Transitions to residential aged care after hospital for people living with dementia

Linked data are essential to examining transitions between hospital and residential aged care settings. The study used the National Integrated Health Services Information (NIHSI) analysis asset, a multi-source enduring linked data asset managed under the custodianship of the AIHW (AIHW 2022). This study used the NIHSI 2018–19, which contains de-identified administrative health (hospitals, Medicare claims and prescriptions), residential aged care, and deaths data from 2010–11 to 2018–19 (NIHSI 2018–19). Further detail on the individual data sets included in the linked database and detailed methods can be found in Technical notes.

Study populations

In total, 705,000 people were included in the study.

The study focused on people who had one or more hospitalisations ending in 2017. See How were hospitalisations defined? for more detail on what was considered a hospitalisation in this study. The year 2017 was chosen because it was the latest year in the data which still allowed for 12-month outcomes to be examined.

People were included in the study if they:

• were aged 65 or over in 2017
• had one or more hospitalisations in 2017
• lived in New South Wales, Victoria, Queensland, South Australia, Tasmania, or Australian Capital Territory from 1 January 2017 until 31 December 2018, or until death. These jurisdictions were selected as complete public hospitals data was available in the linked data.
• had complete key information of interest, such as age and sex.

Identifying people living with dementia

Around 11% of all people included in the study were identified as living with dementia (almost 79,000 people) using dementia diagnosis information available in the linked data. People were identified as living with dementia if they had a record of dementia in the linked data between 1 July 2010 and 31 December 2017.

See Technical notes for a detailed list of the classification codes for dementia in each individual dataset in the linked data.

Using information from multiple datasets over several years ensures that anyone whose dementia diagnosis was not recorded in one dataset at any point in time (for example, during a hospitalisation), may be recorded in other datasets and/or at another point in time.

Identifying people without dementia

About 89% of people included in the study were identified as not having dementia (almost 627,000 people) using the linked data. It is possible that this group incorrectly included some people living with dementia, particularly those who are in the earlier stages of dementia and have had little interaction with the health and aged care systems and those who do not have a diagnosis.