Trial Review

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12625000775415
Ethics application status
Approved
Date submitted
5/07/2025
Date registered
23/07/2025
Date last updated
23/07/2025
Date data sharing statement initially provided
23/07/2025
Type of registration
Retrospectively registered

Titles & IDs
Public title
Validating Intensive Care Risk Scoring Tests in the Oldest Patients admitted to Intensive Care
Scientific title
Validity of ICU Prognostic Risk Stratification Tools in the Oldest Patients: A Comparative Analysis of Clinical Scores in a Multicenter Binational Cohort
Secondary ID [1] 314826 0
None
Universal Trial Number (UTN)
U1111-1324-9135
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Intensive Care 338086 0
Mortality 338087 0
Critical illness 338088 0
Condition category
Condition code
Public Health 334390 334390 0 0
Health service research
Surgery 334391 334391 0 0
Other surgery

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
In this study we will be comparing the performance of four ICU risk prediction tools: Acute Physiology And Chronic Health Evaluation (APACHE) III, its derived risk of death (APACHE3ROD), the Australian and New Zealand Risk of Death (ANZROD), and the Sequential Organ Failure Assessment (SOFA) score in predicting mortality among nonagenarian and centenarian ICU patients.

All these scores are entered as a routine data entry points in the Australian and New Zealand Intensive Care Adult Patient Database. They will be collected directly from the registry. The time period from which patient data will be collected from the database is between 1 Jan 2010 and 31 Dec 2023.

The ANZROD model estimates the risk of hospital mortality for ICU patients in Australia and New Zealand. It uses physiological and clinical data from the first 24 hours after ICU admission to provide a risk-adjusted benchmark for outcomes, allowing meaningful comparisons between ICU.

The APACHE Score is calculated by combining points assigned to abnormal values of various physiological measurements, patient age, and chronic health status collected within the first 24 hours of ICU admission.

The SOFA Score is calculated by scoring the degree of dysfunction across six organ systems (respiratory, cardiovascular, liver, coagulation, kidney, and neurological), with higher scores reflecting greater organ failure severity.

The APACHE3ROD score is calculated by integrating several components collected within the first 24 hours of a patient’s admission to the intensive care unit (ICU). The calculation combines points assigned to the most abnormal values from a broad range of physiological variables (such as vital signs and laboratory results), patient age, and pre-existing chronic health conditions. Additional factors—including the primary reason for ICU admission and the patient’s location prior to ICU entry—are incorporated. This composite score is then entered into a specific mathematical equation, producing an estimated risk of hospital mortality for the individual patient. The APACHE3ROD model is widely used for severity adjustment, benchmarking, and research in critical care outcomes in Australia and internationally

Primary outcome:
The primary outcome will be the comparative predictive performance of the ANZROD model, and SOFA, APACHE3ROD and APACHE III scoring systems for short-term and long-term mortality among nonagenarian and centenarian ICU patients.

Predictive performance will be assessed with three key metrics:

1. Discrimination, measured with the Area Under the Receiver Operating Characteristic curve (AUROC)
2. Calibration, assessed with observed-to-expected mortality comparisons via Standardised Mortality Ratios (SMRs); and
3. Clinical utility, evaluated through ity comparisons via Standardised Mortality Ratios (SMRs); and evaluated through decision curve analysis to quantify net benefit across a range of threshold probabilities.

Data will be extracted directly from the Australian and New Zealand Intensive Care Adult Patient Database for patients admitted to the ICU between 1 Jan 2010 and 31 Dec 2023.
Intervention code [1] 331432 0
Diagnosis / Prognosis
Comparator / control treatment
The reference comparator score will be the APACHE III score.
Control group
Historical

Outcomes
Primary outcome [1] 342073 0
The composite primary outcome will the comparative predictive performance of the ANZROD, SOFA, APACHE3ROD, and APACHE III scoring systems for short-term and long-term mortality among nonagenarian and centenarian ICU patients. Mortality at the ICU, in-hospital, 30-day, and 1-year timepoints was used as the reference for evaluating short and long term score performance.
Timepoint [1] 342073 0
The first primary timepoints for mortality will be during ICU stay, in-hospital, 30-days and 1-year post-ICU discharge.
Secondary outcome [1] 449471 0
Net clinical benefit of the APACHE III risk score.
Timepoint [1] 449471 0
ICU, in-hospital, 30-day, and 1-year timepoints will be used.
Secondary outcome [2] 449947 0
Net clinical benefit (i.e. clinical utility) of the ANZROD model
Timepoint [2] 449947 0
ICU, in-hospital, 30-day, and 1-year timepoints will be used.
Secondary outcome [3] 449948 0
Net clinical benefit (i.e. clinical utility) of the SOFA score
Timepoint [3] 449948 0
ICU, in-hospital, 30-day, and 1-year timepoints will be used.
Secondary outcome [4] 449949 0
Net clinical benefit (i.e. clinical utility) of the APACHE3ROD score.
Timepoint [4] 449949 0
ICU, in-hospital, 30-day, and 1-year timepoints will be used.

