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
ACTRN12620001193965
Ethics application status
Approved
Date submitted
27/07/2020
Date registered
10/11/2020
Date last updated
10/11/2020
Date data sharing statement initially provided
10/11/2020
Type of registration
Retrospectively registered

Titles & IDs
Public title
The adequacy of user seal checking for N95 respirators compared to formal fit testing, a multicentred observational study
Scientific title
The adequacy of user seal checking for N95 respirators compared to formal fit testing in staff working in the intensive care units. A multicentred observational study.
Secondary ID [1] 301700 0
Nil known
Universal Trial Number (UTN)
U1111-1254-8633
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Protection from airborne pathogens 318152 0
COVID-19 318348 0
Condition category
Condition code
Respiratory 316171 316171 0 0
Other respiratory disorders / diseases
Public Health 316736 316736 0 0
Health promotion/education
Infection 316745 316745 0 0
Studies of infection and infectious agents

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
As part of a quality and safety initiative fit testing was provided for staff working in the intensive care units (ICU) at Epworth Richmond, Freemasons and Geelong.

Respirator Allocation
Staff were fit tested on one of the 2 brands of N95 respirators available for use and these were randomly chosen for each participant.
The masks available were either the 3M N95 Healthcare Particulate Respirator and Surgical Mask 1860 Regular or Halyard Fluidshield N95 Particulate Filter Respirator and Surgical Mask Regular.

Sensitivity Testing
Using a 3M Qualitative Fit Test Apparatus, FT-30, a qualitative fit test was performed by a 3M trained tester for each participant.
The subject placed their head within a 3M Test Hood wearing a N95 mask.
The Sensitivity Test agent, Bitter ('Bitrex', Denatonium Benzoate, 3M FT-31), was sprayed directly into the hood device using the 1st 3M nebuliser device. This is a dilute form of the 3M Bitter ('Bitrex', Denatonium Benzoate) Test Solution. Up to 30 sprays were administered. The subject alerted the tester when they could taste the solution and this was recorded on the form. The subject was required to detect the solution prior to commencement of the test, i.e. pass the sensitivity test.
A fail was inability to taste the solution at 30 sprays. The subject was offered the opportunity to return for a re-test on another day during another session. Issues were discussed with subject such as factors affecting taste, for example strong foodstuffs consumed prior to the testing.
The hood was removed and the inside was wiped with an alcoholic wipe.
A pass allowed the subject to proceed to fit testing. They cleaned their face with a face cloth and rinsed their mouth with water. After 10 minutes the subject returned to conduct the test.

User Seal Check (USC)
A user seal check (USC) is a internationally recognised means of assessing gross leakage of N95 respirators. 3M and Halyard each provide their own information on the details of the USC for their N95 respirators.
Using the randomly chosen N95 respirator the staff member conducted an ‘unassisted user seal check’. A written copy of the manufacturers’ instructions for both types of respirators were placed in front of the subject on a table as reference if required.
The tester observed the user seal check and made a note of any errors in their technique. The subject was asked if they felt the user seal was adequate and recorded a pass or fail. If the tester also observed critical errors in the user seal check i.e. poorly fitting mask, the unassisted user check was also recorded as a fail.
A fail resulted in the tester performing an ‘assisted user seal check’ on the subject e.g. making adjustments to the strap and/or mask position. If despite the tester attempting to make adjustments and the fit was obviously inadequate the respirator was recorded as a fail. This assessment identifies any gross leakage which is felt as air leakage on the hands around any point of the respirator (as per manufacturer's instructions).
An assisted user seal check fail resulted in the subject performing a user seal check on the other available mask. If either the unassisted or assisted user seal check failed on the 2nd mask the subject was offered another session to perform testing on the PAPR (CleanSpace® HALO).
Fit testing commenced with a pass of the unassisted user seal check or assisted user seal check.

Fit Test
The subject continued to wear the respirator and placed their head in the same 3M Test Hood.
The 3M Fit Testing Solution Bitter ('Bitrex', Denatonium Benzoate, 3M FT-32) was sprayed in the hood using the 2nd 3M Nebuliser device on regular timed occasions.
The participants conducted a series of 7 breathing exercises, lasting 1 min each.
The test agent is used to detect leakage via the respirator interface seal of the mask by detection of a bitter taste. If during any of the exercises the bitter solution was detected the mask was deemed to have failed and the test was ceased.
For staff failing the initial respirator type the alternative N95 respirator available was then tested (user seal check and fit testing).
The overall time allocated for each fit testing session was approximately 20 min.
Staff failing fit testing on both N95 respirators were offered a repeat session with a Powered Air Purifying Respirator (PAPR), specifically CleanSpace® HALO.

