Understanding the heterogeneity of labor and delivery units: Using design thinking methodology to assess environmental factors that contribute to safety in childbirth

Abstract

Objective:

There is limited research exploring the relationship between design and patient safety outcomes, especially in maternal and neonatal care. We employed design thinking methodology to understand how the design of labor and delivery units impacts safety and identified spaces and systems where improvements are needed.

Study Design:

Site visits were conducted at 10 labor and delivery units in California. A multidisciplinary team collected data through observations, measurements and clinician interviews. In parallel, research was conducted regarding current standards and codes for building new hospitals.

Results:

Designs of labor and delivery units are heterogeneous, lacking in consistency regarding environmental factors that may impact safety and outcomes. Building codes do not take into consideration workflow, human factors and patient and clinician experience. Attitude of hospital staff may contribute to improving safety through design. Three areas in need of improvement and actionable through design emerged: 1) blood availability for hemorrhage management 2) appropriate space for neonatal resuscitation 3) restocking and organization methods of equipment and supplies.

Conclusion:

Design thinking could be implemented at various stages of healthcare facility building projects and during retrofits of existing units. Through this approach, we may be able to improve hospital systems and environmental factors.

Introduction

Giving birth is the most common reason for hospital admission in America¹. Despite advances in modern medicine, the peripartum period presents high-risk situations for mother and neonate. Patients may be at risk because of medical factors and also because of the systems in which they are being treated^2,3. An active error, such as a clinician administering the wrong medication, may be easy to identify. Yet, behind these active failures may exist latent conditions that contribute to morbidity and mortality⁴. These latent conditions have been described as inevitable resident pathogens and may include systemic, environmental factors such as physical resources, protocols, or culture^5,6. Qualitative work may help to identify latent conditions in the labor and delivery space. For example, qualitative investigation to achieve human-centered design solutions for childbirth in low resource settings such as Nigeria and Uganda has led to findings on what local communities need that have been translated into standards of care⁷.

Design thinking methodology may be a powerful tool in conjunction with systems thinking to expand our understanding of problems from a human-centered perspective⁸. Design thinking takes a human-centered approach to solving problems and is chiefly concerned with end users, which in medicine are both clinicians and patients. Taking a human-centered approach means identifying core values of the users and problems that users themselves may or may not be able to recognize such as workarounds, awkward ergonomics, inefficiencies, and process or environmental factors issues that unnecessarily increase cognitive and physical workload. Table 1 outlines the steps in design thinking methodology. Design thinking reveals insights in the synthesis stage when tensions, contradictions, and surprises emerge from extensive research^9,10. Design thinking welcomes outsiders, who are able to see what insiders cannot. The parable of a cat asking a fish, “How’s the water?” and the fish replying, “What water?” illustrates the importance of investigating what others may take for granted.

Table 1.

Characteristics of design thinking

Characteristic	Definition
UNDERSTAND (empathy)	Investigators are immersed in learning. The goal is to develop background knowledge of the environment and its users. What problems do they face? Watch how people behave and interact, and observe physical spaces and places in several settings. Investigators will talk to people about what they are doing, ask questions, and reflect on what they see. Insights are developed.
DEFINE (the problem)	Investigators create need statements that define the problems identified.
IDEATE (brainstorm)	The ideation phase is critical. Investigators brainstorm myriad ideas and suspend judgment. Quantity is encouraged.
PROTOTYPE (make)	Investigators convey an idea quickly and test it with users. Prototyping is rough and rapid. A prototype can be a sketch, model, or skit.

^aThese characteristics of design thinking and their definitions were adapted from material released by the Hasso Plattner Institute of Design at Stanford⁹.

We recognize design thinking is not a typical qualitative research methodology used in scientific and biomedical research. However, design thinking is gaining traction in the health care profession and is already being used in prominent medical schools as part of their curriculum for medical students and residents. Two of our authors have taught classes to university students, including medical school students, using design thinking methodology^11–14.

There is limited research exploring the relationship between design and patient safety outcomes, especially in maternal and neonatal care^{15, 16}. To address this gap, we conducted a multi-site observational study of labor and delivery units in California. Using design thinking methodology, we aimed to understand user needs, determine the prevalence of human-centered design, and identify areas of design that impact safety, care, and workflow.