Eligibility
Key inclusion criteria
Inclusion criteria will be patients aged >90 years who require an admission to the ICU for any indication.
Minimum age
90 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
None

Study design
Purpose
Natural history
Duration
Longitudinal
Selection
Defined population
Timing
Retrospective
Statistical methods / analysis
All statistical analyses will be conducted using R software, version 4.4.3 (R Foundation for Statistical Computing, Vienna, Austria).

Data management, descriptive statistics, and modeling procedures will be performed using validated, peer-reviewed R packages.

Baseline characteristics will be compared between patients who died within 1 month of ICU admission and those who survived. The Wilcoxon rank-sum test will be used for continuous variables, and the Chi square test will be used for categorical variables. P values will be calculated for all comparisons to identify significant clinical and demographic differences between groups.

Discriminative ability will be assessed using the area under the receiver operating characteristic curve (AUROC), with 95% confidence intervals derived using DeLong’s method (pROC package).

Pairwise comparisons between models will be performed using DeLong’s test, with Bonferroni correction applied to adjust for multiple comparisons, where appropriate. Calibration will be evaluated using standardized mortality ratios (SMRs), calculated as the ratio of observed to expected mortality at each pre-specified timepoint.

Expected mortality will be based on model-predicted risk. Ninety-five percent confidence intervals will be estimated using Poisson exact methods. An SMR of 1 will indicate perfect calibration, and values 1 will suggest underestimation.

Clinical utility will be assessed using decision curve analysis (DCA). Net benefit curves will be generated for each model across a threshold probability range of 0 to 0.5 using appropriate statistical packages. Each model will be compared to default strategies of treating all or no patients to evaluate its net benefit at different decision thresholds. All statistical tests will be two-sided, and a P value <0.05 will be considered statistically significant, unless otherwise specified.

Recruitment
Recruitment status
Completed
Date of first participant enrolment
Anticipated
Actual
14/04/2025
Date of last participant enrolment
Anticipated
Actual
21/06/2025
Date of last data collection
Anticipated
Actual
24/06/2025
Sample size
Target
40910
Accrual to date
Final
40910
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 28184 0
Austin Health - Austin Hospital - Heidelberg
Recruitment hospital [2] 28185 0
The Alfred - Melbourne
Recruitment postcode(s) [1] 44396 0
3084 - Heidelberg
Recruitment postcode(s) [2] 44397 0
3004 - Melbourne

Funding & Sponsors
Funding source category [1] 319386 0
Government body
Name [1] 319386 0
Austin Health
Country [1] 319386 0
Australia
Primary sponsor type
Government body
Name
Austin Health
Address
Country
Australia
Secondary sponsor category [1] 321871 0
None
Name [1] 321871 0
Address [1] 321871 0
Country [1] 321871 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 317960 0
Alfred Hospital Ethics Committee
Ethics committee address [1] 317960 0
Ethics committee country [1] 317960 0
Australia
Date submitted for ethics approval [1] 317960 0
15/02/2024
Approval date [1] 317960 0
30/05/2024
Ethics approval number [1] 317960 0
Project No: 253/24

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 142694 0
Prof Laurence Weinberg
Address 142694 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC, 3084
Country 142694 0
Australia
Phone 142694 0
+61 413244770
Fax 142694 0
Email 142694 0
[email protected]
Contact person for public queries
Name 142695 0
Laurence Weinberg
Address 142695 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC, 3084
Country 142695 0
Australia
Phone 142695 0
+61 413244770
Fax 142695 0
Email 142695 0
[email protected]
Contact person for scientific queries
Name 142696 0
Laurence Weinberg
Address 142696 0
Department of Anaesthesia, Austin Health, 145 Studley Road, Heidelberg, VIC, 3084
Country 142696 0
Australia
Phone 142696 0
+61 413244770
Fax 142696 0
Email 142696 0
[email protected]

Data sharing statement
Will the study consider sharing individual participant data?
No
No IPD sharing reason/comment: As this is an observational study, patients have not provided consent for their data to be shared.



What supporting documents are/will be available?

No Supporting Document Provided


Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.