Evaluation of new and conventional N95 fit-testing protocols on equivalence and reproducibility: a study protocol Authors: S.C. Lam, I.T.S. Yu, M. O’Donoghue, K.K. Leung, A.C.Y. Chong, L.L.K. Sung, A. Odetayo, L.K.P. Suen Published: 2026 ABSTRACT SummaryAn N95 respirator is designed to fit tightly and filter airborne particles. Healthcare workers (HCWs), especially those handling aerosols, must complete fit testing to choose the right mask and receive proper training. Traditionally, Occupational Safety and Health Administration (OSHA)’s eight-step protocol was used for N95 fit testing, but in 2019, a shorter four-step version was introduced, cutting testing time. Although the OSHA claims both protocols are equivalent, there is limited empirical support and no research on the reproducibility of the conventional method.The proposed cross-sectional observational study will evaluate the equivalence (i.e. agreement) and reproducibility of the fit-testing results of three different designed respirators tested among Chinese HCWs using both new and conventional fit-testing protocols.The kappa agreement between the new and conventional protocols can serve as an important point of reference for fit-testing policies. If there is a high level of equivalence, both fit-testing protocols could be simultaneously implemented as they would be expected to produce similar results. In contrast, if the conventional protocol is more stringent than the new protocol, conservatively, it would be necessary to exercise caution and suspend implementation of the new protocol. Additionally, the reproducibility of respirator fit testing must be determined. Examining the reproducibility could eliminate uncertainty regarding the ability of a respirator to maintain a proper fit in a real clinical setting.Trial registrationThis study was registered to ISRCTN registry: ISRCTN15612510. The study was retrospectively registered on 28th May 2025; corresponding link: https://doi.org/10.1186/ISRCTN15612510. Background Background Global outbreaks of contagious respiratory illnesses have heightened awareness of occupational safety among healthcare workers (HCWs) [1,2]. An N95 filtering tight-fitting half-facepiece respirator (hereafter, N95 respirator) is a specialised mask designed to securely fit the face and efficiently filter airborne particles [[3], [4], [5]]. For an ill-fitting respirator, the average infiltration of an ambient aerosol was 33%, compared with only 4% for well-fitting respirators [6]. Leakage (i.e. penetration) can occur when there is a gap between the respirator and the wearer’s face. This gap (i.e. face-seal leakage) [7] may enable the infiltration of airborne contaminants into the wearer’s respiratory area, compromising the effectiveness of protection. To ensure sustainable occupational safety, reputable authorities such as the National Institute for Occupational Safety and Health, World Health Organization, Centers for Disease Control and Prevention, and Hospital Authority in Hong Kong require HCWs to undergo N95 fit testing before using an N95 respirator, particularly if they are performing aerosol-generating procedures [1,[3], [4], [5],8]. N95 fit testing facilitates both the selection of a suitable respirator and the training of its user. Quantitative fit testing is an established technique used to ascertain whether a closely fitting respirator adequately fits the wearer. An electronic device is used to quantify the proportions of specific air particles within and outside the breathing area when the user is wearing the respirator. The resulting ratio, known as the fit factor (FF), indicates the extent of leakage [[3], [4], [5], [6],9]. The primary consideration when assessing the fit of an N95 respirator for a population is the fit rate, defined as the average fit-testing pass rate. Before 2019, quantitative N95 fit testing was conducted worldwide using an eight-step protocol, namely Occupational Safety and Health Administration (OSHA) 29CFR 1910.134 [10]. This protocol was established and implemented 20 years ago by the OSHA under the United States Department of Labour, and it involves quantitative fit testing using an ambient aerosol condensation nuclei counter [[3], [4], [5], [6],9,10]. According to the fit-testing procedures documented in 1910.134 App A, an HCW must perform eight exercises: 1) normal breathing, 2) deep breathing, 3) side-to-side head movement, 4) up-and-down head movement, 5) talking (a standard set of passage was provided for reading), 6) grimacing, 7) bending over, and 8) normal breathing [4]. These exercises replicate the physical movements of HCWs in a clinical environment. For every