Architecture

This article was previously published in the Summer 2010 edition of HERD Journal. For more information, visit http://www.herdjournal.com.

It is estimated that by the year 2015, $200 billion will have been spent on hospital design and construction across the United States (Ulrich, Quan, Zimring, Joseph, & Choudhary, 2004). Current healthcare design trends include an emphasis on patient- and family-centered care, improvements in the quality and cost-effectiveness of care delivery, the integration of new technologies, and adherence to new design guidelines intended to promote safe, sustainable, and healing environments for patients and staff (Agency for Healthcare Research & Quality, 2005).

With the growth in the healthcare design market come opportunities to create innovative design solutions to address current as well as long-standing healthcare delivery issues. Improving the work environment of staff, specifically nurses, is an important component of the healthcare delivery equation.

The Work Environment of Nurses

As a profession, nursing has become better able to accommodate high turnover volumes, increased patient acuity, and variable staffing requirements. However, a shortage of qualified nurses is a growing concern in the United States. Although the demand for nurses is expected to grow by 40% between 2000 and 2020, the number of new nurses entering the field is expected to increase by only 6%. The resulting shortage of nurses, combined with an increased turnover rate of approximately 20% per year among the aging nursing workforce (Joint Commission, 2002), will result in an alarming situation that must be addressed by hospitals within the next 10 years (Hendrich, Fay, & Sorrells, 2004).

There is growing awareness of the importance of improving the work environment of nurses. According to a Peter D. Hart Research Associates (2001) survey of registered nurses, the primary reason nurses leave healthcare, other than retirement, is to find a job that is less stressful and physically demanding. This issue was directly addressed in the Institute of Medicine report, Keeping Patients Safe: Transforming the Work Environment of Nurses (2004). The report notes, “It is also essential to foster collaboration of nursing staff with other health care personnel in identifying high-risk and inefficient work processes and workspaces and (re)designing them for patient safety and efficiency” (p. 227). Providing a work environment that supports both the physical (functional) and psychological needs of nursing staff is crucial to improving patient care and nursing efficiency.

A growing body of research suggests a relationship between characteristics of the physical environment and the reduction of stress and fatigue among healthcare staff (Berry & Parish, 2008; Joseph, 2006; Ulrich et al., 2004). Evidence suggests that a supportive physical work environment, along with factors such as high autonomy, low work pressure, and supervisor support, may increase job satisfaction and reduce burnout among nurses (Mroczek, Mikitarian, Vieira, & Rotarius, 2005; Tyson, Lambert & Beattie, 2002). It would logically follow that increased job satisfaction and decreased burnout may improve nursing staff retention, reduce turnover, and increase patient satisfaction.

The Nursing Station

The nursing station is the heart and soul of nursing care activities in a hospital or long-term care facility. It is typically the primary work area assigned to a specific unit. The nursing station usually includes unit reception along with records storage and charting work areas (Office of Space Analysis, 2006). Walking distances, accessibility, visibility, and ease of supervision have been identified as crucial issues in nursing station design (Hamilton, 1993).

Advances in information technology have enabled nurses to move away from traditional centralized paper-charting stations to smaller, decentralized workstations and charting substations located closer to or inside patient rooms. Nursing units of the past were designed with one centralized nursing station. Advances in information technology have enabled nurses to move away from traditional centralized paper-charting stations to smaller, decentralized work stations and charting substations located closer to or inside patient rooms. Coupled with mandated private rooms in new hospital construction, these changes create longer patient care units and therefore longer distances for nurses to walk.

Healthcare is a multidisciplinary practice in which nurses, doctors, respiratory therapists, pharmacists, and other support staff need a place to interact. A centralized nursing station can provide space for collaboration among caregivers. Wakefield (2002) identified a number of other social functions that may be facilitated by the centralized nursing station. The results suggested that centralized nursing stations represent a symbolic barrier separating the public and private spaces of a unit and that they operate as a mechanism of social control where individuals are expected to formally seek permission to enter the unit before proceeding to patient rooms. From a pivotal location within the unit, the nursing station also serves to maintain workflow and is the setting for frequent social interaction and formal and informal teaching and learning activities. Research suggests that some of these important social functions may be affected in decentralized nursing station designs (Tyson et al., 2002).

More nursing unit designs are incorporating decentralized nurses’ stations outside patient rooms so that staff members are distributed around the unit closer to the point of care. This can improve visibility of the patient and reduce time spent walking. Ulrich and colleagues (2004) concluded that decentralized nursing stations allow nurses to interact more closely with patients. Hendrich, Fay, and Sorrells (2004) demonstrated that decentralized nursing stations reduced staff walking and increased patient care time, especially when supply spaces were also decentralized and placed near the nursing stations.

Although decentralized nursing stations may improve patient visibility and reduce walking distance, nurses report feeling more isolated from their colleagues and losing the sense of team connection in comparison to centralized nursing stations (Tyson et al., 2002). This implies that there are important social behavior components of nursing station design that should be considered.

Table 1. Unit/Nursing Station Characteristics

Unit Characteristics	Hospital A Unit 1	Unit 2	Hospital B Unit 1	Unit 2	Hospital C Unit 1	Unit 2
Nursing Unit Design	Y-shape	Y-shape	Single corridor, partly radial	Single corridor	Double corridor	Double corridor
Nursing Station Design	Centralized, open	Decentralized converted	Centralized, open	Decentralized corridor space	Centralized, enclosed	Decentralized, alcoves
Unit Patient Profile	Surgery specialties	Oncology surgery	Pulmonary surgery	Neuro-ortho surgery	Medical surgery	Heart surgery
Year of Design	2004	1980s	1974	1994	1972	2004
No. of Patients/Nursing Unit	36	36	15	44	26	18
No. of Patients/Nursing Station	36	11	15	9	26	2
Nurse-Patient Ratio
Day	3:5	3:5	3:4	3:4	3:5	3:4
Evening	3:5	3:5	3:4	3:4	3:5	3:4
Night	5:6	5:6	4:5	4:5	5:6	5:6

Nursing stations are often noisy, crowded, busy spaces where nurses consult with physicians, specialists, and nurse colleagues. Additional sound sources include alarms, pagers, telephones, and other technical equipment. The decentralized nursing station is intended to keep nurses closer to their patients, but as a result nurses and physicians may choose to meet and consult in hallways, within earshot of other patients and staff. Patients identified talking in the hallways as one of the most distressing sources of noise (Topf, 1985).

