A Noise-Sensor Light Alarm Reduces Noise

in the Newborn Intensive Care Unit

Ying-Ju Chang, Ph.D., R.N.,1 Ya-Jung Pan, B.S.N., R.N.,2 Yuh-Jyh Lin, M.D.,3
Yan-Zen Chang, B.E.,4 and Chyi-Her Lin, M.D.3

ABSTRACT

This one-group pre-post test design was to evaluate sound distribution and
sudden peak noise frequencies (SPNs) and the associated events after using a noise-sensor

Downloaded by: UPSTATE Medical University. Copyrighted material.

light alarm in a tertiary neonatal intensive care unit (NICU). The alarm is activated as the
sound level reaches
65 A-weighted decibel (dBA). The environmental sound level was
monitored continuously for a period of 1 week before and 1 month after using the alarm.
The mean sound level in the incubator of patients receiving ventilator support before and
after using the device were 58.0  0.6 and 56.4  0.7 dBA (t ¼ 8.619; p < 0.001), whereas
those at the radiant heated bed were 58.0  2.4 and 58.1  2.0 dBA (t ¼ 0.715; p ¼ 0.476).
The percentage of observation time of sound levels < 58 dBA increased by 28% in the
incubator and 4% at the radiant heated bed (p < 0.001). Episodes of SPN decreased from
630 to 185 times/d in the incubator and from 2069 to 748 times/d at the radiant heated
bed after using the device. The noise-sensor light alarm effectively reduces sound level
and episodes of SPN in the NICU. This may alleviate stress of noise for newborns with
critical illness.

KEYWORDS: Noise, alarm, newborn intensive care unit

Noise is regarded as an important source of
stress in neonatal intensive care units (NICUs).1 The
fetus receives low-pitched sounds derived mainly from
maternal rhythmic heart beating and intestinal peristal-
sis. In contrast, newborns in the NICU are subjected to
more intense, arrhythmic sounds.2 Monitoring physio-
logical variations in newboms with critical conditions,
using life support equipment, and performing caregiving
activities inevitably makes noise. Exposure to excessive
noise has been associated with infants’ heart rate accel-
eration, bradycardia, decline in oxygenation, increase in
muscle tension, blood and intracranial pressure, sleep
disturbance, agitation, and auditory disorders.3–9 These
responses aggravate infants’ energy consumption, induce
physiological instability and can affect growth and de-
velopment. In addition, recent studies reveal that the
sound intensity not only has an adverse effect on the
infants but also can interfere with caregivers’ communi-
cation and job performance.10
In 1974, the American Academy of Pediatrics
(AAP) recommended that noise levels exceeding 58 dB
be avoided in the NICU.1 Furthermore, in 1997, due to
the adverse impacts of noise on newborns, the AAP also
recommended that the ideal sound level in the NICU

1
Department of Nursing, College of Medicine, National Cheng Kung
University, Taiwan, Republic of China; 2Nursing Department, Chi-
Mei Foundation Hospital, Taiwan, Republic of China; 3Department
of Pediatrics, National Cheng Kung University Hospital, Taiwan,
Republic of China; 4Photoelectronics Application Department, Semi-
conductor Process Equipment Division, Mechanical Industry Research
Laboratories, Industrial Technology Research Institute, Taiwan, Re-
public of China.
Address for correspondence and reprint requests: Dr. Chyi-Her Lin,
Department of Pediatrics, National Cheng Kung University, Taiwan,
No. 1 Ta-Hsueh Road, Tainan, Taiwan, 701, Republic of China.
Am J Perinatol 2006;23:265–272. Copyright # 2006 by Thieme
Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001,
USA. Tel: +1(212) 584-4662.
Published online: May 9, 2006.
DOI 10.1055/s-2006-941455. ISSN 0735-1631.
265
266 AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 23, NUMBER 5 2006

is < 45 dB.11 For preserving better sleep of infants,

Philbin et al12 recommended that continuous sound in
any patient’s space should not exceed an hourly equivalent
sound level of 50 A-weighted decibels (dBA) and the
duration of sound level exceeding 55 dBA should only be
10% of the time or 6 minutes in an hour.
Studies have shown that human-related factors
accounted for a major source of noise in the NICU, such
as conversations among the staff, bumping metalware,
and forcefully opening and closing trash can lids.4,13,14
These findings indicate that modification of staff behav-
ior may be the most effective way to attenuate NICU
sound. Researches have documented that staff behaviors
can be modified by education.15–17 However, it is diffi-
cult to remember to prevent noise all the time, and staff
may need to be reminded when the sound level increases.
Few studies have focused NICU noise reduction using
an alarm reminder. The objective of this study was to

