Trial Review

Technical difficulties have been reported by some users of the search function and is being investigated by technical staff. Thank you for your patience and apologies for any inconvenience caused.

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
ACTRN12621001567819
Ethics application status
Approved
Date submitted
13/09/2021
Date registered
18/11/2021
Date last updated
1/11/2022
Date data sharing statement initially provided
18/11/2021
Date results information initially provided
1/11/2022
Type of registration
Retrospectively registered

Titles & IDs
Public title
Developing a Decision Support System at Emergency Room triage (DESSERT) for predicting health outcomes
Scientific title
Developing a Decision Support System at Emergency Room triage for predicting health outcomes
Secondary ID [1] 304580 0
20-862
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Overcrowding 322475 0
Hospital admission 322476 0
Acute hospital representation 322477 0
Mortality 322478 0
Emergency department triage 324131 0
Hospital admission 324132 0
Condition category
Condition code
Public Health 320115 320115 0 0
Health service research
Public Health 321375 321375 0 0
Epidemiology

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
A decision support system at emergency room triage. No predefined intervention(s)/exposure.

We will access patient medical records with no active involvement from participant required. All medical records of adult patients presenting at Waitemata District Health Board (WDHB) hospitals’ emergency departments (ED) in Auckland, New Zealand during the 5-year study period from 2016 to 2021 will be included in this retrospective cohort study. The duration of follow-up of each participant is up to 6 months post-index ED presentation. The first four years’ retrospective data of this cohort (from July 2016 to June 2020) will be used to develop a Decision Support System at the time of ED triage for predicting hospital admission and long ED length of stay. The prospectively collected cohort data from July 2020 to June 2021 will be used to validate the performance of this Decision Support System. The actual period of the development and validation cohorts may now vary due to the general health recommendations during the global pandemic.

We will use two methods to develop a Decision Support System (prediction model). One is applying machine learning techniques to develop a Decision Support System (prediction model) with the highest predictive ability for hospital admission and longer ED length of stay. The following prediction algorithms will be investigated in this study, logistic regression, support vector machines, Naive Bayes algorithm, decision trees, random forest, gradient boosting and deep learning. In addition, we will also use the traditional logistic regression to develop this Decision Support System for predicting health outcomes. To ensure high statistical and clinical significance, only variables with a p value of <0.05 will be included in multivariable logistic regression analyses to form the proposed Decision Support System. The modelling performance of different prediction models will be assessed by receiving operator characteristic (ROC) curve with a bootstrapping method using 10,000 replicates to calculate 95% confidence intervals, to assess the predictive performance for predicting health outcomes at the time of ED triage.

Our research group involved experienced triage nurses, ED clinicians, Geriatrician, epidemiologists and biostatisticians. We will have regular meetings (1-2 hours every 2 months for 24 months) to discuss the study progress and findings. There was no further planned surveys/focus groups/interviews or other interactions with staff or patients. However, we will seek advice, when needed, from Emergency Medicine Specialist, Operations Manager Emergency Department, ED clinicians, Triage nurses and other researchers.
Intervention code [1] 320932 0
Early Detection / Screening
Intervention code [2] 322060 0
Not applicable
Comparator / control treatment
No control group.
Control group
Uncontrolled

Outcomes
Primary outcome [1] 328925 0
Hospital admission (Yes/No). Hospital admission is assessed by accessing patient medical record. Hospital admission will include patients who were died in ED or longer stay in ED (>=12 hours) or admitted to inpatient ward. Data-linkage to medical records is used for assessing this outcome.
Timepoint [1] 328925 0
Assessed from the time of the index ED presentation (we will check if the index ED presentation lead to following hospital admission by assessing patient medical record)
Primary outcome [2] 329146 0
ED length of stay (data-linkage to medical records is used for assessing this outcome)
Timepoint [2] 329146 0
From index ED presentation to discharge from ED.
Primary outcome [3] 329147 0
Mortality
Timepoint [3] 329147 0
in 28 days post-index presentation
Secondary outcome [1] 400882 0
Mortality (primary outcome)
Timepoint [1] 400882 0
in 7 days post-index presentation
Secondary outcome [2] 401635 0
Acute hospital representation assessed by accessing patient medical record
Timepoint [2] 401635 0
in 28 days post-index presentation
Secondary outcome [3] 401636 0
Acute hospital representation (triage representation). If a patient represents to ED triage after the index ED presentation, this patient will be counted as acute hospital representation (or called triage representation) (data-linkage to medical records). Data-linkage to medical records is used for assessing this outcome.
Timepoint [3] 401636 0
in 7 days post-index presentation
Secondary outcome [4] 401637 0
Total length of stay assessed by accessing patient medical record
Timepoint [4] 401637 0
of the index ED presentation - including ED length of stay and inpatient length of stay (if they have)

Eligibility
Key inclusion criteria
1) Waitemata District Health Board (WDHB) hospitals’ ED presentations
2) Had Australasian Triage Scale (ATS)
3) Adults visits (aged 18 or over)
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
1) No ATS information
2) dead on ED arrival
3) inconsistency/unreasonable data

