Content uploaded by Mariana Figueiro
Author content
All content in this area was uploaded by Mariana Figueiro on Jul 17, 2015
Content may be subject to copyright.
A novel night lighting system for postural
control and stability in seniors
MG Figueiro
a
, L Gras
b
, RQi
a
, P Rizzo
a
, M Rea
b
and MS Rea
a
a
Lighting Research Center, Rensselaer Polytechnic Institute, 21 Union Street, Troy, NY 12180, USA
b
The Sage Colleges, 45 Ferry Street, Troy, NY 12180, USA
Received 21 June 2007; Revised 16 August 2007; Accepted 21 August 2007
Injuries resulting from falls in seniors are a significant health concern. Visual
information is important for postural control in seniors and postural control is
affected by the light level. Moderate ambient illumination might be effective for
postural control, but might compromise subsequent sleep efficiency and quality
for seniors. In addition to safety concerns, a night lighting system must also be
accepted by seniors. Researchers investigated if a self-luminous, night lighting
system that provided horizontal and vertical cues could positively affect postural
control in older subjects and evaluated its acceptance among seniors. The results
showed that the night lighting system was effective for maintaining postural
control in the critical, early phase of the sit-to-stand task and that it was preferred
over conventional night lights.
1. Background
Injuries resulting from falls in seniors (65 years
or older) are a public health concern.
1
Falls
and related injuries are one of the major
threats to the independence of older adults and
are associated with increased functional
impairment, disability, mortality and place-
ment in more controlled environments.
Postural control involves two functional
aspects: postural orientation (ability to estab-
lish and maintain the body position with
respect to the environment) and postural
stability (ability to maintain the body position
in equilibrium).
2
Information about the envi-
ronment from the visual, proprioceptive and
vestibular systems affects postural control.
Age-related changes to these systems result in
impaired balance control, and thus can lead
to increase risk of falls commonly found
among seniors.
3
Visual information provides
a spatial reference for self-position and loca-
tion of obstacles within a person’s surround-
ings. Removal of visual cues by closing the
eyes has been shown to result in increased
body sway.
4,5
Also low ambient light levels
have also been shown to reduce postural
control.
6
Although visual capabilities
decrease as we age,
7
the dependence on
visual information for maintenance of pos-
tural stability seems to increase with age due
to parallel age-related changes in the vestib-
ular and somatosensory systems.
8–10
Postural
stability has been shown to relate to measures
of visual acuity and contrast sensitivity, with
visual contribution to postural stability being
greater for those who have fallen before
than for those with no history of falls.
5,11
Address for correspondence: Mariana G Figueiro, Lighting
Research Center, Rensselaer Polytechnic Institute, 21 Union
Street, 3rd floor, Troy, NY 12180, USA.
E-mail: figuem@rpi.edu
Lighting Res. Technol. 2008; 40: 111–126
ß The Chartered Institution of Building Services Engineers 2008 10.1177/1477153507084198
Persons with severe deficits in the vestibular
or the somatosensory systems have also been
shown to rely on visual cues more often than
normal persons, and they will lose their
balance if visual cues are removed by eye
closure.
12
Similarly, patients who have lost
their central vision rely more on non-visual
systems to maintain balance
13
and those with
simulated impaired vision, either in the form
of defocus or simulating cataract, showed
reduced postural stability.
14,15
The acceptance of any technology, no
matter how useful it is purported to be,
determines whether it will have a real benefit
to users. Moderate to bright ambient light
levels have been shown to improve postural
control over dim lighting in seniors, but the
layout of a bedroom, for example, maybe such
as not to have a light switch conveniently
located next to the bed.
16
Moreover, bright
overhead lighting as commonly found in
senior residences can disturb sleep.
16
Consequently, seniors in these facilities often
try to reach the bathroom at night without
turning the lights on at all, increasing their
likelihood of falling. In principle then, a useful
and practical night lighting system would not
only be demonstrably better at reducing the
risk of falls but it would also be easy and
convenient to turn on at night and would not
initiate light levels so bright as to be disturbing
to residences. Moreover, such a system should
be cost effective so that low-income seniors
can take advantage of its functional benefits.
The first goal of the present study was to
investigate the effectiveness of a novel self-
luminous light emitting diode (LED) night
lighting system that provided linear spatial
orientation cues plus low ambient illumina-
tion for increasing postural control in healthy
seniors. A second related and important goal
was to determine if the night lighting system
would be acceptable to seniors in a mock-up
of a single-occupant, senior living facility
bedroom. As shown and discussed below,
the results suggest that a night-lighting system
that provides perceptual cues can improve
postural stability and control in healthy older
adults under laboratory conditions. The same
night-lighting system seemed to be well
accepted among healthy older adults; further
investigations of the utility and acceptance of
the lighting system for more seriously debili-
tated seniors need to be conducted.