Downloaded by: UPSTATE Medical University. Copyrighted material.

assess the benefits of a bedside noise-sensor light alarm
on reducing the sound level and sudden peak noise
(SPN) episodes in the NICU.
Figure 1 The appearance of the noise-sensor light alarm. The
red bulb lights when the noise level exceeds 65 dBA.
METHODS

Study Design Noise-Sensor Light Alarm

A one-group pre-post test design was conducted from
September 2002 to February 2003 in a 20-bed NICU of
a tertiary care center in southern Taiwan. The sound
level inside incubators and at radiant-heated beds, both
of which were combined with ventilators, were moni-
tored continuously for a period of 7 days before and
1 month after introduction of the noise-sensor light
alarm system. The same model of incubator was used
throughout the study. During the monitoring period,
day, evening, and night shifts of 5 weekdays were
grouped, respectively, and then one sample from each
shift was selected randomly. Sources of SPN in these
24 hours during pretest and posttest periods were ob-
served directly by three trained research assistants.
The assembly structure and circuit diagram of the noise-
sensor light alarm are shown in Figs. 1 and 2, respec-
tively. The operation principle of the noise-sensor light
alarm uses microphone and amplifier circuits to convert
the environmental noise (the sum of acoustic energy
from a human voice, equipment noise, air conditioner
noise, etc.) to electrical potential. This electrical poten-
tial, when it is higher than the set point of potential, will
drive the alarm signal of the red light-emitting diode
(LED) lamp.
This invention was developed by one of the
authors (Y.-Z.C.) for this study, and costs approximately
$200 U.S. Becuse human ears are more sensitive and
susceptible to the adverse effect of high-frequency noise

Figure 2 The circuit diagram of the noise-sensor light alarm. VCC, electrical potential; X1, microphone; R1-11, electric resistance; C1-2,
electric capacitance; D1, diode; U1-2, integrated circuit; LED, light-emitting diode.

