Patterns of Disability Among Older Australians: Results from Microsimulation Dynoptasim

Sophie Pennec, School of Demography/The Australian National University
Heather Booth, School of Demography/The Australian National University

DYNOPTASIM models ageing and disability in the Australian population, focussing on dementia, depression, impaired vision, impaired hearing, impaired mobility and falls. This dynamic microsimulation model is based on the harmonised longitudinal DYNOPTA dataset (the combined data from nine longitudinal studies of ageing in Australia). The first wave of the harmonised dataset is used to derive the DynoptaSIM base population of 47,000 Australians aged 45+ and living in the community, which represents 1% of the 1996 household population by sex, age and geography. The variables included in the model include 13 risk factors. Socio-demographic variables include age, sex, partnership, education and occupation. Behavioural variables include social contact, physical activity, smoking, alcohol consumption. Biomedical variables include BMI, hypertension, diabetes and stroke. The distributions of these variables in the base population have been validated against 1996 census and survey data. Monthly transition probabilities were derived from the DYNOPTA dataset, using the first two or three waves in regression analyses. Validation of the simulated population was against 2006/2011/2016 census and survey data. National sex-specific mortality data and forecasts are used, adjusted for dementia status. The model output provides age- and sex-specific incidence, prevalence and duration measures for each disability. Patterns of change are examined, including changing sex differentials in disability. Hypothetical scenarios also enable examination of the disability-delaying effects of modifiable risk factors, its potential impact on life expectancy, on disability free life expectancy.

Presented in Session 1233: Mortality and Longevity