Staff members perceive increased sound levels at the nursing station as stressful (Bayo & Garcia, 1995; Morrison, Haas, Shaffner, Garrett, & Fackler, 2003). Blomkvist, Erikson, Theorell, Ulrich, and Rasmanis (2005) examined the effects of improving acoustic conditions in a coronary critical care unit. During periods of lower noise, staff members observed many positive outcomes, including improved speech intelligibility, lower perceived work demand, and lower perceived pressure. In a literature review, noise control has been identified as a significant way to improve hospital environments for both staff and patients (Ulrich et al., 2004). Yet little is known about how sound levels compare in centralized and decentralized nursing stations.

Study Purpose

Evidence-based findings of the effects of nursing station design on nurses’ work environment and work behavior are essential to improve conditions for and retention of this fundamental member of the healthcare delivery team. Improved understanding of the trade-offs presented by centralized and decentralized nursing station design has the potential to provide useful information for future nursing station layouts. Such information will be critical to bridge the understanding of the nurse environment “fit.”

The purpose of this exploratory study was to investigate how nursing station design (i.e., centralized and decentralized nursing station layouts) affected nurses’ use of space, patient visibility, noise levels, and perceptions of the work environment. Specifically, the study addressed the following research questions:

• Does nursing station design affect:
• The functional use of space?
• Patient visibility?
• Noise levels?
• Nurses’ perceptions of their work environment (both the physical and psychosocial environments)?

Method

The study used an exploratory design with both quantitative and qualitative methods. Quantitative data-gathering techniques included place- and person-centered space use observations, patient visibility assessments, sound level measurements, and an online questionnaire regarding perceptions of the work environment. Qualitative data on the effects of nursing station design on nurses’ health and work environment were gathered through focus group interviews.

A pilot study was conducted at two U.S. hospitals to improve the research methodology. Observation, sound level, and survey data collection techniques were tested and revised during the pilot study phase of the research. Data collected from the pilot sites were not used in the final data analysis.

Setting and Sample

A convenience sample of six nursing units located in three U.S. hospitals was selected for the study. Each of the three participating hospitals provided access to two medical/surgical inpatient nursing units-one with a centralized nursing station design and the other with a decentralized nursing station design. Sample hospitals were chosen based on the following criteria: similar service profile, 400 to 650 inpatient beds, 15 to 35 beds per nursing unit, and diverse U.S. geographic locations. Hospitals that fit the selection criteria were initially contacted and screened for interest by telephone. A hospital liaison was recruited at each participating hospital upon acceptance into the study to organize on-site activities. Unit type and design varied widely in the sample (Table 1).

Table 2. T-Test Results: Place Observations of Hospital A

Behavior (minutes observed)	M	SD	t	P
Note: Observations = 20 df = 18 * significant .05 **significant .001
Phone
Centralized Decentralized	11.6 4.1	6.8 4.3	3.0	0.008*
Computer
Centralized Decentralized	19.5 10.3	1.1 6.7	4.3	0.0004**
Paper Administration
Centralized Decentralized	13.3 9.2	6.1 7.6	1.3	0.201
Visual Patient Check
Centralized Decentralized	1.4 0.5	2.5 1.6	.99	0.360
Responding to Patient Calls
Centralized Decentralized	0.1 0.0	0.3 0.0	1.0	0.331
Consulting Medical Staff
Centralized Decentralized	12.2 6.4	4.8 3.6	3.1	0.007*
Direct Patient Care
Centralized Decentralized	0.3 0.0	0.7 0.0	1.4	0.18
Handling Supplies
Centralized Decentralized	2.5 5.4	1.5 2.5	-3.2	0.005*
Family Contact
Centralized Decentralized	0.6 0.0	0.7 0.0	2.7	0.014*
Social Interaction
Centralized Decentralized	4.6 1.8	4.8 1.9	1.7	0.103

In this study, a centralized nursing station was defined as a single room or area used by nurses and other patient-care staff to supervise unit reception and admissions and to administer healthcare services. A decentralized nursing station was defined in similar terms, excluding the reception and admission functions. Units in which there was more than one nursing station (of any size or configuration) were classified as decentralized.

Table 3. T-Test Results: Place Observations of Hospital B

Behavior (minutes observed)	M	SD	t	P
Note: Observations = 20 df = 18 * significant .05 **significant .001
Phone
Centralized Decentralized	7.1 1.5	6.5 2.1	2.6	0.017*
Computer
Centralized Decentralized	15.7 5.1	4.9 5.2	4.7	0.0002**
Paper Administration
Centralized Decentralized	19.1 6.0	1.4 7.0	5.8	0.0002**
Visual Patient Check
Centralized Decentralized	0.5 0.0	1.3 0.0	1.2	0.229
Responding to Patient Calls
Centralized Decentralized	0.6 0.1	0.5 0.3	2.6	0.020*
Consulting Medical Staff
Centralized Decentralized	7.9 1.1	6.3 1.5	3.3	0.004*
Direct Patient Care
Centralized Decentralized	0.6 0.0	1.6 0.0	1.2	0.245
Handling Supplies
Centralized Decentralized	4.0 3.8	2.5 2.6	0.2	0.863
Family Contact
Centralized Decentralized	0.2 0.1	0.6 0.3	0.4	0.660
Social Interaction
Centralized Decentralized	3.1 0.7	2.5 1.6	2.6	0.020*

Hospital A

Built in the 1960s, this 400-bed hospital is the largest not-for-profit healthcare system in its region. It operates in partnership with the nearby state university, delivering care through an integrated health network.

Unit 1: Y-Shaped Design With Open Centralized Nursing Station

The large, triangular, centralized nursing station in Hospital A contains three entry points and is partially enclosed by a modular wall system with glass panel tops that surround two sides of the station (Figure 1). The nursing station is carpeted and divided into a reception area (Figure 2), a central work area for nurses, and a separate computer area for physicians. A glass-enclosed room at the back of the nursing station is equipped with work stations. In all, there are approximately 30 work stations in the nursing station.