than to low-frequency noise,3 the A-weighted decibel
scale was applied to monitor noise levels in the NICU.
Based on a previous study indicating that the mean loud
conversation or monitor alarm sound level was approx-
imately 65 dB,15 the light alarm was then set to an
electrical potential threshold induced by noise at
65 dBA. When the environmental noise was > 65 dBA,
a red LED light of the alarm was automatically turned on
to remind the staff to lower their voices or modulate the
source of noise. After repetitive comparisons and calibra-
tions with a sound meter (TES 1336 Soundmeter; TES
Electrical Electronic Corp., Taiwan, Republic of China)
during a testing period, the alarm was found to have a
high sensitivity of 94% and specificity of 92%.
The NICU consisted of two separated areas. A
previous study found that the area near the nurse station
with three opened rooms and a total of 14 beds had a
higher noise level than the other area.13 Therefore, this
area with more noise was investigated. A noise-sensor
light alarm was installed on the central wall. The
researcher informed the NICU staff of the purpose of
explained the alarm and the meaning when the red light
was turned on.
Measurement of Environmental Sound Level
The environmental sound level was measured using the
TES 1336 Soundmeter with accuracy an of  1.5 dBA.
The sound meter’s microphone was placed in the in-
cubator and at the radiant-heated bed within 10 cm of an
infant’s ear. A computer with data logging software was
used to record the level of sound monitored second by
second.
Measurement of SPN
Consisting of the set point of the noise-sensor light
alarm, SPN in this study was defined as any abrupt noise
event
65 dBA. Three research assistants were trained
by one of the authors (Y.-J.P.) to observe and record
sources of the SPN (interrater reliability was 87%). The
researcher first summarized possible noise sources into a
number of categories and used abbreviated terms to
represent each of them. During bedside observation
periods, the corresponding categorical terms were noted
on a chronological table when peak noise events oc-
curred. To allow easy recall of data correlating to
particular noise events from the computer records, the
research assistant used timers synchronized with the
computer and noted the time a peak noise occurred
with the precision the 1-second level.
Statistical Analysis
Environmental sound level data were initially acquired
second by second and aggregated by minute and hourly
NOISE-SENSOR LIGHTALARMIN THE NICU/CHANG ET AL 267
for comparisons. The hourly mean sound levels of pretest
and posttest periods were compared with paired t sta-
tistics. The sources and frequencies of peak noise were
categorized and recorded directly from bedside observa-
tions. Mean sound levels of SPN were calculated by
drawing data from synchronized records in the computer.
RESULTS
During the study periods the hourly mean sound level
inside the incubator was 58.0  0.6 (standard deviation)
and 56.4  0.7 dBA in the pretest and posttest, respec-
tively. The paired t test shows that the difference
between the two periods was significant (t ¼ 8.617;
p < 0.001). The hourly mean sound level on the radi-
ant-heated bed was 58.0  2.4 and 58.1  2.0 dBA in the
pretest and posttest, respectively, and no significant
difference was found (t ¼ 0.715; p ¼ 0.476). The envi-
Downloaded by: UPSTATE Medical University. Copyrighted material.
ronmental sound levels were divided into four intensity
groups based on the pretest mean of the radiant-heated
bed (58 dBA) and with an increment criterion of 2 SD of
the sound level (approximately 5 dBA). The respective
percentage of the total observation time, based on the
aggregated noise level per minute, at each level was
calculated (Table 1). The percentage of sound levels
< 58 dBA in the incubator increased by 28%, whereas
the percentages for the other three groups decreased
after using the alarm. The proportional test showed that
these differences were all significant (p < 0.001). Similar
change patterns were found on the radiant-heated bed.
The percentage of sound levels < 58 dBA increased,
whereas the percentage of levels for the other three
groups decreased. Although amounts of change were
small, they were all statistically significant (p < 0.001).
The peak period of sound levels across 24 hours
was further examined by averaging hourly sound level
throughout a day. In the incubator, as shown in Fig. 3, a
higher sound level phase occurred once every 3 to
4 hours and the sound levels from 10 A.M. to 4 P.M.
were higher than that in remaining hours before the
light alarm was installed. However, the hourly mean
sound levels throughout the day were distributed mostly
at lower values after the device was installed. The phase
of peak sound level was found in the period from 8 A.M. to
Table 1 Percentages of Total Observation Time by
Sound Levels and Beds with a Ventilator
Incubator Radiant-Heated Bed
Sound Level
(dBA) Pretest Posttest Pretest Posttest
< 58 52.0 80.2* 56.2 60.3*
58–62 44.7 17.8* 37.2 34.8*
63–67 2.4 1.5* 4.8 4.0*

68 0.9 0.4* 1.8 0.9*
Total 100 100 100 100
*Results of proportional test with p < 0.001.
268 AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 23, NUMBER 5
Figure 3 Noise distribution of the incubator of patients receiving a ventilator support during a 24-hour period.
1 P.M. For the radiant-heated beds, neither a difference
in hourly mean sound levels nor a clear periodic peak
phase tendency was observed before and after the alarm
was installed. Similarly, the sound levels from 8 A.M. to
4 P.M. were relatively higher than that in the remaining
hours (Fig. 4).
Direct observation of SPN episodes in 24 ran-
domly selected hours of the weekday showed that an
SPN inside the incubator before and after the light alarm
was installed occurred 630 and 185 times, respectively.
The four primary sources of SPN before light alarm use
were nursing activities (57.5%), monitor alarms (19.2%),
staff conversations (10.3%), and oxygen supply proce-
dures (6.3%). These items remained the main sources of
SPN after the device was installed; however, the occur-
rence of SPN was reduced by 71.5% (Table 2). An SPN
at the radiant-heated bed with ventilator before and after
light alarm use occurred 2067 and 748 times, respec-
tively. The four primary sources of SPN before light
Figure 4 Noise distribution on the radiant-heated bed of patients receiving a ventilator support during a 24-hour period.
2006
Downloaded by: UPSTATE Medical University. Copyrighted material.
alarm installation were conversations (51.3%), monitor
alarms (18.7%), nursing activities (9.7%), and movement
of carts or equipment (4.3%). These sources did not
change after the light alarm was installed; however,
occurrences of SPN decreased by 63.8% (Table 3).
DISCUSSION
Data from this study indicate that a noise-sensor light
alarm intervention has a positive effect on reducing
environmental sound in the NICU. In the incubators
with ventilators, the percentage of observation time of
sound levels > 58 dBA was decreased after introduction
of the sensor-light alarm system. Although the mean
sound level was only reduced by 2 dBA, a reduction of
3 dBA reflects a 50% of noise reduction perceived by the
human ear because decibel measurements use a logarith-
mic scale to measure sound pressure.11,15 Therefore,
the findings were clinically significant in environmental
 NOISE-SENSOR LIGHTALARMIN THE NICU/CHANG ET AL 269