Methods

We convened a multidisciplinary team of clinicians and non-clinicians to be engaged in this project. The team included clinicians in pediatrics, maternal-fetal medicine, obstetrics, and nursing, as well as experts in engineering, design, and human factors. We conducted a multi-site observational study using design thinking methodology to analyze data and make cross case comparisons.

Design thinking methodology aims to define and prioritize problems based on patterns identified in the research. Design thinking relies empathizing with users through interviewing users, and observing or shadowing users in their environment to identify patterns. Through pattern recognition, we are able to identify the most important problems (a process similar to other qualitative techniques, like grounded theory).

Members of the study team contacted physicians, medical directors and unit managers at labor and delivery units. The inclusion criterion required for participation was having a labor and delivery unit located at a hospital in California. Thirteen units were invited to participate via phone, email, or in-person invitation, and the first ten units that accepted our invitation were included in the study. The participating ten units represent a convenience sample. Institutional refusal was the only exclusion criteria.¹⁷

The Institutional Review Board at the host institution approved the project protocol on July 13, 2015 and we have renewed every year since. We consented our participants by providing them with a Research Information Sheet, allowing them a chance to read it over and ask questions before we embarked on the site visit. This sheet included a description of our study, estimated time involvement, risks and benefits, a clause stating that no payments would be made in exchange for study participation, and a statement of subject’s rights. The sheet also included contact information for both our senior author and an independent contact at the host institution.

The team performed site visits from July 2015 to February 2017. We used design thinking methodology to assess: 1) the overall labor and delivery unit 2) individual patient rooms (e.g. labor, operating, antepartum and postpartum) 3) staff working areas 4) other areas (e.g. break rooms, locker rooms, storage). Data were obtained through observations, photographs, tours, and interviews with clinicians. On average we spoke to three clinicians per hospital. At each unit, we interviewed the doctors or nurses in leadership roles who gave the tour. We also interviewed a convenience sample of on-duty clinicians. While we made formal quantitative measurements,¹⁷ we made the decision to keep our qualitative observations open-ended, unstructured, and observational and therefore did not follow formal interview guides. Because we were touring operational units with patients present, we did not record our interviews in order to ensure patients’ privacy. We avoided observing direct patient care or taking photographs in the presence of patients or protected health information. To protect hospital privacy, hospital names were kept confidential and data was saved on a password-protected server administered by the host institution. During and immediately after each site visit, team members took notes on what they observed. Materials were securely circulated electronically to all team members. Design thinking is an iterative process, meaning we were able to understand more about labor and delivery spaces and further define aspects of design in need of improvements based upon our findings from each site visit. After each site visit, we shared our notes at monthly team meetings and discussed findings with team members who were not able to be present for the visit. This dialogue informed subsequent site visits and the synthesis of key ideas for this study.

During the same time period in which we conducted the site visits, we researched how hospitals are constructed and designed. As a supplement to site visits and clinician interviews, we spoke with individuals with knowledge of building codes and hospital design. These unstructured discussions were intended to give the site visits context and provide us with a more comprehensive understanding of the top-down factors that influence hospital design, namely building codes and hospital construction guidelines. Of note, we identified three members of the Facilities Guidelines Institute (FGI), an independent, not-for-profit organization that provides guidance for planning, design and construction of hospitals. One of the members of FGI we spoke with is a surgeon and an architect and a founder of FGI.

To provide structure to the observational data we collected, we’ve grouped the findings under broad questions about labor and delivery unit space and design. The following questions were developed based on collaborative discussion between clinicians and non-clinicians before during and after site visits: 1) What factors influence hospital design? 2) How prevalent is human-centered design? 3) What aspects of design could be improved to impact safety, patient care, and clinician workflow? With regard to question 3, we used design thinking to prioritize and define problems based on patterns identified in the research. We identified three specific areas that may significantly influence maternal and neonatal morbidity and mortality.