exercise except grimacing, an individual FF of each exercise contributes to the overall FF, which ranges from 0 to 200. Exercises 1, 2, and 8, which involve static physical movements, generally have higher individual FF, both in practice and in theory, than other exercises [5]. These scores may contribute to an overall increase in the FF. Given these findings and the possibility of overestimating overall FF, a new four-step protocol [10] was introduced, eliminating the three static exercises (1, 2, and 8), and retaining the four more intense head and body movements. In theory, this new four-step fit-testing protocol should require stricter criteria for passing than the conventional protocol. However, it is unclear why the OSHA has claimed that the equivalence of the new and conventional protocols has been demonstrated [10]. There remains a dearth of empirical evidence to substantiate this claim. Impacts of the new four-step protocol The literature focusses on crucial factors pertaining to N95 respirator use, namely the user seal check, fit rate, and real-time leakage [[3], [4], [5], [6],9,[11], [12], [13], [14], [15], [16], [17]]. Research involving American (N = 1887) and Canadian (N = 1271) HCWs has revealed fit rates of 90%–100% for several respirator brands and models developed in Western countries [11,12]. Locally, several investigations have examined the fit rates for different types of N95 respirators used by Chinese HCWs in hospitals [3,4,9]. Unlike studies involving Western populations, the fit rates (successful passes) were consistently low, ranging from 55% to 69%. Our team suggested the need to ensure a proper fit between N95 respirators and the facial dimensions (length and width) of a certain population [13]. Furthermore, we verified that the user seal check (i.e. a wearer’s self-subjective evaluation by positive and negative pressure check) was not a sufficient substitute for quantitative fit testing [[3], [4], [5]]. Real-time leakage (also named face-seal leakage) refers to the immediate and ongoing measurement of the amount of air that escapes around the edges of a fitted N95 respirator while it is worn in clinical settings. The reliability of the real-time protection provided by respirators during clinical procedures, particularly those involving a wide range of body movements, has been questioned in various local studies [14,16,17]. Our team observed significant increases in the leakage of particles larger than 0.3, 0.4, 1.0, and 4.0 μm within properly fitting respirators prior to and during the processes of suctioning, nasogastric tube insertion, and resuscitation (P < 0.001) [14,16]. However, obtaining these results required the prior identification of a fit respirator, which was achieved using the conventional eight-step fit testing [[3], [4], [5], [6],9,[11], [12], [13], [14], [15], [16], [17]]. The introduction of new four-step protocols could lead to results that contradict the literature if the equivalence of the two protocols is uncertain. Reproducibility, which is closely associated with replicability and repeatability, forms the basis of the scientific method [18]. Reproducibility in a study means it can generate consistent and reliable results repeated, whether through an experiment, observational study, or statistical analysis of collected data. In the context of classical measurement theory, reproducibility can be considered to represent stability, specifically in relation to test–retest reliability during instrument validation [[19], [20], [21], [22], [23]]. It is unclear whether repeated fit testing of a respirator according to the conventional eight-step protocol would yield the same result (i.e. FF) as the initial fit testing when performed over a duration of 2 weeks or longer. Examining the reproducibility of respirator fit testing could eliminate uncertainty regarding the ability of a respirator to maintain a proper fit in a real clinical setting. Nevertheless, if the reproducibility of fit-testing results remains unsatisfactory despite the implementation of various established guidelines to ensure the similarity of fit-testing settings (refer to Figure 1), this will necessitate further investigation into methods to mimic the fit-testing environment and clinical setting associated with respirator use. In summary, the reproducibility of fit testing according to either the conventional or new protocol remains unaddressed.Figure 1The flow of respirator fit testing and data collection. The flow of respirator fit testing and data collection. Private and hospital authority institutions in Hong Kong provide HCWs with N95 respirators from two to three primary