2. Experiment 1: Postural control
2.1. Methodology
Postural control was assessed by using a
standardised sit-to-stand (STS) test (EquiTest
System
TM
). In this test, the subject is asked to
shift the body’s centre of gravity forward
from an initial position over the seat to a
location centred over the base of support
(feet), followed by extension of the body to an
erect standing position. A data acquisition
system integrated with sensors in the appara-
tus floor measures the subject’s time to stand
as well as his or her pitch and yaw when
moving from sitting to standing.
Sway velocity (SV) and left/right (L/R)
weight symmetry were used as measures of a
person’s ability to transfer from sitting to a
standing position because these measures are
affected by the visual information parame-
trically studied here. SV is the velocity of the
postural sway of a human subject during
standing or movement tasks. SV documents
control of the centre of gravity over the base
of support during the rising phase and for 5 s
thereafter and is expressed in degrees per
second. L/R weight symmetry is measured in
terms of the difference in the percentage of
body weight on each foot while subjects are
transferring from sitting to standing during
a trial.
Following standard assessment procedures,
subjects were asked to sit on blocks with their
feet on two force plates. The subject’s feet were
placed parallel with one foot on each force
plate. Subjects were instructed to rise from the
blocks as they usually do when rising from
112 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
a chair. A researcher gave the command ‘go’
and the task was initiated. After data acqui-
sition, subjects were instructed to sit back
down on the blocks. Subjects completed three
STS trials under every experimental condition.
Two lighting systems were developed
for the study. One lighting system was
comprised of three linear arrays of
amber LEDs (LINEARFLEXSIDE/615/OS/
LM11A/A;
max
¼ 615 nm, OSRAM
SYLVANIA). The arrays consisted of two
vertical strips of 140 LEDs and one horizontal
strip of 68 LEDs, with a connected power
totalling 16.2 W when operated at the highest
intensity. The three LED arrays were affixed
to a wood frame of the same dimensions and
height as a door frame 2.1 m high 1.02 m
wide (82
00
h 40
00
w). Subjects faced the door
frame, approximately 1 m (3 ft) away, during
the STS test, but the LEDs could not be viewed
directly; rather, they were hidden behind the
simulated door frame providing a sharp
luminous border between the wood fame and
the wall. At the highest level, the LEDs
provided 10 lx at the subjects’ corneas when
they were standing on the STS test apparatus.
To test the efficacy of different LED inten-
sities, the output of the LED arrays was
reduced to 3 lx, 1 lx and 0.3 lx at the subjects’
corneas with calibrated resistors. The door
frame had two different tilts and one correct
orientation for a two-session experiment. In
the first session, the door frame was tilted 4.38
to the left or placed correctly with respect to
gravity and consistently with other architec-
tural cues within the laboratory. In the second
session, the door frame was tilted 4.38 to the
right or placed correctly (Figure 1). In all
sessions, the experimenter guarding the sub-
jects was standing at their left side.
The other lighting system was comprised of
four conventional, 7-W incandescent night
lights, each with a single clear bulb encased in a
plastic lens. Three conventional night lights
were located at the junction of the floor and
the wall adjacent to the STS test apparatus; the
fourth was located across the room at the
junction of the floor and wall. Although the
number of night lights was greater than what is
typically found in residences, these night lights
were needed to provide relatively uniform,
dim ambient illuminance equal to that pro-
vided by the LED night light system at the
lowest illuminance, 0.3 lx, at the subject’s
corneas when the subject was standing at the
STS test apparatus. Only one lighting system
was used on a given trial. When the conven-
tional night lights were used, subjects also
faced the door frame during the STS test but
the LED lighting system was not energised.
Subject fatigue was a major concern for
this experiment, so the planned experimental
design utilised Latin Squares, a classic strategy
for increasing experimental efficiency while
minimising potential confounding between the
presentation order of the experimental condi-
tions and the levels of the independent vari-
ables.
17
Three ‘self-contained’ Latin Squares
were used in each of two sessions. Each Latin
Square counterbalanced the presentation
order of the four LED illuminances (10 lx,
3 lx, 1 lx and 0.3 lx at the cornea) and the two
door frame positions (correctly oriented versus
tilted, either left in the first session or right in
the second session). Again, to minimise the
number of trials, the presentation order of the
conventional night light (NL) was fixed in
the middle of every Latin Square to minimise
confounding from either learning or fatigue.
To limit very long sessions the experimental
trials were divided into two sessions. All of the
trials when the door frame was tilted to the left
were completed in the first session and all of
the trials when the door frame was tilted to the
right were completed in the second session.
Twelve subjects (all ages 65 years or older)
were recruited to complete the entire experi-
ment, but some of the recruited subjects
decided they could only complete the first
session; only eight subjects participated in
both sessions. Several more subjects had to be
recruited at different dates to complete the
Night lighting system for postural control
113
Lighting Res. Technol. 2008; 40: 111–126
experimental design. Six females and six males
participated in session 1 and nine females and
three males participated in session 2. Upon
arrival, subjects were asked to sign a consent
form approved by The Sage Colleges’ and
Rensselaer’s Institute Review Boards (IRBs).