Findings for the first two phases of the design thinking process (Table 1: 1. understanding or developing empathy 2. defining the problems), are reported in the Results. Illustrations are utilized to display observations while protecting institutional and patient privacy. Findings are presented as existing environmental factors and as specific problems that may impact safety, care, and workflow. For this paper, our team chose to address issues we believe are of the highest priority for clinical care and actionable through design.

Results

We visited seven private, non-profit hospitals, two academic hospitals, and one public hospital owned by the county where it is located. Seven hospitals were located in a large city, one hospital was located in a small city, and two hospitals were located in a large suburb. The age of the labor and delivery units differed. At least one unit was less than five years old, whereas several others were more than twenty years old. We visited units with maternal levels of care II, III, and, IV. Unit floor plans and the space allocated for labor and delivery rooms and operating rooms varied significantly among facilities. Although much of the medical equipment was similar, the placement of such equipment varied between, and even within, institutions. The following are our findings grouped under the three broad questions identified in the methods:

1). What factors influence hospital design (including new building construction standards and retrofitting)?

Based on our conversations with architects that work on healthcare construction projects, building codes are set mainly to ensure general life safety (e.g. emergency exits, strength of building materials, plumbing, electrical systems). The state of California has its own building codes for the construction of hospitals, skilled nursing facilities, licensed clinics, and correctional treatment centers. These codes are developed, as necessary, by the Facilities Development Division Building Standards Unit, part of California’s Office of Statewide Health Planning and Development. These codes provide minimum requirements for life safety but do not take into consideration workflow, human factors, and patient or clinician experience. Stakeholders we interviewed described the challenges of constantly changing building codes. For example, at the time of design and construction a unit would be up to code; after a time, it may be operational but no longer up to code. Changing building codes created additional barriers to remodeling units. Remodels were sometimes avoided because of the impossibility of making an operational-but-older unit fit current codes. At one facility, a former labor and delivery room was split into five antepartum spaces separated only by curtains. Sensitive conversations and private health information could easily be overheard by patients and family members in adjoining spaces.

Guidelines for Design and Construction of Hospitals and Outpatient Facilities (2018 Edition)¹⁸ was produced by the Facility Guidelines Institute (FGI), a multidisciplinary group organized to better understand the ever-changing needs of clinicians and patients as they pertain to the physical environment. It should be noted that these guidelines apply only to new construction and do not include suggestions for retrofitting existing buildings.

According to the clinicians whom we interviewed, input from patients and clinicians was not uniformly sought during the hospital design process. Most units reported that clinicians were not directly involved in the design phase of the healthcare facility. Some units reported that clinicians were involved in simulations and testing of a unit after it had been built, but that this limited the number of changes that could be made to a design. In contrast, one facility built a simulated labor and delivery unit environment, and tested equipment placement and workflow with all stakeholders prior to construction.

2). How prevalent is human-centered design?

All sites emphasized the importance of “patient-centered care.” It was our observation that this buzz phrase was rarely congruent with true human-centered design, which would encompass all users. Examples of human-centered design were identified at every hospital, but in different capacities. There was no link between the age of the hospital and human-centered design examples we noticed. Notable examples of human-centered design were rare but impressive when found.

In some units, we observed that human-centered design was utilized to compensate for space or design or resource limitations. Notably, an older unit decided to rethink and redesign how cabinets, drawers, and closets were restocked in labor and delivery rooms. By providing a dedicated feedback notebook in a prototype labor and delivery room, the nurse manager encouraged all users (clinicians, technicians, and maintenance staff) to give feedback about the new layout and organization of supplies. By promoting inclusivity and buy-in from all members of the team in the design process, this unit quickly found solutions and removed barriers to adoption.

Another example of human-centered design in a separate hospital, a newer unit at a large hospital, was a elevator dedicated for labor and delivery clinicians that connected the lower floor, housing antepartum rooms and labor and delivery rooms, with the upper floor, where the operating rooms were located. The elevators themselves were very large and featured intentional supply storage. They incorporated hanging, soft storage pockets and containers for items like scrubs, booties, hair covers, masks, and gloves. This was human-centered because it allowed clinicians to spend time in the elevator efficiently, especially accommodating those who arrive to the unit from elsewhere and quickly need to prepare to enter the operating room. Of course, sterility issues would have needed to be considered, but this elevator design seemed to compensate, in part, for the potential time delay in moving patients, staff, and equipment between floors. This innovation is an example of understanding the scenarios that happen and how to streamline and improve safety. Because this hospital was newer, they also had the privilege of thinking ahead and incorporating human-centered design elements into the actual architecture of the building and its infrastructure.

