OPHTHA12065S200

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/360707009
Interventions for the Management of Computer Vision Syndrome: A

Systematic Review and Meta-analysis
Article in Ophthalmology · May 2022

DOI: 10.1016/j.ophtha.2022.05.009
CITATIONS READS
50 249
4 authors, including:
Sumeer Singh Laura Downie

University of Melbourne University of Melbourne
29 PUBLICATIONS 423 CITATIONS 160 PUBLICATIONS 4,730 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Laura Downie on 18 July 2022.
The user has requested enhancement of the downloaded file.

Interventions for the Management of
Computer Vision Syndrome
A Systematic Review and Meta-analysis
Sumeer Singh, MPhil, PhD,1 Myra B. McGuinness, MBiostat, PhD,2,3 Andrew J. Anderson, MSc, PhD,1,*
Laura E. Downie, BOptom, PhD1,*
Topic: To evaluate the efficacy and safety of interventions for treating eye strain related to computer use
relative to placebo or no treatment.
Clinical Relevance: Computer use is pervasive and often associated with eye strain, referred to as computer
vision syndrome (CVS). Currently, no clinical guidelines exist to help practitioners provide evidence-based advice
about CVS treatments, many of which are marketed directly to patients. This systematic review and meta-analysis
was designed to help inform best practice for eye care providers.
Methods: Eligible randomized controlled trials (RCTs) were identified in Ovid MEDLINE, Embase, the
Cochrane Central Register of Controlled Trials, and trial registries, searched from inception through November 23,
2021. Eligible studies were appraised for risk of bias and were synthesized. The certainty of the body of evidence
was judged using the Grading of Recommendations, Assessment, Development, and Evaluation system. Stan-
dardized mean differences (SMDs) were used when differently scaled measures were combined.
Results: Forty-five RCTs, involving 4497 participants, were included. Multifocal lenses did not improve visual
fatigue scores compared with single-vision lenses (3 RCTs; SMD, 0.11; 95% confidence interval [CI], e0.14 to
0.37; P ¼ 0.38). Visual fatigue symptoms were not reduced by blue-blocking spectacles (3 RCTs), with evidence
judged of low certainty. Relative to placebo, oral berry extract supplementation did not improve visual fatigue (7
RCTs; SMD, e0.27; 95% CI, e0.70 to 0.16; P ¼ 0.22) or dry eye symptoms (4 RCTs; SMD, e0.10; 95% CI, e0.54
to 0.33; P ¼ 0.65). Likewise, berry extract supplementation had no significant effects on critical flicker-fusion
frequency (CFF) or accommodative amplitude. Oral omega-3 supplementation for 45 days to 3 months
improved dry eye symptoms (2 RCTs; mean difference [MD], e3.36; 95% CI, e3.63 to e3.10 on an 18 unit scale;
P < 0.00001) relative to placebo. Oral carotenoid supplementation improved CFF (2 RCTs; MD, 1.55 Hz; 95% CI,
0.42 to 2.67 Hz; P ¼ 0.007) relative to placebo, although the clinical significance of this finding is unclear.
Discussion: We did not identify high-certainty evidence supporting the use of any of the therapies analyzed.
Low-certainty evidence suggested that oral omega-3 supplementation reduces dry eye symptoms in symp-
tomatic computer users. Ophthalmology 2022;-:1e24 ª 2022 by the American Academy of Ophthalmology. This
is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Supplemental material available at www.aaojournal.org.
Computer vision syndrome (CVS) describes a group of eye- questions. Segui et al7 developed and validated a
and vision-related problems associated with prolonged com- questionnaire that involved participants specifying the
puter use.1 Computer vision syndrome is reported to affect 75% frequency and intensity of 16 symptoms associated with
to 90% of computer users.2 Its global prevalence is estimated at computer-induced eye strain. A score of 6 or more has
60 million, with one million incident cases each year.3 been reported to be diagnostic of CVS7; this questionnaire
Common symptoms of CVS are eye strain, blurred vision, subsequently has been used widely in clinical studies to
eye dryness, ocular redness, headache, and may also include identify CVS.8e10
nonocular symptoms such as neck and shoulder pain.4 Established risk factors for CVS are extended periods of
Although these symptoms usually are temporary, they may computer use (more than 4 hours daily),11 reflections and glare
not resolve at the end of the working day.5 Given the typical on the computer screen from surrounding lighting,12 low
chronicity of computer use, many of the symptoms of CVS humidity (< 40%), and poor ergonomics with computer
are recurrent, prone to progression, or both.6 use.13 Ocular symptoms in computer users may be associated
In eye care practice, CVS is diagnosed primarily by with reduced blink rate, uncorrected refractive error,
surveying patients using a range of symptom-focused accommodation anomalies, or a combination thereof.14e17
ª 2022 by the American Academy of Ophthalmology https://doi.org/10.1016/j.ophtha.2022.05.009 1

This is an open access article under the CC BY-NC-ND license ISSN 0161-6420/22
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Published by Elsevier Inc.
REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

Ophthalmology Volume -, Number -, Month 2022
Thorud et al18 studied the association between extraocular Eligibility Criteria

symptoms (such as pain around the eye) with computer use
Eligible for inclusion were randomized controlled trials (RCTs) that
using electromyography and photoplethysmography and compared any intervention for managing signs or symptoms of CVS
found a weakness in the orbicularis oculi muscles to be in humans relative to an inactive control treatment, placebo treat-
linked with CVS symptoms. In addition, blue (short- ment, sham treatment, or no treatment. Eligible studies had recruited
wavelength visible) light emitted from computer screens has male or female participants of any age with a diagnosis of CVS, as
been hypothesized to cause eye strain,19 although this defined by the study authors. Published conference abstracts,
remains contentious given the lack of supporting evidence, noneEnglish-language studies, and non-human studies were
the lack of a compelling biological mechanism through excluded. Studies identified on clinical trial registries with sufficient
which blue light might cause eye strain directly, and the details on interventions and outcomes (in the absence of a published
relatively low level of blue light emission from electronic manuscript) were included. The outcomes from the most recent
devices.20 publication for each study were included. For analysis, the in-
terventions were grouped into the following categories: optical aids,
In terms of CVS management, a diversity of interventions complementary medicine and nutritional supplements, artificial
has been investigated. A common clinical approach involves tears, environmental modification, ergonomic adjustment, visual
recommending an optimal ergonomic setup at the computer hygiene (e.g., blinking exercises, advice to follow the 20-20-20
and advising computer users to follow the 20-20-20 rule, rule), binocular vision training, and other interventions. Comple-
which involves viewing an object 20 feet away for a total of mentary medicine and nutritional supplements were divided further
20 seconds every 20 minutes.21 Other interventions that have into 6 subcategories: oral berry extract supplements, polyunsaturated
been considered include optical interventions (i.e., fatty acid supplements, antioxidant supplements, traditional medi-
progressive addition spectacle lenses22 and blue light- cines, combination supplements, and other interventions. Details on
blocking spectacle lenses19), oral antioxidant and nutritional intervention group and relevant comparators are presented for each
supplements,23 oral omega-3 fatty acid supplements,24 study in Appendix S2 (available at www.aaojournal.org).
artificial tears,4 and traditional medicines.25 Optical aids,
such as progressive addition lenses, are claimed to reduce Search Methods
symptoms of CVS by providing an optimal refractive The following electronic databases were searched initially from
correction for intermediate and near working distances.22,26 inception through December 12, 2019, and were updated through
Blue light-blocking spectacle lenses have been promoted to November 23, 2021: Ovid MEDLINE, Embase, the Cochrane Central
reduce symptoms of eye strain by attenuating blue light Register of Controlled Trials, the United States National Institutes of
emitted from computer screens,27,28 despite the lack of a Health Clinical Trials Registry (www.ClinicalTrials.gov), and the
compelling mechanism of action. Oral omega-3 fatty acids World Health Organization International Clinical Trials
Registry Platform (https://www.who.int/clinical-trials-registry-plat-
may reduce ocular surface inflammation by modulating sys- form). Complete search strategies are provided in Appendix S3
temic cytokine production levels to reduce dry eye symptoms (available at www.aaojournal.org). Reference lists of included RCTs
in computer users.29 Artificial tears are considered to lubricate were screened for additional potentially relevant studies.
the ocular surface and to increase tear volume to reduce CVS
symptoms.21 The mechanism by which antioxidant and Selection of Studies
nutritional supplements might reduce eye strain symptoms
is unclear. Despite this, both interventions are advertised Citations retrieved from electronic databases were compiled into an
frequently as potentially useful treatments for CVS.30,31 EndNote library. After removal of duplicate citations, the library
was imported into Covidence (Veritas Health Innovation). Two
Many CVS interventions are marketed directly to patients,
reviewers (S.S. and either L.E.D., Ji-Hyun Lee, or Eve Makrai [the
and so eye care providers need to be able to provide latter two from the Department of Optometry and Vision Sciences,
evidence-based advice to patients enquiring about the effi- The University of Melbourne]) independently screened citation
cacy, or otherwise, of such interventions. Although CVS has titles and abstracts and decided whether they should be included,
been studied extensively and a wide variety of interventions excluded, or potentially included. Studies judged to be definitely or
exist, currently no clinical guidelines exist to inform best potentially eligible progressed to full-text screening. Two re-
practice. The aim of this systematic review was to identify, viewers (S.S. and either L.E.D., Ji-Hyun Lee, or Eve Makrai)
appraise, and synthesize clinical evidence relating to the ef- independently screened the full texts based on predefined eligibility
ficacy and safety of interventions for treating CVS, and criteria. For studies excluded at the full-text screening stage, the
thereby provide support to practitioners in providing primary reason for exclusion was documented. Any disagreements
between review authors were resolved by discussion or by
evidence-based advice to their patients.
consulting a third review author if required (L.E.D or A.J.A).
Methods Data Extraction and Management

