Proper personal protective equipment (PPE) doffing prevents pathogen spread and healthcare personnel (HCP) self-contamination. 1 Previous research has documented that HCP self-contamination is common, with rates ranging from 46% to 90% in empirical studies. Reference Tomas, Kundrapu and Thota2–Reference Osei-Bonsu, Masroor and Cooper4 Furthermore, HCP regularly make critical doffing errors and contaminate themselves and the environment even when they think they doffed proficiently. Reference Tomas, Kundrapu and Thota2,Reference Lim, Cha, Chae and Jo5–Reference Chughtai, Chen and Macintyre8
Complicating matters, HCP may encounter an array of PPE items (eg, gloves, gowns, masks) and designs (eg, gowns with breakaway neck closures vs tape-tab closures), particularly if they work across clinics or facilities. PPE shortages, such as those experienced during the COVID-19 pandemic, may also affect the supply and designs available. Reference Yanke, Zellmer, Van Hoof, Moriarty, Carayon and Safdar7 Despite the variety of PPE designs HCP may encounter, we know little about how HCP think about and strategize doffing unfamiliar PPE.
We investigated the following: (1) how different designs of the same PPE item affect the risk of HCP self-contamination while doffing (simulations 1–3) and (2) how HCP training and experience affect their ability to doff without self-contamination (simulation 4). This research included simulated doffing scenarios followed by think-aloud interviews during which HCP described their doffing strategies and thought processes. We previously published findings from simulations 1–3 regarding factors that influence HCP doffing strategies and general barriers and facilitators to proper doffing. Reference Baloh, Reisinger and Dukes9 However, these initial analyses led us to further explore an important issue that emerged from the data, namely, how HCP perceive and think through the process of doffing unfamiliar PPE. In this manuscript, we share qualitative findings related to how HCP navigated encounters with unfamiliar PPE during simulations and the influence of routine practices, experience with familiar PPE, and training on their doffing strategies.
Methods
Study design and setting
As part of a large mixed-methods study, we conducted simulated PPE doffing scenarios (ie, simulations 1–4) and think-aloud interviews in 2 Midwestern academic hospitals (hospitals A and B) from September 2017 to May 2019. During simulations 1–3, we observed HCP doffing different designs of the same PPE item (ie, 3 mask designs, 2 gown designs, and 2 glove designs). Reference Baloh, Reisinger and Dukes9 During simulation 4, we observed HCP doffing identical full PPE ensembles (ie, mask, gown, and gloves) under 2 different conditions. Table 1 summarizes the simulation scenarios and the PPE used in each. We conducted simulations in clinical education and training facilities at each hospital. The Institutional Review Board at the University of Iowa approved all study activities and participants provided consent before participation.
Table 1. Doffing Simulation Descriptions and PPE Used
Note: PPE, personal protective equipment.
a The “distraction” consisted of a team member casually asking the participant questions (eg, “How has your day been?”) while they doffed.
Sample and data collection
We recruited 70 HCP to participate in simulations through emails sent to staff and information disseminated during staff meetings (hospitals A and B) and through posters placed in rooms where physicians do their documentation (hospital A). Participants in simulations 1–3 (n = 30; 10 per simulation) were recruited from hospital A and included HCP (eg, nurses, physicians, respiratory therapists) and medical and nursing students on clinical rotations. We designed simulation 4 (n = 40) to evaluate the effects of training and experience on doffing. We recruited 20 HCP and medical and nursing students who did not have special training as well as 10 HCP who had completed special isolation unit (SIU) training at hospital A and 10 HCP from the biocontainment unit at hospital B. We excluded HCP who did not use PPE at work and students not on clinical rotation. We assigned participants to 1 of 4 simulations and collected demographic information.
Simulations
We previously described simulation 1–3 procedures. Reference Baloh, Reisinger and Dukes9 Notably, simulation 3 participants donned and doffed standard exam gloves and Doffy gloves, which have a tab at the wrist. Because the Doffy design was novel to all participants, team members shared that the tab was designed as a doffing aid. Participants were then asked to doff once, to watch a brief video that demonstrated proper Doffy glove doffing technique, and to doff a second time. This was the only PPE item for which the study team provided guidance.
