Factors associated with success in transition care services among older people in Australia

The Australian Transition Care Program (TCP) is a national intermediate care service aiming to optimise functional independence and delay entry to permanent care for older people leaving hospital. The aim of this study was to describe the outcomes of TCP and identify demographic and clinical factors associated with TCP 'success', to assist with clinical judgements about suitable candidates for the program.

We conducted a descriptive cohort study of all older Australians accessing TCP for the first time between 2007 and 2015. Logistic regression models assessed demographic and clinical factors associated with change in performance on a modified Barthel Index from TCP entry to discharge and on discharge to community. Fine-Gray regression models estimated factors associated with transition to permanent care within 6 months of TCP discharge, with death as a competing event.

Functional independence improved from entry to discharge for 46,712 (38.4%) of 124,301 TCP users. Improvement was more common with younger age, less frailty, shorter hospital stay prior to TCP, and among women, those without a carer, living outside a major city, and without dementia. People who received TCP in a residential setting were far less likely to record improved functional impairment and more likely to be discharged to permanent care than those in a community setting. Discharge to community was more common with younger age and among women and those without dementia. Nearly 12% of community TCP and 63% of residential TCP users had transitioned to permanent care 6 months after discharge. Entry to permanent care was more common with older age, higher levels of frailty, and among those with dementia.

More than half of TCP users are discharged to home and remain at home after 6 months. However, residential-based TCP may have limited efficacy. Age, frailty, carer status, and dementia are key factors to consider when assessing program suitability. Future studies comparing users to a suitably matched control group will be very helpful for confirming whether the TCP program is meeting its aims.