categories for aerosol-generating procedures, such as suctioning and resuscitation [8]. Following the COVID-19 pandemic, certain classical respirator models (e.g. cup-shaped 3M 1860s and three-panel horizontal flat-fold 3M 1870+) and recently introduced models (e.g. vertical flat-fold OKSafe N95-01 and pouch-type Master & Frank MF-01) were added to the list of available supplies. Nevertheless, the fit rate and real-time leakage of these new respirator types have not been thoroughly assessed among Chinese HCWs. Prior to implementing the new four-step protocol, our research team accumulated extensive expertise and a commendable history of conducting relevant studies. Since 2011, we have proven that the fit rates of 3M and Kimberly–Clark pouch-type respirators are suboptimal for Chinese HCWs due to variations in facial anthropometrics between Chinese and Western individuals [[3], [4], [5],13]. This finding indicated the need to eliminate certain models from the hospital authority’s procurement policy in Hong Kong. It is important to note that the facial dimension data underpinning fit-testing protocol techniques have been derived from limited population samples and may not accurately represent facial characteristics across different ethnic groups or nationalities [24,25]. Additionally, the user seal check was not a sufficient substitute for quantitative fit testing [[3], [4], [5]]. The issue of real-time leakage during clinical procedures for HCWs wearing a ‘fit’ N95 respirator was identified [[14], [15], [16]]. Additionally, the significance of mask usability in ensuring compliance with respirator use was demonstrated [15]. In 2020, we conducted a pioneering study that compared the conventional 3M N95 respirators with the NASK head-loop model. It focussed on evaluating the fit rates and real-time leakage of these two types of respirators. Consequently, there was conclusive proof that the 3M model, which was in short supply [26], could be substituted with locally produced NASK respirators during the COVID-19 pandemic [15]. However, the aforementioned studies and their findings are directly and indirectly based on the conventional eight-step protocol in respiratory fit testing. If the equivalence between the new and conventional protocols is uncertain, the credibility of those findings is compromised to a certain extent. Additionally, the introduction of the new four-step respirator fit-testing protocol alongside the conventional eight-step protocol led different hospital clusters under the Hospital Authority to adopt different protocols. Furthermore, even hospitals under the same cluster adopted different protocols. Therefore, the proposed study aims to provide empirical evidence to inform fit-testing policy by investigating the equivalence and reproducibility of fit-testing results obtained using the new and conventional protocols. Methods/design Aims and hypotheses Study aims 1.To investigate the equivalence of fit-testing results (i.e. pass or fail) of three different designed respirators tested among Chinese HCWs using the new and conventional protocols.2.To evaluate the reproducibility of fit-testing results, specifically the FF, of the three different designed respirators tested among Chinese HCWs using the new and conventional protocols. The null hypothesis will be adopted. 1.To investigate the equivalence of fit-testing results (i.e. pass or fail) of three different designed respirators tested among Chinese HCWs using the new and conventional protocols. To investigate the equivalence of fit-testing results (i.e. pass or fail) of three different designed respirators tested among Chinese HCWs using the new and conventional protocols. 2.To evaluate the reproducibility of fit-testing results, specifically the FF, of the three different designed respirators tested among Chinese HCWs using the new and conventional protocols. The null hypothesis will be adopted. To evaluate the reproducibility of fit-testing results, specifically the FF, of the three different designed respirators tested among Chinese HCWs using the new and conventional protocols. The null hypothesis will be adopted. Study hypotheses We hypothesise that there is no equivalence in the fit-testing results (i.e. pass or fail) of three different respirator designs evaluated among Chinese HCWs using both new and conventional protocols. We further hypothesise that the fit-testing results of the three differently designed respirators are not reproducible. Study design This proposed methodological study will have a cross-sectional observational design with repeated measures, according to the Guidelines for Reporting Reliability and Agreement