Identification and Characterization of Failures in Infectious Agent Transmission Precaution Practices in Hospitals

Key Points

Question

What circumstances lead to active failures in transmission-based precautions to reduce the transmission of infectious agents by health care personnel?

Findings

This qualitative study of 325 patient rooms with precaution signage found frequent and varied active failures in infectious agent transmisison precaution practices by hospital personnel, including violations, mistakes, and slips.

Meaning

Given the broad array of circumstances contributing to active failures in precaution practices that were identified and categorized, behavioral, organizational, and environmental strategies may be needed to reduce the risk of infection transmission and self-contamination.

Abstract

Importance

Using personal protective equipment (PPE) and transmission-based precautions are primary strategies for reducing the transmission of infectious agents.

Objective

To identify and characterize failures in transmission-based precautions, including PPE use, by health care personnel that could result in self-contamination or transmission during routine, everyday hospital care.

Design, Setting, and Participants

This qualitative study involved direct observation inside and outside patient rooms on clinical units from March 1, 2016, to November 30, 2016. Observations occurred in the medical and/or surgical units and intensive care units at an academic medical center and a Veterans Affairs hospital, as well as the emergency department of the university hospital. Trained observers recorded extensive field notes while personnel provided care for patients in precautions for a pathogen transmitted through contact (eg, Clostridium difficile, methicillin-resistant Staphylococcus aureus) or respiratory droplet (eg, influenza). Specific occurrences involving potential personnel self-contamination were identified through a directed content analysis. These occurrences were further categorized, using a human factors model of human error, as active failures, such as violations, mistakes, or slips.

Main Outcomes and Measures

Number and type of failures involving use of transmission-based precautions.

Results

In total, 325 room observations were conducted at 2 sites. At site 1, a total of 280 observations were completed (196 in medical/surgical units, 64 in intensive care units, and 20 in emergency departments). At site 2, there were 45 observations (36 in medical/surgical units and 9 in the intensive care unit). Of the total observations, 259 (79.7%) occurred outside and 66 (20.3%) inside the room. Two hundred eighty-three failures were observed, including 102 violations (deviations from safe operating practices or procedures), 144 process or procedural mistakes (failures of intention), and 37 slips (failures of execution). Violations involved entering rooms without some or all recommended PPE. Mistakes were frequently observed during PPE removal and encounters with challenging logistical situations, such as badge-enforced computer logins. Slips included touching one’s face or clean areas with contaminated gloves or gowns. Each of these active failures has a substantial likelihood of resulting in self-contamination. The circumstances surrounding failures in precaution practices, however, varied not only across but within the different failure types.

Conclusions and Relevance

Active failures in PPE use and transmission-based precautions, potentially leading to self-contamination, were commonly observed. The factors that contributed to these failures varied widely, suggesting the need for a range of strategies to reduce potential transmission risk during routine hospital care.

This qualitative study examines types of precaution practice violations and errors in reducing transmission of infectious agents by hospital personnel in clinical units.

Introduction

Preventing the spread of infectious organisms is a priority for hospitals and health care systems.¹ A prominent concern is that cross-transmission can lead to patients becoming colonized and acquiring infections, such as hospital-onset Clostridium difficile infection and methicillin-resistant Staphylococcus aureus bacteremia,^2,3 as well as infections from other emerging drug-resistant organisms.^4,5 Minimizing risks for health care personnel is also paramount, as illustrated by the self-contamination and subsequent illness of 2 nurses caring for a patient with Ebola virus disease.⁶

Standard and transmission-based precautions are an infection-prevention mainstay for protecting patients and health care personnel.⁷ These precautions include hand hygiene and the use of special clothing or equipment known as personal protective equipment (PPE). However, studies suggest poor compliance with precaution practices involving PPE use.^8,9,10,11 One study¹⁰ found that 34% of health care personnel donned all recommended PPE for droplet precautions and 54% removed the PPE in the correct sequence. Another study⁹ found that less than 50% of personnel removed their PPE in the correct order and only 17% both properly removed and disposed of their PPE in the patient room. A study of compliance with contact precautions found that gown use at room entry ranged from 52% to 70% across 3 hospitals,¹¹ and another study⁸ found a 7% and a 22% rate of full compliance with contact precautions for patients with C difficile infection at 2 hospitals.

