C L I N I C A L
A N D
E X P E R I M E N T A L
OPTOMETRY
cxo_702
510..514
RESEARCH PAPER
Effect of light-emitting diode colour temperature on magnifier
reading performance of the visually impaired
Clin Exp Optom 2012; 95: 510–514
James S Wolffsohn* PhD MBA
MCOptom FAAO
Eshmael Palmer* BSc
Martin Rubinstein*† PhD BSc MCOptom
FAAO
Frank Eperjesi* PhD MBA MCOptom
FAAO
* Aston University, Life and Health Sciences,
Ophthalmic Research Group, Birmingham, United
Kingdom
†
Department of Ophthalmology, Nottingham
University Hospital, Nottingham, United Kingdom
E-mail: [email protected]
Submitted: 7 August 2011
Revised: 10 October 2011
Accepted for publication: 20 October
2011
DOI:10.1111/j.1444-0938.2011.00702.x
Background: As light-emitting diodes become more common as the light source for low
vision aids, the effect of illumination colour temperature on magnifier reading performance was investigated.
Methods: Reading ability (maximum reading speed, critical print size, threshold near
visual acuity) using Radner charts and subjective preference was assessed for 107 participants with visual impairment using three stand magnifiers with light emitting diode
illumination colour temperatures of 2,700 K, 4,500 K and 6,000 K. The results were
compared with distance visual acuity, prescribed magnification, age and the primary
cause of visual impairment.
Results: Reading speed, critical print size and near visual acuity were unaffected by
illumination colour temperature (p > 0.05). Reading metrics decreased with worsening
acuity and higher levels of prescribed magnification but acuity was unaffected by age.
Each colour temperature was preferred and disliked by a similar number of patients and
was unrelated to distance visual acuity, prescribed magnification and age (p > 0.05).
Patients had better near acuity (p = 0.002), critical print size (p = 0.034) and maximum
reading speed (p < 0.001), and the improvement in near from distance acuity was greater
(p = 0.004) with their preferred rather than least-liked colour temperature illumination.
Conclusion: A range of colour temperature illuminations should be offered to all
visually impaired individuals prescribed with an optical magnifier for near tasks to
optimise subjective and objective benefits.
Key words: colour temperature, illumination, light-emitting diode, low vision aid, visually impaired
Illumination is critical to optimising visual
function, particularly in the ageing eye,
and even more so for people with ocular
disease.1 Lighting can improve contrast
sensitivity, but can also cause glare.2 Traditionally, most low vision aids have been
illuminated by tungsten bulbs, as they are
relatively low cost and bright. When compared with light-emitting diodes (LED),
they are relatively inefficient in their use of
power, not as bright and have a much
shorter life span.3 Light-emitting diode
technology, which is available in many
recent magnification devices, overcomes
these drawbacks and is available in a range
of (correlated) colour temperatures;
however, the optimal colour temperature
and the optimal method of prescribing
this clinically are yet to be determined.
Light-emitting diodes are based on semiconductor diodes, which, when switched
on, cause electroluminescence. The
Clinical and Experimental Optometry 95.5 September 2012
510
colour of the light corresponds to the
semiconductor’s energy gaps and thus
the energy of photons released by the electrons in the device. The resulting blue
colour is converted to a white output using
a yellow fluorescent phosphor.
It is controversial whether certain
colours and lamps can be used to improve
reading performance in people with low
vision.4–6 Reading is the visual skill of
extracting and understanding written
© 2012 The Authors
Clinical and Experimental Optometry © 2012 Optometrists Association Australia
LVA and LED colour temperature Wolffsohn, Palmer, Rubinstein and Eperjesi
Radiance (watts/sr/m 2 )
0.04
2700 K
4500 K
6000 K
0.03
0.02
0.01
0.00
400
500
600
700
Wavelength (nm)
Figure 1. Spectral radiance of the 2,700 K, 4,500 K and
6,000 K colour temperature stand magnifier light-emitting
diodes
information from text and is an essential
part of day-to-day life. For many, the loss of
the ability to read is synonymous with a
loss of independence, decreased enjoyment of life and has an obvious influence
on their ability to perform everyday tasks.7
Spitzberg and Goodrich8 evaluated a
range of new stand magnifiers, which
included improved illumination, and
found a subjective preference but no
change in reading speed. A retrospective
analysis of patients attending the low
vision service at Moorfields Eye Hospital,
London, between 1973 and 2003, showed
a rapid increase in the percentage of illuminated hand magnifiers being dispensed.9 The increase was from 10 per
cent in 1993 and 30 per cent in 1998 to 80
per cent of hand magnifiers in 2003.