Studies [27]. The proposed cross-sectional observational study will evaluate the equivalence (also known as agreement) and reproducibility of the fit-testing results of three different designed respirators tested among Chinese HCWs using the new and conventional fit-testing protocols. A cross-sectional design is appropriate for testing the reliability of a protocol (including agreement and reproducibility) and is commonly used in instrument validation [[19], [20], [21], [22], [23]]. Figure 1 presents a flow diagram of the proposed study. Recruitment and procedures Study procedure Recruitment HCWs will be recruited through online advertising and invited to participate. Screening and consent Prospective participants will undergo screening based on the inclusion criteria. Those who fulfil the criteria will be invited to participate in the study and given a consent form that requires a signature and provides an explanation of the study. Baseline data collection A research nurse will gather data on the fit-testing outcomes for all three respirators according to the new and conventional protocols, as well as information on the participants’ sex, age, and body weight (necessary for fit testing). Optional repeat testing Some participants will be randomly invited using a lucky draw to repeat the fit testing of all three respirators two weeks later. Participant compensation As an incentive and compensation for their participation, participants will receive HK$200 (US$25.48) and HK$300 (US$38.22) cash coupons at the first fit-testing session and the repeated session, respectively. Data collection phase The data collection process will have four distinct phases: room and system preparation, registration, a training session, and a fit-testing session. The participants’ demographic information, including their sex, age, and body weight, along with the results of the user seal check and fit testing conducted using both the new and conventional protocols, will be documented in a data sheet. Environment and system preparation To regulate environmental variables that may impact the outcome of fit testing, all data will be collected in a designated, air-conditioned room with an area of 10 m2. The temperature will be maintained below 23°C, while the humidity level will be kept approximately below 75% [[3], [4], [5]]. Comprehensive daily checks of the fit-testing equipment will be conducted to ensure the proper performance of the system. These checks will include evaluating ambient particles, zero particles, valves, and the system itself [16,28]. Registration During registration, all participants will be obligated to provide signed consent and present themselves in a manner consistent with clinical duty (e.g. securing long hair and shaving where applicable). Training session Participants will attend a standardised 30-min training session delivered by a licenced nurse trained in fit-testing procedures. The session will include the following:•Instruction on the correct donning of standardised N95 respirator•Guidance on performing a user seal check•Video demonstration•Hands-on-practice •Instruction on the correct donning of standardised N95 respirator Instruction on the correct donning of standardised N95 respirator •Guidance on performing a user seal check Guidance on performing a user seal check •Video demonstration Video demonstration •Hands-on-practice Hands-on-practice The trained nurses will evaluate the donning procedure and user seal check method of each participant through re-demonstration before proceeding to the next session. Time allocation In addition to the time spent on registration and the training session, which will vary among participants, an additional 60 min will be required to collect the remaining data. Fit-testing session Participants will undergo user seal checks and fit testing for three types of N95 respirators using both new and conventional protocols. The sequence in which each respirator and protocol is tested will be randomly assigned to the participants in a sealed envelope. The aforementioned procedures have been reported and recommended in the literature [[3], [4], [5],[14], [15], [16]]. Study participants and settings The study is being conducted in Hong Kong. Participants will be selected and enlisted through online advertising and invitations to collaborators, including HCWs (e.g. doctors, nurses, and allied healthcare professionals) employed in non-governmental organisations or hospitals, as well as students pursuing healthcare disciplines in institutions. The student-to-staff ratio among participants will be 2:1. The collaborating organisations employ HCWs who directly administer treatment and care to patients. Data will be collected at the Integrative Health Centre located in the School of Nursing, Tung Wah College. The centre is equipped with fit-testing devices, which have been used to provide fit-testing services to HCWs in the community. Eligibility criteria Inclusion criteria The inclusion criteria will be as follows:1.Students pursuing healthcare disciplines or HCWs who are licenced staff members providing direct patient care.2.Age of 18 years or older. 