Quantitative Fit Testing
For a minority of staff who are unable to perform the test due to failure of the sensitivity test with the Sensitivity Test Agent Bitter ('Bitrex', Denatonium Benzoate, 3M FT-31), a further session was offered using the quantitative fit testing with the TSI PortaCount 8040. Briefly this is a particle counter device that counts the number of particles of a predefined particle size inside and outside the mask. This calculates the ratio of particles known as a fit factor (>100 = pass).

In this way the intent is to identify at least one successfully fitted respirator for each staff member.
Intervention code [1] 318006 0
Not applicable
Comparator / control treatment
Reference comparator is the 'fit test'
Control group
Active

Outcomes
Primary outcome [1] 324353 0
The primary outcome measured is the proportion of healthcare personnel failing fit testing on the first N95 respirator type fitted. A failure of fit testing is defined as failing both first attempt (without assistance) and second attempt (with assistance).
Qualitative fit testing failure is based on the ability to taste the test agent, BitrexTM,(Denatonium Benzoate) during any of one of the series of exercises. Based on manufacturer guidance and AS/NZS 1715:2009.
‘Without assistance’ describes the participant performing their own user seal check without assistance from the tester.
‘With assistance’ describes the tester checking the accuracy of the participants user seal check and making any required adjustments to optimise the seal.
Timepoint [1] 324353 0
N95 respirator testing on both mask types within a single session.
Secondary outcome [1] 384867 0
Proportion failing first mask without assistance.
‘Without assistance’ describes the participant performing their own user seal check without assistance from the tester.
Qualitative fit testing failure is based on the ability to taste the test agent, BitrexTM,(Denatonium Benzoate) during any one of the series of exercises. Based on manufacturer guidance and AS/NZS 1715:2009.
Timepoint [1] 384867 0
N95 respirator testing on both mask types within a single session. Those requiring PAPR fitting will be offered a subsequent 2nd session.
Secondary outcome [2] 384869 0
Proportion failing both N95 respirator types
Qualitative fit testing failure is based on the ability to taste the test agent, BitrexTM,(Denatonium Benzoate) during any one of the series of exercises. Based on manufacturer guidance and AS/NZS 1715:2009.
Timepoint [2] 384869 0
N95 respirator testing on both mask types within a single session. Those requiring PAPR fitting will be offered a subsequent 2nd session.
Secondary outcome [3] 384872 0
Failure rates for each mask type (Halyard, Apharetta, Georgia, USA; Sydney, Australia and 3M, St. Paul, Minnesota, USA; Sydney, Australia)
Qualitative fit testing failure is based on the ability to taste the test agent, BitrexTM,(Denatonium Benzoate) during any one of the series of exercises. Based on manufacturer guidance and AS/NZS 1715:2009.
Timepoint [3] 384872 0
N95 respirator testing on both mask types within a single session. Those requiring PAPR fitting will be offered a subsequent 2nd session.
As this is a cross sectional study the timepoint is at the initial testing session for N95 masks or masks.
Secondary outcome [4] 384873 0
Proportion failing both N95 masks and Powered Air Purifying Respirator (PAPR)
Qualitative fit testing failure is based on the ability to taste the test agent, BitrexTM,(Denatonium Benzoate) during any one of the series of exercises. Based on manufacturer guidance and AS/NZS 1715:2009.
Of note, the PAPR device is tested with the same method for N95, however, the motor is switched off during testing as per manufacturer guidance.
Timepoint [4] 384873 0
1st session for N95 respirator testing and 2nd session for PAPR testing.

Eligibility
Key inclusion criteria
Critical care staff trained in the use of personal protective equipment (PPE) including N95 respirators.
Minimum age
18 Years
Maximum age
No limit
Sex
Both males and females
Can healthy volunteers participate?
Yes
Key exclusion criteria
Staff not previously trained in PPE use.
Staff with previous fit testing and known failure to fit one or both of the N95 respirators used for the study.