Noncompliance is documented in numerous studies, but the potential failures and the circumstances surrounding failures in precaution practices are not well described in the literature. Understanding the types of failures and the context in which failures occur is essential for effectively mitigating the risk of pathogen transmission. The objective of this study was to identify and characterize failures in transmission-based precautions, including PPE use, by health care personnel that could result in self-contamination and transmission during routine hospital care.

Methods

Study Design and Setting

From March 1, 2016, through November 30, 2016, we directly observed health care personnel outside and inside patient rooms with signage indicating required precautions for a pathogen transmitted through contact (eg, C difficile, methicillin-resistant S aureus) or respiratory droplet (eg, influenza). Observations occurred on 16 acute care medical and/or surgical units and 4 intensive care units (ICUs) at 2 medical centers, a university-based academic medical center (site 1) and a Veterans Affairs (VA) hospital (site 2), as well as the emergency department at the university-based medical center. This study obtained approval from the institutional review boards of the University of Michigan and the VA Ann Arbor Healthcare System. For the in-room observations, verbal informed consent was obtained from both patients and personnel. For the out-of-room observations, personnel could opt out if they felt uncomfortable with the process.

Data Collection

Data were collected through unstructured field notes. Observers recorded on paper a detailed account of what they saw during the observation period. Eleven observers, including 6 research staff and 5 undergraduate nursing and premed students, received structured didactic and field-based observation training from a medical anthropologist (M.H.). The field notes were reviewed regularly, and feedback was provided to ensure complete and consistent information capture.

Observations took place in 2- to 3-hour blocks, with multiple observations occurring during each period. After arriving in the unit, observers identified patient rooms with transmission-based precaution signs, introduced themselves to unit personnel, and then targeted a single room or several rooms that could be viewed from the same vantage point. Observers recorded information while standing outside and/or inside the rooms; no observations were conducted, however, if the patients or personnel indicated they were uncomfortable with the process. The length of observation time was determined by the observer. If there was no activity in a selected room, the observer would move to another room or continue the observation until the activity or observation period ended.

Observers transcribed their handwritten notes to create an electronic field note immediately after completing an observation period. Each field note contained the unit name, precaution type for each observed room, and the observation time frame, as well as detailed, time-stamped information about health care personnel and their interactions outside and inside the room. This information included personnel role, whether hand hygiene was performed and if and how the PPE was donned, interaction with the patient and patient environment, tasks performed, PPE removal process (doffing), and activities immediately after room exit.

Data Analysis

The analytic approach involved a directed content analysis of the field notes.¹² We developed an initial coding framework based on the Systems Engineering Initiative for Patient Safety model,¹³ which focuses on the work system, processes, and outcomes. Several field notes were coded by 3 study team members with and without clinical hospital experience (S.L.K., M.H., and L.W.) to ensure coding consistency. Thereafter, the field notes were divided among the team members and coded using NVivo qualitative data analysis software, version 10 (QSR International Pty Ltd). After the initial set of codes was applied, code reports were generated and reviewed to confirm coding accuracy and identify additional codes. Independent review of the code reports by multiple team members led to the identification of several commonly occurring events of possible self-contamination or the potential for transmission of an infectious agent, such as unprotected contact with contaminated surfaces. Team members then returned to the field notes to apply these new codes. These occurrences were further categorized, using the Reason model of human error,^14,15 as active failures such as violations, mistakes, or slips.

As discussed by Reason, active failures are unsafe acts committed at the human-system interface, including violations (intentional deviations from recommended practices, procedures, or standards) and errors (“the failure of planned actions to achieve their desired goal” ^14(p81)). Errors include mistakes and slips or lapses. Mistakes are failures of intention and often occur when the plan for dealing with a problem does not achieve the intended outcome. Slips or lapses are failures of execution that generally indicate lack of attention or memory-related failures during the automatic performance of routine tasks.

Results

A total of 325 room observations were conducted across both study sites—280 (196 medical/surgical, 64 ICU, and 20 emergency department) at site 1 and 45 (36 medical/surgical and 9 ICU) at site 2. Of the total observations, 259 (79.7%) occurred outside and 66 (20.3%) occurred inside the room. Specific precaution practices, as required by policy, for each type of precaution are described in Table 1. Most observations (237 [72.9%]) involved contact precautions, 81 (24.9%), contact enteric precautions, and 7 (2.1%), droplet precautions. Standard precautions were required for all patients in addition to the transmission-based precautions.