Simulation 4 procedures were like simulations 1–3 with these exceptions: (1) participants donned and doffed identical PPE ensembles (ie, mask, gown, and gloves) twice, rather than different designs of the same item and (2) participants donned and doffed under 2 conditions (distraction vs nondistraction with the condition order assigned randomly). For the “distraction,” a team member casually asked participants questions (eg, “How has your day been?”) during doffing. Methods used to assess (ie, blacklight) and document (ie, digital camera) baseline fluorescence and self-contamination with Glo Germ fluorescent marker following each episode remained unchanged. As in simulations 1–3, we video-recorded the simulation from 4 angles, and for this report, we used interview data from all 4 simulations.
Interviews
Immediately following simulations, each participant completed an audio-recorded think-aloud interview Reference Beam, Gibbs, Hewlett, Iwen, Nuss and Smith10–Reference Iedema, Hor and Wyer13 during which they watched their recorded doffing episodes and described and reflected on their performances. Interviewers provided brief instructions and probed to encourage participants to expand or clarify responses (Table 2). Qualitative researchers (E.C., K.D., and J.B.) acted as the primary interviewers, but other team members with clinical patient care, infectious disease, or human factors engineering expertise (L.H., J.P., and H.S.) also asked questions.
Table 2. Think-Aloud Interview Script Questions and Probes
a Italicized questions only applied to simulation 4 (ie, distraction or nondistraction). Outcomes related to distraction vs nondistraction are not addressed in this paper.
Data analysis
We transcribed and uploaded interviews to MAXQDA qualitative data management software. 14 Baloh et al Reference Baloh, Reisinger and Dukes9 previously described our codebook development and analyses for simulations 1–3 (first 30 participants). We inductively developed the codes familiarity and training to capture emergent themes related to factors that influence HCP doffing strategies. Given that simulation 4 (final 40 participants) differed from previous simulations in important ways (ie, doffing identical ensembles under 2 conditions), we anticipated that codebook refinement might be necessary.
For simulation 4 data, 2 coders (E.C. and J.P.) applied the existing codebook to 2 transcripts and used open coding and memos to document data that did not fit within existing definitions. After comparing their coding, they added and defined several codes, including routine practice, and then applied the updated codebook to 2 additional transcripts to test fit. One coder (E.C.) coded all simulation 4 transcripts using the final codebook and examined data from simulations 1–3 to identify additional data that met the routine practice code definition. To increase reliability, the second coder (J.P). coded 8 (20.0%) of 40 simulation 4 transcripts and met periodically with the first coder to compare coding. They consistently had high agreement, discussed discrepancies to reach consensus, and documented connections between emergent themes related to doffing unfamiliar PPE in analytical memos.
Results
Table 3 reports participants’ characteristics. When analyzing interview data, we inductively identified 3 interconnected themes, which provide insight into how HCP navigate doffing unfamiliar PPE: (1) (lack of) familiarity with PPE items or designs, (2) influence of routine practice, and (3) training experiences and needs. We describe these themes below and share supporting quotations in Table 4.
Table 3. Study Participants’ Characteristics by Simulation Scenario
Note: PPE, personal protective equipment.
a Mean and range reported for 39 participants for simulation 4 and 69 participants for total; 1 participant did not report age.
b Other, eg, nursing assistant, physician assistant, respiratory therapist, clinical pharmacist, nurse practitioner.
c Prior training on PPE donning reported separately from doffing for simulation 4 and total; 2 simulation 4 participants provided different responses for donning versus doffing training.
d Hospital A (n=10); hospital B (n=10).
Table 4. Think-Aloud Interview Themes and Subthemes with Illustrative Quotations a
Note: HCP, healthcare personnel; N-95, N-95 respirator.
a Quotations have been lightly edited to remove word repetitions (eg, stammering) and verbal hesitations (eg, “um”).