The static blood pressures of potential
subjects were measured prior to the start of
the experiment while subjects were sitting and
immediately after standing. Orthostatic
hypotension is found in 15–35% of the older
population.
18
It is characterised by a decrease
in systolic blood pressure 420 mm Hg or a
fall in diastolic blood pressure 410 mm Hg
within 3 min of standing from a sitting or
supine position. Orthostatic hypotension
symptoms include light-headedness, weak-
ness, blurred vision, fatigue, lethargy and
falls.
18
Although subjects were screened for
orthostatic blood pressure, none of them
showed any of the symptoms, either because
they did not have it or because they were on
blood pressure medication.
2.2. Results
2.2.1. L/R weight symmetry
If visual cues are important for postural
stability, then tilting the door frame should
demonstrably affect the relative weight
applied to each foot in the transition from
sitting to standing. The difference between
the percentages of weight a subject applied to
the two feet when the door frame was tilted
relative to when the door frame was correctly
positioned was calculated for the same light-
ing condition.
Thus, for every subject, lighting condition,
and trial, the L/R weight symmetry percent-
age value for the tilted condition was
(a) (b)
Figure 1 (a) Correctly oriented and (b) tilted door frame positions
114 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
subtracted from the matched L/R weight
symmetry percentage value for the correctly
positioned door frame.
A two-way, mixed design, analysis of
variance (ANOVA) was performed using the
difference between the percentage of weight
applied to the two feet when the door frame
was tilted relative to the percentage of weight
applied to two feet when the door frame was
positioned correctly. The independent vari-
ables were (a) the five lighting conditions
(four LED light levels plus the NL) and
(b) the two door frame position (tilted-left
versus tilted-right) sessions; these were treated
as fixed variables and subject (N ¼ 12) was
treated as a random variable. There was no
statistically reliable difference between light-
ing conditions but the two door frame posi-
tion sessions were significantly different
(P50.005). There was no significant inter-
action between the two independent variables.
Figure 2 shows the average difference
between L/R weight symmetry percentage
values for the tilted-left and for the tilted-
right sessions; for every lighting condition
subjects leaned more to the left when the door
frame was tilted to the left and leaned more to
the right when the door frame was tilted to
the right. These results clearly show an
influence of the visual environment on
postural orientation. It should be noted that
this inference also holds true for the NL.
2.2.2. Sway velocity
SV is a particularly important measure for
this study because it indicates how stable
people are during the STS task. If visual cues
affect SV, then subjects should be less stable
when visual, proprioceptive and vestibular
information about the environment are incon-
gruous. Moreover, it appears that the transi-
tion from sitting to standing is governed by
two relatively distinct operations, initial
rising and then maintaining stabilisation on
one’s feet after rising operation has been
completed.
2
The standard measure of SV
generated by the apparatus used for clinical
assessment is a single number based upon the
average relative movement of a patient every
100 ms for 5 s. Given the two, apparently
different, operations people perform while
making the transition from sitting to standing
a more refined set of STS outcome measures
was developed in cooperation with the appa-
ratus manufacturer. Six sample times of
postural stability were obtained for every
trial; subject movements from 0 to 0.01 s, 1.0
to 1.01 s, 2.0 to 2.01 s, 3.0 to 3.01 s, 4.0 to
4.01 s and 5.0 to 5.01 s were obtained for
analysis.
If visual cues were, in fact, important to
postural stability as a person makes the
transition from sitting to standing, then
subjects should be less stable when visual
cues are in conflict with proprioceptive and
vestibular cues during the initial few seconds
of a 5 s trial than when visual cues are
veridical. Conversely, if visual cues are impor-
tant for maintaining postural stability after
the initial transition to standing has been
completed, then subjects should be less stable
in the final few seconds of a 5 s trial.
2
Obviously, if visual cues are important to
both, subjects should be unstable at all times
during a trial. In any case, if visual cues are
Left Right
(Night light)
Illumination at cornea (lx)
Variation (∆%)
Figure 2 L/R weight symmetry variation for the tilted-left
and the tilted-right sessions
Night lighting system for postural control 115
Lighting Res. Technol. 2008; 40: 111–126
demonstrably important to postural stability
when making the transition from sitting to
standing, it would then be important to
provide older people with veridical environ-
mental visual cues in nursing homes and
senior residents where falls are prevalent.
A four-way, mixed design, ANOVA was
performed on the SV data using the STS times
as the dependent variable; the independent
variables were: (a) the five lighting conditions,
(b) the six sample times during the 5 s trial,
(c) the door frame positions and (d) sequen-
tial sessions (first, when the door frame was
tilted to the left for half the trials and, second,
when the door frame was tilted to the right for
half the trials) were treated as fixed variables;
subject (N ¼ 12) was treated as a random
variable. It should be noted that it was
important to look for interactions between
the independent variables because visual cues
could, as outlined above, differentially affect
the early and late phases of the transition
from sitting to standing. Given the large
number of samples, and to limit examination
of possible spurious statistical significant
results, a probability 1% was taken as the
statistical criterion for Type I errors.