Table 1. Summary of Required Precaution Practices^a.

Precaution Type	Common Disease	Gloves Use	Gown Use	Mask Use	Eye Protection (Mask/Goggles) Use	Equipment Handling	Hand Hygiene Performance
Contact	MRSA, scabies, VRE, CRE	When entering room	For contact with patient or environment, or when entering room	NA	NA	Disinfect before removing from room, or use dedicated equipment	When entering or leaving room
Contact enteric	Clostridium difficile, norovirus	When entering room	For contact with patient or environment, or when entering room	NA	NA	Disinfect with bleach before removing from room, or use dedicated equipment	When entering room; use soap and water before exiting room
Droplet	Influenza, meningococcus, pertussis	When entering room, or if anticipating contact with secretions	For contact with patient or environment, or if anticipating contact with bodily fluids	When entering room, or within 3 ft of patient	Within 3 ft of patient, or if contact with secretions likely	Disinfect before removing from room, or use dedicated equipment	When entering and leaving room
Standard	All patients	For contact with bodily fluids	If anticipating soiling of clothing with bodily fluids	During procedures likely to generate droplets of bodily fluids	During procedures likely to generate droplets of bodily fluids	No special action	Before and after patient contact

Abbreviations: CRE, carbapenem-resistant Enterobacteriaceae; MRSA, methicillin-resistant Staphylococcus aureus; NA, not applicable; VRE, vancomycin-resistant Enterococcus.

^aThere were some differences in how certain precautions were applied across sites.

Active failures were frequently observed across the study sites. More than 280 failures were observed, including 102 violations, 144 process or procedural mistakes, and 37 slips. Each type of active failure is associated with the potential for self-contamination or direct (eg, skin) or indirect (eg, fomite) transmission of an infectious agent, but the circumstances surrounding these events varied widely, even within failure types.

Violations

Observed violations, or deviations from recommended practices, ranged from entering the precaution room without donning PPE to using PPE improperly or incompletely (eg, gowns not tied) and doffing and disposing of PPE outside the room (Table 2). A common scenario was personnel entering a room without any PPE, including gloves, which were required at room entry for all precaution types.

Table 2. Examples of Common Failures in Transmission-Based Precaution Practices.

Failure Type	Example From Field Note
Violation
No contact	Contact precautions, site 1, ICU: Physician walks into the room (no PPE) and inspects the patient and their machines, careful not to touch anything.
No contact	Contact precautions, site 1, non-ICU: Nurse walks back into the room (does not touch door, door handle, or room environment on the way in) and places plastic tub on bed. Nurse does not make contact with anything in the room, walks out, and closes door from the outside using the door handle. Nurse touches pager, does not gel, walks away.
With contact	Contact precautions, site 2, ICU: Nurse enters the patient room and walks directly to the vitals screen and begins to press buttons. No gown or gloves and no gel on entry into room.
With contact	Contact enteric precautions, site 1, ED: Physician approaches room with clipboard and papers in hands, does not gel, does not don gown or gloves. Enters room to talk to patient’s family, sets clipboard down on bedside table, scrubs touch the side of the bedside table. Physician finishes talking to family, picks up clipboard, and exits room; does not gel or wash hands with soap and water; walks away while shuffling through papers on clipboard.
Emergency	Contact precautions, site 1, ICU: There is a loud beeping coming from the patient’s room; everyone rushes to the room and flings the curtain open. They rouse the patient, and the patient’s status returns to normal; a few people went into the room without PPE on to address the emergency.
Mistake
Doffing	Droplet precautions, site 1, non-ICU: Respiratory therapist removes glasses and mask with gloved hands.
Doffing	Contact enteric precautions, site 1, ED: Nurse removes gown by ripping it from the back of the neck, does not carefully remove the gown or gloves, throws them both into the trashcan before exiting, does not use gel (or soap and water) on exiting.
Doffing	Contact precautions, site 2, non-ICU: Nurse brings gloved hands to shoulders and lifts off the neck strap. Then, nurse pulls off the gown and gloves in one motion, and discards into the in-room trash, picks up the food tray, and carries it out of the room.
Doffing	Contact precautions, site 1, ICU: Nurse starts to doff PPE, untying the gown and pulling it over head, when the machine lets out a beep. Nurse promptly puts the gown back over head.
Logistics issue	Contact enteric precautions, site 1, non-ICU: Nurse reaches gloved hand under gown to pull a packaged item out of pocket.
Logistics issue	Contact enteric precautions, site 1, non-ICU: Nurse is inside the room with an untied gown and gloves on, reaches under gown to pull up scrub pants by the pockets, and then helps the patient to the bathroom.
Slip
Touch	Contact precautions, site 2, non-ICU: Health care personnel talking with patient, leaning on bedrail, looks a little flustered and starts to perspire, and uses forearm to wipe face. While wiping face, knocks glasses partially off nose; uses gloved hand to reposition glasses on face.
Device	Contact precautions, site 1, ED: Nurse’s pager goes off, nurse reaches underneath gown to grab pager and read the message, restores pager to underneath gown.