A third example of human-centered design, in another, older unit we visited, was a large format photo book depicting a woman giving birth via C-section. This book was used as an educational tool for both scheduled and emergency C-sections. These easy-to-interpret images were reported to help reduce patients’ fear and provide them with the ability to understand and anticipate future events in their care.

Interestingly, we observed that while a newer unit may be outfitted with cutting edge technologies and clinicians capable of delivering a high level of care, it may not feature human-centered design. For example, the main hallways and operating rooms in one newer institution contained a large amount of equipment that made it difficult for clinicians to maneuver. We also saw a hospital with hallways so wide and long that clinicians and patients had difficulty navigating the unit according to those we interviewed. Other issues that were observed included: clocks that were not synchronized; the absence of digital clocks in operating rooms; unmanaged cords (Fig 1) in both operating rooms and labor and delivery rooms; and four different types of trashcans that were not integrated into the room design, thereby obstructing pathways, cabinets and other equipment.

Figure 1:

Unmanaged cords became a burden for staff when a new skin-to-skin protocol was put into place in the OR

Privacy issues impact patients and clinicians

We also observed variability in human-centered design with regard to patient and clinician privacy. In 2008, FGI recommended single patient rooms as the gold standard in design. In fact, it was the group’s single most discussed issue. We found that 90% of units had or soon would have exclusively private postpartum rooms. However, we observed many triage and recovery room areas with semi-private rooms divided by cloth or plastic curtains. Clinicians who were interviewed expressed concern that these shared spaces could contribute to hospital-acquired infections. They also noted that sensitive discussions, where protected health information is relayed, could potentially be overheard by unintended parties. On the other hand, hard walls and doors were considered by some clinicians to be an impediment to care because they created barriers to efficient transport of patients in urgent situations. We also observed a lack of privacy for clinicians themselves. Several clinicians reflected on privacy issues that impacted their work. They noted there was limited space to have private conversations with coworkers, quietly reflect on a challenging case, or simply rest. In fact, one clinician at a large, high-level care facility remarked that the only location providing personal privacy for clinicians was the restroom stall.

Positive Attitudes Supporting Change

Often, it’s a hospital’s clinicians and staff members who help transform a space to exhibit human-centered design. We found that attitude played a very important role in the functionality of a unit. Informal design thinking by staff led to novel solutions that were used by individuals or incorporated into systems-level improvements. For example, introduction of a new protocol for skin-to-skin care following cesarean section birth required walkways to be cleared in order for neonatal teams to safely initiate skin-to-skin contact between mother and newborn. Despite being told by the hospital’s space planning team that it wasn’t possible to clear the hallways, the unit’s nurse manager and her team took it upon themselves to prototype an actual operating room with the needed clearances. They succeeded within a week. Once the proposed operating room design was approved by hospital administration and physicians, the nursing staff duplicated this effort in the unit’s second operating room.

3). What aspects of design could be improved to impact safety, patient care, and clinician workflow?

After reviewing the data from all 10 units, we identified several clinical scenarios, processes, and protocols where environmental factors impact safety and have the potential for actionable improvement through design. The team consensus was to focus on three specific areas that may significantly influence maternal and neonatal morbidity and mortality: 1) blood availability during postpartum hemorrhage management; 2) appropriate space for neonatal resuscitation; and 3) restocking and organization methods of equipment and supplies. Table 2 shows our recommendations for improving design in these areas.