Theme 1. (Lack of) Familiarity with PPE items or designs
All participants used PPE in their work, but the specific items and designs differed by role, unit, and hospital. Many described encountering a PPE design during the simulation that was unfamiliar or that they used infrequently. For example, most participants reported that they typically used procedure masks with ear loops rather than surgical masks with ties. Hospital A participants typically used disposable gowns with breakaway neck closures and thumb loops, whereas those at hospital B wore reusable cloth gowns with neck ties. Some participants worked in settings that rarely required them to use PPE other than nitrile exam gloves. Multiple participants shared that using unfamiliar PPE was a doffing barrier during simulations.
Specific designs were truly novel for some participants (eg, Doffy gloves, Reference Gleser, Schwab, Solbach and Vonberg15 pouch-style masks with elastic headbands). More often, participants had used certain designs during their education and training but not during their recent practice. Some stated that they used different PPE designs when they worked in other facilities or when hospitals obtained new designs due to shortages or changes in purchasing agreements. Participants reported encountering the greatest variation in mask and gown designs.
Although some participants believed that they retained the ability to doff infrequently used PPE properly, most who addressed this issue said they were less comfortable with designs they used infrequently. These participants reported that they struggled to draw on muscle memory or to remember their early training when doffing these items during the simulations. A minority shared that they would seek out familiar or preferred PPE designs in other areas if these were not stocked where they worked.
Participants also noted that proper doffing procedures for unfamiliar PPE were not always intuitive. For example, participants at hospital B were uncertain which strategies to use when donning and doffing the gown with a perforated neck closure, thumb loops, and side-tie belt. Those who did not use surgical masks regularly had a similar response to this mask design. Participants in simulations 1–3 also were uncertain about using unfamiliar designs even though they had the opportunity to decipher cues by comparing the current design with the design they just donned and doffed.
In addition to inhibiting proper doffing, unfamiliar PPE designs sometimes required HCP to change doffing order and doffing strategies. To correctly doff the simulation 4 gown, for example, HCP must first don the gown over their heads, place their thumbs through the thumb loops, and don gloves over the loops. HCP should doff by breaking the gown’s neck closure first, then remove the gown and gloves together. Participants felt this process required a different mental approach than that required to remove the gown and gloves separately.
Participants identified several factors that influenced their general doffing strategies, most prominently the desire to avoid self-contamination and patient care demands related to their specific roles. They acknowledged these concerns when doffing any PPE item but noted additional cognitive demands when using unfamiliar PPE.
Theme 2. Influence of routine practice
Participants who encountered unfamiliar PPE during simulations described their approaches to troubleshooting the appropriate doffing sequence and strategy. They particularly noted the role of their day-to-day PPE practices including connected factors such as designs typically worn (and donning/doffing order) and design cues (eg, fasteners or perforations). Participants also cited less obvious factors such as donning and doffing frequency during routine patient care tasks and the specific contexts in which they provided care. Many stated that their routine practices, which they developed through frequent donning and doffing, ingrained habits into their muscle memory. When encountering unfamiliar PPE, they relied on these habits rather than on conscious thought. Some perceived these habits as potentially difficult to change.
The 20 participants (10 per hospital) in simulation 4 with SIU training often drew on this training when formulating strategies for doffing unfamiliar PPE. The specialized training included practice donning and doffing unfamiliar PPE items and more complex ensembles, practice following more rigorous donning and doffing protocols, and coaching with feedback. Several participants described how the rigorous protocols they followed as part of SIU training (and/or subsequent biocontainment unit experience) made them more cautious when doffing and shaped how they conceptualized their risk for self-contamination. They perceived themselves as acting with similar caution in their approach to unfamiliar PPE during simulations. However, SIU-trained participants also acknowledged aspects of donning and doffing in SIU training and the biocontainment unit that differed significantly from the routine practice settings in which they might encounter unfamiliar PPE. These included following steps in a directed donning and doffing process while being coached by an observer, focusing exclusively on donning and doffing, and using specific PPE items and designs. In contrast, participants typically described donning and doffing without coaching while managing competing cognitive demands (eg, conversations, interruptions) during routine practice.