Abbreviations: ED, emergency department; ICU, intensive care unit; PPE, personal protective equipment.

Violations occurred when personnel entered rooms with the apparent goal of talking with patients and families, checking devices, or “dropping and dashing” (running into a room to drop something off and then dashing out), as illustrated in the following notes:

• Nurse walks into room and stays by the doorway to tell the patient something. Walks in further when the patient asks a question. Nurse has no PPE. They talk for a few minutes, nurse leaves, and gels hands. (Contact precautions, site 1, non-ICU)

• Nursing assistant enters the room about three feet with a large bunch of supplies and quickly tosses them onto a counter top, then exits and gels hands. (Contact precautions, site 1, ICU)

Although personnel may enter rooms without PPE and no intent of contact, their encounters often resulted in contact with the patient or the patient environment, as the following notes show:

• Nurse enters without foaming or donning PPE and places water on the patient’s tray table. Nurse then picks up the patient’s call light and turns the lights off, exits, and foams hands. (Contact precautions, site 1, emergency department)

• Physician enters without foaming or PPE and stands near the bedside. Physician touches a button on a machine at the end of the bed with ungloved hand. (Contact precautions, site 1, ICU)

Violations in which direct patient contact was undoubtedly intended were also observed. Whether these events were the result of simple disregard for proper procedure or some other circumstances was not always observable:

• Physician sanitizes hands using the dispenser then enters the room without donning gloves or gown, touches patient’s stomach over the [patient] gown. Physician rests arm on bedside, pulls records out of white coat pocket, reviews, then places back in pocket, exits room and sanitizes hands. (Contact precautions, site 1, non-ICU)

• Nurse gels and is rubbing hands while entering patient room. Nurse is carrying an incontinence bed pad under their underarm, walks directly to the patient bedside, begins to tuck underneath the patient, leans against bedrail with ungloved hands, and begins discussion with patient. [W]hen completed, nurse walks out of room and gels on exit. No gown or gloves were used at any point. (Contact precautions, site 2, non-ICU)

In some cases the reason for the individual’s actions, such as a potential medical emergency, was relatively clear:

• Nurse puts on gloves, then a mask. Grabs a gown from the stash. The patient begins coughing heavily, so nurse hurries into the room while carrying the gown, puts the gown under underarm, and completes a patient-related task. With gown under the underarm, gloves and mask on, nurse quickly hurries out of the room. (Droplet precautions, site 2, non-ICU)

Mistakes

Mistakes, which are errors of intent, were often observed when personnel were dealing with a challenge and used a course of action that led to a failure with probable self-contamination (Table 2). Mistakes were frequently observed during doffing. One characteristic example was of personnel attempting to remove the PPE while holding an object:

• While holding a binder in left arm, social worker pulls the gown over their head and lifts the right side of the gown off so it drapes in front, uses a gloved hand to remove the right side of the gown, then moves the binder to the right side, holding binder against ungowned [now unprotected] section of clothing. Social worker begins to doff the left side of gown. Grabs the inside front section of gown [clean area] and pulls and breaks the back tie. (Contact enteric precautions, site 2, non-ICU)

Other doffing-related mistakes included using an incorrect removal sequence (eg, removing eye protection and mask with gloves still on), using a process that was not consistent with the product (eg, lifting a gown with a perforated neck over one’s head rather than tearing away), and using a rushed or aggressive removal process that often resulted in a presumed contaminated surface touching a clean surface, as these notes show:

• Physician shakes patient’s hand to conclude visit. Then, standing in the center of the room, physician uses gloved hands to lift the gown neck tie off over head. (Contact enteric precautions, site 2, non-ICU)