This review was registered prospectively on the International Two reviewers (S.S. and either L.E.D., A.J.A., Ji-Hyun Lee, or Eve
Prospective Register of Systematic Reviews database (identifier, Makrai) independently extracted study data; any discrepancies
CRD42020164092), was conducted in accordance were resolved by discussion. Information extracted from each
with the principles in the Cochrane Handbook, and is reported eligible study included: study details (author, title, journal, year of
according to the Preferred Reporting Items for Systematic Reviews publication), methodology, participants, interventions, outcomes,
and Meta-Analyses statement (Appendix S1, available at and other details (e.g., sources of funding, conflicts of interest). For
www.aaojournal.org).32 studies in which missing outcome data were identified or additional

Singh et al
Interventions for Computer Vision Syndrome
information was required for statistical analysis, we attempted to than 3 RCTs were included, a fixed-effect model was used;
contact the study authors via e-mail for the information. If a otherwise, a random-effects model was adopted.42,43
response was not received within 4 weeks or if the trial authors Clinical and methodologic heterogeneity were assessed by
were unable to provide the information, the analysis was performed considering trial design, baseline participant characteristics, and
using details in the full-text publication. risk-of-bias judgements. Statistical heterogeneity was assessed
using the I2 statistic, which describes the percentage of total vari-
Risk of Bias Assessment ation across all studies. An I2 value of more than 50% or a P value
for Cochrane’s Q statistic of less than 0.05 was used to define
Risk of bias was assessed for eligible studies according to guidance significant statistical heterogeneity.44,45 For comparisons in which
in the Cochrane Handbook for Systematic Reviews of a meta-analysis was not possible (e.g., only 1 eligible study) or data
Interventions.33 The assessed risk of bias domains were: selection pooling was considered inappropriate (e.g., I2 value of > 50%), a
bias (random sequence generation and allocation concealment), descriptive summary of results was created.
performance bias (masking of study participants and personnel), When sufficient data were available, subgroup analyses were
detection bias (masking of outcome assessors), attrition bias performed for interventions with multiple subcategories (i.e.,
(incomplete outcome data), reporting bias (selective outcome complementary medicines and nutritional supplements), as previ-
reporting), and other sources of bias (such as funding sources ously defined. Because fewer than 10 trials were included for each
and conflicts of interest). Each study was judged to have low, of the intervention categories, funnel plots were not inspected
high, or unclear risk of bias in each domain. Two review authors visually to assess for potential small study effects.46
(S.S. and either L.E.D., A.J.A., Ji-Hyun Lee, or Eve Makrai)
independently performed the assessments, with any Summary of Findings Table
disagreements resolved by discussion.
The certainty of the body of evidence was assessed as high,
moderate, low, or very low for the primary and secondary out-
Data Synthesis and Analysis comes using the Grading of Recommendations, Assessment,
For all analyses in this review, the unit of analysis was defined as Development, and Evaluation (GRADE) approach.47 The reason(s)
the study participant. Whenever possible, data for prespecified for downgrading the certainty estimates are reported in
primary (visual fatigue symptoms and critical flicker-fusion fre- Tables 1 and 2 or in Table S1 (available at www.aaojournal.org).
quency [CFF]) and secondary (quality of life, dry eye symptoms,
amplitude of accommodation, near point of convergence, blink
rate, and overall patient satisfaction with the intervention) out-
Results
comes (Appendix S4, available at www.aaojournal.org) were
extracted preferably as the change from baseline; otherwise, data Characteristics of Included Studies
at the study end point were extracted. This latter measure is a Electronic searches yielded 4891 citations (Fig 1, Preferred
suitable substitute when comparing groups (under the assumption Reporting Items for Systematic Reviews and Meta-Analyses flow
of randomized interventions).34 In the absence of clinical diagram). After removing duplicates (n ¼ 922), title and abstract
guidelines to define best practice for CVS, the most widely screening was performed for 3969 citations. Of these, 296 records
reported outcomes of visual fatigue symptom scores2,5,18,35 and proceeded to full-text screening and 45 trials met the eligibility
CFF19,36e38 were used as primary outcome measures. criteria; 43 were full-text articles,19,20,22e25,48e84 published be-
When studies reported outcomes both before (“pre”) and after tween 1991 and 2021 and 2 studies85,86 were on clinical trial
(“post”) computer use after different durations of an intervention, registries with available data (registered in 201686 and 201885).
we used data reported for the after-computer task at the final study In addition, 24 RCTs listed on clinical trial registries were
visit for analysis. For studies that investigated multiple in- marked as ongoing (Table S2, available at www.aaojournal.org).
terventions, analyses were performed by comparing each inter- Table S3 (available at www.aaojournal.org) lists studies excluded
vention arm with the comparator group. Because different scales after full-text screening, with the primary reason for exclusion.
were used to measure visual fatigue and dry eye scores across the The key characteristics of all eligible studies are provided in
studies, the standardized mean difference (SMD) was used to pool Table S4 (available at www.aaojournal.org). Included RCTs were
data.39 Cohen40 suggests that a SMD of 0.2 represents a small conducted in 11 countries: Japan (n ¼ 11),23,59e62,64,65,73,74,77,80
effect, SMD of 0.5 represents a medium effect, and SMD of 0.8 India (n ¼ 9),24,25,48e51,68,70,71 the United States
represents a large effect. Further, because Cochrane advises not (n ¼ 6),19,52,69,72,85,86 Hong Kong (n ¼ 2),22,55 Korea (n ¼ 1),66
to combine mean change and end point outcome data, only China (n ¼ 1),81 Saudi Arabia (n ¼ 1),84 Australia (n ¼ 1),20
studies with end point outcomes were pooled because this Norway (n ¼ 1),56 Spain, and the United Kingdom (n ¼ 1).82
allowed the maximum number of studies to be included in the Eleven trials did not report the study
meta-analyses.39 For studies that reported only change from location.53,54,57,58,63,67,75,76,78,79,83 Most RCTs (n ¼ 35) used a
baseline data, their individual study results were summarized. parallel arm design,19,20,23e25,48e51,54e71,74,75,77,79e82,84 and 10
For the purpose of the analysis, data of crossover trials were studies used a crossover design.22,52,53,72,73,76,78,83,85,86 In total,
treated as parallel-design studies because no unit of analysis 4497 participants were recruited across the included 45 RCTs;
issue was present.41 If data were not reported as a mean and individual study sample sizes ranged from 10 to 522 participants.
standard deviation, guidance in chapter 7 of the Cochrane Further details about the study characteristics are provided in
Handbook for Systematic Reviews of Interventions was followed Table S5 (available at www.aaojournal.org).
to convert the data into the required form.34
Meta-analyses were performed using Cochrane Review Man- Risk-of-Bias Assessment
ager (RevMan software version 5.4; The Cochrane Collaboration).
Mean differences were calculated for the primary and secondary Figure 2 summarizes risk-of-bias assessments.19,20,22e25,48e86 No RCT
outcomes for each intervention subcategory when deemed clini- was judged to have low risk of bias in all 7 Cochrane domains. The
cally meaningful (e.g., for studies where the intervention, partici- domain with the most studies judged to be at low risk was attrition bias
pants, and underlying clinical questions were similar). When fewer (30 studies17,19,20,22e24,48,49,54e59,61e64,66,68,69,74e76,79e83,86). Domains

4 Table 1. Summary of Findings Table, Including Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) Assessments for the Certainty of the Body of
Evidence, for Optical Interventions for Managing Computer Vision Syndrome
Certainty of the Evidence Estimate (95% Confidence Interval)

(Grading of
Recommendations,
Intervention vs. Assessment, Development, With Placebo or With Active
Comparator Outcome Measure No. of Participants (No. of Studies) and Evaluation) Control Intervention Comments
22
Multifocal vs. Visual fatigue score, measured 352 (4 RCTs) 4222,* very low The pooled summary estimate from 3 Kee et al and Del Mar
single-vision using a Likert scale or studies22,85,86 indicated little to no Segui-Crespo et al82
correction VAS, with follow-up increase in visual fatigue score with measured visual fatigue
lenses ranging from 1 wk to 6 mos the multifocal intervention relative using a Likert scale;
to a single-vision correction: SMD, NCT0358579085 and
0.11 unit higher (0.14 unit lower to NCT0292108786

