mortality (1971–2023) between cancer types with early onset increases (breast, colorectum, uterine, kidney, liver, pancreas, thyroid, prostate, oesophagus) and comparator cancers (lung, cervix, melanoma of the skin). For early onset cancer incidence, there was an increasing cohort effect for both males and females in colorectal, liver, pancreatic, thyroid cancers, and for uterine cancers among women, and for prostate cancer among men. The largest relative increase was in pancreatic cancer among females, and thyroid cancer among males. Decreasing cohort effects were evident for melanoma of the skin and cervical cancer (among women), with no substantial differences across birth cohorts for lung cancer and oesophageal cancer. Period effects (adjusting for cohort effects) were relatively stable, except for prostate cancer for males and cervical cancer for females. Decreasing period and cohort effects for cancer mortality were evident for most cancers and for both males and females. The exceptions were an increasing period effect for thyroid cancer and pancreatic cancer, and cervical cancer and uterine cancer (for females). This study found higher population-level incidence of cancer types associated with early onset increases among later born cohorts, which may indicate cohort-specific differences in risk factors associated with these outcomes.