1.Students pursuing healthcare disciplines or HCWs who are licenced staff members providing direct patient care. Students pursuing healthcare disciplines or HCWs who are licenced staff members providing direct patient care. 2.Age of 18 years or older. Age of 18 years or older. Exclusion criteria HCWs with a documented medical history of chronic respiratory disease or conditions such as asthma, congestive heart failure, or coronary heart disease, as well as those who are currently pregnant, will be excluded [16]. Outcomes and measurements Primary outcomes Primary outcomes include the FF (range: 0–200) and the pass rate of fit testing (pass or fail). A validated system will be used to measure these variables. Measurements Quantitative fit testing will be conducted using a PortaCount Respirator Fit Tester system (model Pro+ 8048, TSI Incorporated, Minnesota, USA; https://tsi.com/), a recognised technology for counting airborne particles [[3], [4], [5],9,16]. This system uses a compact, continuous condensation nuclei counter based on technology acknowledged by the OSHA for quantifying airborne particles of 0.02–1.00 mm in diameter [10]. For the conventional eight-step protocol, OSHA 29CFR 1910.134, the participant will perform eight exercises: normal breathing, deep breathing, side-to-side head movement, up-and-down head movement, talking (a standard set of passages will be provided for reading aloud), grimacing, bending over, and normal breathing [[3], [4], [5], [6],9]. The grimace is an intentional attempt to disrupt the face seal and observe the respirator’s ability to subsequently reset itself. For the new four-step protocol, OSHA Fast-filtering Face, the participants will be required to perform only four exercises: side-to-side head movement, up-and-down head movement, talking, and bending over [10]. Table I compares the new and conventional protocols, including screen captures from the PortaCount system.Table IComparison between a new 4-step protocol and the conventional 8-step protocolTable IN95 fit testingTime (sec)4-step protocolTime (sec)8-step protocol1Approval codeOSHA Fast-filtering faceOSHA 29CFR 1910.1342Exercises•Normal breathing60Yes•Deep breathing60Yes•Head side to side30Yes60Yes•Head up and down30Yes60Yes•Talking30Yes60Yes•Grimace15Yes•Bending over50Yes60Yes•Normal breathing60Yes3Total fit test time2:29 min7:15 min4SafetySameSample of screen captured Table IComparison between a new 4-step protocol and the conventional 8-step protocolTable IN95 fit testingTime (sec)4-step protocolTime (sec)8-step protocol1Approval codeOSHA Fast-filtering faceOSHA 29CFR 1910.1342Exercises•Normal breathing60Yes•Deep breathing60Yes•Head side to side30Yes60Yes•Head up and down30Yes60Yes•Talking30Yes60Yes•Grimace15Yes•Bending over50Yes60Yes•Normal breathing60Yes3Total fit test time2:29 min7:15 min4SafetySameSample of screen captured Comparison between a new 4-step protocol and the conventional 8-step protocol •Normal breathing Normal breathing •Deep breathing Deep breathing •Head side to side Head side to side •Head up and down Head up and down •Talking Talking •Grimace Grimace •Bending over Bending over •Normal breathing Normal breathing An overall FF greater than 100 will indicate a ‘pass’, indicating that the tested respirator fits the wearer properly. The PortaCount Respirator Fit Tester system will be checked daily to ensure consistent performance of the machine and system. Figure 3 displays the fit-tester system, tubing connection, and respirator. The cup-shaped 3M 1860s, three-panel horizontal flat-fold 3M 1870+, and vertical flat-fold OKSafe N95-01 respirator models have been selected due to their extensive usage in local clinical environments and their representation of three distinct designs. The results will provide important information for increasing the generalisability of testing protocols locally and internationally. Descriptions of the N95 respirators selected for the proposed study are presented in Figure 2.Figure 2Introduction of selected N95 respirators for the study. ∗Some data and photos are adapted from Lam et al. (2024).Figure 3Quantitative fit testing system, tubing connection, and