Figure 3 shows the statistically significant
main effect of sample times (P50.0001).
In the initial few seconds of transition,
subjects are swaying more, but seemed more
stabilised during the last few seconds of the 5 s
trial. The data in the first 2 s of the trial period
appear to follow one function whereas for the
subsequent time, there appears to be another,
slightly different function to describe the data.
This change in response functions is consis-
tent with the literature
2
suggesting that there
are two distinct operations performed by
seniors when making the transition from
sitting to standing.
A statistically significant main effect of
door frame tilt (correct orientation versus
tilted, both left and right) was also revealed
in the ANOVA (P50.004). Subjects were less
stable when the door frame was tilted than
when the door frame tilt was consistent with
proprioceptive and vestibular cues.
Figure 4 shows the significant interaction
(P50.001) between sample times and door
frame position (correct orientation versus
tilted, both left and right). Post hoc student
t-tests were used to compare the mean values
for the two door frame positions at the six
sample times. Only the first two sample times,
0–0.01 s and 1.0–1.01 s, showed significant
differences between the two door frame
conditions (t ¼ 0, P50.006; t ¼ 1, P50.003).
The ANOVA also showed a statistically
significant interaction (P50.001) between
sample times and the two sequential sessions.
This interaction was not easily interpreted but
Trial number
Sway velocity (deg/s)
Figure 3 Average sway velocities as a function of time during a 5 s trial
116 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
was hypothesised to have happened either
because different subjects were run in the first
and second sessions or because an experi-
menter was, for subjects’ safety, always
standing close to the left side of the subjects
for both sessions.
3. Experiment 2: User acceptance
3.1. Methodology
The second experiment was conducted in a
mock-up of a single-occupant, senior living
facility bedroom 3 m 2.9 m (122
00
114
00
)
and adjoining private bathroom. The main
objective of this experiment was to compare
the acceptability of the novel night lighting
system used in the first experiment, including
a passive infrared (PIR) motion sensor for
automatically turning the night lighting
system on when subjects get up at night and
automatically turning it off when they
return to bed, with a conventional wall-plug
mounted night light (Figure 5). The bed-
room–bathroom configuration was easily
changed by a hinged wall segment containing
the bathroom door, providing subjects
with and without a view of the bathroom
Trial number
Sway velocity (deg/s)
Figure 4 Significant interaction between sample times and door frame positions
Figure 5 The novel night lighting system and the conventional night light used in the study with a view of the
bathroom door. Only one lighting system was operated at a time when the subjects provided their evaluations
Night lighting system for postural control 117
Lighting Res. Technol. 2008; 40: 111–126
door (Figure 6). Two configurations were
used because both are common in North
American senior living facilities.
Two lighting systems were used in the
study (Figure 5). The novel lighting system
was comprised of three linear arrays of
yellow LEDs (
max
¼ 589 nm) manufactured
by Ningbo SELF Electronics, China. Two
arrays consisted of vertical strips of
72 LEDs each and the other was a hori-
zontal strip of 36 LEDs (180 LEDs total).
The three LED arrays were affixed to a
wood frame forming the simulated bath-
room door 2.1 m 1.02 m (82
00
h 40
00
w).
The novel lighting system was operated at
two power levels, approximately 6.5 W or
approximately 0.65 W, producing either 1 lx
or 0.1 lx at the subject’s eyes when seating
on the edge of the bed. The other lighting
system was a conventional wall-plug night
light with a single, 7 W, clear lamp encased
in a plastic lens. The conventional night
light was located at the junction of the floor
and the wall adjacent to the bed. The
conventional night light produced 0.1 lx at
the subject’s eyes. Only one lighting system
was used on a particular trial.
Twelve subjects, seven males and five
females, ages 65 years and older volunteered
as subjects. All subjects were considered
independently mobile and readily able to
make trips to the bathroom unassisted.
Subjects were obtained mainly using
Rensselaer’s e-mail server, but some were
also recruited through word-of-mouth to
visitors of the Lighting Research Center.
Upon subjects’ arrival at the laboratory,
they were asked by the experimenter to sit,
in turn, at a desk illuminated only with a
desk lamp housing an incandescent lamp;
general room lighting was turned off.
Subjects read and signed a consent form
while a research assistant prepared the first
of the lighting scenarios in the bedroom
mock-up. Then, the experimenter informed
subjects about the general purpose of the
experiment, which was to evaluate lighting
conditions under various scenarios. Subjects
were also informed of the approximate
length of the study (about 20 min) including
the sequence of events and how they would
be asked questions. Subjects were then led
into the bedroom mock-up. The subject was
asked to lie down on the bed and close his
or her eyes for 10 min to adapt to the dark
room. If a second subject accompanied the
first (i.e. a spouse or friend who was
subsequently going to serve as a subject in
the experiment), that subject waited at the
desk outside the mock up and be given dark
glasses to wear to begin adapting to a dark
environment while the first subject com-
pleted the experiment. The experimenter was
positioned behind the partition serving as
the wall separating the simulated bathroom
and bedroom to operate the two lighting
systems.