Despite these differences, participants described specific doffing strategies that they used in the simulation as related to their SIU training such as rolling dirty gown surfaces away from their bodies, breaking (rather than untying) mask ties, and using glove-in-glove technique. This was also true when the PPE provided differed from that used in SIU training or the biocontainment unit. For example, one SIU-trained participant intentionally removed the surgical mask ties during the simulation based on the order (s)he was taught to follow for an N95.
Participants in all simulations reported looking for design cues on unfamiliar PPE. Although we focused primarily on doffing, we observed that participants often surveyed PPE while donning to identify cues such as gown tie placement and neck closure, mask fasteners, and Doffy glove tabs. Participants reported looking for design features that resembled or served the same function as those on familiar PPE to inform their doffing strategy. Nevertheless, recognizable cues did not always point participants to a clear and appropriate strategy. In fact, some participants experimented while donning and/or changed their processes between doffing episodes within the same simulation. The cues provided a starting place. Conversely, participants became confused if they did not identify cues or encountered unfamiliar cues.
Theme 3. Training experiences and needs
Participants described their routine practices as developing over time, shaped by hospital policies, available PPE, protocols, the demands of patient care tasks (eg, frequency of donning and doffing), and importantly, previous training. Many referenced PPE training, including training on multiple designs, that they received during their education, professional training, or employment orientation at a specific hospital. However, given the time elapsed since training, some participants had difficulty recalling the strategies they learned or indeed if training was the source of specific habits. Others described learning different ways to properly doff an item, developing a preferred method, and then using that method with subsequent designs. Though previous training was a touchpoint for participants’ routine practices, a few also noted that Centers for Disease Control and Prevention (CDC) guidelines were available for reference when they were uncertain about donning and doffing protocols.
Participants also recalled that their training provided general guidelines about donning and doffing order, regardless of PPE designs; however, the order participants were taught differed. For example, during simulation 2 (ie, 2 gown designs), a nurse (hospital A) shared, “… always starting with the gown is what we’ve been told.” Conversely, a student (hospital A) assigned to the same simulation stated she was instructed to don gloves first, then the gown. Participants’ comments and strategies indicated that PPE training is not standardized. In addition, some reported that they did not receive either ongoing PPE training that included diverse designs, nor just-in-time training when unfamiliar PPE was introduced. Participants perceived lack of training as a barrier to their ability to doff unfamiliar PPE properly and minimize self-contamination.
Participants felt that training could increase their familiarity with different PPE designs and decrease their likelihood of self-contamination. For example, while watching his performance, a student (hospital A) said, “… just in general, without really kind of any training, doing that I would just assume that I would contaminate myself.” Participants referenced annual competencies required for nursing staff, videos, and posters displayed on units as existing training methods in their hospitals, and they provided feedback to improve training effectiveness. They particularly perceived training that involved physical practice donning and doffing various PPE designs as valuable, compared with either written or verbal instructions or observations of another HCP’s performance.
Discussion
We qualitatively explored how HCP approach unfamiliar PPE within the context of a mixed-methods study examining factors that influence HCP doffing strategies and self-contamination. We did not include this question in the initial study aims; however, it was implicit in that we included multiple designs of each PPE item in simulations 1–3. We recognized the need to explicitly address it in our analyses after it emerged inductively in participants’ comments during simulations and interviews. Our findings indicate that HCP drew on their routine practices with familiar PPE to inform their strategies. Furthermore, our data suggest that these routine practices develop through prior training and in response to hospital policies and patient care contexts. Thus, HCP feedback on training modalities could improve training and, thereby, likely improve HCP doffing strategies and decrease the risk of self-contamination.