respirators. ∗The figure is adapted from Lam et al. (2024), page 9. Introduction of selected N95 respirators for the study. ∗Some data and photos are adapted from Lam et al. (2024). Quantitative fit testing system, tubing connection, and respirators. ∗The figure is adapted from Lam et al. (2024), page 9. Sample size estimation A sample size for a study is determined based on the requirement for kappa agreement [27,29]. Given a null hypothesis kappa (k) value of 0.40, a pass rate of 0.70 [[3], [4], [5]], and a power of 80%, the required sample size was determined to be 85. Assuming a 20% attrition rate, a final sample size of 102 will be needed, similar to the sample sizes reported in clinical studies [21,23]. Furthermore, a conventional formula for calculating the sample size in a reproducibility study related to the intraclass correlation coefficient (ICC) recommends a sample larger than 60 validly paired data points, assuming an expected ICC of 0.80 with a 95% confidence interval of 0.20 and a 20% attrition rate [30]. Data analysis Both descriptive and inferential statistics will be performed using SPSS statistical software (IBM, New York, USA; https://www.ibm.com/contact/global). The demographics, FF, and fit rate will be reported using descriptive statistics such as the mean, standard deviation, percentage, and frequency. Table II illustrates the concept of data preparation for inferential statistical analysis.Table IIContingency table and test statisticsTable IIResultFailed fit testing by the new protocol (4 steps)Passed fit testing by the new protocol (4 steps)Failed fit testing by the conventional protocol (8 steps)A (true positive)B (false positive)Passed fit testing by the conventional protocol (8 steps)C (false negative)D (true negative)Testing itemsMethodsStatistic methods/sampleRequirementEquivalence (agreement)Agreement of results between two protocolsCohen’s kappa statistics/102 participants per respiratork ≥ 0.70Reproducibility (stability)2-week test-retest reliabilityICC/60 participantsICC ≥ 0.75ICC, intraclass correlation coefficient. Table IIContingency table and test statisticsTable IIResultFailed fit testing by the new protocol (4 steps)Passed fit testing by the new protocol (4 steps)Failed fit testing by the conventional protocol (8 steps)A (true positive)B (false positive)Passed fit testing by the conventional protocol (8 steps)C (false negative)D (true negative)Testing itemsMethodsStatistic methods/sampleRequirementEquivalence (agreement)Agreement of results between two protocolsCohen’s kappa statistics/102 participants per respiratork ≥ 0.70Reproducibility (stability)2-week test-retest reliabilityICC/60 participantsICC ≥ 0.75ICC, intraclass correlation coefficient. Contingency table and test statistics ICC, intraclass correlation coefficient. The k value for agreement will be used to indicate the equivalence of the fit-testing results (i.e. pass or fail) obtained for a given respirator using the new and conventional protocols [19,23,29]. A k value in the range of 0.41–0.60 or 0.61–0.80 indicates moderate or substantial agreement, respectively [27,29,30]. According to the American Industrial Hygiene Association specification ANSIZ88.10-2010, however, a k value exceeding 0.70 is recommended for establishing equivalence between two protocols or systems specific to respirator fit testing [31]. Reproducibility will be examined using the 2-week test–retest reliability [19,22,23]. The ICC will be used to compute the coefficients between the FFs determined using the new and conventional protocols for a given respirator worn by the same participant after a 2-week interval. An ICC exceeding 0.75 will indicate satisfactory reproducibility [19]. All tests will be two-sided, and the statistical significance threshold will be set as a P value of 0.05. Data management All data recorded are stored and backed up on an encrypted hard drive. Data entry and management will be completed by one of the research team members and double-checked by a minimum of one other team member. Personal identifiers will be removed upon completion of the study and stored for long-term retention of the research data. Data will be retained for a minimum of 3 years. Plan to disseminate research findings The research findings will be shared with the Infection Control Branch and Occupational Safety and Health of the Hospital Authority and the Centre of Health Protection of the Hong Kong SAR Government to inform them and provide a point of reference. The findings will also be shared through presentations at both local and international conferences and in reputable journals focussed on infection control and occupational