Factors associated with discharge destination from the acute general medical ward

Ms Aruska D’Souza1,2, Dr Catherine Granger1,2, Mr Cameron Patrick2, Ms Jacqueline Kay1, Associate Professor Catherine Said2,3

1Melbourne Health, Parkville, Australia, 2University of Melbourne, Parkville, Australia, 3Western Health, Sunshine, Australia


Factors associated with discharge destination from the acute general medical ward

To investigate the association between patient factors (including physical and cognitive function) on hospital admission with discharge destination.

This was a prospective, single-site observational study of general medical patients referred to physiotherapy. Patients from residential care were excluded.  Data collected included demographics (Blaylock Risk Assessment Screening Score, BRASS), co-morbidities (Charlson Comorbidity Index, CCI), pre-morbid physical function, current physical function (de Morton Mobility Index, DEMMI and Alpha Functional Independence Measure, AlphaFIM) and cognition (Rowland Universal Dementia Assessment Scale, RUDAS), all measured within 72 hours of initial Physiotherapy contact.

Between July 2016 and August 2017, 417 patients were recruited (53% female, median age 81 years [IQR 76 – 86]. 245 patients were discharged directly home; 172 were not discharged home, of whom 140 were discharged to subacute. Patients going directly home had higher median functional, mobility and cognitive scores. Fifty-four patient factors were associated with discharge destination. Data were partitioned into training, validation and test sets to provide unbiased estimates of sensitivity, specificity, receiver operative characteristic (ROC) curve and area under curves (AUC) with confidence intervals (CI). Models best associated with discharge destination were “DEMMI and toilet transfers” (AUC 83.8%, 95% CI 76.4 – 91.2) and “AlphaFIM and walking independence” (AUC 81.5, 95% 95% CI 73.2 – 89.7)

Significance of the findings to allied health:
These models provide the basis for developing tools which can be used to rapidly identify discharge destination in this complex patient group.


Ms D’Souza is a senior physiotherapist who graduated from La Trobe University in 2010 and has worked in the public health system for almost eight years. She currently works at the Royal Melbourne Hospital. Ms D’Souza was the successful recipient of the Mary Elizabeth Watson Early Career Fellowship in Allied Health and commence a PhD part time at the University of Melbourne. Her PhD topic aims to investigate discharge from acute general medical wards. She is the lead researcher in a systematic review, two observational studies and a qualitative study. Ms D’Souza was successful in a poster presentation at the 2017 national Australian Physiotherapy Association conference and a table top discussion at the 2018 Australian Association of Gerontology conference.