Table 2 Peak Noise Distribution in the Incubator with a Ventilator before and 1 Month after Using Noise-Sensor Light
Alarm
Frequency (%) Noise Level (dBA)*
Source Pretest Posttest Pretest Posttest

Caregiving (rubbing, moving, or wrapping object) in the incubator 362 (57.5) 98 (53.0) 69.3  4.1 68.5  3.5
Alarm of monitors 121 (19.2) 14 (7.6) 71.4  4.6 67.8  3.6
Staff conversation 65 (10.3) 17 (9.2) 68.2  3.1 68.8  4.6
Handling equipment of oxygen supply 40 (6.3) 4 (2.2) 72.9  7.3 66.7  2.7
Other (moving cart or chair, closing trash can lid or drawers, 32 (5.4) 50 (26.7) 67.6  3.7 67.4  3.0
bumping or dropping objects, cleaning apparatus and containers,
tearing plastic or paper bag, reading charts, opening, closing,
bumping, and cleaning the incubator, on or off monitors)
Total 630 (100.0) 185 (100.0) 69.5  4.3 68.1  3.4
*Data expressed as mean  1 standard deviation.

support for newborns. On the radiant-heated beds, the as high as 70 to 80 dB when a ventilator or other oxygen
mean sound level did not change after the noise-sensor supply devices were incorporated.22–24 These study re-

Downloaded by: UPSTATE Medical University. Copyrighted material.

light was introduced. Only a small increase in
the percentage of observation time that sound levels
< 58 dBA was found. This finding differed from the
experimental result in the incubator. This was probably
because a radiant-heated bed is an open space without
any shielding, unlike the environment in an incubator.
Thus, there might be more uncontrollable noise sources.
In addition, the noise-sensor light threshold was set
at 65 dBA. The red light turned on only if a sound
impulse exceeded the threshold. Therefore, this might
have modified the clinical personnel’s awareness of more
intense noise and made them less sensitive to the noise
< 65 dBA.
Previous studies indicated that sound levels in the
NICU ranged from 45 to 135 dB and the mean sound
level varied considerably from 54 to 80 dBA.11,13,18–21
Long et al14 found that incubators could partially block
noise and preserve a quieter internal space than the
external environment by 5 to 18 dB. Other studies
reported that sound levels inside incubators could reach
sults revealed that the sound level in the NICU has great
variation. The daily mean sound level may not be
sufficient to describe the actual clinical situation.
Through continuous monitoring, this study used per-
centages of observation time for specific noise level
ranges to display newborns’ exposure to noise. This is
another way to determine the sound context of the
NICU and provides a new method for assessing the
effects of the noise-sensor light alarm.
Similar to the results from previous reports, this
study found that the highest sound level, whether in
incubators or radiant-heated beds, occurred in the day-
time,13,24–26 suggesting that noise is associated primarily
with staff activities. Thus, analysis of staff activities in
the NICU can be a useful basis for noise reduction.
The most significant effect from the noise-sensor
light was to reduce peak noise occurrences. Abrupt
noise may awaken infants and causes startle responses,
which are stressful to the vulnerable infants.12 Because
SPN events in the NICU occurred mainly due to

Table 3 Peak Noise Distribution at the Radiant-Heated Bed with a Ventilator before and 1 Month after Using the
Noise-Sensor Light Alarm
Frequency (%) Noise Level (dBA)*
Source Pretest Posttest Pretest Posttest

