 |
 Introduction
While empirical studies have addressed the appropriate
design of healthcare facilities in general, there is very little
research-based information regarding the design of supportive
facilities for persons with AIDS. HIV/AIDS environment studies have
been limited to a few postoccupancy evaluations (e.g., Shepley and
Wilson 1999), case studies (e.g., Chambers and Guerin 1993), and
research on the appropriateness of AIDS-dedicated facilities (e.g.,
Rothman and Tynan 1990). Another topic that has not been thoroughly
addressed is nursing unit size and configuration as well as related
decisions based on staffing ratios and recent precedent. Although
the average number of beds on a typical unit is 25 to 30 beds
(Velsey and Egbert 1994), unit size may range widely-from as few as
six beds in intensive care to as many as 48 in a general medical
unit. In addition, the overall physical dimensions of units are
wide-ranging due to the impact of patient density (single-,
double-, or multiple-bed rooms). AIDS units tend to have more
single-bed rooms (to support infection control). The increased area
per patient results in a larger nursing unit and, potentially,
decreased staff efficiency. Even if two-bed rooms meet the social
and medical needs of some patients, these rooms lack the room
assignment flexibility and infection control efficiency of
single-bed rooms.
The objective of this study was to compare two nursing unit designs
in facilities for persons with AIDS. The two dependent variables
were noise level and distance walked by nurses. There were two
hypotheses. The first was that nursing staff would walk more in the
large rectangular units. This has been suggested in previous
studies (e.g., Trites et al. 1970). The second hypothesis was that
noise levels would be lower in the more compact radial design
because this design would use fewer staff. In addition, because
patients are visible from the nursing station, they might be less
likely to create noise by using the nurse call system. Although
studies have measured noise levels on nursing units, these levels
have not been compared for differing unit configurations. If both
the hypotheses are supported, radial plans can be presumed more
effective in these ways.
Walking
Nurses spend a lot of time walking. A study of 17 nurses
participating in 37 shifts indicated that the average nurse in a
general ward walked approximately 6,260 meters (3.89 miles) and
8,260 meters (5.13 miles) in an intensive care ward (Bauer and
Knoblich 1978). In a nursing home study, Burgio et al. (1990) noted
that almost 28.9 percent of nursing staff time was spent walking.
This came second to patient-care activities, which accounted for
56.9 percent of observed behavior. It follows that if less time is
spent walking, more time will be available to spend with patients
or on activities that would improve patient care. Seelye (1982)
noted that travel distance and patient contact are the two most
important issues affecting nursing care.
In addition to the anecdotal reports of reduced walking in cluster
units (e.g., Fisher 1982), there are related studies on the
relationship between unit design and walking. A study at Rochester
Methodist Hospital in Minnesota (Sturdavant 1960) compared two
intensive care units, one with a radial design and the other with a
single-corridor design. Sturdavant found that fewer trips were made
to patient rooms in the cluster units due to the increased visual
supervision of the patient from the nursing station, while the
average time spent with patients was equivalent (39 percent to 40
percent). Increased patient visibility also enabled the head nurse
to participate more in patient care in the radial unit than in the
rectangular unit. Staff expressed greater satisfaction with the
radial unit as well.
Noise
Several studies have found links between noise and stress-related
physiological changes (e.g., Glass, Singer, and Friedman 1969). In
healthcare environments, inappropriately high noise levels can
adversely affect patients and staff (Falk and Woods 1973). More
specifically, increased noise level is a stressor for
immunocompromised patients (Griffin 1986). Although the maximum
sound level recommended for a hospital is 45 decibels, Hilton
(1985) recorded levels that range from 50 to 80 decibels in an
acute nursing unit. Lack of noise control has been identified as an
element that contributes to a patient's sense of helplessness. In
addition, noise levels may contribute to the occurrence of
medication and charting errors by the staff in a patient
unit.
