How do dementia risk differences between birth cohorts affect future incidence predictions: A microsimulation study

Abstract Background In a recent coordinated analysis across population‐based cohorts, we have shown that the age‐specific dementia incidence has declined by 13% per decade since 1990. Here, aim to incorporate these trends into projection of future burden, using microsimulation models can synthesise complex information, take time account, and estimate long‐term population wide intervention effects. Method Based on Microsimulation Screening Analysis (MISCAN) model from cancer research, developed model. It synthesises data Rotterdam Study with published disease stage durations several mostly clinical cohorts. We changes in demographics defining unique birth life tables for every 10‐year cohort between 1910 1980. Three trend scenarios were calibrated fit observed Study: 1) no risk differences cohorts; 2) linear decline rate cohort; 3) two nonlinear declines result same cumulative average as trend. Projections Netherlands until 2050 calculated scenario. Result The simulated 10 million individuals, representative Dutch 1980 was able well. Figure 1 shows projections implementing aforementioned assumptions. Assuming results highest projections, lowest, lie between. By 2050, compared decline, 40% 25‐35% fewer cases. Conclusion Assumptions about play crucial role incidence, subsequently prevalence mortality. is well capable projecting incidence. An application projects 25‐40% lower than currently assumed. Modelling effect interventions level provides essential information policy makers researchers.