Unit 2: Y-Shaped Design With Enclosed Decentralized Nursing Stations

Hospital A’s decentralized nursing stations are transformed patient rooms located in the center of each corridor (Figure 3). The basic structure of these rooms has not been altered: windows, doors, vinyl flooring, and wall surfaces are all original (Figure 4). These nursing stations function as work stations for nurses and physicians and are equipped with one wall-mounted desk comprising three work stations. Medical supplies are located along the wall opposite the desk, and clean supplies are stocked in the area formerly used as a patient closet. There is also a small round table with two chairs near the window. The original centralized nursing station on this unit is still intact and is used primarily by the receptionist, the charge nurses, and other allied medical staff.

Hospital B

Originally constructed in the 1930s, this community-owned hospital contains nearly 500 inpatient beds and is the largest medical facility in its region, providing a variety of healthcare services. The hospital has achieved Magnet Status from the American Nurses Credentialing Center.

Unit 1: Single-Corridor/Radial Design With Open Centralized Nursing Station

An open, square, centralized nursing station is located in the center of the nursing unit. The nursing station contains two entry points (Figure 5) with a continuous, stand-up work station on three sides and a seated work station on the fourth. There is sufficient space for approximately six to eight people, including the receptionist (Figure 6). Nurses, physicians, and allied staff have access to this area. In addition, there are two work stations located in the corridor of this unit. The entire nursing unit floor is carpeted and the ceilings are low.

Unit 2: Single-Corridor Design With Decentralized Open Nursing Station Corridor Room

The decentralized nursing stations in Hospital B are located in the center of the corridor in each of the unit’s four wings (Figure 7). Each nursing station contains computer work stations with cabinets hung above and below the counter area (Figure 8). Additionally, there are two separate computer work stations located nearest the corridor. A separate supply room is located at the back of the nursing station. Each of the nursing stations also contains an ice machine and refrigerator. Small charting work stations are located in the corridors outside each patient room. The reception desk is centrally located on the floor, serving all four nursing unit wings.

Hospital C

Hospital C is a 650-bed community-based teaching hospital. It is a regional referral center for patients who require specialized medical care; more than 100 medical specialties and subspecialties are represented at this facility.

Unit 1: Double-Corridor Design With Enclosed Centralized Nursing Station

The centralized nursing station in Hospital C is located in the middle of the racetrack unit design (Figure 9). It is an enclosed space with glass windows and three entry points. The nursing station is divided into two different areas by a floor-to-ceiling partition wall. The larger side of this area contains eight work stations and is used by physicians, paramedical staff, the charge nurse, and the unit secretary. The smaller side of the station is used by the nurses (Figure 10), and it contains five work stations with storage provided for three computers on wheels. The nursing station appeared crowded with furniture and equipment.

Unit 2: Double-Corridor Design With Open Decentralized Nursing Station Alcoves

The 10 decentralized nursing stations on this unit are located in recessed alcoves along the double corridor (Figures 11 and 12). There is one nursing station for every two patient rooms, each equipped with two to three work stations. The nursing stations face the patient rooms; generous windows allow a direct view of the patients. The work stations are easily accessible, allowing for flexibility. Nurses are not assigned to a specific nursing station area but may use any of the alcoves provided.

An important issue emerging from the study concerned the absence of a collective definition of centralized and decentralized nursing stations. It became apparent that the interpretation of nursing station varied widely across healthcare systems, ranging from one or two converted patient rooms to frequently placed hallway alcoves. Each nursing unit observed contained a unique manifestation of each type of unit (centralized or decentralized).

All registered nurses, advanced practice nurses, and licensed practical nurses working on the sample units were eligible to participate in the study survey. The two hospitals with the highest survey response rate were selected for focus group interviews. All nurses who volunteered to participate in the focus groups were included. Three to four nurses from each nursing unit participated in the focus group interviews.

Data Collection

Prior to data collection, an investigator agreement was signed with each institution to ensure compliance with their Institutional Review Board regulations.

Functional Space Use-Observations

One researcher (a graduate student in environmental psychology) conducted both place-centered and person-centered observations in each nursing station. To protect the privacy of patients, the researcher did not enter patient rooms.

During place-centered observations, the frequency and duration of predetermined nursing station activities and behaviors occurring in the nursing station were observed and recorded using a digital timer and visual data charts. Total observation time for each unit was 200 minutes, broken down to include four 20-minute time periods for each morning and afternoon shift and two 20-minute time periods for the night shift (10 observation periods per nursing station).

For person-centered observations within the nursing stations, three nurses were randomly selected and followed for 20-minute periods during morning, afternoon, and night shifts. The frequency and duration of the same predetermined behavior categories used during place-centered observations were recorded. Observation time for each unit was 180 minutes.

Visibility Status

Visibility status was assessed by counting the number of patient beds and patient rooms in direct view of the nursing station and the number of monitors used in the nursing station. Photographs were taken of all observed nursing stations for later verification of data. No patients or identifiable personnel were included in the photographs.

Sound Levels

Sound levels in each nursing station were measured using a portable decibel meter (Extech Model #407738) during all nursing shifts and shift changes. The device was placed so as not to interfere with nurses’ work procedures, and it did not display readings in view of staff. Maximum, minimum, and mean decibel [dB(A)] levels were collected for each shift covering high- and low-activity periods (60 to 99 readings per 3-hour observation period). Equivalent dB(A) and peak dB(A) levels were measured by placing meters a minimum of 12 inches from a direct sound source (e.g., a telephone).

Questionnaire

An online questionnaire assessing nurses’ experienced workplace demand, control, and support was distributed to all nursing staff working in the sample units. The questionnaire was based on Karasek and Theorell’s (1990) Demand-Control-Support instrument and modified to include questions about the physical environment. The demand-control-support model (Figure 13) (Karasek & Theorell, 1990) was chosen because it has been validated from a large body of workplace research over the past 20 years.

The underlying assumption of the model is that the psychological demands of a job/work position and the ability to exercise control/decision latitude in the position can be used to predict job-related stress. A work environment characterized by high demands and high control results in challenging, stimulating work. Although the stress response is elevated in this case, work is regarded as pleasurable. The combination of high demand and low control in the workplace increases stress and subsequent vulnerability to physical disease. Low demand and low control may result in “passive” work performance, and low psychosocial demands coupled with high control may lead to “low-strain” or relaxed work. The social support component of the model acts as an important buffer that generally compensates for any adverse effects associated with “high strain” and the negative stress experienced. Social support may result in an increased sense of control and the ability to manage stress.

