Disability occurrence and proximity to death

Purpose: This paper aims to assess whether disability occurrence is related more strongly to proximity to death than to age. Method: Self reported disability and vital status were available from six annual waves and a subsequent 12-year mortality follow-up of the Dutch GLOBE longitudinal study. Logit and Poisson regression methods were used to study associations of disability occurrence with age and with proximity to death. Results: For disability in activities of daily living (ADL), regression models with proximity to death had better goodness of fit than models with age. With approaching death, the odds for ADL disability prevalence and incidence rates increased 20.0% and 18.9% per year, whereas severity increased 4.1% per year. For the ages younger than 60, 60-69 and older than 70 years, the odds for ADL disability prevalence increased 6.4%, 16.0% and 23.0% per year. Among subjects with asthma/COPD, heart disease and diabetes increases were 25.1%, 19.5% and 22.72% per year. Functional impairments were more strongly related to age. Conclusions: The strong association of (ADL) disability occurrence with proximity to death implies that a substantial part of the disability burden may shift to older ages with further increases in life expectancy.