0.37 unit higher). Del Mar Segui- quantified visual fatigue
Crespo et al82 reported no using a VAS. Estimates are
significant difference between presented as SMD.
multifocal and single-vision
correction: MD, 1.50 units lower
(3.60 units lower to 0.60 unit
higher).
Dry eye symptom score, 44 (1 RCT) 4222,y very low One study (NCT0292108786) reported NCT0292108786 measured
measured using a no significant difference between dry eye symptoms using the
questionnaire or rating intervention arms. MD relative to Contact Lens Dry Eye
scale at the 1-wk follow-up. placebo, 0.70 unit lower (2.62 units Questionnaire 8 (out of
lower to 4.02 units higher). 37).
Adverse event rate, measured 180 (3 RCTs) 4422,z low The pooled summary estimate from 3 None
at follow-up ranging from 1 studies82,85,86 indicated little to no
wk to 6 mos. difference in adverse event rate
with multifocal relative to a single-
vision correction: RR, 2.09 (0.20 to
21.45).
Residual Visual fatigue symptom score, 12 (1 RCT) 4222,x very low Wiggings et al72 reported a significant Wiggings et al72 quantified
astigmatism measured using a difference between intervention visual fatigue symptoms
corrected in questionnaire or rating arms, but did not provide using a Likert scale.
CLs vs. not scale at 25 minutes of quantitative data.
corrected in follow-up.
CLs
Table 1. (Continued.)

(Grading of
Recommendations,
Intervention vs. Assessment, Development, With Placebo or With Active
Comparator Outcome Measure No. of Participants (No. of Studies) and Evaluation) Control Intervention Comments
k 20
Blue light Visual fatigue score, measured 166 (3 RCTs) 4422, low Singh et al reported no significant Singh et al20 quantified visual
blocking vs. using a Likert scale or difference between intervention fatigue using a VAS.
noneblue light VAS, with follow-up arms; MD relative to placebo, 9.76 Dabrowiecki et al83 and
blocking lenses ranging from < 1 to 5 days. units higher (33.95 units lower to Lin et al19 measured visual
53.47 units higher). Two other fatigue using a Likert scale.
studies19,83 reported no significant
Singh et al
difference between intervention

arms.
CFF, measured in Hertz, with 156 (2 RCTs) 4422,{ low Singh et al20 reported no significant None
follow-up of 2 hrs. difference between intervention
arms; MD relative to placebo, 1.13

Hz lower (3.00 Hz lower to 0.74 Hz

higher). Lin et al19 reported a
significant difference between
intervention arms, but did not
provide numeric data.
Dry eye symptom score, 2 RCTs (130) 4442,# moderate Singh et al20 reported no significant Singh et al20 quantified dry
measured using a Likert difference between intervention eye symptoms using a VAS.
scale or VAS, with follow- arms; MD relative to placebo, 1.31 Dabrowiecki et al83
up ranging from < 1 to 5 units lower (10.39 units lower to measured dry eye symptoms
days. 7.77 units higher). Dabrowiecki using a Likert scale.
et al83 reported no significant
difference between intervention
arms, but data were reported as daily
average change, rather than change
from baseline or end point data.
Amplitude of 1 RCT (120) 4442,** moderate Singh et al20 reported no significant None
accommodation, measured difference between intervention
in diopters (D), with a arms; MD relative to placebo, 0.30
follow-up of 2 hrs. D lower (0.71 D lower to 0.11 D
higher).
Near point of convergence, 1 RCT (120) 4442,** moderate Singh et al20 reported no significant None
measured in centimeters, difference between intervention
with a follow-up of 2 hrs. arms; MD relative to placebo, 0.29
cm higher (0.28 cm lower to 0.86
cm higher).
Blink rate, measured as the 1 RCT (120) 4442,** moderate Singh et al20 reported no significant None
number of blinks per difference between intervention
minute, with a follow-up of arms; MD relative to placebo, 1.37
2 hrs. blinks/min higher (0.44 blinks/min
lower to 3.18 blinks/min higher).
Adverse events, measure at 2 1 RCT (120) 4442,** moderate Singh et al20 reported no adverse events. None
hrs of follow-up.
5
with the greatest number of studies considered to have a high

CFF ¼ critical flicker-fusion frequency; CLs ¼ contact lenses; D ¼ diopters; hrs ¼ hours; MD ¼ mean difference; mos ¼ months; RCT ¼ randomized controlled trial; RR ¼ relative risk; SMD ¼ standardized
*Downgraded 2 levels because of risk of bias and 1 level because of inconsistency, because participants and outcome assessors were not masked in 2 studies,22,85 1 study82 had industry sponsorship, and 1
Downgraded 1 level for each of risk of bias and inconsistency, because 1 study19 had industry sponsorship and 1 study20 showed no intergroup differences, whereas the other study19 reported a positive effect
risk were performance bias, relating to a lack of masking of
participants, personnel, or a combination thereof (17
studies19,22,25,51e57,68,70,71,76,84,85,87); detection bias, pertaining to the
masking of outcome assessors (15 studies22,25,51e57,67,68,76,83e85); and
other bias (of 16 studies judged to have high risk of other bias, in 15
studies,19,23,59,60,62,64,65,69,74,76,77,79e82 this was because of industry
funding and in 1 study,48 it was because of baseline differences between
participant groups).
Effects of Interventions. Tables 1 and 2 provide summaries of
findings for the optical aids and complementary medicine and
nutritional supplement intervention categories, respectively. A
sufficient number of studies (n 2) was identified to perform
meta-analyses for 2 categories of interventions: (1) optical aids
and (2) complementary medicines and nutritional supplements.
Intervention Category: Optical Aids. Ten trials (Table 3)
investigated optical aids, consisting of multifocal contact lenses,85,86
Downgraded 2 levels for risk of bias, because in 1 study,85 participants and outcome assessors were not masked and 1 study82 had industry sponsorship.
progressive addition (multifocal) spectacle lenses,22,56,82 single-

vision cylindrical lenses,72 single-vision addition lenses,78 and blue
light-blocking spectacle lenses,19,20,83 relative to control single-
vision distance optical corrections,22,56,82,85,86 placebo lenses,72,78 or
noneblue light-blocking spectacle lenses.19,20,83
Downgraded 2 levels for risk of bias, because in 1 study83 outcome assessors were not masked and 1 study19 had industry sponsorship.
1. Primary outcome. Eight studies measured subjective visual

fatigue using a Likert scale19,22,72,82,83 or visual analog scale.20,85,86
Pooling comparable data from 3 studies22,85,86 provided no
difference in visual fatigue with use of a multifocal relative to a
Downgraded 3 levels because of imprecision, because the outcome is based on only 1 study86 that had a small sample size.
single-vision distance correction (Fig 3; 3 studies, 262

participants; SMD, 0.11; 95% confidence interval [CI], e0.14 to
0.37; I2 ¼ 7%; P ¼ 0.38). Del Mar Segui-Crespo et al82 reported
no difference for the change in visual fatigue scores with
**Downgraded 1 level because of imprecision, because results derived from a single study, at low risk of bias.20
multifocal compared with single-vision lenses (mean difference

[MD], e1.50 units; 95% CI, e3.60 to 0.60 units; P ¼ 0.16). The
certainty of evidence for this outcome was judged to be very low.
Downgraded 1 level because of risk of bias, because in 1 study,83 outcome assessors were not masked.
In addition, Wiggins et al72 provided very low certainty evi-

dence for reduced visual fatigue symptoms when correcting re-
sidual astigmatism in contact lens wearers compared with having
uncorrected residual astigmatism (Table S6, available at
www.aaojournal.org). For the other studies in this
category,19,20,83 no difference was found for the overall change
in visual fatigue19,20 or average daily change in visual fatigue83
with blue light-blocking lenses compared with noneblue light-
Only outcome measures evaluated by at least 1 study are listed in the table.
blocking lenses (Table S6).

