Predictors of mortality in men and women aged 90 and older: a nine-year follow-up study in the Vitality 90+ study

Background: information about the predictors of mortality among the oldest-old is limited. Also possible gender differences are poorly known. Objective: to examine the predictors of mortality among individuals aged 90 and older, focusing on differences between men and women. We also analysed gender differences in survival at different levels of mobility and activities in daily living (ADL). Design: this 9-year follow-up study is part of the Vitality 90+ study, a population-based study of people aged 90 and older. Subjects: all inhabitants aged 90 and older in the area of Tampere, Finland were contacted, irrespective of health or dwelling place. The study population consisted of 171 men and 717 women. Methods: data were collected with a mailed questionnaire asking questions concerning ADL and mobility, self-rated health, chronic conditions and socio-economic factors. The participation rate was 79%. Cox regression enter models were used for the analysis. Results: older age, male gender, disability in ADL and mobility, poor self-rated health and institutionalisation increased the risk of mortality in the total study group. In age-adjusted Cox regression models, ADL and mobility were stronger predictors in men than in women (gender interactions, P< 0.001). Among those who were partly but not totally dependent in ADL or mobility women survived longer than men. Conclusion: the same health indicators that are important at younger old age also predict mortality in the oldest-old. Disability increases the likelihood of death more in men than women. At a very old age, women survive longer with moderate disability than do men.