Staff conversation 1062 249 (33.3) 68.4  2.9 68.9  3.9

Alarm of monitors 386 (18.7) 108 (14.4) 69.5  3.1 69.3  3.6
Caregiving (rubbing, moving or wrapping object) on the radiant warmer 201 (9.7) 39 (5.2) 67.6  2.1 71.4  4.6
Moving cart, chair, or equipment 90 (4.3) 33 (4.4) 69.3  2.7 67.9  4.9
Opening or closing trash can lid 54 (2.6) 33 (4.4) 68.0  3.1 68.8  4.1
Others (bumping objects or stainless cart, handling equipment of 276 (13.4) 286 (38.3) 69.3  2.8 68.3  3.7
oxygen supply, cleaning the warmer bed, on or off monitors,
tearing plastic or paper bag, closing drawers at the station,
staff walking or running, cleaning apparatus and containers,
reading charts, dropping objects, phone or pager ringing, hanging)
Total 2069 748 (100.0) 68.6  2.8 68.6  3.9
*Data expressed as mean  1 standard deviation.
270 AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 23, NUMBER 5
human-related factors such as caregiving procedures
and staff conversations, these sources have favorable
potential for modification. The staff members involved
in this study were not informed about what peak noise
sources were detected during the pretest period. They
only received immediate feedback via the light alarm,
making them aware of noise occurrences after installa-
tion of this alarm. One month after noise-sensor light
alarm introduction, SPN events were reduced by 70 and
63% in incubators and radiant-heated beds, respec-
tively. Similar results were shown in a previous study
displaying noise intensities from a variety of caregiving
activities on videotape, in-service education for noise
prevention, and a 3-month program of behavior mod-
ification for staff in the ICU. After education, the mean
sound level in the unit decreased,15,27 and the peak
sound events > 80 dBA were reduced by 28%.27 How-
ever, some studies found that education programs may
not be effective in noise reduction. Although staff
members might be aware of the sources of noise, they
may not be aware of when to control the noise.18,27 This
indicates the importance of an immediate alarm for
noise reduction. A similar recommendation of a sound-
detection device that sends a signal to a vision alarm
when the sound level is considered too high has been
addressed for the purpose of infants’ developmental
supportive care.28
Although the mean sound level of incubators
after using a noise-sensor light alarm was reduced, it is
not ideal. This is probably because the light alarm was
set to a high threshold (
65 dBA). It is possible that
if the noise-sensor light were set to a threshold of
60 dBA and the sound level values under this threshold
are simultaneously displayed to the staff, the observa-
tion time for environmental sound levels > 58 dBA
would become longer. This issue deserves further
study. The current study indicated that staff caregiving
activities and conversations were the main sources
of peak noise, in agreement with previous stud-
ies.13,14,24,29 Therefore, further studies focusing on
the context and factors involving these peak noise
events will be crucial to sound reduction. If a staff
education program on sound reduction is conducted
with noise-sensor light use, the combination may
create a synergistic effect.
Reducing sound levels in the NICU is a way to
assist and support newborn development and should also
be the common objective of NICU personnel. The
noise-sensor light alarm described in this study de-
creased NICU sound level and episodes of SPN. How-
ever, whether long-term use of noise-sensor light can
bolster the effect of noise reduction or lead to a state of
neglect requires further investigation. Furthermore, this
study is limited by the fact that the observation and
manipulation were done in a single NICU. The results
could be confounded by the number of patients, person-
2006
nel, and ventilators used in the unit during the pretest
and posttest periods. Whether the noise-sensor light
alarm could have the similar effects in other settings
warrants further study.
ACKNOWLEDGMENTS
This study was sponsored by the Cheng-Hsing Medical
Foundation and Chinese Premature Baby Foundation.
The authors thank the staff in the NICU of NCKUH
for their support in the noise-sensor light alarm shape
design and data collection.
REFERENCES
1. American Academy of Pediatrics Committee on Environ-
mental Hazards. Noise pollution: neonatal aspects. Pediatrics
1974;54:476–479
Downloaded by: UPSTATE Medical University. Copyrighted material.
2. Gottfried AW, Hodgman JE. How intensive is newborn
intensive care? An environmental analysis. Pediatrics 1984;74:
292–294
3. Catlett AT, Holditch-Davis D. Environmental stimulation of
the acutely ill premature infant: physiological effects and
nursing implications. Neonatal Netw 1990;8:19–26
4. DePaul D, Chambers SE. Environmental noise in the
neonatal intensive care unit: implications for nursing practice.
J Perinat Nurs 1995;8:71–76
5. Osfeld B, Krawciw N, Burke S, et al. Impact of environment
on the high risk infant: the effect of monitor alarm. Pediatr
Res 1990;27:218A
6. Trapanotto M, Benini F, Farina M, et al. Behavioral and
physiological reactivity to noise in the newborn. J Paediatr
Child Health 2004;40:275–281
7. Van Reempts PJ, Wouters A, De Cock W, Van Acker KJ.
Clinical defense response to cold and noise in preterm
neonates after intrauterine conditions associated with chronic
stress. Am J Perinatal 1996;13:277–286
8. Wharrad HJ, Davis AC. Behavioral and autonomic responses
to sound in pre-term and full-term babies. Br J Audiol 1997;
31:315–329
9. Zahr LK, Balian S. Responses of premature infants to routine
nursing interventions and noise in the NICU. Nurs Res 1995;
44:179–185
10. Thomas KA, Martin PA. NICU sound environment and the
potential problems for caregivers. J Perinatol 2000;20:S94–
S99
11. American Academy of Pediatrics Committee on Environ-
mental Hazards. Noise: a hazard for the fetus and newborn.
Pediatrics 1997;100:724–727
12. Philbin MK, Robertson A, Hall JW. Recommended permis-
sible noise criteria for occupied, newly constructed or renovated
hospital nurseries. The Sound Study Group of the National
Resource Center. J Perinatol 1999;19:559–563
13. Chang Y-J, Lin C-H, Lin L-H. Noise and related events in a
neonatal intensive care unit. Acta Paediatr Taiwan 2001;42:
212–217
14. Long JG, Lucey JF, Philip AG. Noise and hypoxemia in the
intensive care nursery. Pediatrics 1980;65:143–145
15. Elander G, Hellstrom G. Reduction of noise levels in
intensive care units for infants: evaluation of an intervention
program. Heart Lung 1995;25:376–379