Singh et al20 reported no difference for the change in CFF with
use of blue light-blocking lenses compared with noneblue light-
blocking lenses (MD, e1.13 Hz; 95% CI, e3.00 to 0.74 Hz;
P ¼ 0.240). In contrast, Lin et al19 described a less negative change
in CFF (i.e., less visual fatigue) with high blue light-blocking
mean difference; VAS ¼ visual analog scale; wk ¼ week.
lenses compared with both low blue-light blocking and

noneblue light-blocking lenses (Table S6). The GRADE certainty
of the evidence assessments for no effect on visual fatigue and little
or no effect on CFF (blue light-blocking vs. noneblue light-
blocking lenses) were both judged to be low.
with the blue light-blocking intervention.
2. Secondary outcomes. Three studies considered dry eye

symptoms. The study with ClinicalTrials.gov identifier
NCT0292108786 reported very low certainty of evidence for no
difference in symptom scores between multifocal and single-
study86 had divergent effects.
vision contact lenses after 1 week (44 participants; MD, 0.70

units; 95% CI, e2.62 to 4.02 units; P ¼ 0.68). Two studies20,83
provided moderate certainty evidence for no improvement in dry
eye symptoms with blue light-blocking lenses compared with
noneblue light-blocking lenses (Table S6).
For the other secondary outcomes, Singh et al20 reported no
difference between blue light-blocking and clear spectacle lenses
for the change from baseline in near point of accommodation
(MD, e0.30 diopter [D]; 95% CI, e0.71 to 0.11 D; P ¼ 0.15), near
{
k
#
y
z
x

Table 2. Summary of Findings Table, Including Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) Assessments for the Certainty of the Body of
Evidence for Complementary Medicine and Nutritional Supplements for Managing Computer Vision Syndrome

(Grading of
No. of Recommendations,
Intervention vs. Participants Assessment, Development, With Placebo or With Active
Comparator Outcome Measure (No. of Studies) and Evaluation) Control Intervention Comments
Oral berry extract Visual fatigue score measured 322 (7 RCTs) 4422,* low The pooled summary estimate from 7 In the 7 studies, visual fatigue
Singh et al
vs. placebo using a Likert scale or VAS studies23,60,64e66,74,77 indicated no was quantified using either
with follow-up ranging difference in visual fatigue score with a Likert scale60,66,77 or
from 4 to 12 wks. the berry extract relative to the VAS.23,64,65,74 Estimates
placebo supplement: SMD, 0.27 unit are presented as SMD.
lower (0.70 unit lower to 0.16 unit
higher).

CFF, measured in Hertz (HZ), 193 (3 RCTs) 4422,y low The pooled estimate from 3 studies23,64,74 Doses ranged from 60 to 160

with follow-up ranging indicated little to no decrease in CFF mg/day. Two studies23,64
from 4 to 12 wks. with the oral berry extract relative to evaluated bilberry extract
placebo: MD, 0.35 Hz lower (1.39 Hz and 1 study74 evaluated
lower to 0.69 Hz higher). maqui berry. Estimates are
presented as MD.
Quality of life measured using 74 (1 RCT) 4222,z very low One study74 reported no significant Yamashita et al74 measured
a questionnaire or rating difference between intervention arms: quality of life using the Dry
scale at 4 wks of follow-up. MD relative to placebo. 1.00 unit Eye Questionnaire Score
higher (1.36 units lower to 3.36 units (out of 100).
higher).
Dry eye symptom score, 248 (5 RCTs) 4422,x low The pooled estimate from 4 studies23,60,74,77 Dry eye score was quantified
measured using a Likert indicated no improvement in dry eye using either a Likert
scale or VAS, with follow- symptoms with the oral berry extract scale60,66,74,77 or VAS23
up ranging from 4 to 8 wks. relative to placebo: SMD, 0.10 unit (out of 100). Estimates are
lower (0.54 unit lower to 0.33 unit presented as SMD.
higher). Park et al66 reported no
improvement in dry eye score with
bilberry extract compared with
placebo: 0.69 unit lower (1.41 units
lower to 0.03 unit higher).
Amplitude of 119 (2 RCTs) 4422,k low The pooled estimate from 2 studies23,64 Both studies evaluated oral
accommodation, measured indicated little to no decrease in bilberry extract. Estimates
in diopters (D), with amplitude of accommodation with the are presented as MD.
follow-up ranging from 8 to oral berry extract relative to placebo:
12 wks. MD, 0.06 D lower (0.64 D lower to
0.53 D higher).
Adverse event rate, measured 8 RCTs (440) 4222,{ very low Eight studies23,60,64e66,74,77,79 reported no None
at follow-up ranging from 4 adverse events.
to 12 wks.
(Continued)
7
8

(Grading of
#
Oral omega-3 fatty Dry eye symptom score, 978 (2 RCTs) 4422, low The pooled summary estimate from 2 Both studies measured dry eye
acids vs. measured using a Likert studies24,48 indicated a reduction in dry symptoms using the Dry
placebo scale or VAS, with follow- eye score with oral omega-3 fatty acids Eye Questionnaire and
up ranging from 45 days to relative to placebo: MD, 3.36 units Scoring System (out of

3 months. lower (3.63 units lower to 3.10 units 18).101 Estimates are
lower). presented as MD.
Adverse event rate, measured 978 (2 RCTs) 4422,** low Two studies24,48 did not explicitly report None
at 45 days to 3 months of adverse events, but did mention that
follow-up. participants dropped out of the omega-
3 fatty acid supplement group because
of gastric intolerance.
Oral carotenoid Visual fatigue score measured 39 (1 RCT) 4222,yy very low Nagaki et al63 reported significant None
supplement vs. using a questionnaire or differences between intervention arms,
placebo rating scale at 4 wks of but did not provide quantitative data.
follow- up.
CFF, measured in Hertz, with 74 (2 RCTs) 4222,zz very low The pooled estimate from 2 studies63,69 The dose of carotenoids
follow-up ranging from 4 indicated a significant improvement in ranged from 5 mg/day63 to
wks to 6 mos. CFF with the oral berry extract relative 24 mg/day.69 Estimates are
to placebo: MD, 1.55 Hz higher (0.42 presented as MD.
Hz higher to 2.67 Hz higher).
Amplitude of 26 (1 RCT) 4222,xx very low One study63 reported no significant None
accommodation, measured difference between the intervention
in diopters at 4 wks of arms: MD relative to placebo, 0.50 D
follow-up. higher (0.56 D lower to 1.56 D
higher).
Adverse event rate measured 26 (1 RCT) 4422,kk low Nagaki et al63 reported no adverse events. None
at 4 wks of follow-up.
Traditional Visual fatigue score measured 60 (1 RCT) 4222,{{ very low One study57 reported no significant Joshi and Ujwale57 measured
medicine vs. using a Likert scale or VAS difference between the intervention visual fatigue using a VAS
placebo at 60 days of follow-up. arms: MD relative to placebo, 0.70 unit (out of 10).
lower (1.52 units lower to 0.12 unit
higher).
Adverse event rate measured 270 (3 RCTs) 4422,## low Three studies49e51 reported no adverse None
at follow-up ranging from 1 events.
mo to 6 wks.

(Grading of
81
Combination of Visual fatigue score measured 473 (4 RCTs) 4222,*** very low Kan et al reported a significant difference All studies quantified visual
oral using a questionnaire or between interventions: MD relative to fatigue using a Likert scale.
supplements vs. rating scale, with follow-up placebo, 2.83 units lower (3.56 units
placebo ranging from 4 to 12 wks. lower to 2.10 units lower). Kawabata
and Tsuji58 reported no significant
difference between the intervention
Singh et al
arms: MD relative to placebo, 0.40 unit

lower (1.20 units lower to 0.40 unit
higher). Kondo and Sawano59 reported
no significant difference between the
intervention arms: MD relative to

placebo, 0.40 unit lower (2.20 units

lower to 1.40 units higher). Kizawa
et al80 reported no significant
difference between the intervention
arms: MD relative to placebo, 0.00
units (0.65 unit lower to 0.65 unit
higher).
Dry eye symptom score, 363 (3 RCTs) 4422,yyy low Kan et al81 reported a significant difference All studies quantified dry eye
measured using a Likert between interventions: MD relative to symptoms using a Likert
scale or VAS, with follow- placebo, 0.52 unit lower (0.74 unit scale.
up ranging from 4 to 12 lower to 0.30 unit lower). Kawabata
wks. and Tsuji58 reported no significant
difference between interventions: MD
relative to placebo, 0.58 unit lower
(1.80 units lower to 0.64 unit higher).
Kizawa et al80 reported a significant
difference between interventions: MD
relative to placebo, e1.00 units lower
(1.81 units lower to 0.19 unit lower).
Amplitude of 110 (1 RCT) 4222,zzz very low One study59 reported no significant None
accommodation, measured difference between the intervention
in diopters at 12 wks of arms: MD relative to placebo, 0.29 D
follow-up. lower (0.94 D lower to 0.36 D higher).
Blink rate, measured as the 13 (1 RCT) 4222,xxx very low Yagi et al73 reported no difference between None
number of blinks per the study intervention arms, but did
minute, at 2 wks of follow- not provide quantitative data.
up.
Adverse event rate measured 453 (3 RCTs) 4222,kkk very low Two studies80,81 reported no adverse events. None
at follow-up ranging from 6 Kondo and Sawano59 reported 15
to 12 wks. adverse events in the combination
supplement group and 19 in the
placebo group; all were deemed
unrelated to the interventions.
(Continued)
9
10 Table 2. (Continued.)