Although the percentage of stand magnifiers remained stable over this period, the
proportion with internal illumination
increased from approximately 50 to 80 per
cent. In the final sample taken in 2003,
LED-illuminated magnifiers were noted
for the first time, constituting approximately 80 per cent of the illuminated
hand magnifiers and 30 per cent of the
illuminated stand magnifiers prescribed.
The aim of the present study was to
assess the effect of varying colour tempera-
ture of LED-illuminated stand magnifiers
on near visual function in the visually
impaired.
METHODS
Participants were recruited from the
outpatients eye department at Queens
Medical Centre, University of Nottingham, and enrolled in the study following
informed consent after explanation of the
nature and possible consequences of
the study. The research was approved by
the local research ethics committee and
adhered to the tenets of the Declaration of
Helsinki. Inclusion criteria were ophthalmological confirmation of visual impairment and a visual acuity of 6/9 Snellen or
worse in the better eye, ability to read
English fluently and the clinical need for a
magnifier to perform fine reading tasks
(near visual acuity of N8 or worse at the
individual’s habitual working distance). A
total of 107 participants (61 women and
46 men) were assessed, with an age range
of 37 to 96 years (median 82 years). The
principal causes of visual impairment were
age-related macular degeneration (AMD;
91), diabetic retinopathy (8), retinal
detachment (2) and other ocular conditions (6). Visual acuities ranged from 6/9
© 2012 The Authors
Clinical and Experimental Optometry © 2012 Optometrists Association Australia
to 0.5/60 (logMAR equivalent: +0.20 to
+2.10 logMAR). Eighty-five of the participants were experienced users of optical
magnifiers, with the remainder having
been prescribed magnifiers in the low
vision clinic assessment that immediately
preceded the data gathering session for
the present study.
Magnifier heads (A Schweizer Gmbh,
Forchheim, Germany) ranged from
+8.00 D to +28.00 D in 4.00 D steps with
additional powers of +39.00 D, +48.00 D
and +56.00 D available. The standmagnified heads minimise external light
(which was constant for all conditions)
from illuminating the task and are of
white plastic to maximise even illumination. The colour temperatures, trialled in
counterbalanced order, were nominally
2,700 K, 4,500 K and 6,000 K. The illuminance of the battery-powered handle LED
was adjusted so that illuminance in the
visible spectrum was within 50 lux of
1800 lux, as measured with a CA 810
Luxmeter (Chauvin Arnoux, Slough, UK).
A PR-650 SpectraScan SpectraColorimeter
using SpectraWin v2.15a software (Photo
Research Inc, Chatsworth, CA, USA) was
used to generate spectral outputs for each
LED imaged directly along their primary
axes (Figure 1). Subjects ranked each of
the three colour temperature illuminations (best, in between or worst performance) on ease of reading, comfort and
overall preference and any resulting comments were recorded.
Each participant’s refractive correction
was optimised as part of their low vision
assessment and their logMAR visual acuity
measured for each eye. The need for
optical magnification for fine print
reading was confirmed during the low
vision assessment and the power of
the stand magnifier head used to assess
reading speed selected to match the magnification of the device prescribed during
the low vision assessment.10 In combination with the stand magnifier, participants
used their preferred spectacle correction,
which was distance single vision for nine,
near single vision for 21, bifocals for 42,
varifocals for eight and unaided for 27 participants. The choice of spectacle lens type
will have affected the effective magnifica-
Clinical and Experimental Optometry 95.5 September 2012
511
LVA and LED colour temperature Wolffsohn, Palmer, Rubinstein and Eperjesi
2,700 K
4,500 K
6,000 K
Maximum reading speed (words per
minute)
104⫾ 50
104⫾ 50
104⫾ 52
logRAD score
0.29⫾ 0.30
0.29⫾ 0.28
0.30⫾ 0.30
Critical print size (logMAR)
0.50⫾ 0.25
0.49⫾ 0.25
0.49⫾ 0.26
-0.31⫾ 0.26
-0.30⫾ 0.27
-0.29⫾ 0.30
logMAR to logRAD difference
Table 1. Reading metrics with colour temperature of the illuminated stand magnifier
(n = 107)
tion of the magnifier but this was consistent between the colour temperatures
assessed for each individual.