Study design
Purpose
Screening
Duration
Cross-sectional
Selection
Defined population
Timing
Prospective
Statistical methods / analysis
The primary endpoint, number of mask-fitting failures will be assessed as the proportion of staff members experiencing failures, accompanied by 95% confidence interval. The latter will be estimated using the exact binomial (Clopper-Pearson) method (Armitage et al, 2002). A sample size for planning analysis (Cumming, 2012) was undertaken by Epworth HealthCare biostatistician A/Prof Dean McKenzie, employing Stata 16 (Stata Corporation, College Station, Texas, 2019). Precision or half-width of the corresponding exact binomial 95% confidence intervals was defined as 7.5%. The failure rate is expected to range from zero to 20%, a study of 204 nursing students observing a failure rate of 18.6% in quantitative fit testing of two 3M N95 masks.
A sample size of at least 150 healthcare workers would allow a precision of 7.5% for any error rate from zero to 20%.
The lower and upper limits of exact binomial 95% confidence interval for a range of possible failure rates, for a sample size of 150, is shown below. For a point estimate of zero failures, for example, the 95% confidence interval would be 0% to 2.4%., (Hanley & Lippman-Hand, 1983). For a point estimate of 20% failures, the 95% confidence interval would be 13.9% to 27.3%. In both cases, the 95% confidence interval is no wider than the point estimate plus or minus 7.5%.
number of errors (out of 150)= 0, % = 0.0, 95% CI = 0.0 to 2.4
number of errors (out of 150)= 2, % = 1.3, 95% CI = 0.2 to 4.7
number of errors (out of 150)= 5, % = 3.3, 95% CI = 1.1 to 7.6
number of errors (out of 150)= 10, % = 6.7, 95% CI = 3.2 to 11.9
number of errors (out of 150)= 15, % = 10.0, 95% CI = 5.7 to 16.0
number of errors (out of 150)= 25, % = 16.7, 95% CI = 8.3 to 19.8
number of errors (out of 150)= 30, % = 20.0, 95% CI = 13.9 to 27.3
As described above, the primary outcome of number of failures in fitting masks will be estimated using exact binomial 95% confidence intervals. Failures will be reported as frequencies and percentages, dimensional variables such as age will be reported as means and standard deviations or, in the presence of skewness, medians and interquartile ranges (the difference between the 25th and 75th percentiles). Appropriate statistical graphics such as bar charts and box plots will be presented.
Overall failures and proportion failing first mask fitting will be compared on clinically relevant variables such as mask type (Halyard and 3M), gender , staff (doctor, nurse), recency of prior mask use, type of training will be analysed using Poisson regression with robust standard errors in order to generate relative risks (McKenzie & Thomas, 2020). Analysis of number of attempts at mask fitting before eventual passing or failing will be analysed using appropriate methods for count data such as negative binomial regression (McKenzie et al, 1998). Possible clustering due to different types of masks fitted on the same person will be taken into account using generalized estimating equation (GEE) or mixed methods / multilevel regression (Singer & Willett, 2003).
A commonly suggested ‘rule of thumb’ in ascertaining the number of variables into regression models for binary data is at least 10 events per variable, where events are defined as the less frequent binary outcome (Austin & Steyerberg, 2015). As the error rate is expected to be less than or equal to 20% of 150 or 30 staff members, a maximum of three predictor variables or potential confounders could therefore be entered into regression models. The comparison of the performance of potential predictors of failure such as mask type, with and without potential confounders such as gender, will inform larger studies.
All statistical analyses will be conducted by, or under the direct supervision of, Epworth HealthCare biostatistician(s), using standard professional statistical software such as Stata 16 (Stata Corporation, College Station, Texas, 2019) or higher. The level of statistical significance will be set at 0.05, 2-tailed in all analyses. 95% confidence intervals will be calculated throughout. No interim analysis is planned.

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
Actual
29/05/2020
Date of last participant enrolment
Anticipated
30/11/2020
Actual
Date of last data collection
Anticipated
30/11/2020
Actual
Sample size
Target
150
Accrual to date
187
Final
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 17022 0
Epworth Richmond - Richmond
Recruitment hospital [2] 17023 0
Epworth Freemasons (Victoria Parade) - East Melbourne
Recruitment hospital [3] 17024 0
Epworth Hospital Geelong - Waurn Ponds
Recruitment postcode(s) [1] 30691 0
3121 - Richmond
Recruitment postcode(s) [2] 30692 0
3002 - East Melbourne
Recruitment postcode(s) [3] 30693 0
3216 - Waurn Ponds

