Trial Review

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

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
ACTRN12623001130651
Ethics application status
Approved
Date submitted
20/09/2023
Date registered
2/11/2023
Date last updated
2/11/2023
Date data sharing statement initially provided
2/11/2023
Type of registration
Prospectively registered

Titles & IDs
Public title
Un-Necessary Tests in Emergency Departments (UNTIE), a Stepped-wedge Cluster Randomised Trial
Scientific title
A Step-wedged cluster randomised trial comparing audit feedback vs observation in improving appropriate pathology test ordering practices in Emergency Departments
Secondary ID [1] 306770 0
Nil Known
Universal Trial Number (UTN)
U1111-1297-0386
Trial acronym
UNTIE
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Abdominal pain severe
325797 0
Trauma 325798 0
Back pain atraumatic (requiring admission) 325799 0
Cellulitis (requiring admission)
 325800 0
Chest pain 325801 0
Snake bite 325802 0
Confusion
 325803 0
Cerebrovascular accident
 325804 0
Diabetic Ketoacidosis
 325805 0
Fever for investigation 325806 0
Shortness of breath 325807 0
Fractures 325808 0
Fractures minor for theatre >55 year old
 325809 0
Gastrointestinal bleed
 325810 0
Pneumonia 325811 0
Liver disease
 325812 0
Oncology patients (febrile neutropenia)
 325813 0
Overdose (significant)
 325815 0
Vaginal bleeding 325816 0
Condition category
Condition code
Emergency medicine 323133 323133 0 0
Other emergency care

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
This is a ten week step-wedged randomized control trial involving the generation of a novel UNTIE score for each of the 35 emergency department presentations covered by the study. The generation of these scores will be throughout the study period (irrespective of intervention phase) to provide baseline data collection. UNTIE scores are calculated as an UNTIE U and UNTIE N score. An UNTIE N score is the percentage of tests listed as ‘Perform’ for the specific presentation which were actually performed within the ED visit. A UNTIE-N score of <100% indicates potential under-utilization of necessary tests. An UNTIE U score is a percentage calculated as: NUMERATOR: number of tests actually performed within the ED visit, which are NOT listed in the ‘Perform’ or ‘Consider or Ask Supervisor’ categories for the specific presentation, divided by: DENOMINATOR: the number of tests listed within the ‘Perform’ category. The denominator is designed to capture the complexity of respective presenting problems. A UNTIE-U score of >0% indicates potential over-utilization of tests.. We will automate this to generate indicators on a fortnightly basis.

The determination of 'Perform', 'Consider' or 'Ask supervisor categories' are based on a published joint statement by the Australian College of Emergency Medicine (ACEM) and the Royal College of Pathologists of Australia (RCPA) 2018 titled Pathology Testing in the Emergency Department guideline. This guidelines covers the 35 emergency department presentations reviewed in our study and provides guidance on investigation use for each of these presentations. This guidance forms the foundation for the calculation of the various UNTIE scores.

Each site will then be randomized to an audit feedback intervention based on prospectively collected data on emergency department UNTIE score performances. The audit feedback intervention will involve fortnightly interactions with clinical staff including medical, nursing and technical and clinical assistant workforces outlining department performance for test ordering and encouraging patterns of behavior more compliant with suggested guidelines. Staff performing the audit will be a combination of local champions and research staff who are familiar with the models of care functioning within each emergency department. The audit feedback process will be multimodal and involve face to face presentations, recorded presentations, routine emails, and the use of posters and visual stimuli in appropriate clinical spaces to prompt behavioral change. All audit feedback data will be recorded in a study specific database. Data on ongoing performance will then be recorded over the duration of the intervention period. Every ten weeks an additional department will be included in the randomisation process.
Intervention code [1] 323230 0
Behaviour
Intervention code [2] 327251 0
Treatment: Other
Comparator / control treatment
Routine care with standard investigations as per usual practice for that particular site. Routine care would involve standard investigation ordering based on current practice and clinician preference, There will be no intervention via audit feedback using the UNTIE score for this period. This is a step-wedged randomised controlled trial. There is no active intervention in the control group but the control group is not historical.

A baseline retrospective data set for each site over 3 years will be used (2020 - 2023) will also be taken into account.
Control group
Active

Outcomes
Primary outcome [1] 330905 0
Outcome Measurement: The primary outcome is UNTIE-U, This will be generated blinded to an ED’s intervention/comparator status, and generated individually for each of the 35 selected ED presentations.