Table 2:

Design recommendations for improving safety and experience in labor and delivery units

	Baseline Requirement.	Better	Best
Blood availability during postpartum hemorrhage	(A) Storage of blood products in same building	(B) Storage of blood products on labor & delivery floor	(B) and (C) Prototyping new protocols for blood delivery to find most efficient method; OR (D) Storage of blood products outside the operating room is optimal if patient volume supports this kind of system
Appropriate space for neonatal resuscitation	(A) Necessary resuscitation equipment should be easily accessible and not block other supplies	(A) and (B) Dedicated space for clinicians to surround the infant warming table and perform resuscitation	(A), (B), and (C) Updated resuscitation equipment and supplies (on par with the NICU)
Restocking and organization methods of equipment and supplies	(A) Standardized equipment and supplies in all rooms	(A) and (B) Testing different restocking methods to discover which is optimal (e.g. nurse vs. technician restocking room)	(A) (B) and (C) Designing storage solutions for all basic equipment and supplies by incorporating input from all users

Blood Availability During Postpartum Hemorrhage

In eight of the ten hospitals visited, blood was available in the same building where labor and delivery was located. Average time for retrieval from time of call was reported to be seven minutes. At one hospital, those interviewed reported that blood was only available in a building three blocks down the street. Average time for retrieval at this hospital was 14 minutes from the time the call is placed. In contrast, at another hospital, blood was available directly outside the operating rooms in a small refrigerator. These supplies were constantly rotated in order to maintain a fresh supply of blood and blood products. Average time for retrieval at this institution was reported to be one minute. A larger supply was also available one floor below.

Regarding blood availability, distance was not the only factor cited by stakeholders. Indeed, the level of personnel experience also was cited by some hospitals to have an impact on getting blood to hemorrhaging patients on time. While most hospitals had a massive transfusion protocol in place, the contents, initiation steps and retrieval logistics to get the blood to hemorrhaging women were variable. In a notable example, one hospital, after understanding that blood retrieval times were variable, prototyped multiple ways to get blood. They found that sending a technician as a runner via a particular stairwell was faster than having the blood bank send blood in a separate container through the tube system, or the blood bank delivering the blood to the labor and delivery unit via courier.

Appropriate Space for Neonatal Resuscitation

Neonatal resuscitation equipment was allocated to a corner of most labor and delivery rooms. With the exception of two newer hospitals that had a dedicated alcove for this equipment, the location of this equipment was not standardized within and across institutions. Not only was resuscitation equipment pushed against a wall, blocking glove boxes, light switches and other equipment (Fig 2), its placement did not allow for three or more clinicians to perform complex neonatal resuscitation. We noticed that the resuscitation beds allocated to the labor rooms were older models, often handed down from the NICUs when newer, more modern equipment became available.

Figure 2:

Often, there is not enough space in a labor room or operating room for a pediatrics team to perform neonatal resuscitation on a baby in respiratory distress.

The location of the neonatal resuscitation equipment in the operating rooms was also highly variable between hospitals. In some rooms neonates were resuscitated next to the door, so if the patient required transfer to the NICU the path was relatively unobstructed. In certain rooms, however, neonatology teams were so close to the door they could be struck when others entered or exited. At other hospitals, neonatology teams had to maneuver around equipment and pass through the sterile field in order to travel to the NICU. Two hospitals, one old and one new, had what was called a pass-through window in the operating room. Here, a neonate born via cesarean section is handed from one clinician to another through a small sliding window installed in the wall. This window is the passageway for transferring the newborn to a separate resuscitation room adjacent to the operating room.

Restocking and Organization Methods of Equipment and Supplies

We observed a range of tactics for stocking supplies. Most frequently, hospitals had a dedicated technician who restocks the labor and delivery rooms after each birth. Some hospitals required a nurse to double check the restocking done by the technician before admitting a patient. One hospital’s workflow was to have the attending nurse restock the room. It was found that when the attending patient’s primary nurse owns the labor and delivery process from beginning to end, the responsibility becomes personal, leading to fewer restocking errors. They reported that this system decreased restocking errors by a margin large enough to justify eliminating restocking technicians altogether. Although costlier per hour than a technician, nurses were more informed about what was used during their patient’s stay.