Sources of noise in an inpatient unit include the movement of
people and equipment, computer printers, telephones, staff and
patient conversations, patient noises, public address system
messages, and alarms. Haslam (1970) listed the following primary
sources of noise, in order of their level of irritation to
patients: conversations between staff, visitors, and patients;
sounds from patients in distress; noises from mechanical devices;
and sounds from television and radio. One of the most disturbing
noises to patients has been identified as talking in the hallway
(Topf 1985). Redding, Hargest, and Minsky (1977) noted that three
physicians talking in the hallway generated 68 dB. High noise
levels in the corridors may be a result of long corridors and
increased distance from the nurses' station. Staff may tend to have
miniconferences in the hallways instead of walking all the way back
to the nurses' station. Also, the staff cannot see patients (as is
common in rectangular designs), patients make noise to get staff
attention. Bells and the rattling side rails generate 78 dB and 80
dB at 10 feet, respectively (Redding, Hargest, and Minsky
1977).
Relationships have also been found between noise and physiological
responses. Kryter (1985) identified the following six
responses:
- Vasoconstriction of the peripheral blood
vessels
- Altered breathing rate
- A modification in galvanic skin response
- Skeletal and muscular tension
- Gastrointestinal motility changes
- Blood and urine chemical modifications.
Relative to hospital populations, Marshall
(1972) identified a correlation between acoustical stimulation and
pulse rate in a critical care unit. In addition, Minckley (1968)
found that patients exposed to higher noise levels are more likely
to perceive higher pain levels as measured by requests for pain
medication.
Many inpatients complain that it is difficult
to sleep. This is likely due to the number of loud noises in a
nursing unit as well as the impact of being in an unfamiliar
environment. Helton, Gordon, and Nunnery (1980) found a correlation
between interrupted sleep and negative psychological status such as
hallucinations, combativeness, and disorientation in an intensive
care unit. Snyder-Halpern (1985) also reported evidence of physical
and psychological alterations when noise interferes with patients'
sleep.
Cluster or radial units are usually smaller
than corridor units. If smaller units are associated with less
noise and increased staff efficiency, better patient care may be
provided when the number of beds per unit is limited. In the case
of general diagnosis patients, better patient care may result in
shorter hospital stays and perhaps reduced medical costs. In the
case of AIDS/hospice facilities, where length of stay is less of a
consideration because many patients die on the unit, less noise and
increased efficiency may increase staff productivity and enhance
the patient's sense of well-being.
 |
| Figure 1 |
|
 |
| Figure 2 |
|
 |
| Figure 3 |
|
 |
| Figure 4 |
|
 |
| Table 1 |
|
 |
| Table 2 |
|
 |
| Table 3 |
Linear Corridor versus Radial Configuration
A thorough study regarding nursing unit configuration was
undertaken in 1970 by Trites et al. This study compared three unit
designs: radial, double-corridor (racetrack), and single-corridor
(L-shaped). The main finding was that radial designs were more
successful from the perspective of functionality and staff
preference. Nurses traveled significantly less, and the time was
converted into more time spent with patients. Double-corridor
designs were rated second and single-corridor designs were rated
last in terms of both these variables. Most of the medical staff
preferred the radial units and felt that configuration enhanced the
quality of patient care. In addition, it was noted that staff
members had relatively more accidents on single-corridor units.
Preference for radial designs has also been demonstrated elsewhere
(Macdonald et al. 1981).
Methodology
The two nursing unit sites selected for comparison had the
following characteristics in common: location within a few miles of
one another in the same large city, management by the same health
organization, and a focus on care for HIV/AIDS patients. Although
the units had much in common, they differed dramatically in plan
configuration. One unit had a double-corridor rectangular floor
plan with the nurses' station/support spaces in the core and
patient rooms on the perimeter wall (see Figure 1). The other unit
was circular in configuration with the patient rooms radiating
around the circumference and the nurses' station at the center (see
Figure 2).
The rectangular unit accommodated 24 beds and
was approximately 18,500 square feet including support areas. The
radial unit was one of five 10-bed units and approximately 8,100
square feet. The number of staff per square foot in each nursing
unit was comparable. Staffing on the rectangular unit ranged
between five and 10 nurses on a single shift, while staffing on the
radial unit was typically two nurses.