Reading speed was assessed with the
Radner reading charts, which have been
shown to provide reliable measures of
reading performance.11,12 They consist of
paragraphs sized from 1.2 to -0.2 logMAR
in 0.1 logMAR steps with the same number
of words (13), word length, number of
syllables, position of words, lexical difficulty and syntactical complexity. All
subjects were allowed to use whichever
method of viewing (monocular or binocular) and eye-to-magnifier lens distance
that was most comfortable for them, while
maintaining the magnifier head flat on
the reading chart. Each participant’s
viewing style was monitored by observation for consistency across the different
types of LED. The participants were asked
to read each chart sentence aloud, as precisely and quickly as possible, without
repeating words or correcting errors. The
time taken to read the sentence was
recorded and any mistakes, including
words that were omitted or misread and
the number of syllables in that word, were
noted. As outlined in the Radner chart
manual, the patient’s reading acuity was
calculated as:
logRAD score
= logRAD for lowest line read +
(0.005 × syllables of incorrectly
read words ).
The maximum reading speed was an
average of all reading speeds before the
critical print size (CPS), which was calcu-
lated as the visual acuity at which the
reading speed reduced by more than two
standard deviations below the maximum
reading speed. The logMAR to logRAD
difference was calculated by subtracting
the distance visual acuity of the best eye
from the calculated logRAD score.
Maximum reading speed, CPS and
logRAD score with each of the colour temperatures was compared with repeated
measures analysis of variance and correlated with distance visual acuity, magnifier
power and age. The difference between experienced compared with nonexperienced optical magnifier users and
AMD compared with subjects with other
causes of visual impairment were compared with a one-way analysis of variance.
Chi-squared was used to assess how the
clinical performance was related to subjective preference in colour temperature illumination. With 107 subjects and using the
repeatability of the Radner chart,12 the
study was 80 per cent powered to detect a
0.6 words per minute (wpm) change in
maximum reading speed, a 0.003 change
in logRAD score (which also applies to the
logMAR to logRAD difference) and a 0.18
logMAR change in CPS.13
RESULTS
The corrected maximum reading speed
(F = 0.037, p = 0.964), logRAD score (F =
0.413, p = 0.662), CPS (F = 0.024, p =
0.977) and the logMAR to logRAD difference (F = 0.662, p = 0.517) did not differ
with colour temperature (Table 1). There
was no difference between new and expe-
Clinical and Experimental Optometry 95.5 September 2012
512
rienced magnifier users in corrected
maximum reading speed (F = 0.046, p =
0.831), logRAD score (F = 0.449, p =
0.773), CPS (F = 1.657, p = 0.196) and
logMAR to logRAD difference (F = 0.442,
p = 0.644). There was an interaction
between the maximum reading speed and
illumination colour temperature, with
experienced users reading faster with the
6,000 K and slower with the 2,700 K colour
temperature illumination than new users
(F = 4.884, p = 0.009).
Maximum reading speed, logRAD score
and CPS achieved with the magnifiers
were correlated to distance visual acuity
and prescribed magnification with all
three colour temperature illuminations
but were unrelated to age other than
maximum reading speed, which decreased with increasing age (Table 2). The
logMAR to logRAD difference was also
correlated with distance visual acuity and
prescribed magnification but was only significantly negatively correlated with age
when the 4,500 K colour temperature illumination was used.
The visually impaired ranked ease of
use, visual comfort and overall performance in the same order for each colour
temperature in all cases. The preference
for colour temperature was evenly spread
with 36 participants preferring 2,700 K,
34 preferring 4,500 K and 36 preferring
6,000 K colour temperature illumination.
The 2,700 K colour temperature was rated
as the worst by 45 subjects compared with
28 subjects with 4,500 K and 33 subjects
with 6,000 K, although this difference was
not statistically significant (Chi-squared =
4.321, p = 0.115). Colour temperature illumination preference and dislike were
unrelated to distance visual acuity (F =
0.429, p = 0.621; F = 1.176, p = 0.313,
respectively), prescribed magnification (F
= 0.120, p = 0.887; F = 2.139, p = 0.123)
and age (F = 0.094, p = 0.910; F = 0.396, p
= 0.674). Patients had better near acuity
(0.28 ⫾ 0.53 logRAD versus 0.34 ⫾ 0.61
logRAD, p = 0.002), CPS (0.47 ⫾ 0.45
logMAR versus 0.52 ⫾ 0.53 logMAR, p =
0.034), maximum reading speed (108 ⫾
100 wpm versus 99 ⫾ 96 wpm, p < 0.001)
and logMAR to logRAD difference (-0.31
⫾ 0.53 logMAR versus -0.26 ⫾ 0.61
© 2012 The Authors
Clinical and Experimental Optometry © 2012 Optometrists Association Australia
LVA and LED colour temperature Wolffsohn, Palmer, Rubinstein and Eperjesi
Distance visual acuity
Prescribed magnification
Age
Maximum reading speed
r = -0.558 to -0.602
r = -0.473 to -0.543
r = -0.209 to -0.252
p < 0.001
p < 0.001
p < 0.05
logRAD score
r = 0.536 to 0.611
r = 0.301 to 0.404
r = -0.067 to 0.048
p < 0.001
p < 0.005
p > 0.45
Critical print size
r = 0.476 to 0.546
r = 0.298 to 0.348
r = -0.048 to 0.101
p < 0.001
p ⱕ 0.005
p > 0.30
logMAR to logRAD difference
r = -0.484 to -0.539
r = -0.223 to 0.299
r = -0.128 to -0.250
p < 0.001
p ⱕ 0.005
4,500°K p < 0.01
Table 2. Correlation between reading metrics and distance visual acuity, prescribed magnification
and age (n = 107)
logMAR, p = 0.004) with their preferred
rather than their least-liked colour temperature illumination.