Organization of supplies also varied. At some hospitals, supplies were scattered among numerous carts (some not even in the room, but outside in the hall), cupboards, and built-in drawers. At other hospitals, custom built-ins for certain products were standard in every labor and delivery room. All units had delivery carts in each room, containing common supplies used during vaginal delivery. One hospital was notable for its hybrid cart design, novel in its containment of supplies all in one place immediately accessible to clinicians. It houses both the standard obstetrics equipment for vaginal delivery and an epidural box with anesthesia equipment necessary to perform an epidural or emergency cesarean section. This particular hospital’s solution eliminated the need for clinicians to take time to travel to a central location for the labor analgesic medications, which can disrupt workflow.

Another reorganization effort that improved workflow was mentioned above in the “How prevalent is human-centered design?” sub-section. At this older unit, the charge nurse led a three-month effort to change how medical supplies were stored in labor and delivery rooms and the nursery. They purged the unit of anything extraneous and then re-organized supplies to improve standardization. One example from the redesign effort involved installing walk-in closets in labor and delivery rooms with custom racks that stored labor aids for patients.

Discussion

Our aim was to use design thinking to investigate how design impacts safety and experience and better understand the needs of users in the labor and delivery unit environment. In contrast to the more traditional systems thinking that puts users at the sharp end, design thinking fosters human-centered design which places users at the center¹⁹ (Fig 4). It works to cultivate empathy and identify latent conditions, and it begins to address the question of how design can influence safety, workflow, and quality of care.

The 10 California labor and delivery units included in this observational study varied in size and delivery volume, but were all maternal care levels II, III or IV²⁰. By no means exhaustive, our exploration showed variability in design and inconsistent consideration for human factors. Despite the heterogeneity, we observed common patterns in building standards and retrofitting. Design was frequently constrained by the actual buildings or geographic sites. Frequently changing building codes ensure basic human safety measures, but do not emphasize workflow, human-centered design, or efficiency. Units are built to meet current codes and have fixed floor plans with little flexibility to accommodate changing equipment and long-term fluctuations in patient volume. In many older units, retrofitting and/or remodeling may be needed; however, there are no specific recommendations for environmental improvements. Moreover, space and financial limitations often hinder renovation plans.

The Center for Health Design identified the physical environment as a key latent condition that impacts safety in the healthcare setting. The report proposed focusing on patient safety issues during the pre-design phase of a healthcare facility building project²¹. In 2014, Safety Risk Assessments (SRAs) were incorporated into FGI’s Guidelines for Design and Construction of Health Care Facilities, recognizing a need to foster a more proactive approach to patient and clinician safety that takes into consideration environmental factors²². Indeed, we did visit one site where they simulated the labor and delivery environment prior to construction with a variety of stakeholders present. This is a highly desirable process, and any initial investment in space and other resources to conduct these simulations is highly likely to be recouped in the form of fewer costly revisions later in the construction process.

Encouragingly, as design becomes less of an afterthought, we envision sharing tested and proven designs as they apply to a unit’s physical environment. For example, solutions for the organization of supplies could be recommended and potentially integrated into the architecture itself. This kind of foresight may reduce errors and save time. One way to employ a human-centered approach toward improving healthcare design would be to create a national database that could house examples of design, both successes and failures. Our broader vision is that design thinking creates a framework for better outcomes.

Following an Institute of Medicine Result report “To Err is Human”³, there was a substantial increase in the number of patient safety publications with the most frequent subject of study being organizational culture²³. Indeed, our site visits confirmed that attitude matters and organizational culture impacts the physical work environment and clinician experience. Successful design changes come from a culture of inclusion where every voice matters and buy-in from all users decreases psychological barriers to innovation. Finally, champions in each unit can lead critical thought processes, create prototypes, and test ideas that improve the labor and delivery environment. We gave two examples of an organization’s members working together to find solutions: 1) one older hospital prototyped new labor and delivery rooms, and there was a book staff members were encouraged to use to provide their feedback on the design in order to find solutions; and 2) the team that led the effort to create the appropriate walkway clearances in operating rooms in order to enable a new skin-to-skin protocol.

Our observations revealed many problems that could be addressed with design thinking. Three issues described here—1) blood availability during postpartum hemorrhage 2) appropriate space for neonatal resuscitation 3) restocking and organization methods of equipment and supplies—had strong ties to safety and were actionable through design.