Measurement Apparatus
Sound levels were measured using a noise-logging dosimeter with an
operating range of 30 dB to 125 dB. The dosimeter was downloaded
every 36 hours, producing noise-level histograms and other data,
including thresholds (the percentage of time that sound exceeded a
specified level), peak levels, and averages. The apparatus
consisted of the measurement device and a small
microphone.
Walking was measured using electronic
(battery-operated) pedometers that logged the number of steps
taken. These pedometers were attached to the back of the nursing
staff's belts. The effectiveness of the pedometers was pretested on
a running track to confirm their intradevice mechanical accuracy,
which was found to be between 97.7 percent and 99.7 percent.
Procedure
A survey of the physical characteristics of the nursing stations
identified variables that might significantly affect noise
measurement data. While the ceilings were of comparable height and
finish (acoustical tile), the floor finishes were not (see Figures
3 and 4). The radial unit was carpeted, while the rectangular unit
had resilient flooring. The carpet had sound reduction indices of
.03 to .08 at 125 Hz and .72 to .80 for 4000 Hz (Templeton 1993).
On the other hand, the radial unit also had a low-volume radio
turned on throughout most of the study. (A soft-volume radio has an
average decibel level of 35 dB [Waterfall 1929].) Although these
two characteristics of the radial unit may have been
self-canceling, it was decided that only highly significant
differences in sound level would reliably indicate variability in
noise levels.
In each case, the dosimeter was located near
the core of the nurses' station on vertical casework. Ten- to
30-minute gaps occurred in the data when the dosimeter was reset
every 36 hours. These gaps were staggered to avoid the repeated
absence of data during the same time interval. The dosimeter
recorded decibel levels every minute, and the levels were averaged
for every five minutes. The starting and ending times were recorded
to allow the data to be entered by time of day. This method reduced
the data to 288 points while also allowing for analysis by time and
day of the week. The data retrieval spanned from 00:00 (midnight)
to 23:55 (11:55 p.m.).
Noise data were collected at the rectangular
and radial units for 174 and 140 hours, respectively, and recorded
on consecutive weeks. Walking measures were obtained during the
same period that noise measurements were made. Nursing staff were
pretrained in the use of the pedometers. When the study period
began, each nurse set his or her pedometer and recorded the start
time. The pedometers were removed at the end of the work shift, and
the time that they were returned was also recorded.
Results
Walking Behavior
Steps per minute per work shift were
calculated for each subject. The data were broken down to steps per
minute rather than steps per work shift because shifts varied in
duration from one hour to 16 hours and 40 minutes. Nurses working
in the rectangular configuration took an average of 7.9 steps per
minute, and those in the radial configuration took an average of
4.7 steps per minute. Using a conversion factor of 2.6 feet per
step, the nurses traveled an average of 20.5 feet per minute and
12.22 feet per minute, respectively. The maximum distance walked
per segment was approximately 6.87 miles (33 feet per minute) in
the rectangular design and 1.97 miles (9.66 feet per minute) in the
radial design. A t-test showed the difference in walking distance
between the two units to be statistically significant (p<.01).
Tables 1 and 2 demonstrate the different patterns between the two
units.
Noise Levels
The five-minute totals were averaged over the duration of
the study. For example, the average decibel level at 6 a.m. over
seven consecutive days was approximately 58 dBs. Each entry on
Table 3 represents averages at five-minute intervals over six to
seven days. The total average, combining all data entries, was
60.98 dB for the rectangular design and 60.65 dB for the radial
design. A t-test assuming unequal variances indicated no
significant difference between the two units. Commonalities in
noise patterns were observed, however. Noise level tended to drop
at night (9 p.m. to 6 a.m.) relative to the day (9 a.m. to 6 p.m.).