(Grading of
{{{ 62
Other oral Visual fatigue score measured 173 (2 RCTs) 4422, low Morita et al reported no difference with Both studies quantified visual
interventions using a questionnaire or oral probiotics relative to placebo: MD fatigue using a VAS.
(green tea, rating scale, with follow-up relative to placebo: 3 units lower
probiotic, and ranging from 8 to 12 wks. (13.06 units lower to 7.06 units
taurine higher). Maeda-Yamamoto et al61
supplements) reported no difference among
vs. placebo Sunrouge, Yabukita, and placebo green

tea extracts.
CFF, measured in Hertz, with 84 (2 RCTs) 4222,### very low Morita et al62 reported no difference with None
follow-up ranging from 12 oral probiotics relative to placebo: MD
days to 8 wks. relative to placebo: 0.03 Hz higher
(1.60 Hz lower to 1.66 Hz higher).
Zhang et al75 reported no difference
with a taurine supplement relative to
placebo: MD relative to placebo, 4 Hz
lower (10.65 Hz lower to 2.65 Hz
higher).
Dry eye symptom score, 59 (1 RCT) 4222,**** very low Morita et al62 reported no difference with Morita et al62 measured dry
measured using a Likert oral probiotics relative to placebo: MD eye symptoms using a
scale or VAS at 8 wks of relative to placebo, 2.20 units higher Likert scale.
follow-up. (7.11 units lower to 11.51 units
higher).
Amplitude of 77 (2 RCTs) 4222,yyyy very low Morita et al62 reported no difference with None
accommodation, measured oral probiotics relative to placebo: MD
in diopters, with follow-up relative to placebo, 0.27 D higher
ranging from 8 to 12 wks. (1.17 D lower to 1.71 D higher).
Maeda-Yamamoto et al61 reported an
increase in amplitude of
accommodation with Sunrouge green
tea extract, relative to a placebo, but
did not provide quantitative data.
Adverse events 114 (1 RCT) 4442,zzzz moderate Maeda-Yamamoto et al61 reported no
adverse events.
CFF ¼ critical flicker-fusion frequency; D ¼ diopters; MD ¼ mean difference; mos ¼ months; NA ¼ not applicable; RCT ¼ randomized controlled trial; SMD ¼ standardized mean difference; VAS ¼ visual
analog scale; wks ¼ weeks.
Only outcome measures evaluated by at least 1 study are listed in the table.
*Downgraded 2 levels because of risk of bias, because 1 study60 had incomplete outcome data and selective reporting bias and 6 studies23,60,64,65,74,77 had industry sponsorship.
y
Downgraded 2 levels because of risk of bias because all 3 studies23,64,74 had industry sponsorship.
z
Downgraded 1 level for risk of bias and 2 levels for imprecision because 1 study74 had industry sponsorship and a small sample size.
x
Downgraded 2 levels because of risk of bias because 1 study60 had incomplete outcome data and selective reporting bias and 4 studies23,60,74,77 had industry sponsorship.
k
Downgraded 1 level for each because of risk of bias and imprecision because 2 studies23,64 had industry sponsorship and relatively wide confidence intervals.
{
Downgraded 3 levels because of risk of bias because 6 studies had selective reporting bias23,60,65,66,74,79 and 7 studies had industry sponsorship.23,60,64,65,74,77,79
#
Downgraded 2 levels because of inconsistency resulting from the high interstudy heterogeneity and 1 study48 having baseline differences.
**Downgraded 2 levels because of indirectness because 2 studies24,48 did not explicitly report adverse events.
yy
Downgraded 3 levels because of imprecision because 1 study63 had a small sample size.
Singh et al
point of convergence (MD, 0.29 cm; 95% CI, e0.28 to 0.86 cm;
Downgraded 2 levels for risk of bias and 1 level for imprecision because in 1 study57 the method of allocation concealment was not reported, the participants and outcome assessors were not masked, and it
Downgraded 1 level for each of risk of bias, inconsistency, and imprecision: 1 study69 had industry sponsorship, 1 study63 showed no intergroup differences with wide confidence intervals, one study69
P ¼ 0.32), and blink rate (MD, 1.37 blinks/minute; 95% CI, e0.44
to 3.18 blinks/minute; P ¼ 0.14). The GRADE certainty of the
Downgraded 1 level for risk of bias and 2 levels for imprecision because 1 study62 had industry sponsorship and wide confidence intervals and this and another study75 had small sample sizes.
evidence for each of these outcomes was judged to be moderate.
3. Adverse events. Four trials20,82,85,86 reported adverse event
data. No adverse events were reported for 2 studies reporting
Downgraded 1 level for risk of bias and 2 levels for imprecision because 1 study62 had industry sponsorship and wide confidence intervals and another study61 had a small sample size.
spectacle lens interventions (blue light blocking [2 hours]20 and
multifocal lenses [6 months]82). Although 1 study85 reported no
Downgraded 1 level for each of risk of bias and imprecision because in 1 study,51 the participants and outcome assessors were not masked, and this study had a small sample size.
adverse events in 2 contact lens-wearing groups after 1 week, in

another study,86 2 participants in a multifocal contact lens group
and 1 participant in a control group (single-vision contact lens)
showed mild anterior eye adverse events after 1 week. Pooling
data from 3 studies82,85,86 provided low certainty evidence for no
****Downgraded 1 level for risk of bias and 2 levels for imprecision because 1 study62 had industry sponsorship, wide confidence intervals, and a small sample size.
difference in the risk of an adverse event between multifocal and

Downgraded 2 levels for risk of bias and 1 level for imprecision because this study59 had selective reporting bias, industry sponsorship, and small sample size.
single-vision lenses (Fig S1, available at www.aaojournal.org; 3

studies; 180 participants; relative risk, 2.09; 95% CI,
0.20e21.45; P ¼ 0.53).
Intervention Category: Complementary Medicines and
Nutritional Supplements. Subcategory: Berry Extract. Eight
trials23,60,64e66,74,77,79 investigated forms of oral berry extract
compared (bilberry extract, n ¼ 623,60,64,65,77,79; bog bilberry,
Downgraded 3 levels because of risk of bias because 1 study59 had selective reporting bias and 3 studies59,80,81 had industry sponsorship.
Downgraded 1 level for each of risk of bias and imprecision because 1 study62 had industry sponsorship and wide confidence intervals.
n ¼ 166; and maqui berry, n ¼ 174) relative to a placebo.

Table 4 summarizes the characteristics of these studies.
***Downgraded 3 levels for risk of bias because 1 study59 had selective reporting bias and 3 studies59,80,81 had industry sponsorship.
1. Primary outcomes. Seven studies evaluated subjective vi-

sual fatigue.23,60,64e66,74,77 Combining their data gave low-
certainty evidence for no improvement in visual fatigue score

with oral berry extract supplementation relative to placebo (Fig 4; 7
Downgraded 2 levels because of imprecision, because the outcome is based on 1 study63 that had a small sample size.
studies; 322 participants; SMD, e0.27 units; 95% CI, e0.70 to

0.16 units; I2 ¼ 70%; P ¼ 0.22); significant heterogeneity was
observed because of divergent effects noted in 1 study77 (Fig S2,
available at www.aaojournal.org). Pooling data from 3
studies23,64,74 provided low certainty for no difference in CFF
reported a positive effect with the carotenoid intervention, and both studies63,69 had a small sample size.
with oral berry extract supplementation relative to placebo at the

study end points (Fig 5; 193 participants; MD, e0.35 Hz; 95%
Downgraded 1 level because of imprecision because results were obtained from a single study.61
CI, e1.39 to 0.69 Hz; I2 ¼ 29%; P ¼ 0.51).

2. Secondary outcomes. Yamashita et al74 provided very low
certainty evidence for no difference in quality-of-life score after 4
weeks of using oral maqui berry supplementation relative to placebo
Downgraded 2 levels for risk of bias because 2 studies80,81 had industry sponsorship.
(74 participants; MD, 1.00 units; 95% CI, e1.36 to 3.36 units; P ¼
0.41). Pooling comparable data from 4 studies23,60,74,77 showed no
decrease in dry eye symptoms with oral berry extract supplements
relative to placebo (Fig 6; 4 studies, 198 participants; SMD, e0.10
units; 95% CI, e0.54 to 0.33 units; I2 ¼ 50%; P ¼ 0.65). Park
et al66 reported no difference for the change in dry eye scores with
bilberry extract supplement compared with placebo (MD, e0.69
units; 95% CI, e1.41 to 0.03 units; P ¼ 0.06). The certainty of
evidence for this outcome was judged to be low.
Pooling data from 2 studies23,64 provided low certainty
evidence for no difference in accommodative amplitude at the
study end points with oral bilberry extract relative to placebo
supplementation (Fig 7; 2 studies; 119 participants; MD, e0.06
D; 95% CI, e0.64 to 0.53 D; I2 ¼ 0%; P ¼ 0.85).
3. Adverse events. All 8 studies23,60,64e66,74,77,79 reported no
adverse events in either the berry extract or placebo groups. The
certainty of the evidence for this outcome was considered very low.
Subcategory: Polyunsaturated Fatty Acids. Two parallel-
arm trials24,48 evaluated oral long-chain omega-3 fatty acids
(eicosapentaenoic acid and docosahexaenoic acid) compared with
had a small sample size.
placebo.
1. Secondary outcomes. Pooling data from these studies,24,48
which had very high statistical heterogeneity (I2 ¼ 89%; Fig 8),
showed moderate certainty for improved dry eye symptoms with
oral omega-3 supplementation after 45 days to 3 months of
follow-up (2 studies; 978 participants; MD, e3.36 units; 95% CI,
e3.63 to e3.10 units; I2 ¼ 89%; P < 0.00001) relative to placebo.
{{{
yyyy
zzzz
kkk
###
yyy
zzz
xxx
{{
kk
##
zz
xx
11