• Physician uses a circling motion in the air to twist the gown and gloves into a bundle and puts the bundle in the trash. (Contact precautions, site 1, ICU)

Finally, a variety of situations led to procedurally suboptimal actions, creating a transmission opportunity or self-contamination event and were thus classified as a mistake. Examples included challenges with badge-enforced login system for in-room computers, use of pockets to transport and/or retrieve medications or supplies, and difficulty with certain logistical situations, as indicated in the following notes:

• Nurse walks to computer and reaches inside gown with gloved hand to grab ID badge (gloved hand touches front of scrubs, inside of gown, ID badge). Scans ID badge on computer. (Contact precautions, site 1, ICU)

• Using gloved hand, physician reaches underneath gown to get stethoscope from white coat pocket. Places on patient chest and begins evaluation. When finished using stethoscope, wraps around neck and hangs over gown. Gown was removed without disturbing stethoscope placement around neck. (Contact precautions, site 2, non-ICU)

• The health care worker takes gown and gloves and discards into trash can. Turns to exit but the room door is closed, pulls down sweater sleeves to cover hands, uses the sleeve to open the door and exits, gels hands immediately after exiting the room. (Contact/multidrug-resistant organism precautions, site 2, non-ICU)

Slips

The third type of commonly observed active failure comprised highly automatic behaviors or slips (Table 2). Perhaps the most innocuous type of failure, slips routinely resulted in a high likelihood of personnel self-contamination. Some of the most frequently observed slips involved personnel touching their face with contaminated gloves or external parts of a gown and subconsciously or automatically using their personal devices:

• Nutritionist is standing and leaning on the overbed table with gloved hands. During the conversation, nutritionist uses a gloved hand to push hair behind ear, then places hand back on the table, then uses a gloved hand to push eyeglasses higher up on face. (Contact precautions, site 2, non-ICU)

• Respiratory therapist moves the oxygen sensor on the patient, wipes own forehead with gowned forearm, and then moves the oxygen sensor again. (Contact enteric precautions, site 1, ICU)

• In the middle of a discussion with patient, social worker’s phone rings. With a gloved hand, social worker reaches behind gown to grab the phone from pants pocket, answers the phone, holding the phone to face. After the phone call ends, reaches behind gown to store the phone back in pants pocket. (Contact enteric precautions, site 2, non-ICU)

Discussion

Transmission-based precautions and PPE use are recommended strategies for reducing the potential spread of infectious diseases in health care settings. Our findings identify several active failures in PPE use and precaution practice during routine hospital care that could result in self-contamination or transmission. Commonly observed failure types were violations (deviations from recommended operating practices or procedures), mistakes (failures of intention), and slips (failures of execution). Our analysis also revealed the relatively broad array of contributing factors or circumstances within each failure type, suggesting that a range of strategies may be necessary to reduce potential transmission risk during clinical care.

One type of violation was room entry without PPE by health care personnel who were rushing to respond to a perceived patient emergency. Prioritizing serious patient needs, which may be of greater consequence than the potential transmission of many common infectious organisms, cannot and should not be modified in most cases. A more typical violation, however, was room entry by personnel with no PPE or partially donned PPE to complete certain tasks, such as talking with patients or delivering supplies. Such violations appeared to be relatively deliberate given that precaution signs and PPE supplies were in view and accessible before room entry. Intent cannot necessarily be observed, but often it appeared that personnel did not anticipate touching the patient or patient environment by, for example, placing their hands behind their back. Nonetheless, although some personnel completed their task without contact, there were frequent intentional and inadvertent touches involving the patient or patient environment. This form of violation could suggest a lack of knowledge or a calculated determination by personnel about the potential risk (to self or to other patients) vs the perceived expenditure of time and resources associated with PPE use.¹⁶ Thus, although education could be a useful strategy, a re-evaluation of precaution policies and practice may also be needed. This re-evaluation includes consideration of strategies that allow for more latitude, such as defining an area in the patient room where PPE is not required.¹⁷ A question for further study is whether eliminating precaution requirements for select endemic organisms¹⁸ might reduce their perceived burden and thus violations. Nevertheless, minimizing violations remains an important topic given the emergence of new organisms and the possible unknown risks to those exposed.