UNTIE-U (UNTIE-Unnecessary) is a percentage calculated as:
1..NUMERATOR: number of tests actually performed within the ED visit, which are NOT listed in the ‘Perform’ or ‘Consider or Ask Supervisor’ categories according to clinical guidelines for the specific presentation, divided by:
2.DENOMINATOR: the number of tests listed within the ‘Perform’ category.
The denominator is designed to capture the complexity of respective presenting problems.

An UNTIE U score for all 35 emergency department presentations will be monitored throughout the audit feedback intervention to determine the presence of any behavioral change.

Data for calculation of UNTIE U score will be obtained from the centralized pathology database and linked to patient records locally.
Timepoint [1] 330905 0
Fortnightly review over a 50 week period, with a new site added to the intervention arm at 10 week intervals in a Step-wedged manner. .
Secondary outcome [1] 408022 0
Our secondary outcome is UNTIE-N. This will be generated blinded to an ED’s intervention/comparator status, and generated individually for each of the 35 selected ED presentations.

UNTIE-N (UNTIE-Necessary) is defined as the percentage of tests listed as ‘Perform’ according to clinical guidelines for the specific presentation, which were actually performed within the ED visit.

Data for calculation of UNTIE N score will be obtained from the centralized statewide pathology database and linked to patient records
Timepoint [1] 408022 0
Fortnightly review over a 50 week period, with a new site added to the intervention arm at 10 week intervals.

Eligibility
Key inclusion criteria
All consecutive visits at the four Western Sydney Local Health District ED’s and Nepean Hospital over each time period, where the patient presentation maps to one of the 35 presenting problems covered in the guideline.
- Abdominal pain severe (upper/epigastric)
- Abdominal pain severe (lower)
- Back pain atraumatic
- Cellulitis
- Chest pain suspected IHD
- Chest pain - suspected PE
- Confusion
- Syncope
- Stroke
- Diabetic Ketoacidosis
- Fever of unknown cause
- Fever in returned traveler
- Fractures neck of femur or other major long bone
- Fractures requiring theater
- Gastrointestinal bleeding
- Jaundice for investigation
- Liver disease
- Oncology patients (febrile neutropenia)
- Overdose
- Per vaginal bleeding in 1st trimester
- Pneumonia
- Pyelonephritis
- Renal colic
- Renal disease
- Seizures (1st episode)
- Seizures (recurrent)
- Septic joint
- Sepsis
- Snake bite
- Shortness of breath (asthma)
- Shortness of breath (suspected acute pulmonary oedema)
- Shortness of breath (suspected chronic obstructive airways disease)
- Trauma (major)
- Trauma (minor)
- Warfarin therapy
Minimum age
No limit
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
Not attending emergency department
Presenting with a diagnosis beyond the 35 mapped presentations

Study design
Purpose of the study
Treatment
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Allocation is not concealed
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Centralized randomization by computer program performed by an independent statistician. Trial site notified of intervention group 1 week prior to intervention.
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Parallel
Other design features
This is a stepped-wedge cluster randomised control trial. The order of intervention will be randomised, and determined by an independent external statistician. There will be five clusters (one ED per cluster), and five steps each correlating with the change of medical terms standardised across all sites as dictated by HETI NSW. Dates for step commencement are chosen to coincide with the five-term year for Junior Medical Officers. UNTIE-U and UNTIE-N will be generated and analysed monthly for all sites, to inform sites with audit-feedback sessions. In order to avoid anticipatory effects, sites are not notified of its intervention/comparator status until two weeks prior to intervention commencement (to allow for scheduling of the introductory presentation).
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
The stepped-wedge data will be analysed using generalised linear mixed models, with UNTIE-U/N indicators for individual presentations as the unit of observation. A priori, the data will be modelled with random effect for cluster and fixed effect for each step. Additional random effects considered include type of presentation, and level of the attending medical officer (Junior Medical Officer/Registrar/Consultant). The possibility of lag between intervention and effect will be assessed by considering total length of intervention at the ED, as a potential effect modifier. The possibility of a ‘rising tide’ secular trend (that is, test utilisation improved over time due to some other external factor), will be explored by adjusting for calendar time. All analysis will be carried out using Stata/SE 12.1 (Stata Corp, TX, USA). Following the principle of intention-to-treat, observations will be classified according to the randomised intervention dates, regardless of whether the intervention occurred on time.