In our previous work, HCP tried to balance doffing PPE safely to reduce self-contamination with patient care needs and the demanding clinical environments in which they work. Reference Baloh, Reisinger and Dukes9 However, we found that HCP had clear design preferences, suggesting that HCP do not view different designs of the same PPE item as interchangeable. Reference Herlihey, Gelmi and Flewwelling16 HCP who replicate, or adapt, their routine practices and training to unfamiliar PPE may use inappropriate doffing processes and contaminate themselves, particularly if they wear more PPE items during a given patient care episode than usual. Reference Zamora, Murdoch, Simchison and Day17
Previous research has likewise demonstrated that different PPE designs are associated with different rates of self-contamination. Reference Guo, Li and Wong18–Reference Hajar, Mana, Tomas, Alhmidi, Wilson and Donskey20 Hospitals that switch out a specific design for another or do not provide training when new PPE is introduced place the onus on busy HCP to do their own research or to trouble-shoot in the moment. However, the research on training HCP to doff PPE is limited and existing guidelines and recommendations vary. Reference Chughtai, Chen and Macintyre8,Reference Doll, Feldman and Hartigan21,Reference Krein, Mayer and Harrod22 CDC guidelines present general donning and doffing instructions but may not include specific designs such as gowns with different neck closures or nondisposable gowns. 23 In part, this is likely due to the commercial availability of multiple designs, whereas CDC guidelines must cut across all designs. Just-in-time training may help HCP learn to use new PPE designs if buying contracts change or shortages occur, but our findings indicate that HCP could benefit from refresher training sessions, and from training involving actual donning and doffing, rather than written or visual components alone.
Finally, HCP reported that encounters with unfamiliar designs were not unusual, even before 2020. The COVID-19 pandemic exacerbated this problem, given severe PPE shortages and that HCP needed to wear items that were not part of their prior routine PPE ensembles (eg, eye protection, respirators), as well as the introduction of extended PPE use. Given the likelihood of future pandemics caused by respiratory pathogens, HCP strategies for doffing unfamiliar PPE (particularly when used in ensembles involving multiple PPE items) have important implications for infection prevention.
This study had several limitations. We observed HCP in simulated rather than real-world settings. Participants knew they were being observed, which may have affected their behavior (ie, Hawthorne effect). Simulations did not incorporate the patient care tasks that HCP routinely perform while doffing. Given the role of muscle memory, real-world observations might illuminate how these tasks affect practices in ways HCP are and are not aware of. As with any self-reported data, participants’ descriptions of their thought processes and routine behaviors were subject to misrepresentation or recall bias. However, the strength of this approach lies in the valuable insights that participants provide into the observed behaviors. We mitigated potential recall bias by conducting interviews immediately following simulations. Finally, we recruited from only 2 hospitals in similar states, and participants within a hospital were exposed to similar PPE designs, protocols, and training. Nevertheless, HCP brought diverse perspectives due to their different disciplines, professional- and PPE-specific training, and employment histories. We did not recruit based on demographic characteristics (eg, age and sex); however, future research should examine whether these factors affect HCP perceptions of PPE and PPE doffing.
Healthcare facilities should consider the challenges inherent in doffing unfamiliar PPE when introducing new designs and not assume that different designs are interchangeable or that optimal donning and doffing methods are intuitive. In cases of shortages or rapid changes in PPE stock, just-in-time training may help HCP adapt to unfamiliar PPE. However, HCP also require ongoing training that emphasizes hands-on practice using appropriate doffing techniques and practice doffing both routinely used and newly introduced PPE designs.
Acknowledgments
We thank the healthcare personnel who participated in this study for their time and insights. We also acknowledge study staff member Sandra Cobb for her contributions to this research. Halyard (Alpharetta, GA) provided gowns, masks, and nonsterile nitrile gloves. IP Gloves GmbH (Aachen, Germany) provided Doffy gloves. Neither company had any input into the study design, data collection, data interpretation, or the manuscript preparation.
The findings and conclusions reported here are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention, National Institutes of Health, or the University of Iowa.
Financial support
This work was supported by the Centers for Disease Control and Prevention (grant no. CK000448-01) and the National Center for Advancing Translational Sciences (grant no. KL2 TR003108).
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
Open access