Unlike violations, which are a deliberate breach in practice or protocol, mistakes generally result from faulty decisions or planning. Mistakes often occurred during PPE doffing, whereby faults in removal technique could have led to the transmission of an organism to health care personnel, their clothing, or other objects that were transported outside the patient room.¹⁹ Factors contributing to these mistakes may include knowledge deficits (wrong sequence) or perceived time pressures to move on to the next task or patient (careless removal process). Other mistakes were associated with difficulty in handling certain logistical situations and less-than-ideal environmental or equipment factors, such as gaining access to in-room computers or juggling items while removing the PPE. Poor planning was another contributor, especially the storage and retrieval of items in pockets after donning the PPE and during direct patient care. As such, reducing mistakes requires not only tackling possible knowledge gaps but developing guidance and strategies for dealing with specific environmental and logistical challenges.

The most difficult to address and yet concerning type of failure is the slip. Slips are associated with automatic behaviors and attention-related failures that occur during the performance of routine tasks.¹⁴ Health care personnel were observed, with seemingly little awareness, pushing up glasses or wiping sweat from their face with contaminated gloves or gowns. Another common scenario was the automatic response invoked by a device, which resulted in personnel reaching with gloved hands underneath their gown and into a pocket to retrieve the device, respond to a message, and then replace the device back in their pocket. These automatic responses usually appeared to be instinctual actions and not intentional breaches, but there was a high likelihood of potential transmission. Preventing a reflexive action is inherently difficult and requires a high level of awareness and cognitive input. Use of full-face protection might be 1 strategy for combating some of these occurrences but given the resource, comfort, and workload implications, might best remain reserved for certain organisms. Similarly, slips caused by the use of devices could be prevented by banning them in precaution rooms. Increasing reliance on technology, however, could make this a difficult policy to implement and enforce. Video-reflexive techniques, involving video recording and review by health care personnel to facilitate their own recognition of infection risks, might be another potential strategy for promoting awareness of slips and mindful behavior while caring for patients in precautions.²⁰

Our findings are consistent with the issues involving appropriate PPE use identified by other studies.^8,10,11 Similarly, our results suggest that previously identified factors, including knowledge and risk perceptions,^21,22,23 likely contribute to the observed failures. Improving the use of PPE often requires skills training and correcting knowledge deficits,^24,25 but our in-depth assessment broadens existing knowledge, including potential targets, contributing factors, and strategies for intervention and improvement. Some failures are the result of intentional violations, whereas complex tasks rely more frequently on cognition, which is negatively affected by error-producing conditions (eg, multitasking, time pressures) that can lead to mistakes and slips.^26,27 Consequently, reducing active failures in precaution practices requires behavioral, organizational, and environmental solutions. Some examples include gown redesign to improve functionality; environmental cues to prompt appropriate actions, such as proper doffing sequence; changes to the built environment to ensure sufficient space and access to supplies during care delivery; and identifying more effective strategies for dealing with specific challenges, such as in-room computer log-ins and cognitive load issues.

Limitations

This qualitative study has limitations. First, data were collected through observation, and health care personnel were informed that they were being observed. Although lack of blinding could result in some personnel altering their actions, research shows that observer-induced bias tends to be limited.²⁸ In addition, changes in behavior owing to being observed would, if anything, be expected to decrease (rather than increase) the number of observed failures. Second, observations were conducted by several observers, and the data collected could be affected by observer background and perceptions. To combat this potential issue, all observers received the same training and were instructed to record exactly what they saw without interpretation. Third, there is potential for misclassification bias because the intent behind an error (ie, if intentional, the error is a violation) must often be inferred, which is not necessarily observable. Finally, the observations occurred at 2 affiliated medical centers. Although our findings are specific to these 2 sites, our data allow an assessment of the degree to which similar issues might occur at other institutions. In addition, there are differences in policies, recommended practices, and products between sites, but similar issues were observed across sites in general.

Conclusions

Active failures in PPE use and transmission-based precautions, potentially leading to self-contamination, were commonly observed. These failures included violations in practice, procedural mistakes, and unintentional slips. However, our assessment of active failures was not intended to call attention to the failed actions; instead, we wanted to identify the challenges faced by health care personnel that need to be addressed to promote effective PPE use. The broad array of contributing factors in each type of failure suggests that some circumstances may be more modifiable than others and that a range of strategies—behavioral, organizational, and environmental—may be needed to reduce the transmission risk during routine hospital care.