Funding & Sponsors
Funding source category [1] 306137 0
Hospital
Name [1] 306137 0
Epworth HealthCare
Country [1] 306137 0
Australia
Primary sponsor type
Hospital
Name
Epworth HealthCare
Address
Epworth Hospital Richmond
89 Bridge Road
Richmond
Victoria
3121
Country
Australia
Secondary sponsor category [1] 306605 0
Individual
Name [1] 306605 0
Dr Jonathan Barrett
Address [1] 306605 0
Intensive Care Department
Epworth Richmond Hospital
89 Bridge Road
Richmond
Victoria
3121
Country [1] 306605 0
Australia

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 306485 0
Epworth HealthCare Human Research Ethics Committee
Ethics committee address [1] 306485 0
89 Bridge Road
Richmond
Victoria
3121
Ethics committee country [1] 306485 0
Australia
Date submitted for ethics approval [1] 306485 0
05/06/2020
Approval date [1] 306485 0
26/06/2020
Ethics approval number [1] 306485 0
EH2020-580

Summary
Brief summary
The current novel coronavirus (COVID-19) pandemic has highlighted the importance of safety for healthcare personnel particularly in relation to PPE. There are few studies in healthcare comparing user seal checks to the reference standard of formal fit testing.
Better evidence is urgently required to guide current practice.
Epworth ICUs use of one of two brands of disposable N95 respirators or a Powered Air Purifying Respirator (PAPR).
Staff receive training in the use of the various masks, and are expected to perform a user seal check on each occasion.
Fit testing has not previously been performed.
This study aims to assess the adequacy of the user seal check, compared to the reference standard of fit testing for various models of N95 respirators.
Trial website
Trial related presentations / publications
Public notes
This study satisfies the conditions set out in section 5.1.18 of the NHMRC National Statement on Ethical Conduct in Human Research (2007) and has been ethically reviewed at an institutional level. This intervention is a quality and safety initiative using research methodology and hence implementation occurred prior to ethics review. Ethics review was hence required for publication not for commencement of the intervention itself.
The above project has been considered by the Research Development and Governance
Unit (RDGU) and has demonstrated no apparent risk beyond minor inconvenience to participants. The study was reviewed by the Institutional Review Board and the study was approved as low risk process and given approval as a low risk quality and safety study / health services research.
1st staff member tested on 29 May 2020 as part of quality and safety initiative for staff during the COVID-19 pandemic. Ethics approval received a month later. Therefore ethics approval after commencement but before completion of data collection. The intent is collection of data until the sample size is achieved. Although the initial intention was to prospectively register on both a trials website and with HREC the intervention could not wait for these steps to occur. We are satisfied if that means the listing now needs to be retrospective rather than prospective, although we would note that it is a mixture of both. this just reflects the reality of the need for urgent evaluation during the COVID pandemic.

Contacts
Principal investigator
Name 103590 0
Dr Helen Cass
Address 103590 0
Epworth HealthCare
Epworth Hospital Richmond
89 Bridge Street
Richmond
Victoria
3121
Country 103590 0
Australia
Phone 103590 0
+61 3 9506 3193
Fax 103590 0
Email 103590 0
[email protected]
Contact person for public queries
Name 103591 0
Jonathan Barrett
Address 103591 0
Epworth HealthCare
Epworth Hospital Richmond
89 Bridge Street
Richmond
Victoria
3121
Country 103591 0
Australia
Phone 103591 0
+61 3 9506 3193
Fax 103591 0
Email 103591 0
[email protected]
Contact person for scientific queries
Name 103592 0
Jonathan Barrett
Address 103592 0
Epworth HealthCare
Epworth Hospital Richmond
89 Bridge Street
Richmond
Victoria
3121
Country 103592 0
Australia
Phone 103592 0
+61 3 9506 3193
Fax 103592 0
Email 103592 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
Our institution does not yet have a formal data sharing policy.
The primary author is happy to share study protocols, ethics approval and potentially data on request pending institutional approval post request.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
8566Study protocol    380133-(Uploaded-22-07-2020-15-32-13)-Study-related document.docx
8568Ethical approval    380133-(Uploaded-22-07-2020-13-08-07)-Study-related document.pdf



Results publications and other study-related documents

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

Documents added automatically
SourceTitleYear of PublicationDOI
Dimensions AIThe adequacy of user seal checking for N95 respirators compared to formal fit testing: A multicentred observational study2022https://doi.org/10.1016/j.aucc.2022.08.012
N.B. These documents automatically identified may not have been verified by the study sponsor.