print & web 4C=FPO
Figure 1. PRISMA flow diagram showing studies included in this systematic review (1991e2021). CVS ¼ computer vision syndrome; RCT ¼ randomized
controlled trial.
2. Adverse events. Although neither study explicitly reported sesame oil) relative to placebo (60 participants; MD, e0.70 units;
adverse events, participants dropped out of the omega-3 fatty acid 95% CI, e1.52 to 0.12 units; P ¼ 0.09).
supplement group because of gastric intolerance (n ¼ 648 and n ¼ 2. Adverse events. Three studies49e51 involving 270 partici-
2624). Neither study reported if dropouts occurred in the placebo pants reported no adverse events in the traditional medicine or
group. The certainty of evidence for this outcome was judged to placebo groups over 1 month51 to 6 weeks.49,50 The certainty of
be low. evidence for this outcome was judged to be moderate.
Subcategory: Oral Carotenoid Supplements. Nagaki Subcategory: Combination Supplements. Five
et al63 and Stringham et al69 evaluated oral carotenoid supplements studies58,59,73,80,81 investigated combination oral supplements
relative to placebo. relative to placebo (Table S4).
1. Primary outcomes. Nagaki et al63 provided low certainty 1. Primary outcome. visual fatigue score. Four
evidence for reduced visual fatigue with oral carotenoid relative studies58,59,80,81 reported on visual fatigue; however, data were not
to placebo supplements; numeric data were not provided. Pooling pooled because the studies evaluated different supplement
data from the two studies63,69 provided very low certainty of an combinations (Table S4). Kan et al81 reported reduced visual
improvement in CFF with oral carotenoid supplementation fatigue symptoms with a combination supplement of lutein ester,
relative to placebo (Fig 9; 2 studies; 74 participants; MD, 1.55 zeaxanthin, and extracts of blackcurrant, chrysanthemum, and
Hz; 95% CI, 0.42 to 2.67 Hz; I2 ¼ 0%; P ¼ 0.007). goji berry after 4 weeks (303 participants; MD, e2.83 units;
2. Secondary outcomes. Stringham et al69 did not measure any of 95% CI, e3.56 to e2.10 units; P < 0.001) relative to placebo.
the secondary outcomes. Nagaki et al63 provided very low certainty Three studies58,59,80 reported no improvement in symptoms of
evidence for no difference in amplitude of accommodation after 4 visual fatigue with combination supplements compared with
weeks of carotenoid supplementation relative to placebo (26 par- placebo (Table S6). The certainty of the evidence for this
ticipants; MD, 0.50 D; 95% CI, e0.56 to 1.56 D; P ¼ 0.35). outcome was judged to be very low.
3. Adverse events. Only Nagaki et al63 (n ¼ 26) reported 2. Secondary outcomes. Kan et al81 reported improved dry eye
adverse events, finding none in either the carotenoid supplement symptom scores with the combination supplement after 4 weeks
or placebo groups. The certainty of evidence for this outcome relative to placebo (303 participants; MD, e0.52 units; 95% CI,
was judged to be low. e0.74 to e0.30 units; P < 0.001). Likewise, Kizawa et al80
Subcategory: Traditional Medicines. Six parallel-arm reported improved dry eye symptom scores with a combination
RCTs25,49e51,57,68 investigated traditional medicines relative to a supplement after six weeks (40 participants; MD, e1.00 units;
placebo; study details are summarized in Table S4. 95% CI, e1.81 to e0.19 units; P < 0.02) relative to placebo.
1. Primary outcome. Joshi and Ujwale57 provided very low Further, one study58 reported no difference in dry eye symptom
certainty evidence for no difference in visual fatigue with use of scores with combination supplements compared with placebo
Tila Taila Padabhyanga traditional medicine (foot massage with (Table S6).
12

Singh et al
Very low certainty evidence was found for no difference in

amplitude of accommodation with the combination supplement
(Table 2) investigated by Kondo and Sawano59 relative to placebo
(110 participants; MD, e0.29 D; 95% CI, e0.94 to 0.36 D; P ¼
0.38). Likewise, very low certainty evidence was found for no
difference in eye blink rate between the combination supplement
evaluated by Yagi et al73 relative to a placebo supplement.
3. Adverse events. Two studies80,81 reported no adverse events
in either the combination supplement or placebo groups. Kondo
and Sawano,59 studying 110 participants, reported 15 adverse
events in the combination supplement group and 19 in the
placebo group over 12 weeks; all were self-reported and deemed
unrelated to the intervention by the study medical investigator. The
certainty of evidence for this outcome was judged to be very low.
Subcategory: Other. Three parallel-arm trials61,62,75
investigated other types of complementary medicines or
nutritional supplements relative to a placebo supplement. The
interventions that were evaluated were Yabukita (group I) and
Sunrouge (group II) green tea extracts,61 an oral probiotic,62 and
an oral taurine supplement.75
1. Primary outcomes. Two studies reported visual fatigue
data.61,62 Morita et al62 reported no difference with probiotics
relative to placebo (59 participants; MD, e3.00 units; 95% CI,
e13.06 to 7.06 units; P ¼ 0.56). Maeda-Yamamoto et al61 found
no difference among Sunrouge or Yabukita green extracts, and a
placebo; numeric data were not reported.
For CFF, Morita et al62 reported no difference between oral
probiotics and placebo (59 participants; MD, 0.03 Hz; 95% CI,
e1.60 to 1.66 Hz; P ¼ 0.97). Zhang et al75 observed no
difference between taurine and placebo supplements (25
participants; MD, e4.00 Hz; 95% CI, e10.65 to 2.65 Hz; P ¼
0.24).
The certainty of evidence for visual fatigue scores was judged
to be low, and that for CFF was judged to be very low.
2. Secondary outcomes. Morita et al62 provided very low
certainty evidence for no difference in dry eye symptom score
between oral probiotics and placebo (59 participants; MD, 2.20
units; 95% CI, e7.11 to 11.51 units; P ¼ 0.64).
Two studies62,75 evaluated accommodative amplitude. Morita
et al62 found no difference with oral probiotics relative to
placebo (59 participants; MD, 0.27 D; 95% CI, e1.17 to 1.71 D;
P ¼ 0.71). Maeda-Yamamoto et al61 performed a subgroup
analysis and described that, in a group younger aged less than 45
years, an increase in amplitude of accommodation was found
with Sunrouge green tea extract (n ¼ 8) relative to a placebo
(n ¼ 10); no numeric data were provided. The certainty of
evidence for this outcome was judged to be very low.
3. Adverse events. Maeda-Yamamoto et al,61 studying 114
participants, reported no adverse events in the green tea extract
or placebo groups. The certainty of evidence was judged to be
moderate.
Intervention Category: Artificial Tears. Guillon et al76
investigated 2% povidone preservative-free lubricating eye drops
using 3 dosing regimens compared with no intervention.
Visual fatigue was measured, although no numeric data were
provided. The study provided very low certainty evidence for
print & web 4C=FPO
reduced symptoms with the eye drops using all 3 instillation fre-
quencies compared with no intervention.
Secondary outcomes and adverse events were not reported.
Intervention Category: Environmental Modifications. Two
parallel-arm studies investigated environmental modifications,54,55
comprising a moist cool air device54 and computerized risk
Figure 2. Risk-of-bias assessment for included randomized controlled trials, assessment systems.55
assessed using the Cochrane Risk of Bias tool. Green, red, and yellow circles 1. Primary outcome. Hirayama et al54 found no difference
indicate judgements of low, high, and unclear risk of bias, respectively. in visual fatigue score with a moist cool air device after 5
13

14
Table 3. Key Characteristics of Randomized Controlled Trials Evaluating Optical Aids for the Management of Computer Vision Syndrome
Recruited Intervention
Study Intervention Subcategory Participant Population Intervention(s) (Frequency) Comparator (Frequency) Sample Size Duration
ClinicalTrials.gov identifier, CLs Symptomatic electronic MF (F: daily wear, daily SV (F: daily wear, daily 23 1 wk
NCT0292108786 device users disposable schedule) disposable schedule)
ClinicalTrials.gov identifier, CLs Symptomatic electronic MF (F: NR) SV (F: NR) 45 1 wk
NCT0358579085 device users
Dabrowiecki et al83 Spectacles Radiology residents Blue light-blocking SVL Noneblue light-blocking 10 5 days
(F: 8:00 AMe5:00 PM/day) SVL (F: 8:00 AMe5:00 PM/

day)