After sitting up from a prone position on
the bed subjects were sequentially presented
the novel lighting system producing 1 lx at the
eye (system power 6.5 W) or the conven-
tional night light producing 0.1 lx at the eye
(7 W); subjects saw the two lighting condi-
tions in a counterbalanced order (Figures 6(a)
and (e)). After they had seen both lighting
conditions, including enough time for the
novel night lighting system to turn on and
then turn off automatically, the experimenter
orally asked the subjects to respond verbally
to the following questions.
Q1. I prefer the night light on the baseboard
more than the other lighting system framing
the door. (5-point scale; 1 ¼ strongly disagree,
5 ¼ strongly agree)
1
2 3 4 5
Q2a. I like the on motion sensor feature of the
lighting system framing the door.
118 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
Figure 6 (a) Scenario 1: High illuminance from the novel lighting system with a view of the bathroom door (1 lx at the
eye, 6.5 W); (b) Scenario 2: Low illuminance from the novel lighting system with a view of the bathroom door (0.1 lx at
the eye, 0.65 W); (c) Scenario 3: Moderate illuminance from the novel lighting system with, no view of the bathroom
door (0.3 lx at the eye, 6.5 W); (d) Scenario 4: Very low illuminance from the novel lighting system with no view of the
bathroom door (0.05 lx at the eye, 0.65 W); (e) Scenario 5: Conventional night light with a view of the bathroom door
(0.1 lx at the eye, 7 W)
Night lighting system for postural control 119
Lighting Res. Technol. 2008; 40: 111–126
(5-point scale; 1 ¼ strongly disagree,
5 ¼ strongly agree)
1
2 3 4 5
Why? ____ (for any additional comments)
___________________
Q2b. I like the off motion sensor feature of
the lighting system framing the door.
(5-point scale; 1 ¼ strongly disagree,
5 ¼ strongly agree)
1
2 3 4 5
Why? ____ (for any additional comments)
___________________
Subjects were next asked to comment on
how well they could discern several objects on
the floor posing as potential trip or fall
hazards under five lighting scenarios
(Figures 6(a)–(e)). Three low-contrast objects
(an electrical cord, a pair of slippers, and a
copper penny) were placed on the carpet
between the bed and bathroom. Immediately
after experiencing each of the lighting scenar-
ios, presented to different subjects in a
random order, subjects were asked to
answer the following question.
Q3: I can easily see all three hazards on the
floor (5-point scale, strongly disagree to
strongly agree).
1
2 3 4 5
Finally, to close the session, the subjects
were asked the following question.
Q4: Compared to the conventional night light
on the baseboard, I like the lighting system
framing the door with the motion sensor
feature. (5-point scale; 1 ¼ strongly disagree,
5 ¼ strongly agree)
1
2 3 4 5
Subjects were then escorted out of the
bedroom mock-up, and back to the desk and
chair to receive a gift card as a thank you for
participating in the experiment.
3.2. Results
Figure 7 shows the mean responses to
questions Q1, Q2a, Q2b and Q4 comparing
lighting Scenario 1 (the novel lighting system
framing the door and providing 1 lx at the
eye, 6.5 W) to lighting Scenario 5 (the con-
ventional wall-plug night light providing
0.1 lx at the eye, 7 W). One-sample t-tests
were conducted to evaluate subjects’ replies to
each of the questions Q1, Q2a, Q2b and Q4.
For each t-test, it was assumed that if there
were no real difference between Scenarios 1
and 5, the mean response from the subjects
would be equal to 3 (neither agree nor
disagree). Thus, sample means statistically
greater than or less than 3 would indicate
reliable differences between the responses to
the two lighting scenarios.
In general, subjects preferred the novel
lighting system framing the door to the
conventional wall-plug night light. However,
there was no statistically reliable difference
between lighting Scenarios 1 and 5 for Q1,
which was answered prior to experiencing all
of the other lighting scenarios. Interestingly,
Q4, asking essentially the same question, but
asked of subjects after experiencing all of the
other lighting scenarios, showed a statistically
120 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
reliable preference (P50.05) for the novel
lighting system over the conventional wall-
plug night light. Insight into this change in
subjects’ attitudes toward the two lighting
systems may be revealed in the answers to
Q2a and Q2b. Answers to these two questions
clearly demonstrated a very positive response
to the ‘on’ and ‘off ’ motion sensor feature of
the novel lighting system. Responses to both
of these questions using one-sample t-tests
were statistically different than chance (more
positive toward the motion sensor option)
with a probability 50.001.
In general, older subjects saw objects on
the floor better as illuminance increased
(0.1 lx to 1.0 lx); however, a one-way analysis
of variance (ANOVA) did not reveal any
statistically reliable differences among the
lighting scenarios. Figure 8 shows the mean
responses to Q3 for each lighting scenario.