Power was calculated using the Hemming and Girling estimation implemented in Stata, based on our original 4-site design. The Ministry has asked us to add a fifth site and the power of this new design will be higher. Input included: hypothesized UNTIE-U (0.25 and 0.1875), 4 steps, 1 cluster randomised at each step, average cluster size of 5498, significance level of 0.05. The IntraCluster correlation for UNTIE-U is unknown, and we varied it from 0.01 to 0.99. With a total of 109,960 observations (5498x 4EDs x 5periods), the power we have for detecting the hypothesised change was estimated to be >0.99, regardless of choice of IntraCluster correlation. (As suggested, this has been checked by our statistical consultant)

Recruitment
Recruitment status
Not yet recruiting
Date of first participant enrolment
Anticipated
12/02/2024
Actual
Date of last participant enrolment
Anticipated
27/01/2025
Actual
Date of last data collection
Anticipated
27/01/2025
Actual
Sample size
Target
100000
Accrual to date
Final
Recruitment in Australia
Recruitment state(s)
NSW
Recruitment hospital [1] 24199 0
Nepean Hospital - Kingswood
Recruitment hospital [2] 24200 0
Westmead Hospital - Westmead
Recruitment hospital [3] 24201 0
Auburn Hospital & Community Health Services - Auburn
Recruitment hospital [4] 24202 0
Blacktown Hospital - Blacktown
Recruitment postcode(s) [1] 39732 0
2747 - Kingswood
Recruitment postcode(s) [2] 39733 0
2145 - Westmead
Recruitment postcode(s) [3] 39734 0
2144 - Auburn
Recruitment postcode(s) [4] 39735 0
2148 - Blacktown

Funding & Sponsors
Funding source category [1] 311110 0
Government body
Name [1] 311110 0
NSW Health
Country [1] 311110 0
Australia
Primary sponsor type
Government body
Name
Western Sydney Local Health District
Address
Hawkesbury Road Westmead Australia 2145 NSW
Country
Australia
Secondary sponsor category [1] 312460 0
None
Name [1] 312460 0
Address [1] 312460 0
Country [1] 312460 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 310640 0
Western Sydney Local Health District Human Research Ethics Committee
Ethics committee address [1] 310640 0
WSLHD Office, Westmead Hospital Campus Institute Road, Westmead NSW 2145 PO Box 533, Wentworthville NSW 2145
Ethics committee country [1] 310640 0
Australia
Date submitted for ethics approval [1] 310640 0
27/07/2017
Approval date [1] 310640 0
02/02/2018
Ethics approval number [1] 310640 0
AU RED HREC/17/WMEAD/274

Summary
Brief summary
Hypothesis: We hypothesise that audit feedback will be more effective than observation in improving appropriate pathology test ordering practices in Emergency Departments
 
Population: All/consecutive visits at the four western Sydney local health district's (WSLHD) emergency departments (ED) and Nepean Hospital over each 10 weeks time period, where the patient presentation maps to one of the 35 presenting problems covered in the guideline.

Intervention: Automated generation of UNTIE U and UNTIE N scores for each presentation generated by reviewing initial pathology testing ordered. After site randomisation to intervention, data will then be presented to clinical staff in the form of an audit feedback process involving multimodal investigations. UNTIE scores will continue to be collected and department performance reviewed in response to the audit feedback process

Comparison: Usual practice- there will be no contact between the research team and an ED that is not in an intervention phase.

Outcome: The primary outcome is UNTIE-U, secondary outcome UNTIE-N. Both will be generated blinded to an ED’s intervention/comparator status.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 118362 0
Prof Clement Loy
Address 118362 0
Westmead Hospital, Cnr Hawkesbury Road and, Darcy Rd, Westmead NSW 2145
Country 118362 0
Australia
Phone 118362 0
+61 411641912
Fax 118362 0
Email 118362 0
[email protected]
Contact person for public queries
Name 118363 0
Pramod Chandru
Address 118363 0
Westmead Hospital, Cnr Hawkesbury Road and, Darcy Rd, Westmead NSW 2145
Country 118363 0
Australia
Phone 118363 0
+61 419798902
Fax 118363 0
Email 118363 0
[email protected]
Contact person for scientific queries
Name 118364 0
Pramod Chandru
Address 118364 0
Westmead Hospital, Cnr Hawkesbury Road and, Darcy Rd, Westmead NSW 2145
Country 118364 0
Australia
Phone 118364 0
+61 419798902
Fax 118364 0
Email 118364 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
This was a condition of data release to the researchers. All data is stored on intra-hospital networks with data analysis occurring within the hospital network. Also we are anticipating over a million patient encounters with multiple data points for each encounter, making data sharing not practical.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
18492Ethical approval    383826-(Uploaded-03-03-2023-23-05-00)-Study-related document.docx



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.