None listed
Conditions
Brief summary
Patients admitted to General Medicine often have complex health needs, and some are unexpectedly readmitted to hospital soon after going home. This study aims to see whether using an electronic tool that estimates a patient’s risk of re-admission, together with a more consistent approach to discharge planning, can reduce unplanned re-admissions within 28 days. The tool uses information already collected during routine care and supports, but does not replace, clinical judgement. In this trial, both groups receive the same formalised, standardised model of care; the intervention group additionally receives the AI-generated re-admission risk score to support prioritisation. Patients will not be asked to do anything extra as part of the study, and no consent form is required because the study is low risk and part of standard care. All patient information will be handled securely and confidentially.
Interventions
The intervention is an AI generated re-admission risk score, generated in the Electronic Medical Record in addition to usual care. This will be available as a clinical decision support tool to the multidisciplinary team (doctors, nurses & allied health) and will be visible in their patient lists and on a dashboard. The readmission score with also include contributing factors explaining the score to clinicians (example demographics, medications & hospital utilisation). The re-admission risk score
The intervention is an AI generated re-admission risk score, generated in the Electronic Medical Record in addition to usual care. This will be available as a clinical decision support tool to the multidisciplinary team (doctors, nurses & allied health) and will be visible in their patient lists and on a dashboard. The readmission score with also include contributing factors explaining the score to clinicians (example demographics, medications & hospital utilisation). The re-admission risk score will be recalculated 4 times per day and can change based on the patients progress. Some variables are fixed (such as past hospital utilisation) while others can change (patient vital signs and weight). The re-admission prediction score will be displayed from admission until discharge while admitted under general medicine. The average length of stay for general medicine is 6 days, however for each individual patient the score will be available throughout the admission. The readmission score is calculated within the medical record, and no data will leave the secure server for processing. Doctors, nurses and allied health professionals will be given 1 hour of face-to-face training to understand the features included in the readmission prediction score, 2 weeks prior to the first participant enrolment. The intervention will occur over a 4-month period, recruitment may be extended if the required sample size has not been achieved.
Sponsors
Melbourne Health
Study design
Allocation
Randomised controlled trial
Intervention model
Parallel
Primary purpose
Prevention
Masking
Open (masking not used)
Eligibility
Sex/Gender
All
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Patients at the Royal Melbourne Hospital will be eligable to participate if they are discharged under Acute Medical Unit (AMU) or Medical Units 1-4 (MU1-4).
Exclusion criteria
Patients admitted under AMU or MU1-4 but discharged by another inpatient unit
Outcome results
None listed