The resulting survey used in this study contained 35 questions regarding nurses’ perceptions of demand, control, and support in the work environment and perceptions of their nursing station and unit design. Nurses evaluated questionnaire items on a nine-point Likert scale. Questionnaires were completed online after nurses read and signed participation consent forms. Two follow-up letters were sent via e-mail to increase participation.

Focus Groups

Nurses from the centralized and decentralized nursing stations with the highest questionnaire return rates were invited to participate in two focus group interviews. One focus group had five attendees and the other, eight. Both focus group interviews lasted 2 hours and were held in a closed meeting room on a single occasion at each hospital in September 2007. No hospital administrators or representatives were present during the focus groups. Two moderators conducted the focus groups, and sessions were recorded and transcribed with the participants’ permission.

The focus group instrument consisted of seven semi-structured questions supported by an interview guide. Questions focused on how nursing unit design might affect the overall health, perceived professional status, quality of patient care, professional work needs, communication, multitasking, workflow, and efficiency of nursing staff. To capture the rich detail of the nurses’ experiences and perceptions, concept definitions were not provided in the interview guide. To further develop and clarify responses, follow-up questions were posed. At the end of each group session, nurses were asked to draw a picture of a nursing station and a nursing unit that would provide optimal support for their work.

Data Analysis

Quantitative data were aggregated and analyzed using descriptive statistics and independent-sample t-tests with an emphasis on the differences between centralized and decentralized units. Focus group data were analyzed using a qualitative content method (Granheim & Lundman, 2004). The analysis process involved identifying content, condensing statements, grouping statements with similar content together, identifying subcategories of meaning, and interpreting results.

Results Functional Space Use Observations

The results of place-centered observations of space use in the study units are presented in Tables 2 through 5. The first and second most frequent behavior in all units was either paper administration or computer use. Consultations between medical staff were the next (third) most frequently observed behavior in four of the six study units. Telephone calls and handling supplies also appeared in the top five most frequent activities observed on all units.

Nurses working in the centralized unit of Hospital A spent more time performing tasks in all behavior categories compared to nurses working in the decentralized unit of Hospital A, except in handling supplies (Table 2). Differences between observed activities in centralized and decentralized nursing units in Hospital A were tested using independent-sample t-tests. Significant differences were found for the phone (centralized M = 11.6, SD = 6.8; decentralized M = 4.1, SD = 4.3; t(18) = 3.0, p = .008); computer (centralized M = 19.5, SD = 1.1; decentralized M = 10.3, SD = 6.7; t(18) = 4.3, p = .0004); consultation with medical staff (centralized M = 12.2, SD = 4.8; decentralized M = 6.4, SD = 3.6; t(18) = 3.1, p = .007); handling supplies (centralized M = 2.5, SD = 1.5; decentralized = 5.4, SD = 2.5; t(18) = -3.2, p = .005); and family interaction (centralized M = 0.6, SD = 0.7; decentralized M = 0.0, SD = 0.0; t(18) = 2.7, p = .014) behavior categories.

Table 4. T-Test Results: Place Observations of Hospital C

Behavior (minutes observed)	M	SD	t	P
Note: Observations = 20 df = 18 * significant .05 **significant .001
Phone
Centralized Decentralized	12.3 1.6	6.8 2.2	4.8	0.0002**
Computer
Centralized Decentralized	19.5 7.3	1.6 4.8	7.7	<0.0001**
Paper Administration
Centralized Decentralized	20.0 10.6	0.0 5.3	5.7	0.0003**
Visual Patient Check
Centralized Decentralized	0.0 0.1	0.0 0.3	1.0	0.330
Responding to Patient Calls
Centralized Decentralized	1.0 0.0	0.8 0.0	3.9	0.001*
Consulting Medical Staff
Centralized Decentralized	10.1 4.0	5.9 4.3	2.6	0.016*
Direct Patient Care
Centralized Decentralized	0.1 0.0	0.3 0.0	1.0	0.330
Handling Supplies
Centralized Decentralized	6.7 0.9	3.7 1.4	4.6	0.0002**
Family Contact
Centralized Decentralized	0.9 0.0	1.9 0.0	1.5	0.154
Social Interaction
Centralized Decentralized	4.3 5.3	0.9 1.6	1.9	0.07

Nurses working in the centralized unit of Hospital B spent more time on tasks in every behavior category than nurses working in the decentralized unit (Table 3). Independent-sample t-tests revealed significant differences in the amount of time spent on the phone (centralized M = 7.1, SD = 6.5; decentralized M = 1.5, SD = 2.1; t(18) = 2.6, p = .017), computer (centralized M = 15.7, SD = 4.9; decentralized M = 5.1, SD = 5.2; t(18) = 4.7, p = .0002); paper administration (centralized M = 19.1, SD = 1.4; decentralized M = 6.0, SD = 7.0; t(18) = 5.8, p = .0002); responding to patient calls (centralized M = 0.6, SD = 0.5; decentralized M = 0.1, SD = 0.3; t(18) = 2.6, p = .020); consulting with medical staff (centralized M = 7.9, SD = 6.3; decentralized M = 1.1, SD = 1.5; t(18) = 3.3, p = .004); and social interaction (centralized M = 3.1, SD = 2.5; decentralized M = 0.7, SD = 1.6; t(18) = 2.6, p = .020).