The average noise level at night was 60.25 dB on the radial unit
and 59.63 dB on the rectangular unit, and the average day levels
were 62.56 dB on the radial unit and 62.02 dB on the rectangular
unit. Saturday peaks were also noted on both units (79 dB in the
radial unit and 86 dB in the rectangular unit).
Discussion
Nursing staff in the radial unit walked significantly less than
staff in the rectangular unit. This has been confirmed by other
studies (Trites et al. 1970). The Trites study and this study were
also similar in that they both found a large variance between
individuals. The extreme variation in the data is likely a result
of job descriptions. Differences in walking distance have been
identified in previous studies (Burgio et al. 1990), where it was
found that LPNs (licensed practical nurses) spent 15.8 percent of
their day walking, whereas RNs (registered nurses) spent 24.1
percent of their day walking. The Trites findings suggest that a
decrease in the percentage of time spent walking correlated to an
increase in the percentage of time spent in patient care
activities.
Regarding noise levels, the lack of
significant variation between the two designs might be misleading.
Possible confounding variables in the field study included
differences in staffing levels and nursing station size; variations
in acoustical finishes of the nursing stations; the presence of a
radio in one unit; and possibly, the amount of time nurses spent in
the nurses' station.
The range of approximately 52 dB to 67 dB recorded in this project
is consistent with findings by Hilton (1985), who reported levels
ranging from 50 dB to 80 dB. The fact that the noise levels in the
rectangular and radial units proved similar may suggest another
hypothesis-that the noise levels in most nursing units will fall
into that range, regardless of unit size, due to any of the
following factors:
- Relative increase in time spent in the nursing
station as patient rooms become more proximate. Although small
units would have fewer staff, the staff could spend more time in
the station because they wouldn't be walking as much, and they
would have less reason to leave the station as patients would be
more readily seen.
- Increased size of the nursing station as the
unit gets larger. Larger nursing stations will provide acoustical
distribution of noise.
- Social stimulation requirements. The use of the
radio in the radial unit may have been a result of increased need
for stimulation due to the decrease in staff numbers in the
station.
- Community size. Observers in this study noted
that the smaller radial unit had a much less formal ambience than
the rectangular unit, which may have accounted for increased
talking among staff.
Conclusions
The intention of this study was to examine noise levels
and walking behavior in radial and rectangular AIDS/HIV facility
designs. The data from this research support previous studies that
indicate that a dense radial nursing unit design requires less
walking on the part of staff. If less walking is required, staff
may be able to spend more time directly caring for or supervising
patients. Noise levels, on the other hand, were similar for the two
designs. In future studies, noise levels for other unit densities
should be compared to verify the relationship between nursing unit
size and noise levels. If noise levels are comparable between
units, they may be relatively stable across nursing unit
configurations, and efforts to reduce noise should therefore focus
on operational issues and acoustical finishes.
Although the relatively small size of radial units may reduce
staffing assignment flexibility, the decrease in walking would help
to support both direct and indirect patient supervision. If the
trend in increased patient acuity in inpatient care continues, it
is likely that fewer patients will be assigned to each nurse, thus
justifying clustered/radial units.
Staff that had worked in both types of units and participated in
this project indicated that security (relative to theft and control
of intruders) on the dense radial unit was greater than on large
rectangular units and expressed a strong preference for the radial
unit. Data from previous studies suggest that nursing staff and
patients prefer the radial units.
The obvious limitations of the smaller units are increased
construction and staffing costs. Another shortcoming of radial
units may be problems with regard to wayfinding. Although standard
rectangular units are also confusing, their unidirectional symmetry
is less difficult to interpret than the bidirectional symmetry of
the radial units. This lack of wayfinding support could be easily
rectified by providing a window view. All the radial units surveyed
in this study lacked windows from the nurses' station to the
outside. Although the use of exterior walls for unit windows rather
than for patient rooms would decrease the number of beds around a
nursing station, the window space could be multifunctional and be
used as a small lounge area.
In conclusion, this study suggests that radial units require less
walking. If radial/cluster units are to be recommended for
hospitals, however, efforts must be undertaken to enhance their
wayfinding characteristics, and acoustical finishes and staffing
costs need to be balanced against the need for improved care
environments.