4. Discussion
The findings presented here indicate that a
novel night lighting system that provides
visual cues has a potential to positively
impact postural orientation and stability in
older subjects and, after being placed in a
simulated living environment with a motion
sensor feature, is well accepted by seniors.
Results from the first experiment showed that
tilting the door frame induces older subjects
to change the weight distribution on two feet.
By tilting the door frame to the left, they
leaned to the left and by tilting the door frame
to the right, they leaned to the right. These
visual cues also affected SV during the initial
phase of the STS test. Initially, the tilted door
frame makes older subjects less stable, but
toward the end of the trial, there is little if any
difference in SV when the door frame was
tilted or when it was positioned consistent
with proprioceptive and vestibular informa-
tion. Older subjects are particularly affected
by visual cues during the initial phase of the
transition from sitting to standing, but by the
end of the trial, they appear to rely more on
proprioceptive and/or vestibular cues for
maintaining an erect standing position.
In addition to improving postural stability
and control, results from the subjective
evaluation showed that the novel night
lighting system was preferred over the con-
ventional night lights, probably because of
the motion-sensor feature. The results also
suggest that the novel night lighting system
providing 1 lx at the subjects’ eyes while
seated on the edge of the bed allowed subjects
to see objects on the floor better than the
Q1 Q2a Q2b Q4
Strongly
agree
Agree
Neutral
Disagree
Strongly
disagree
Figure 7 Responses to Q1, Q2a, Q2b and Q4
Night lighting system for postural control 121
Lighting Res. Technol. 2008; 40: 111–126
conventional wall-mounted night light provid-
ing 0.1 lx. At the same illuminance at the eye
(but less power requirement), subjects could
see the objects on the floor slightly better with
the novel night lighting, but this difference
was not statistically significant. Regardless,
the novel night lighting can either deliver the
same amount of light than conventional night
lights for less power or more light (and better
light distribution) for the same electrical power
during operation. Since the cost of owning
a lighting system is usually dominated by
energy costs, it is deemed reasonable to com-
pare the two lighting systems in a simulated
living environment based on connected load. It
should be noted, however, that the novel night
lighting system was controlled by a motion
sensor so even for the same operational power,
the total energy savings would be much
greater. Since this novel night lighting system
is not a commercially available product, it was
not possible to make a satisfactory life-cycle
cost analysis. It was estimated that the
operating cost of the conventional night light
for a period of 6 years would be about $24.00
(energy use and lamp replacement). The
assumption was that the 7W night light
would be on for 8 h per night during 2190
days (or 6 years), resulting in an energy use of
approximately 124 kWh. The energy cost
would be estimated at $12.00, assuming the
cost of the kWh is 10 cents. We also assumed
that lamp replacement at every 750 h would
cost $0.50 cents each, adding about $12.00 to
the total operating cost of the conventional
system. Given the low maintenance costs and
the low energy requirements of the novel night
lighting system, it was estimated that the novel
night lighting system could cost approximately
$26.00 for a 6-year break-even cost, compared
to a $2.00 initial price for the conventional
night light.
6. Conclusions
Falling and sleep disruption are perhaps the
largest health-related problems associated
with senior living facilities.
1
Falls often
occur as seniors transition from a supine or
seated position to standing. Although turning
on bright lights at night is not the only cause
of sleep disruption in older adults, room lights
can certainly aggravate the problem.
16
High
1.0 lx 0.1 lx 0.3 lx 0.05 lx 0.1 lx
illuminance,
with view
Strongly
agree
Agree
Neutral
Disagree
Strongly
disagree
Low
illuminance,
with view
High
illuminance,
no view
Low
illuminance,
no view
Night
light
Figure 8 Older subjects’ responses to Q3
122 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
Further, because older people take longer to
dark-adapt, recovery from bright light expo-
sure at night can also increase the risk for falls
if they turn the room lights off
before reaching bed. The findings of this
study indicate that visual spatial cues strongly
affect postural orientation and stability.
These findings also demonstrate that a night
lighting system that uses occupancy sensors
and provides visual cues is well accepted by
older adults and can be an energy-efficient,
high-benefit solution. It is important to note,
however, that the present studies were not
performed in a real-life situation, where
subjects are awaking from sleep or where
other distractions, such as emergency lighting
and lighting for night-time staff, are present.
The results of this study strongly suggest that
the benefits of the novel night lighting system
would be even greater in field applications.
Moreover, these findings may be even more
important for persons with severe deficits in
the vestibular or the proprioceptive systems,
or with patients who have had a stroke,
because they have been shown to rely on
visual cues more often than normal persons.
Naturally, however, further and more exten-
sive investigations are very much needed to
solidify any generalisations about the utility
and acceptance of this novel night lighting
system, particularly with those having serious
disabilities, such as vertigo or dementia.