Participants with AMD were older than
the other subjects with visual impairment
(82.0 ⫾ 7.0 years versus 64.4 ⫾ 16.1 years)
but there was no difference in reading performance (p > 0.05) or subjective preference (Chi-squared p > 0.20) with any of
the colour temperature illuminations.
DISCUSSION
Most measures of reading ability did not
differ with colour temperature illumination of the magnifier. In general, there
was no difference between new and experienced low vision aid users in reading
ability or subjective preference to the
colour temperature of the illumination,
suggesting that it is unlikely that users
adapt to the illumination colour temperature of their magnifier. Despite reading
size metrics being unaffected by age, the
reading speed was greater and both the
smallest size read and the CPS were better
in the subjects who had better distance
visual acuity and lower amounts of prescribed magnification. The distance visual
acuity and the prescribed magnification
are obviously related (0.577, p < 0.001).
This highlights that although enlargement
using stand magnifiers can enhance visual
acuity, there are significant drawbacks
such as a more restricted working distance, optical aberrations from the high-
powered lenses and a restricted field of
vision, which impact on the achieved
reading ability.14 The logMAR to logRAD
difference was significantly associated with
the prescribed magnification as expected,
as optical enlargement should allow
smaller words to be read and hence drop
the logRAD score; however, the benefit
achieved (on average 0.3 logMAR or three
logMAR lines at 1.26 times enlargement
per line, hence 2.0 times) was lower than
the stated magnification (on average 3.8 ⫾
1.7 times). This will relate partly to the
restrictions of optical magnifiers already
described along with the individual’s
ocular pathology, which is known to affect
distance and near visual acuities to different degrees.15 There was no difference in
reading ability or subjective preference
between participants with AMD and those
with other causes of visual impairment in
the present study. This population, as with
the low vision population as a whole,
mainly had visual loss due to AMD and
hence the results can be confirmed only
for this population, although the results
seemed to be unaffected by the condition
causing low vision. In addition, reading on
the near chart involves lowercase letters
and word recognition, which results in a
poorer visual acuity than that measured at
the same distance with an upper case
chart.16 Age-related maculopathy is known
to affect colour discrimination17 but the
lack of correlation with visual acuity,
which can be taken as an indirect measure
© 2012 The Authors
Clinical and Experimental Optometry © 2012 Optometrists Association Australia
of the severity of the age-related maculopathy, suggests that an acquired colour
defect is unlikely to affect the LED colour
temperature preferences found in the
present study. The decrease in reading
speed with age agrees with previous
studies.18,19
All colour temperature magnifier illuminations were equally liked or disliked,
with that preference projected onto all
aspects of the magnifiers, namely, ease of
use, visual comfort and overall performance. While the strength of preference
of one colour temperature illumination
over another was not directly quantified,
comments from patients indicate that only
seven did not perceive much difference
between them. The preference could not
be predicted from participant age, cause
or severity of visual impairment. Some
commented on differences in glare and
reflections, apparent intensity and contrast, apparent text size and letter doubling or jumbling with the different colour
temperatures. Hence, media opacities, to
which glare and reflections are generally
attributed, might influence subjective
preference for the colour temperature of
the illumination of the magnifier more
than the primary cause of visual impairment. Whatever the cause of these subjective differences, patients did read smaller
print and were faster with their preferred
rather than least-liked temperature illumination, despite the relatively modest
change in spectral radiance of the LED
Clinical and Experimental Optometry 95.5 September 2012
513
LVA and LED colour temperature Wolffsohn, Palmer, Rubinstein and Eperjesi
from 2,700 K to 6,000 K (Figure 1).
Although one must be wary of a placebo
effect influencing the results with studies
involving subjective rating, patients were
given no reason to believe one colour temperature should out-perform another.