Study design
Purpose
Screening
Duration
Longitudinal
Selection
Defined population
Timing
Retrospective
Statistical methods / analysis
All statistical analyses will be performed using SAS version 9.4, SAS enterprise miner and Python. The retrospectively collected ED presentations will be used for model training and validation. The ability of each of the potential predictors will be assessed by two methods. One is applying machine learning techniques to develop a Decision Support System with the highest predictive performance for primary and secondary outcomes. The following prediction algorithms will be investigated in this study, logistic regression, support vector machines, Naive Bayes algorithm, decision trees, random forest, gradient boosting and deep learning. In sensitivity analyses, we will use traditional logistic regression to develop this Decision Support System for predicting outcomes. To ensure high statistical and clinical significance, only variables with a p value of <0.05 are included in multivariable logistic regression analyses to form the proposed Decision Support System. The modelling performance of different prediction models will be assessed by calibration plot and receiving operator characteristic (ROC) curve with a bootstrapping method using 1,000 replicates to calculate 95% confidence intervals, to assess the predictive performance for predicting primary and secondary outcomes at the time of ED triage.

Recruitment
Recruitment status
Completed
Date of first participant enrolment
Anticipated
Actual
26/05/2021
Date of last participant enrolment
Anticipated
1/12/2021
Actual
1/12/2021
Date of last data collection
Anticipated
1/01/2022
Actual
1/01/2022
Sample size
Target
400000
Accrual to date
Final
530165
Recruitment outside Australia
Country [1] 24113 0
New Zealand
State/province [1] 24113 0
Auckland

Funding & Sponsors
Funding source category [1] 308945 0
Government body
Name [1] 308945 0
The Health Research Council of New Zealand (HRC)
Country [1] 308945 0
New Zealand
Funding source category [2] 309834 0
Commercial sector/Industry
Name [2] 309834 0
Precision Driven Health (PDH)
Country [2] 309834 0
New Zealand
Primary sponsor type
Government body
Name
The Health Research Council of New Zealand (HRC)
Address
Level 3/110 Stanley Street, Grafton, Auckland 1010
Country
New Zealand
Secondary sponsor category [1] 309867 0
None
Name [1] 309867 0
Address [1] 309867 0
Country [1] 309867 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 308831 0
Northern B Health and Disability Ethics Committee
Ethics committee address [1] 308831 0
133 Molesworth Street, Thorndon, Wellington 6011
Ethics committee country [1] 308831 0
New Zealand
Date submitted for ethics approval [1] 308831 0
01/02/2021
Approval date [1] 308831 0
27/04/2021
Ethics approval number [1] 308831 0
21/NTB/17

Summary
Brief summary
Emergency department overcrowding is a major global healthcare issue. The consequences are well-established, usually affecting patients (poor outcomes), staff (stressed) and healthcare system (long length of stay). Without increases in the number of EDs and staff, an effective way is to optimise the use of existing resources. This study intends to develop a decision support system at ED triage time, to predict hospital admission and longer ED length of stay by using a wide range of routinely collected big data (DHBs Health Records System). This system has the potential to meet the ED health target of a ‘shorter stay’ and ‘lower hospital admission rates’ by accurately identifying high-risk patients at an early stage of ED and making more effective interventions for them. If so, this decision support system can be widely used by ED triage assessors in the near future, with the potential to improve the quality of acute care.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 112050 0
Dr Zhenqiang Wu
Address 112050 0
124 Shakespeare Road, North Shore Hospital, Takapuna, Auckland 0620, New Zealand
Country 112050 0
New Zealand
Phone 112050 0
+64211531391
Fax 112050 0
Email 112050 0
zhenqiang.wu@auckland.ac.nz
Contact person for public queries
Name 112051 0
Dr Zhenqiang Wu
Address 112051 0
124 Shakespeare Road, North Shore Hospital, Takapuna, Auckland 0620, New Zealand
Country 112051 0
New Zealand
Phone 112051 0
+64211531391
Fax 112051 0
Email 112051 0
zhenqiang.wu@auckland.ac.nz
Contact person for scientific queries
Name 112052 0
Dr Zhenqiang Wu
Address 112052 0
124 Shakespeare Road, North Shore Hospital, Takapuna, Auckland 0620, New Zealand
Country 112052 0
New Zealand
Phone 112052 0
+64211531391
Fax 112052 0
Email 112052 0
zhenqiang.wu@auckland.ac.nz

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
Not approved by HDEC


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
13174Study protocol  zhenqiang.wu@auckland.ac.nz Please contact PI for more information.
13175Ethical approval  zhenqiang.wu@auckland.ac.nz Please contact PI for more information.
13176Statistical analysis plan  zhenqiang.wu@auckland.ac.nz Please contact PI for more information.
13177Analytic code  zhenqiang.wu@auckland.ac.nz Please contact PI for more information.


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.