Acknowledgements
The authors would like to express their
gratitude to Jerry Mix and Dorene Maniccia
from Watt Stopper/Legrand, as well as Mark
Makwinski from Wiremold/Legrand who
supported this study. Shawn Jiang from
EquiTest, Dennis Guyon, Marylou
Nickleson, Huiying Wu, and Robert Wolsey
from the LRC, Audrey Hummer, Krystal
Flatebo and Jessica Knapp from The Sage
Colleges are gratefully acknowledged for the
technical assistance.
References
1 Falls and hip fractures among older adults.
In: Center for Disease Control (CDC).
Retrived 1 November 2006. from http://
www.cdc.gov/ncipc/factsheets/falls.htm
2 Shumway-Cook A, Woollacott M. Motor
Control: Theory and Practical Applications.
2001. 2nd edn, Baltimore, Lippincott Williams
and Wilkins.
3 Black A, Wood J. Vision and falls. Clin.
Experim. Opt. 2005; 88: 212–22.
4 Paulus WM, Straube A, Brandt T. Visual
stabilization of posture: physiological stimulus
characteristics and clinical aspects. Brain 1984;
107(pt4): 1143–63.
5 Turano K, Rubin G, Hedman S, et al. Visual
stabilization of posture in the elderly: fallers vs.
nonfallers. Optom. Vis. Sci. 1994; 71(12):
761–9.
6 Brooke-Wavell K, Perrett L, Howarth P,
Haslam R. Influence of the visual environment
on the postural stability in healthy older
women. Gerontology 2002; 48(5): 293–7.
7 Figueiro M. Lighting the way: a key to
independence. Guidelines for Designing
Lighting for Older Adults. 2001. Retrived 1
November 2006, from http://www.lrc.rpi.edu/
programs/lightHealth/AARP/index.asp
8 Pyykko I, Jantti P, Aalto H. Postural control
in elderly subjects. Age Ageing 1990; 19(3):
215–21.
9 Park J, Tang Y, Lopez I, et al. Age-related
changes in the number of neurons in the
human vestibular ganglion. J. Comp. Neurol.
2001; 431(4): 437–43.
10 Skinner H, Barrack R, Cook S. Age-related
decline in proprioception. Clin. Orthop. Relat.
Res. 1984; April (184): 208–11.
11 Lord SR, Clark RD, Webster IW. Visual
acuity and contrast sensitivity in relation to
falls in an elderly population. Age Ageing 1991;
20(3): 175–81.
12 Paulus WM, Straube A, Brandt T. Visual
postural performance after loss of somatosen-
sory and vestibular function. J. Neurol.
Neurosurg. Psych. 1987; 50(11): 1542–45.
Night lighting system for postural control 123
Lighting Res. Technol. 2008; 40: 111–126
13 Turano KA, Dagnelie G, Herdman SJ. Visual
stabilization of posture in persons with central
visual field loss. Invest. Ophthalmol. Vis. Sci.
1996; 37(8): 1483–91.
14 Anand V, Buckley JG, Scally A, et al.
Postural stability changes in the elderly
with cataract simulation and refractive blur.
Invest. Ophthalmol. Vis. Sci. 2003; 44(11):
4670–75.
15 Lee HK, Scudds RJ. Comparison of balance in
older people with and without visual impair-
ment. Age Ageing 2003; 32(6): 643–9.
16 Figueiro M, Rea M. New research in the light
and health field is expanding the possibilities for
LED lighting in healthcare environments: CIE
Midterm Meeting Conference Proceedings.
Leon, Spain, 2005.
17 McGuigan F. Experimental Psychology:
Methods of Research. Upper Saddle River, NJ,
Prentice Hall, 1997.
18 Sahni M, Lowenthal D, Meuleman J.
A clinical, physiology and pharmacology
evaluation of orthostatic hypotension in
the elderly. Int. Urol. Nephrol. 2005; 37(3):
669–74.
Discussion
Comment 1:
P T Stone (47 Loughborough Road, Quorn,
Leicestershire, LE12 8DU)
This excellent experiment has provided a most
insightful analysis of the postural behaviour
of elderly subjects when in transition from
lying down to standing up and has yielded
moderate support for the validity of a novel
design aid. However, for a crucial evaluation
of this system two other conditions need to be
examined through user acceptance trials. The
first of these is to test elderly subjects in the
real condition of having been asleep and just
woken up. As the authors state ‘older subjects
are particularly affected by visual cues during
the initial phase of transition from sitting to
standing’ so from waking to sitting up is
obviously an important moment for recogni-
tion of environmental cues In the work
reported here only ‘healthy’ subjects, who
rested for 10 min but were fully awake, were
employed. The second condition is concerned
with the ability of people having visual
disabilities to see the system adequately.
There is a strong correlation between aging
and low vision and such factors as loss of
contrast sensitivity and lenticular light scat-
tering and reduced visual fields might alter the
visibility of the lighting system.
