Published Papers - Abstract 836

Harris ML, Dolja-Gore X, Kendig H & Byles JE. First incident hospitalisation for Australian women aged 70 and beyond: a 10 year examination using competing risks. Archives of Gerontology and Geriatrics, 2016; 64: 29-37

There are increasing concerns regarding high hospital use among older adults and the capacity to manage the economic impact of the ageing population trend on healthcare systems. First hospitalisation in old age may act as a catalyst for ongoing intensification of health problems and acute care use. This study examined factors associated with first incident hospitalisation in women aged over 70, accounting for the health inequalities associated with geographic location. Survey data from 3780 women from the 1921 to 1926 cohort of the Australian Longitudinal Study on Women’s Health were matched with the Admitted Patients Data Collection and National Death Index. Days to first event (hospitalisation or death) were modelled using competing risks methods. A total of 3065 (80.3%) women had at least one hospital admission. More than half of the top 15 reasons for first hospitalisation were related to cardiovascular disease, with atrial fibrillation the most common. Proportional subdistribution hazards models showed that first hospital admission was driven by enabling and need factors including asthma/bronchitis diagnosis (HR = 1.16; p = 0.047), private health insurance (HR = 1.16; p = 0.004) more than two prescribed medications in previous month (HR = 1.31; p = 0.001), more than four general practitioner visits in previous year (HR = 1.50; p = 0.034), lower physical functioning (HR = 0.99; p < 0.001) and living in an inner regional area (HR = 1.17; p = 0.003). First overnight hospitalisation was primarily related with potentially preventable and treatable chronic diseases. Primary and secondary strategies aimed at chronic disease generally, and better chronic disease management particularly for cardiovascular and respiratory diseases, may play a vital role in disease prevention or delay in readmissions among this population

Return