Premium
The development of a microsimulation model to predict the future burden of dementia and effects of public health interventions
Author(s) -
Brück Chiara,
Wolters Frank J,
Ikram M Arfan,
Kok Inge MCM de
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.040855
Subject(s) - microsimulation , dementia , psychological intervention , gerontology , population , cohort , quality adjusted life year , incidence (geometry) , psychology , medicine , demography , environmental health , psychiatry , risk analysis (engineering) , engineering , cost effectiveness , disease , sociology , physics , optics , pathology , transport engineering
Background Microsimulation models can synthesise complex information, evaluate uncertainties, estimate long‐term and population wide effects of (hypothetical) interventions, and take time trends in risk factors into account. No microsimulation models have been developed so far that include all stages from cognitively healthy to severe dementia and evaluate risk factor changes and the effects of (hypothetical) interventions. We developed a microsimulation model that predicts how dementia incidence and mortality will develop over the coming decades in The Netherlands, taking into account different risk factor scenarios and interventions. Method Based on the well‐known Microsimulation Screening Analysis (MISCAN) model from cancer research, a dementia microsimulation model was developed. It synthesises dementia incidence data and survival probabilities from the large population‐based Rotterdam Study cohort (Ikram et al., 2017) with dementia severity stage duration estimates from the literature (Vermunt et al., 2019). The model simulates the life histories of a large number of individuals, each of whom can develop mild cognitive impairment (MCI) which can then progress to mild dementia and moderate/severe dementia and finally death. Based on these life histories, the number of dementia cases, dementia deaths, (quality adjusted) life years of both patients and caregivers, and costs can be evaluated by age and gender. Finally, the model can estimate the effect of different interventions on the life histories and subsequently the outcome measures. Result We were able to develop the MISCAN‐Dementia model, as depicted in Figure 1. MISCAN‐Dementia consists of three modules: a demography module, a natural history module, and a screening module. Given this structure, the model can be used to evaluate long‐term benefits and harms of important interventions at different disease stages: 1) changes in risk factors (i.e. onset of cognitive decline); 2) earlier detection and treatment by population screening (i.e. early, pre‐dementia stage); and 3) care and cure (i.e. dementia stage). Conclusion The dementia microsimulation model developed in this study predicts how the burden of dementia will develop over the coming decades, including incidence, mortality, quality of life, and costs. In addition, modelling the effect of interventions on a population level will provide essential information for researchers and policy makers.