Table 5. T-Test Results: Place Observations of Hospitals A, B, and C

Behavior (minutes observed)	M	SD	t	P
Note: Observations = 60 df = 58 * significant .05 **significant .001
Phone
Centralized Decentralized	10.3 2.4	6.7 3.1	5.75	<0.0001**
Computer
Centralized Decentralized	18.2 7.6	3.4 5.6	8.61	<0.0001**
Paper Administration
Centralized Decentralized	17.5 8.6	4.6 6.6	5.93	<0.0001**
Visual Patient Check
Centralized Decentralized	0.6 0.2	1.7 0.9	1.23	0.224
Responding to Patient Calls
Centralized Decentralized	0.6 0.03	0.7 0.2	4.16	0.0001**
Consulting Medical Staff
Centralized Decentralized	10.1 3.8	5.7 3.8	4.90	<0.0001**
Direct Patient Care
Centralized Decentralized	0.3 0.0	0.1 0.0	1.84	0.07
Handling Supplies
Centralized Decentralized	4.4 3.4	3.1 2.8	1.32	0.192
Family Contact
Centralized Decentralized	0.6 0.03	1.2 0.2	2.36	0.022*
Social Interaction
Centralized Decentralized	4.0 1.1	4.2 1.7	3.4	0.0012*

Behavior observations in Hospital C revealed that nurses in the centralized unit spent more time performing all activities compared to nurses working in the decentralized unit (Table 4). Independent-sample t-tests revealed significant differences for the phone (centralized M = 12.3, SD = 6.8; decentralized M = 1.6, SD = 2.2; t(18) = 4.8, p = .0002); computer (centralized M = 19.5, SD = 1.6; decentralized M = 7.3, SD = 4.8; t(18) = 7.7, p = < .0001); paper administration (centralized M = 20.0, SD = 0.0; decentralized M = 10.6, SD = 5.3; t(18) = 5.7, p = .0003); responding to patient calls (centralized M = 1.0, SD = 0.8; decentralized M = 0.0, SD = 0.0; t(18) = 3.9, p = .001); consulting with medical staff (centralized M = 10.1, SD = 5.9; decentralized M = 4.0, SD = 4.3; t(18) = 2.6, p = .016); and handling supplies (centralized M = 6.7, SD = 3.7; decentralized M = 0.9, SD = 1.4; t(18) = 4.6, p = .0002).

An analysis of place-centered observations in centralized and decentralized units across all three hospitals was performed using independent-sample t-tests (Table 5). Nurses in all centralized units spent more time performing all behavior categories compared to nurses working in decentralized units. Significant differences were found for the phone (centralized M = 10.3, SD = 6.7; decentralized M = 2.4, SD = 3.1; t(58) = 5.75, p = < .0001); computer (centralized M = 18.2, SD = 3.4; decentralized M = 7.6, SD = 5.6; t(58) = 8.61, p = < .0001); paper administration (centralized M = 17.5, SD = 4.6; decentralized M = 8.6, SD = 6.6; t(58) = 5.93, p = < .0001); responding to patient calls (centralized M = 0.6, SD = 0.7; decentralized M = 0.03, SD = 0.2; t(58) = 4.16, p = .0001); consulting with medical staff (centralized M = 10.1, SD = 5.7; decentralized M = 3.8, SD = 3.8; t(58) = 4.9, p = < .0001); family contact (centralized M = 0.6, SD = 1.2; decentralized M = 0.03, SD = 0.2; t(58) = 2.36, p = .022); and social interaction (centralized M = 4.0, SD = 4.2; decentralized M = 1.1, SD = 1.7; t(58) = 3.4, = .0012).

During person-centered observations, nurses in five of six units spent the most time engaged in direct patient care. The mean number of minutes spent on patient care ranged from 9.1 minutes (SD = 6.7) in the centralized unit of Hospital B to 13.1 minutes (SD = 4.6) in the decentralized unit of Hospital A. Paper administration was the next most frequently observed behavior in four of six units. Independent-sample t-tests comparing person-centered behavior in centralized and decentralized units in each hospital revealed no significant differences.

Across hospitals, person-centered observations revealed that nurses spent the most time on tasks associated with direct patient care. Nurses working in decentralized units spent slightly more time engaged in direct patient care compared to nurses working in centralized units (centralized M = 10.7, SD = 5.4; decentralized M = 11.5, SD = 4.4). Independent-sample t-tests indicated that this difference was not significant, t(58) = -0.62, p = 0.538. Paper administration was the second most frequently observed behavior across hospitals (centralized M = 6.0, SD = 5.8; decentralized M = 5.4, SD = 4.6; t(58) = .44, p = .662). Computer work and handling supplies ranked as the third or fourth most frequently observed behavior. Nurses working in decentralized nursing stations were observed consulting medical staff more frequently than nurses working in centralized nursing stations (centralized M = 1.5, SD = 1.7; decentralized M = 2.7, SD = 2.5). T-test results indicated that this difference was significant, t(58) = 2.11, p = .039.

Table 6. Responses to Open-Ended Question on Nursing Unit and Nursing Station Design

Category	t	Comments
Note: n = 26
Space issues	11	A lot of dead space on unit Not enough space for team work in station Shortage of storage space (f = 2) No adequate space for visitors Not enough charting space Not enough space to prepare meds (f = 2) Too little space for nurses and doctors Small areas create clutter (f = 2)
Layout, Walking distances	7	Everything is too spread out Design supports teamwork and staff cohesion Not nurse friendly; a lot of walking (f = 2) Difficult to perform tasks that require extra hands Long corridor confusing patients Efficient, meeting patient needs
Supply areas	6	Poor planning/organization of supply areas (f = 3) Supply areas need to be stocked properly (f = 3)
Sound levels	6	Loud ice machines Difficult to speak on the phone due to loud calls Difficult to discuss confidential information High noise level in nursing station due to chatting The open station can be noisy at times Difficult to hear staff assist alarms
Flow, Workflow	4	Frequent interruptions while charting (f = 4)
Visibility Monitoring	4	Critical patients close to nursing station is good Closer monitoring from station is good No outside views on the unit Cannot see the front desk from the station
Furniture Ergonomics Equipment	3	Broken chairs No adjustable chairs or keyboards Not enough computers in nursing station
Crowding	2	A gathering place-not enough space Doctors monopolize the computer stations
Climate	1	Too cold during night shifts on unit

Table 7. Summary of Response Categories From Focus Group Interviews

Category	Design Subcategory
Provide nurse-specific support Teamwork-oriented nursing stations Nursing units for family and patient needs Improve acoustic conditions in nursing stations	Maintain nurses’ sense of control Minimize distance between nursing stations and patient Centralize information and communication areas More work and traffic space in nursing stations Flexible nursing station furnishings	Acknowledge nurses’ professional privacy Centrally located, enclosed meeting rooms Separate nursing station areas for charting Reception desk areas	Enable nurses to stay separate and connected Teamwork-oriented nursing stations Clearly defined nursing station spaces Differentiated nursing task work areas Supplies close but separate from nursing station

Visibility

Visibility of patient beds and patient rooms from the nursing stations differed significantly among observed units. There were no indications that either centralized or decentralized nursing station designs resulted in superior visibility. Rather, variations related to the placement of nursing stations in relation to patient rooms (i.e., the number of patient beds visible from a nursing station) and the type of monitoring system in use (i.e., the number of electronic vital sign monitors used).