The difference was an improvement of 19
per cent in near acuity, 10 per cent in CPS,
nine per cent in maximum reading speed
and 18 per cent in logMAR to logRAD
difference, and hence appears to be clinically as well as statistically significant.
Thus, magnifiers that offer a range of illumination colour temperatures will benefit
a larger number of patients. Therefore, it
should be considered best practice that a
range of illuminations of different colour
temperatures is offered to all visually
impaired individuals who are prescribed
an optical magnifier for near tasks to optimise their reading ability.
GRANTS AND FINANCIAL SUPPORT
This research was sponsored in part by
Optima Low Vision Services Ltd and A
Schweizer GmbH.
REFERENCES
10. Wolffsohn JS, Eperjesi F. Predicting prescribed
magnification. Ophthalmic Physiol Opt 2004; 24:
334–338.
11. Burggraaff MC, van Nispen RMA, Hoek S, Knol
DL, van Rens GHMB. Feasibility of the Radner
Reading Charts in low-vision patients. Graefes Arch
Clin Exp Ophthalmol 2010; 248: 1631–1637.
12. Stifter E, Konig F, Lang T, Bauer P, RichterMuksch S, Velikay-Parel M, Radner W. Reliability
of a standardized reading chart system: variance
component analysis, test-retest and inter-chart reliability. Graefes Arch Clin Exp Ophthalmol 2004; 242:
31–39.
13. Faul F, Erdfelder E, Lang AG, Buchner A.
G*Power 3: a flexible statistical power analysis
program for the social, behavioral, and biomedical
sciences. Behav Res Methods 2007; 39: 175–191.
14. Wolffsohn JS, Eperjesi F. The effect of relative
distance enlargement on visual acuity in the visually impaired. Clin Exp Optom 2005; 88: 97–102.
15. Nguyen NX, Trauzettel-Klosinski S. Effectiveness
of magnifying low vision aids in patients
with age-related macular degeneration. NeuroOphthalmology 2009; 33: 115–119.
16. Gupta N, Wolffsohn JS, Naroo SA. Comparison of
near visual acuity and reading metrics in presbyopia correction. J Cataract Refract Surg 2009; 35:
1401–1409.
17. Arden GB, Wolf JE. Colour vision testing as an aid
to diagnosis and management of age related
maculopathy. Br J Ophthalmol 2004; 88: 1180–1185.
18. Legge GE, Ross JA, Isenberg LM, Lamay JM. Psychophysics of reading. XII. Clinical predictors of
low-vision reading speed. Invest Ophthalmol Vis Sci
1992: 33: 677–687.
19. Ahn SJ, Legge GE. Psychophysics of reading. XIII.
Predictors of magnifier-aided reading speed in low
vision. Vision Res 1995; 35: 1931–1938.
1. Cornelissen FW, Bootsma A, Kooijman AC.
Object perception by visually-impaired people
at different light levels. Vision Res 1995; 35: 161–
168.
2. Lindner H, Rinnert T, Behrens-Baumann W. Illumination conditions of visually impaired people
under private domestic circumstances—clinical
study on 91 patients. Klin Monatsbl Augenh 2001;
218: 774–781.
3. Jacob B. Lamps for improving the energy efficiency of domestic lighting. Lighting Res Tech 2009;
41: 219–228.
4. Eperjesi F, Fowler CW, Evans BJW. Do tinted
lenses or filters improve visual performance in low
vision? A review of the literature. Ophthalmic Physiol
Opt 2002; 22: 68–77.
5. Eperjesi F, Fowler CW, Evans BJW. Effect of light
filters on reading speed in normal and low vision
due to age-related macular degeneration. Ophthalmic Physiol Opt 2004; 24: 17–25.
6. Eperjesi F, Fowler CW, Evans BJW. The effects of
coloured light filter overlays on reading rates in
age-related macular degeneration. Acta Ophthalmol
Scand 2004; 82: 695–700.
7. Ryan EB, Anas AP, Beamer M, Bajorek S. Coping
with age-related vision loss in everyday reading
activities. Educ Gerontol 2003; 29: 37–54.
8. Spitzberg LA, Goodrich GL. New ergonomic stand
magnifiers. J Am Optom Assoc 1995; 66: 25–30.
9. Crossland MD, Silver JH. Thirty years in an urban
low vision clinic: changes in prescribing habits of
low vision practitioners. Optom Vis Sci 2005; 82:
617–622.
Clinical and Experimental Optometry 95.5 September 2012
514
© 2012 The Authors
Clinical and Experimental Optometry © 2012 Optometrists Association Australia