16. Hale DR. Noise in the hospital: a quality improvement
approach. J Nurs Admin 1996;26:4–8
17. Philbin MK, Gray L. Changing levels of quiet in an intensive
care nursery. J Perinatol 2002;22:455–460
18. Horgan MJ, Eston C. Sources of environmental noise:
perceptions by NICU staffs. Pediatr Res 1999;45:201A
19. Levy GD, Woolston DJ, Browne JV. Mean noise amounts in
level II vs level III neonatal intensive care units. Neonatal
Netw 2003;22:33–38
20. Thomas K. How the NICU environment sounds to a preterm
infant. Am J Matern Child Nurs 1989;14:249–251
21. Weibley T. Inside the incubator. Am J Matern Child Nurs
1989;14:96–100
22. Bess FH, Peek B, Chapman J. Further observations on noise
levels in infant incubators. Pediatrics 1979;63:100–106
23. Blennow G, Svenningsen N, Almqist B. Noise levels in infant
incubators (adverse effects?). Pediatrics 1974;53:92–32
NOISE-SENSOR LIGHTALARMIN THE NICU/CHANG ET AL 271
24. Chen H-F, Chang Y-J. Noise distribution of an incubator
with nebulizer at neonatal intensive care unit in southern
Taiwan. J Nurs Res 2001;9:25–32
25. Moore MM, Nguyen D, Nolan SP, et al. Intervention to
reduce decibel levels on patient care units. Am Surgeon 1998;
64:894–900
26. Robertson A, Cooper-Peel C, Vos P. Peak noise distribution in
the neonatal intensive care unit. J Perinatol 1998;18:361–364
27. Kahn DM, Cook TE, Carlisle CC, et al. Identification and
modification of environmental noise in an ICU setting. Chest
1998;114:535–540
28. Bowie BH, Hall RB, Faulkner J, Anderson B. Single-room
infant care: future trends in special care nursery planning and
design. Neonatal Netw 2003;22:27–34
29. Gorski P, Hole W, Leonard C, Martin J. Direct compu-
ter recording of premature infants and nursery care: distress
following two interventions. Pediatrics 1983;72:198–202
Downloaded by: UPSTATE Medical University. Copyrighted material.
Downloaded by: UPSTATE Medical University. Copyrighted material.