Del Mar Segui-Crespo M Spectacles Pre-presbyopic, symptomatic PALs with low near add SVLs (F: when using 90 6 mos
et al82 computer users (F: when using electronic electronic displays)
displays)
Horgen et al56 Spectacles Experienced VDT users 1: Essilor Interview lenses SVLs (F: NR) 158 1 yr
(F: NR)
2: Zeiss Gradal room distance
lenses (F: NR)
3: Technica American
Optical lenses (F: NR)
Kee et al22 Spectacles Computer users > 2 hrs/day Zeiss PALs (F: NR) Zeiss SVLs (F: NR) 64 1m
Lin et al19 Spectacles Adults who did not perform 1: High blue light-blocking Noneblue light-blocking 36 2 hrs
VDT work for 1 hr SVLs (F: 1 2-hr session) SVLs (F: 1 2-hr session)
beforehand 2: Low blue light-blocking
SVLs (F: 1 2-hr session)
Singh et al20 Spectacles Symptomatic computer users 1: Blue light-blocking SVLs 1: Noneblue light-blocking 120 2 hrs
with positive deception SVLs with positive
(F: 1 2-hr session) deception (F: 1 2-hr
2: Blue light-blocking SVLs session)
with negative deception 2: No blue light-blocking
(F: 1 2-hr session) SVLs with negative
deception (1 2-hr
session)
Wiggins et al72 Spectacles Soft CL users Subjective residual SVLs with þ0.12 D (F: 1 12 25 min
astigmatism corrected with 25-min session)
trial lenses (F: 1 25-min
session)
Yammouni and Evans78 Spectacles Symptomatic computer users 1: Low add þ0.50 SVLs Plano lenses (F: NR) 107 < 1 day
(F: NR)
2: Low add þ0.75 SVLs
(F: NR)
3: Low add þ1.25 SVLs
(F: NR)
CL ¼ contact lens; D ¼ diopters; F ¼ frequency; hr ¼ hour; min ¼ minute; MF ¼ multifocal; mos ¼ months; NR ¼ not reported; PAL ¼ progressive addition lens; SV ¼ single vision; SVL ¼ single-vision
lens; VDT ¼ visual display terminal; wk ¼ week; yr ¼ year.
Singh et al
print & web 4C=FPO
Figure 3. Forest plot comparing active optical aid (spectacle or contact lens) versus control. The outcome was visual fatigue or asthenopia score, measured
using a subjective questionnaire or visual analog scale. Data are reported at the end of the study follow-up period: from 1 week (NCT0358579085 and
NCT0292108786; multifocal contact lenses) to 1 month (Kee et al22; progressive addition spectacle lenses). For Kee et al,22 data were combined from 2
separately reported age groups (18e35 years [n ¼ 19] and 30e40 years [n ¼ 45]). For NCT03585790,85 the presented data were calculated from the
median and interquartile range. For Kee et al,22 mean scores were converted to ensure that higher scores indicated more severe symptoms for
consistency. For all studies presented here, higher scores indicate more severe symptoms. CI ¼ confidence interval; IV ¼ inverse variance; SD ¼
standard deviation; VAS ¼ visual analog scale.
days relative to no intervention (20 participants; MD, e0.20 generated by a software system relative to participants who
units; 95% CI, e2.70 to 2.30 units; P ¼ 0.88). Ho et al55 received a delayed intervention after 2 weeks (111
reported reduced visual fatigue in participants who received participants; MD, e1.47 units; 95% CI, e2.35 to e0.59 units;
workplace modification advice per the recommendations P ¼ 0.001). After modifying the software to be web based, no
Table 4. Key Characteristics of Randomized Controlled Trials Evaluating Formulations of Oral Berry Extract
Intervention(s) (Dose; Comparator (Dose; Recruited Intervention

Study Participant Population Frequency) Frequency) Sample Size Duration
Kosehira et al79 Symptomatic computer users Bilberry extract (dose: 240 Oral placebo (dose: NR; 109 12 wks
mg; F: 1 capsule/day) F: 1/day)
Liang et al60 Symptomatic computer users Bilberry extract (dose: 160 Oral placebo (dose: NR; 21 6 wks
mg; F: 1 capsule/day) F: 1/day)
23
Ozawa et al Symptomatic VDT users Bilberry extract (dose: 160 Oral placebo containing 88 8 wks
mg; F: 3 capsules, 1/day) starch, crystalline cellulose,
and calcium stearate (dose:
NR; F: 3 capsules, 1/day)
Okamoto et al64 Symptomatic VDT users Bilberry extract (dose: 60 mg; Oral placebo, consisting of 47 12 wks
F: 1 capsule/day) edible oil and corn starch
(dose: 555 mg; F: 1 capsule/
day)
Okamoto et al65 Symptomatic VDT users Bilberry extract (dose: 60 mg; Oral placebo containing 44 12 wks
F: 1 capsule/day) edible oil and corn starch
(dose: 555 mg; F: 1/day)
Park et al66 Computer users > 2 hrs/day Vaccinium uliginosum extract Oral placebo, containing 60 4 wks
(DA9301) tablet (dose: lactose (dose: NR; F: NR)
1000 mg; F: 1/day)
Sekikawa et al77 Symptomatic VDT users Bilberry extract capsule (dose: Placebo capsule (dose: NR; 32 6 wks
43.2 mg; F: 1/day) F: 1/day)
Yamashita et al74 Symptomatic computer user MaquiBright capsule dose: Oral placebo (dose: 180 mg; 74 4 wks
120 mg dextrin plus 60 mg F: 1/day)
MaquiBright (21 mg
anthocyanins plus 15mg
delphinidins plus 4 mg
delphinidin-3,5-O-
diglucoside; F: 1/day)
F ¼ frequency; mg ¼ milligrams; NR ¼ not reported; VDT ¼ visual display terminal; wks ¼ weeks.
15

print & web 4C=FPO
Figure 4. Forest plot comparing oral berry extract (bilberry and maqui berry) versus placebo supplementation. The outcome was visual fatigue or asthenopia
score, measured using a subjective questionnaire or visual analog scale. For Park et al,66 (bilberry, 1000 mg/day), Okamoto et al,64 (bilberry, 60 mg/day), and
Sekikawa et al77 (bilberry, 43.2 mg/day), data are reported at the end of the study follow-up period, which were week 4,66 week 6,77 and week 12.64 For
Ozawa et al23 (bilberry, 3 160 mg capsules/day), Liang et al60 (bilberry, 160 mg/day), Okamoto et al65 (bilberry, 60 mg/day), and Yamashita et al74
(dextrin 120 mg/day plus maqui berry extract 60 mg/day), data are reported after VDT load at the end of the follow-up period, which were week 4,74
week 6,60 week 8,23 and week 12.65 Presented data were calculated from the standard error of the mean for Ozawa et al23 and from median and
interquartile range for both Yamashita et al74 and Sekikawa et al.77 For Yamashita et al,74 mean scores were converted to ensure that higher scores
indicated more severe symptoms for consistency. For all studies presented here, higher scores indicate more severe symptoms. CI ¼ confidence interval;
IV ¼ inverse variance; SD ¼ standard deviation; VAS ¼ visual analog scale.
significant improvement in visual fatigue score was observed in Intervention Category: Ergonomic Adjustment. Robertson
participants who received the work modification advice et al67 investigated office ergonomic training compared with
immediately compared with those who received it after a minimal training. None of the primary or secondary outcomes
2-week delay (75 participants; MD, e0.78 units; 95% were measured, nor were adverse events documented.
CI, e1.83 to 0.27 units; P ¼ 0.15). The certainty of evidence Intervention Category: Visual Hygiene. Alrasheed and
was judged to be very low for this outcome. Alghamdi84 investigated whether the 20-20-20 rule reduced
2. Secondary outcomes. Hirayama et al54 provided very low symptoms associated with computer use compared with a
certainty evidence for no difference in dry eye symptom score or placebo treatment involving advice to drink water.
blink rate with a moist cool air device relative to no intervention for 1. Primary outcome. Alrasheed and Alghamdi84 provided
5 days (Table S6). very low certainty evidence for no improvement in visual fa-
3. Adverse events. Hirayama et al54 (n ¼ 20) reported no tigue symptoms with the 20-20-20 rule relative to advice to
adverse events with either the moist cool air device or no drink water, at the end of a 20-day follow-up period (40 par-
intervention. The certainty of evidence was judged to be very low ticipants; MD, 0.80 units; 95% CI, e0.59 to 2.19 units;
for this outcome. P ¼ 0.26).
print & web 4C=FPO
Figure 5. Forest plot comparing oral berry extract (bilberry and maqui berry) versus placebo supplementation. The outcome was critical flicker-fusion
frequency, measured in Hertz. For Okamoto et al64 (bilberry, 60 mg/day) and Yamashita et al74 (dextrin 120 mg/day plus maqui berry extract 60 mg/
day), data are reported at the end of the follow-up period, which are week 474 and week 12.64 For Ozawa et al23 (bilberry, 3 160 mg capsules/day),
data are reported after visual display terminal load at the end of the 8-week follow-up period. Presented data were calculated from the standard error of the
mean. For all studies presented here, more negative values indicate a greater improvement in visual fatigue. CI ¼ confidence interval; IV ¼ inverse variance;
SD ¼ standard deviation.
16