Sound Levels

Sound levels measured in all nursing stations exceeded the recommended levels of 45 dB(A) for continuous background noise in daytime and 35 dB(A) for nighttime set by the World Health Organization (WHO) and the Occupational Safety and Health Administration (OSHA), but they did not differ significantly between centralized and decentralized nursing station designs (Figure 14). Recommended levels were exceeded on all nursing units during all shifts. Sound sources within the nursing stations that produced the highest dB(A) levels included ice machines, doors shutting, pneumatic transport stations, alarms and pagers, telephones and fax machines, voices, medication refrigerator doors, and fans.

Perceptions of the Work Environment Questionnaire

Of the 190 nurses who worked on the six units, 57 returned a usable questionnaire. The response rate for individual units ranged from 5% (Unit 1 in Hospital C) to 31% (Unit 1 in Hospital A). Nearly half (45%) of responding nurses were between 20 and 30 years old, 51% were between 31 and 60 years old, and only 5% were older than 60 years. Most of the respondents were registered nurses (94%). Work experience was generally high; 42% of the nurses had worked in nursing for 10 years or more, and only 14% had worked in nursing for less than 1 year. A comparison of means, standard deviations, and t-tests for questions regarding the demand-control-support model revealed only small variances between centralized and decentralized nursing station designs. The low response rate for the online survey (5%-31% of nurses) was not considered sufficiently representative to draw reliable conclusions.

Two open-ended questions soliciting comments from respondents about nursing unit and nursing station design were included in the survey. Comments obtained from 26 respondents (42%) were qualitatively analyzed (i.e., counted and grouped together by topic) (Table 6). The most frequently identified category of comments (f = 10) focused on insufficient space available for various activities (e.g., teamwork, storage, visitors, medication preparation, charting). The next most frequently identified issues included layout and walking distances (f = 7), supply areas (f = 6), and sound levels (f = 6). Nurses linked high sound levels to their inability to carry out work tasks accurately and efficiently. Nurses working on both centralized and decentralized nursing stations made similar comments.

Focus Groups

Results from the focus group interviews highlighted four important themes regarding nursing station design and the associated design implications (Table 7): (1) provide nurse-specific support; (2) maintain nurses’ sense of control; (3) acknowledge nurses’ professional privacy needs; and (4) enable nurses to stay separate and connected.

During the focus groups, among the many interesting comments made by nurses on the subject of nursing station design and work environment were:

• “Patients and families come over to see the nurses at the station. It’s a comfort to them, so that helps the nurse out, too. The anxiety of the patient is lowered.”
• “At the centralized nursing station you are valued as a team member more. At the decentralized station you feel alone.”
• “The physicians come on the floor on the decentralized unit and get off the floor without being seen by a nurse, and then you have to call them and they say they were just there. Then that becomes a nurse dissatisfier now too because I get to take the brunt of it.”
• “We are working on a concept of bringing the nurse more to the bedside…but there are certain things you cannot do at the bedside and need to be done in a place that is separate.”

In nurses’ drawings of the optimal nursing station, a separate reception desk was often placed adjacent to the nursing station to handle first contact with family members and other individuals that did not require nurses’ direct involvement. The choice of one centralized nursing station combined with a handful of small, decentralized work stations was seen in all but one drawing. The most critical patients were placed closest to the nursing station to maximize view and accessibility. The importance of patient placement was emphasized in all drawings. Nursing stations were often depicted as semi-enclosed by transparent materials for privacy and confidentiality, allowing nurses to visually monitor patients and stay connected to other staff members. Nurses also recognized the importance of the relationship between their work areas and the location of supplies. One nurse suggested that supply areas be differentiated with a specific location for different categories of supplies.

Discussion

Data from earlier studies of the work environment of nurses have suggested that nurses spend a considerable amount of time performing administrative duties and much less time providing direct patient care (Hendrich et al., 2004; Tucker & Spear, 2006). Observation data from this research support this evidence. Nurses in this study working in both centralized and decentralized units spent the most time on phone, computer, and paper administration tasks.

More nursing unit designs are incorporating decentralized nurses’ stations outside patient rooms so that staff members are distributed around the unit closer to the point of care, potentially improving patient visibility, increasing patient care time, and decreasing staff fatigue by reducing walking distances (Joseph, 2006; Ulrich et al., 2004). Analysis of place-centered observations in centralized and decentralized units within each hospital indicated that significantly more time was spent on phone, computer, and paper administration in each of the centralized units compared to decentralized units. It appears that decentralizing nursing stations successfully decreased and redistributed these tasks throughout the nursing unit. However, during place-centered observations, there was no significant difference in direct patient care behavior in the decentralized nursing units.

In this study, neither centralized nor decentralized nursing station designs resulted in superior visibility. Rather, variations were related to the placement of nursing stations in relation to patient rooms (i.e., the number of patient beds visible from a nursing station) and the type of monitoring system in use (i.e., the number of electronic vital sign monitors).

In previous research, nurses working in decentralized units reported feeling more isolated from their colleagues and losing a sense of team connection compared to nurses working in centralized nursing stations (Tyson et al., 2002). In this study, significantly less time was spent consulting with other medical staff on each of the decentralized units compared with the centralized units (within hospitals and across hospitals). In addition, observed social interactions were significantly higher in centralized units (across hospitals) than in decentralized units.

According to the Demand-Control-Support model (Karasek & Theorell, 1990), the social support component of the model acts as an important buffer that generally compensates for any adverse effects associated with “high strain” and any negative stress experienced. Social support may result in an increased sense of control and the ability to manage stress. In this study, fewer opportunities for staff consultation and social interaction in decentralized nursing units may decrease social support and increase stress among nurses working in these units. This may have consequences for staff collaboration, informal learning, and social support as well for the quality of patient care.