First, hemorrhage remains a leading cause of maternal morbidity and mortality in the U.S. The Centers for Disease Control and Prevention report that 11.5 percent of pregnancy-related deaths from 2011 to 2014 were a result of hemorrhage²⁴. Managing hemorrhage requires a coordinated multidisciplinary effort by obstetrics, anesthesiology, nursing, laboratory, and transfusion medicine²⁵. From the perspective of the hemorrhaging woman and the clinicians taking care of her, reliable and quickly accessible blood products are paramount. That said, blood products are an expensive and time-intensive resource with unique risks and challenges. The range of needs during obstetric hemorrhage—from no transfusion, to limited transfusion, to massive transfusion—makes it difficult to predict how much blood, if any, to keep in labor and delivery units. We propose that all units consider ease of access to blood supplies. New and old units may want to reassess what practices are in place and simulate new ways of accessing supplies to ensure that protocols are as efficient as possible.

Secondly, we noted inadequate space for neonatal resuscitation. Labor rooms have not been intentionally designed to incorporate state of the art NICU equipment with reduced footprints and consideration for human factors²⁶. Most hospitals in the U.S. use space within the delivery room to resuscitate newborns (83%)²⁷, and only a small fraction (15%) have a dedicated resuscitation room²⁶. The ideal design of a labor and delivery or operating room would easily accommodate any kind of birth and resuscitation.

Thirdly, we observed examples of restocking methods and organization of supplies that reduced error and improved workflow. One hospital found it was better for nurses to restock their patients’ rooms instead of technicians; at another, an empowered nurse manager completely reorganized equipment and supplies, leading to improved efficiency and reduced clinician frustration during emergencies. Easy access to fully stocked supply carts and cabinets is important and may help to prevent operational failures. Our findings are supported by another study that shows how standardizing the location of emergency equipment in an ICU using a drawer divider improved the reliability, restocking times, and restocking workload for nurses²⁸. By taking time to understand design and implement organizational improvements to the environment, patient outcomes may also improve.

Our study has several limitations. Sample size was small. Because we studied labor and delivery units that were accessible to us and relied on speaking with clinicians at each unit who were available, our results are influenced by convenience bias. Because we prioritized patient privacy, we did not audio record interviews with clinicians and relied on notes and team discussions. Therefore, we were unable to formally analyze interviews using traditional qualitative methodologies. Furthermore, while we did not observe or interview patients, they are important users of labor and delivery units. We had in our minds an indirect view of the patient perspective, either from firsthand experience as a patient who gave birth in a labor and delivery unit or through the lens of a clinician reporting on patient experiences in her unit. We acknowledge this biased viewpoint and recognize our study is limited in that it does not represent the views of patients at the sites we visited. Future work in this area should directly examine patient perspectives of labor and delivery unit space and design. Gathering the opinions of women and families, while beyond the scope of this particular study, is something that our Safety Learning Lab for Neonatal and Maternal Care is committed to investigating.²⁹

Our work was limited in several other ways. Because our team included clinicians who work in labor and delivery units, we acknowledge our experience may have influenced the questions we asked during site visits and the problems we may have been primed to find. In order to minimize any bias in our observations, we included non-obstetric healthcare professionals on the team. We tried to make observations across sub-specialties as well. For example, we had a neonatologist looking at products, protocols, and systems that might affect obstetric care, clinicians inquiring into the workflow of nursing care, and vice versa. Furthermore, clinicians on our team could never fully understand the workflow of the insiders working in a unit that was not theirs. Simultaneously, a strength and limitation, non-clinicians on our team could not fully comprehend the workflow from a clinical perspective.

Our study has the potential to form a foundation for further research in labor and delivery unit design. Approaching healthcare design as a science recognizes its capacity to affect the fundamental system and culture of an organization, and may lead to sustainable improvements that make hospitals safer places for both patients and clinicians. We believe it would be beneficial for architects, engineers, and designers to communicate regularly with hospital staff and integrate fieldwork to identify a hierarchy of needs for any healthcare facility building project. Our ultimate goal is to initiate more interest in healthcare environment design with the intent of improving patient safety and outcomes.

Figure 3:

Systems design places providers and patients at the sharp end, whereas human-centered design has them as the focal point