health. The goal will be to raise global awareness about the newly developed four-step fit-testing protocol and promote the importance of ensuring reproducible results when testing N95 respirators. Discussion In clinical environments, N95 respirators are used to restrict respiratory virus transmission. The evaluation of the fit-testing protocols will serve as a valuable reference for occupational safety professionals and in clinical settings. In Hong Kong, some public hospitals have implemented the new fit-testing protocol, while others have maintained the conventional protocol. Regrettably, in some instances, individuals have successfully passed N95 fit testing using the new protocol but failed when testing the same respirator using the conventional protocol and vice versa. The kappa agreement between the new and conventional protocols can serve as an important reference point for fit-testing policies as it provides a standardised measure of consistency and reliability between testing methods. If there is a high level of equivalence, both fit-testing protocols could be simultaneously implemented as they would be expected to produce similar results. However, if the new protocol identifies leakage more rigorously than the conventional protocol, it would be necessary for all HCWs to repeat fit testing using the new protocol to guarantee occupational safety in clinical settings. In contrast, if the conventional protocol is more stringent than the new protocol, conservatively, it would be necessary to exercise caution and suspend implementation of the new protocol. Further research should be conducted to thoroughly examine the justification of the new protocol. In addition, investigating the reproducibility of respirator fit testing can resolve uncertainty regarding a respirator’s capacity to maintain an appropriate fit in an actual clinical environment. These expected results will have substantial clinical and policy implications regarding the readiness and responses of hospitals to current and future pandemics and epidemics. Some limitations deserve discussion. First, the current study does not include all N95 respirators that are available in the market. The focus of this study is to evaluate the FF of N95 respirators for the specific design (i.e. 3M 1860s, 3M 1870+, and OKSafe N95-01), rather than promoting one respirator over another. Second, this study is conducted in a controlled simulated environment. It is anticipated that the result of the FF may be underestimated or considered minimal due to uncontrolled factors, such as clinical environment and different personnel involved. CRediT authorship contribution statement S.C. Lam: Writing – review & editing, Writing – original draft, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. I.T.S. Yu: Methodology, Investigation, Funding acquisition. M. O’Donoghue: Methodology, Investigation, Funding acquisition. K.K. Leung: Methodology, Investigation, Funding acquisition. A.C.Y. Chong: Validation, Project administration, Methodology, Investigation, Funding acquisition. L.L.K. Sung: Writing – original draft, Data curation. A. Odetayo: Writing – review & editing, Writing – original draft, Methodology. L.K.P. Suen: Writing – review & editing, Methodology, Investigation, Funding acquisition, Conceptualization. Ethics statement The study was approved by the Tung Wah College Research Ethics Committee (ref no: REC2024221) according to the ethical principles of the Declaration of Helsinki. The participants will provide their written informed consent to participate in this study. Consent for publication Written informed consent was obtained from the individuals for the publication of any identifiable images or data included in this article. Availability of data materials The data will not be made available to the public due to participant confidentiality and institutional restrictions, but are available from the corresponding author on reasonable request. Funding sources The authors declare financial support was received for the research, authorship, and/or publication of this article. The study is funded by The Health and Medical Research Fund, the Health Bureau, The Government of the Hong Kong Special Administrative Region (Ref No 24230052). The funders have no role in the study design, data collection, data analysis, manuscript preparation, or decision to publish. Conflict of interest statement The authors declare that they have no competing interests. The first and corresponding author is an Editor for Infection Prevention in Practice. Editorial policies were followed to ensure editorial independence, and the manuscript was handled by separate Editors.