References
Bauer, H., and K. Knoblich. 1978. Ein beitrag zur
erfassung der laufleistung stationar tatiger krandenschwestern
(Recording of walking performance of nurse working in hospitals).
Zum Gesellschaft Hygiene 24 (7):539-540.
Burgio, L., B. Engel, A. Hawkins, K. McCorick, and A. Scheve. 1990.
A descriptive analysis of nursing staff behaviors in a teaching
nursing home: Differences among NAs, LPNs and RNs. The
Gerontologist 30:107-112.
Chambers, S., and D.A. Guerin. 1993. AIDS hospice unit: Design
criteria and prototypes. Journal of Interior Design 19
(2):41-50.
Falk, S., and N. Woods. 1973. Hospital noise levels and potential
health hazards. The New England Journal of Medicine
215:774-781.
Fisher, S. 1982. Design reduces nurses' walking, encourages
patients to visit with each other. Journal - American Health Care
Association 8 (March):40-43.
Glass, D., J. Singer, and I. Friedman. 1969. Psychic cost of
adaptation to an environmental stressor. Journal of Personality and
Social Psychology 12:200-210.
Haslam, P. 1970. Noise in hospitals: Its effect on the patient. The
Nursing Clinics of North America 5 (4):715-724.
Helton, M., S. Gordon, and S. Nunnery. 1980. The correlation
between sleep deprivation and the intensive care unit syndrome.
Heart and Lung 9 (3):646-648.
Hilton, B. 1985. Noise in acute patient care areas. Research in
Nursing and Health 8:283-291.
Kryter, K. D. 1985. The effects of noise on man. 2d ed. Orlando,
Fla.: Academic Press.
Macdonald, M. R., J. J. Schentag, W. B. Ackerman, and M. A. Walsh.
1981. ICU nurses rate their work places. Hospitals 55 (2):115-116,
118.
Marshall, L. 1972. Patient reaction to sound in an intensive
coronary care unit. Community Nursing Research 5:81-92.
Minckley, B. 1968. A study of noise and its relationship to patient
discomfort in the recovery room. Nursing Research 17:247-250.
Redding, J. S., T. S. Hargest, and S. H. Minsky. 1977. How noisy is
intensive care? Critical Care Medicine 5 (6):275-276.
Rothman, D. J. and E. A. Tynan. 1990. Advantages and disadvantages
of special hospitals for patients with HIV infection. The New
England Journal of Medicine 323:764-768.
Seelye, A. 1982. Hospital ward layout and nurse staffing. Journal
of Advanced Nursing 7:195-201.
Shepley, M. M., and P. Wilson. 1999. Designing for persons with
AIDS: A postoccupancy study at the Bailey-Boushay House. Journal of
Architectural and Planning Research 16 (1):17-32.
Snyder-Halpern, R. 1985. The effect of critical care unit noise on
patient sleep cycles. Critical Care Quarterly 7 (4):41-51.
Sturdavant, M. 1960. Intensive nursing service in circular and
rectangular units. Hospitals, JAHA. 34:46-48, 71-78.
Templeton, D. 1993. Design acoustics. In Acoustics in the Built
Environment, edited by D. Templeton. Oxford:
Butterworth-Heinemann.
Topf, M. 1985. Personal and environmental predictors of patient
disturbance due to hospital noise. Journal of Applied Psychology 21
(6):717-733.
Trites, D., F. Galbraith, M. Sturdavant, and J. Leckwart. 1970.
Influence of nursing-unit design on the activities and subjective
feelings of nursing personnel. Environment and Behavior 2
(3):303-334.
Velsey, D. W., and C. F. D. Egbert. 1994. Healthcare facilities. In
Architectural Graphic Standards, edited by J. R. Hoke. New York:
John Wiley, 827-830.
Waterfall, W. 1929. A loudness scale. Engineering News Record
102:60-62.
|