Another consideration regarding postural
orientation is the possibility of autokinetic
movement. It has been shown experimentally
that the removal of the spatial framework
that occurs in a dark room can cause a source
of light to appear to drift, either to left or
right. There are wide individual differences
among subjects in the extent of this drift but
it is by no means a rare experience. This effect
could apply to people with diminished vision,
the framework of lights appearing to move or
float and thus causing apprehension and
uncertainty about an individual’s balance
and orientation. This point could be pursued
during a user acceptance trial.
Comment 2:
E Noell-Waggoner (Center of Design for an
Aging Society. 9027 NW Bartholomew Drive,
Portland, OR 97229, United States of
America)
For the 1.45 million people living in nursing
homes in the United States, any effort to
improve lighting to meet their needs, rather
than just the needs of the staff, will certainly
be a welcome change. The common complaint
of nursing home residents is not enough light
during the day and too much light at night.
Currently there are no federal requirements
for lighting, leaving each state to develop
their own. There is great variation among
state lighting requirements; some are based
124 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126
on the Illuminating Engineering Society of
North America, RP-28 Lighting and the
Visual Environment for Senior Living, while
others are vague leaving the decision up to the
administrator.
The novel lighting system has two very
important features, activation by a motion
sensor and defining the location of the bath-
room door by light. Fifty percent of people
living in nursing homes and assisted living
facilities are 85 years and older. I think the
subjects should be more closely matched in age
to that of the residents in order to determine
whether or not this system would improve
postural control and stability in this age group.
Future studies should be conducted in facilities
during the night, since the increased number of
night-time falls may be associated with sleep
medication and anti-depressants, rather than
the issues addressed in this study.
Regarding the methodology, I would ques-
tion whether or not the postural assessment
procedure used was appropriate for residents
of care facilities. Placing the feet parallel
may be possible for strong healthy people.
However, due to lower extremity impairment,
many older people place one foot in front of
the other and use their arms to push themselves
up to a standing position. If space allows, they
will place one foot under the bed using their
leg to push up to a standing position.
Although the comparison of energy use for
the two systems is interesting, the initial cost
of installation for a lighting system is a much
greater factor for care facilities. Development
of an inexpensive, easy to install lighting
system will insure greater acceptability by the
owner/manages of these facilities. Those who
might benefit most from a night lighting
systems are those living in private homes.
Currently, only 6.5 percent of persons age
65 years or older live in care facilities. Also,
the number of older people choosing to age-
in-place at home is increasing. These people
have the will and the means to address their
personal needs.
Reply to comments by MG Figueiro,
L Gras, R Qi, P Rizzo, M Rea and MS Rea
We would like to thank Mr Stone and Mrs
Noell-Waggoner for their constructive com-
ments. We agree with both discussants that
the findings presented here can be expanded
to include testing in other situations as well
as to include different populations, such as
adults older than 85 years and those with low
vision. We hope to be able to install these
lights in real-life applications and determine
whether their long-term use can help reduce
falls risk in older people living at home and
in more controlled environments. We are
currently expanding our research to include
patients who have had a stroke. We are
investigating whether we can use the tilted
lights to change midline orientation of
patients who have had a stroke and promote
the use of their involved side.
1
Regarding
Mr Stone’s concerns about autokinetic move-
ments, we agree that there is a small possibil-
ity that this may happen, but because the
novel nightlights are linear light sources that
provide enough light in the space to reveal the
spatial framework of the room, and because
these nightlights are designed to reinforce
the spatial framework by delineating the door
frame rather than remove any spatial frame-
work from the space, we believe the possibil-
ity of autokinetic movement is very much
reduced with this night lighting system. With
regard to Mrs Noell-Waggoner’s question
about the postural assessment procedure, it
is important to point out that the person’s feet
did not need to be parallel, only the force
plate was parallel. The subjects had ample
room for various foot placements on the force
plate and none of the subjects required the use
of their upper extremities to push off the
surface for the testing. Moreover, we used
standard testing procedures to measure sway
velocity so we believe that the assessment
procedure was also relevant to residents of
care facilities. Finally, we agree with
Night lighting system for postural control
125
Lighting Res. Technol. 2008; 40: 111–126
Mrs Noell-Waggoner that the initial cost of
the product may still be a barrier for its
widespread implementation, but we hope
that, as these products are manufactured in
larger scale, the final cost will be much
reduced. It is also important to note that
the cost of light emitting diodes has been
dropping sharply in the last few years, which
will certainly reduce the overall cost of the
product. Moreover, less expensive light
sources, such as linear incandescent lamps
can be used, though the life of the product
would be reduced. The ease of installation is
not a barrier because the product can be
easily installed around the door frame and
with the wall plugs does not need additional
wiring.
Reference
1 Gras L, Flatebo K, Knapp J, Qi N, Figueiro
M. The use of a lighted tilted frame to assist
in midline orientation for subjects who had a
stroke. Abstract presented at the International
Conference of Aging, Disability and
Independence (ICADI). February 2008,
St. Petersburg, FL.
126 MG Figueiro et al.
Lighting Res. Technol. 2008; 40: 111–126