Nursing stations are often noisy, crowded, busy spaces. Sources of sound include alarms, pagers, telephones, and other technological equipment as well as frequent and multiple conversations. It makes intuitive sense to assume that smaller, decentralized nursing stations would be quieter than busy centralized nursing stations. However, sound levels measured in all nursing stations in this study exceeded recommended levels set by the WHO and OSHA, and they did not differ significantly between centralized and decentralized nursing station designs. Factors potentially contributing to increased sound levels in the units included the type of equipment placed within the nursing station, the degree of design enclosure, and the sound absorption qualities of materials in the space.

Noise control is one important way to improve hospital environments for nurses, as suggested by previous research studies (Blomkvist et al., 2005; Dube, Barth, Cmiel, Cutshall, Olson, et al., 2008; Marqués Sanchez, Calle Pardo, Sanchez, Gelado, & Garcia, 2008). Higher sound levels have been associated with greater stress, annoyance, emotional exhaustion, and burnout among nursing staff (Morrison et al., 2003). Focus group results revealed that improved acoustic conditions are an important factor identified by nurses, particularly when performing cognitive tasks such as charting and giving reports. Observation data revealed that nurses spent considerable time performing office-type duties. The relationship between decreased noise levels and medical errors caused by sound distraction deserves further attention.

Results from the observational studies indicated that contact between nurses and family members rarely occurred within nursing stations; rather, it occurred in patient rooms or corridors unless specific family meeting rooms were available. Nurse-supportive design characteristics were associated with family-friendly environments in focus group interviews, suggesting a holistic patient care perspective among nurses. A family-friendly setting may lower anxiety levels among families and staff, creating a more supportive work environment for nurses.

The patient-centered care and Planetree models (Arneill & Frasca-Beaulieu, 2003) initiated the design of open, barrier-free nursing stations accessible to patients and their families. However, results from this study suggest a more complex problem that calls for designs that enable nursing staff to be both separate and connected to patients, families, and healthcare colleagues. These seemingly conflicting needs emerged as an important design issue that was reflected in nurses’ drawings of semi-enclosed nursing stations. Transparent materials were included in nurses’ drawings to provide for privacy and confidentiality while allowing nurses to monitor patients visually and stay connected to staff team members. The idea of the “hybrid” nursing station model with bedside nursing stations and a collaborative centralized nursing, or teaming, station may offer an alternative solution.

Sufficient space to maintain good workflow in nursing stations appeared to be of considerable importance to nurses in the study. The provision of separate spaces for charting in the nursing station to avoid compromising efficiency was emphasized in survey comments. It appears that the nursing stations in this study may hinder the execution of some nursing tasks. Results indicated that clearly defined nursing station spaces allow nurses differentiated work areas, thereby improving efficiency. In previous research, efficiency and workflow issues in nursing station design have been related primarily to walking distances (Hendrich et al., 2004; Tucker & Spear, 2006). Results from this study indicate that the addition of a reception desk to facilitate first contact with family and assist visitors would support the nurses’ need for professional and personal space while minimizing interruptions. This concept was also illustrated in nurses’ drawings of a separate reception desk adjacent to the nursing station.

The generalizability of these findings is limited because of the small sample size, and it suggests the need for more in-depth studies. There is clearly a need to replicate this research and to gather longitudinal data examining in further detail the effects of nursing station design on nurses’ health and behavior. It is quite possible-if not likely-that each unit’s patient profile, as well as acuity, plays an important role in nursing station design. Nurses working in a psychiatric unit are likely to have very different design needs compared to nurses on an emergency unit. Nursing station design should reflect the unique clinical needs of each unit.

Although this study attempted to control organizational factors that might confound data analysis (i.e., hospital size, type and age, organizational structure, nursing unit type, and nurse-patient ratio), there may be unspecified factors not included in the model that also correlate with behavioral or health characteristics (e.g., ambient environment, cultural and organizational characteristics). Future research using a larger sample would increase the significance of these findings and permit the inclusion and control of additional variables. An examination of the cumulative effects of environmental factors might also provide useful information. Future research must address the reality that design changes that benefit patients could conflict with staff needs or organizational goals.

Focus group results highlighted the need for an environment that supports multitasking and flexibility to maintain nurses’ sense of control in daily work routines. This may imply that standardized nursing stations designs are insufficient and do not meet nurses’ workplace needs. Unique design solutions rather than generic solutions can be achieved only by better understanding the complexity and variation of stress sources among practice areas.

In light of the current trend toward more flexible hospital care, truly adaptable nursing units are gaining momentum. Nursing stations may be increasingly transformed into information centers used mainly to enter, process, and retrieve data. In light of the current trend toward more flexible hospital care, truly adaptable nursing units are gaining momentum. Nursing stations may be increasingly transformed into information centers used mainly to enter, process, and retrieve data. New types of information and communication areas include docking stations for portable computers between patient rooms and movable charting stations on wheels. These smaller and more sophisticated nursing stations reflect advanced technologies in nursing. On the other hand, face-to-face consultations will continue to be an important part of the profession. The hybrid nursing model, in which decentralized nursing stations are coupled with centralized meeting rooms for consultations among staff, may strike a balance between the increase in computer duties and the continued need for communication and consultation, thereby addressing the conflicting demands of technology and direct patient care.

The success of any environmental transformation requires a corresponding change in organizational operations. Without appropriate organizational change, the potential benefits of a new environment may be neutralized (Tyson et al., 2002). This is particularly important to nurses, because poor organizational or environmental fit may contribute to stress and burnout. Further investigation of the interactions between environmental conditions and organizational culture and the way in which the physical attributes of nursing stations may prevent nurses’ stress and fatigue is needed.

Ideally, nursing stations should be designed with input from the staff that will be using them. As healthcare culture continues to change, it becomes increasingly important to employ a collaborative planning process in which nurses are actively involved with designers and architects. Tapping the specialized, front-line experience of nurses enables architects and designers to shape nursing stations to be spaces where nurses can perform their duties comfortably, efficiently, and safely in the context of a changing healthcare arena.

A growing body of evidence demonstrates that healthcare work is practiced most effectively when practitioners work highly interdependently in well-functioning teams, with the active participation of patients and families (McCarthy & Blumenthal, 2006). Evidence-based knowledge of the effects of nursing station design on nurses’ work environment and work behavior is essential to improve conditions for and increase the retention of these fundamental members of the healthcare delivery team. Results of this study raise a cautionary flag regarding best practices for nursing station and nursing unit design. Of particular concern is the sense of isolation and loss of social support that may occur among nursing staff working in decentralized units. A better understanding of the trade-offs presented by centralized and decentralized nursing station design has the potential to provide useful information for nursing station layouts in the future.

Acknowledgment

This article was based on research funded by Ellerbe Becket and Knoll, Inc.

Dr. Zborowsky is Director of Healthcare Education and Research at Ellerbe Becket, Inc., in Minneapolis, Minnesota. Dr. Bunker-Hellmich is Research Specialist at Ellerbe Becket, an AECOM company. Ms. Morelli is with Högskolan i Gävle in Gefle, Sweden. Dr. O’Neill is Senior Director, Workplace Research, at Knoll Inc., in East Greenville, Pennsylvania. Corresponding Author: Lou Bunker-Hellmich, PhD, Ellerbe Becket, Inc., an AECOM company. 800 LaSalle Avenue, Minneapolis, MN 55402 (lou_bunkerhellmich@ellerbebecket.com)

References

1. Agency for Healthcare Research and Quality. (2005). The hospital built environment: What role might funders of health services research play? (Publication No. 05-0106-EF). Rockville, MD: U.S. Government Printing Office.
2. Arneill B., & Frasca-Beaulieu K. (2003). Healing environments: Architecture and design conducive to health. In S. Frampton & L. Gilpin (Eds.), Putting patients first: Designing and practicing patient-centered care. San Francisco, CA: Jossey-Bass.
3. Bayo M., & Garcia A. (1995). Noise level in an urban hospital and workers’ subjective responses. Archives of Environmental Health, 50 (3), 247-251.
4. Berry L., & Parish J. (2008). The impact of facility improvements on hospital nurses. Health Environments Research & Design Journal, 1 (2), 5-13.
5. Blomkvist V., Eriksen C. A., Theorell T., Ulrich R. S., & Rasmanis G. (2005). Acoustics and psychosocial environment in coronary intensive care. Occupational and Environmental Medicine, 62 (3), e1.
6. Dube J. A., Barth M. M., Cmiel C. A., Cutshall S. M., Olson S. M., Sulla S. J., … Holland D. E. (2008). Environmental noise sources and intervention to minimize them: A tale of 2 hospitals. Journal of Nursing Care Quality, 23 (3), 216-231.
7. Granheim U. H., & Lundman B. (2004). Qualitative analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24 (2), 105-112.
8. Hamilton D. K. (1993). Unit two thousand: Patient beds for the future (Call no. 1995 G-208). Houston, TX: Watkins Carter Hamilton.
9. Hendrich A. L., Fay J., & Sorrells A. K. (2004). Effects of acuity-adaptable rooms on flow of patients and delivery of care. American Journal of Critical Care, 13 (1), 35-45.
10. Institute of Medicine. (2004). Keeping patients safe: Transforming the work environment of nurses. Washington, DC: National Academies Press.
11. Joint Commission on the Accreditation of Healthcare Organizations.(2002). Health care at the crossroads: Strategies for addressing the evolving nursing crisis. Retrieved May 24, 2010, from http://www.jointcommission.org/NR/rdonlyres/5C138711-ED76-4D6F-909F-B06E0309F36D/0/health_care_at_the_crossroads.pdf
12. Joseph A. (2006).The role of the physical and social environment in promoting health, safety and effectiveness in the healthcare workplace. Retrieved May 24, 2010, from http://www.healthdesign.org/research/reports/documents/CHD_Issue_Paper3.pdf
13. Karasek R., & Theorell T. (1990). Healthy work: Stress, productivity and reconstruction of working life. New York, NY: Basic Books.
14. Marqués Sanchez M., Calle Pardo A., Sanchez D., Gelado Y., & Garcia M. (2008). Nurses’ perceptions of noise levels in hospitals in Spain. Journal of Nursing Administration, 38 (5), 220-222.
15. McCarthy D., & Blumenthal D. (2006). Committed to safety: Ten case studies on reducing harm to patients. New York, NY: Commonwealth Fund.
16. Morrison W. E., Haas E. C., Shaffner D. H., Garrett E. S., & Fackler J. C. (2003). Noise, stress and annoyance in a pediatric intensive care unit. Critical Care Medicine, 31 (1), 113-119.
17. Mroczek J., Mikitarian G., Vieira E. K., & Rotarius T. (2005). Hospital design and staff perceptions: An exploratory analysis. Health Care Manager, 24 (3), 233-244.
18. Office of Space Analysis, University of Michigan.(2006). Departmental space inventory survey-Room type codes & definitions. Retrieved December 12, 2006, from http://www.umich.edu/∼ofa/Space/RmTyp800.htm#E
19. Peter D. Hart Research Associates.(2001). The nurse shortage: Perspectives from current direct care nurses and former direct care nurses. Washington, DC: Federation of Nurses and Health Professionals.
20. Topf M. (1985). Personal and environmental predictors of patient disturbance due to hospital noise. Journal of Applied Psychology, 21 (6), 717-733.
21. Tucker A., & Spear S. (2006). Operational failures and interruptions in hospital nursing. Health Services Research, 41 (3), 643-662.
22. Tyson G. A., Lambert G., & Beattie L. (2002). The impact of ward design on the behavior, occupational satisfaction and well-being of psychiatric nurses. International Journal of Mental Health Nursing, 11 (2), 94-102.
23. Ulrich R., Quan X., Zimring C., Joseph A., & Choudhary R. (2004). The role of the physical environment in the hospital of the 21st century: A once-in-a-lifetime opportunity. Retrieved November 6, 2006, from http://www.healthdesign.org/research/reports/physical_environ.php
24. Wakefield A. (2002). The changing “shape” of the nursing station. Contemporary Nursing, 13, 148-157.

Healthcare Design 2010 November;10(11):50-78