Singh et al
print & web 4C=FPO
Figure 6. Forest plot comparing oral berry extract (bilberry and maqui berry) versus placebo supplementation. The outcome was dry eye symptom score,
measured using a questionnaire or rating scale. For Yamashita et al74 (dextrin 120 mg/day plus maqui berry extract 60 mg/day) and Sekikawa et al77 (bilberry,
43.2 mg/day), data are reported at the end of the study follow-up period, which were week 466,74 and week 6.77 For Yamashita et al74 and Sekikawa et al,77
data were calculated from the median and interquartile range. For Ozawa et al23 (bilberry, 3 160 mg capsules/day) and Liang et al60 (bilberry, 160 mg/day)
data are reported after visual display terminal load at the end of the follow-up period, which were week 660 and week 8.23 For Ozawa et al,23 data were
calculated from the standard error of the mean. For all studies presented here, higher scores indicate more severe dry eye symptoms. CI ¼ confidence
interval; IV ¼ inverse variance; SD ¼ standard deviation; VAS ¼ visual analog scale.
2. Secondary outcomes. This study84 provided very low with yoga relative to recreational activity (118 participants; MD,
certainty evidence for reduced dry eye symptoms with imple- e0.80 units; 95% CI, e1.04 to e0.56 units; P < 0.0001).
menting the 20-20-20 rule relative to placebo (40 participants; Galinsky et al52 described less visual fatigue with supplementary
MD, e2.60 units; 95% CI, e4.51 to e0.69 units; P ¼ 0.008). Also, breaks compared with conventional breaks, but did not provide
very low certainty evidence was found for no difference in blink numeric data. Both studies were judged to provide very low
rate between the study groups at the study end point (40 partici- certainty evidence for this outcome. Neither study measured any
pants; MD, 0.85 blinks; 95% CI, e3.47 to 5.17 blinks; P ¼ 0.70). of the secondary outcomes or documented adverse events.
3. Adverse events. Adverse events were not reported in this
study.
Intervention Category: Other. Four studies investigated other Discussion
interventions (Table S4).52,53,70,71 Galinsky et al53 evaluated the
effects of supplementary workplace rest breaks compared with
conventional workplace breaks. Another study from the same
This systematic review investigated the efficacy and safety
group52 evaluated supplementary breaks with stretching exercises of interventions for CVS. Forty-five RCTs, with sample
compared with conventional breaks. Two parallel-arm studies by sizes ranging from 10 to 522 participants and post-inter-
the same lead author, Telles and Naveen70 and Telles et al,71 vention follow-up periods ranging from less than 24 hours
evaluated the effects of yoga relative to no exercise. to 1 year, were identified. Of these trials, most investigated
Relevant to the primary outcome, two studies reported visual complementary medicines or nutritional supplements (58%)
fatigue data.52,71 Telles et al71 described reduced visual fatigue or optical aids (22%). Overall, quantitative data relating to
print & web 4C=FPO
Figure 7. Forest plot comparing oral bilberry extract versus placebo supplementation. The outcome was amplitude of accommodation, measured in diopters
(D). For Okamoto et al64 (bilberry, 60 mg/day), data are reported at the end of the 12-week follow-up period. For Ozawa et al23 (bilberry, 3 160 mg
capsules/day), data are reported after visual display terminal load at the end of the 8-week follow-up period. Presented data were calculated from the
standard error of the mean. For both Okamoto et al64 and Ozawa et al,23 more negative values indicate a greater improvement. CI ¼ confidence interval;
IV ¼ inverse variance; SD ¼ standard deviation.
17

print & web 4C=FPO
Figure 8. Forest plot comparing oral omega-3 fatty acid versus placebo supplementation. The outcome was dry eye symptoms, measured using a ques-
tionnaire or rating scale. For both studies (Bhargava et al48: oral omega-3 fatty acid dose, 4 180 mg eicosapentaenoic acid [EPA] plus 120 mg docosa-
hexaenoic acid [DHA] capsules/day; Bhargava et al24: 8 180 mg EPA þ 120 mg DHA capsules/day), data are reported at the end of the follow-up period
(45 days for Bhargava et al24 and 3 months for Bhargava et al48). Higher scores indicate more severe dry eye symptoms, measured with the Dry Eye Scoring
System (out of 18).101 The symptom severity schema is: 0e6, mild; 6.1e12, moderate; and 12.1e18, severe. CI ¼ confidence interval; IV ¼ inverse
variance; SD ¼ standard deviation.
the efficacy or safety of CVS interventions were limited. personnel to the intervention, including 33% where outcome
The certainty of evidence for the pre-specified outcome assessors were not masked. Larger effect sizes have been
measures, across all interventions, was judged to be mod- associated with studies without masking relative to those
erate, low, or very low. The two most common reasons for with masking.89 Confidence in estimates from studies that
downgrading the certainty of the evidence were risk of bias were neither single nor double masked thus is reduced. A
(participant and outcome assessor not being masked or in- further consideration is that 1 in 3 included studies
dustry funding) and imprecision (small study sample size). received industry funding, a factor that has been
This uncertainty about the potential effect(s) of interventions associated with more favorable conclusions in relationship
on outcomes relevant to CVS should be considered when to the efficacy of an intervention compared with trials not
making clinical management decisions. receiving industry funding.90
This review found very low certainty evidence for no effect Of the 45 included RCTs, two-thirds did not provide trial
on visual fatigue with a multifocal, compared with single- registration details, resulting in limited capacity to assess
vision distance, optical correction based on findings from 3 selective outcome reporting. Two-thirds of included trials
RCTs over intervention durations of 1 week to 6 months. Low did not report sample size calculations, raising concerns
certainty was found for no reduction in visual fatigue with oral about whether these studies were powered adequately to
berry extract supplementation compared with placebo over 4 detect the intended treatment effects.
to 12 weeks based on data from 7 RCTs. Also, low certainty
evidence was found for berry extract supplementation not Overall Completeness and Applicability of the
improving CFF relative to placebo. Regarding the trial Evidence
follow-up periods relative to the time that might be expected
for particular interventions to modulate outcome measures, Optical aids are used commonly to manage CVS. A recent
the exposure period for optical aids likely is appropriate given survey of 372 Australian optometrists found that most
that the time to derive any benefit is relatively short. However, practitioners (87%) used spectacle lens corrections for CVS
for nutritional supplementation, a minimum follow-up of 12 management;91 we expect that a proportion of these
weeks might be necessary based on their proposed systemic corrections likely involved multifocal lenses for pre-
mechanism of action.88 presbyopic patients. The findings in this review suggest
Substantial risk of performance, detection, and other that such multifocal lens prescribing is unlikely to provide
biases were identified in several studies. Specifically, 38% benefit relative to single-vision correction for managing
of eligible studies did not mask participants or study CVS symptoms.
print & web 4C=FPO
Figure 9. Forest plot comparing oral carotenoid versus placebo supplementation. The outcome was critical flicker-fusion frequency measured in Hertz. For
Nagaki et al63 (carotenoid supplement, 5 mg/day) and Stringham et al69 (macular carotenoid supplement, 24 mg/day), data are from the end of the study
follow-up period (4 weeks63 and 6 months69). For both studies, more positive values indicate a greater improvement. CI ¼ confidence interval; IV ¼ inverse
variance; SD ¼ standard deviation.
18

Singh et al
The potential use of blue light-blocking spectacle lenses High statistical heterogeneity was evident for the
for treating CVS has gained attention in recent years.92,93 pooled analysis of symptom scores for the 2 studies24,48
Despite this, only 3 RCTs investigating these lenses were by the same research group evaluating omega-3 fatty
identified.19,20,83 All 3 studies reported no benefit on eye acid supplements, although the direction of effect was
strain with blue light-blocking lenses compared with no consistent. This finding may relate to use of different
blue light-blocking lenses. Lin et al19 did not provide omega-3 doses in the studies. In practice, practitioners
numeric data, but described a benefit (i.e., less negative also should be mindful of potential adverse events asso-
change in CFF following computer use) with obviously ciated with oral omega-3 fatty acid supplementation.103
brown-colored, ‘high’ blue light-blocking lenses Bhargava et al24 experienced greater participant dropout
compared with both more traditional ‘low’ blue light- because of gastric intolerance in the group prescribed
blocking and no blue light-blocking lens interventions. the higher dose of omega-3 fatty acids (n ¼ 26; 1440
However, use of CFF as a measurement of visual fatigue mg eicosapentaenoic acid plus 960 mg docosahexaenoic
has been questioned because not all studies measuring CFF acid per day) compared with the group receiving the
have found a correlation between a decrease in CFF and lower dose48 (720 mg eicosapentaenoic acid plus 480 mg
symptoms of eye strain.94e96 Furthermore, the findings of docosahexaenoic acid per day). Gastric intolerance is a
Lin et al19 should be viewed in the context of the study commonly documented side effect of oral omega-3
receiving industry funding, not being prospectively supplementation.99,104
registered, having a moderate sample size (n ¼ 36), and For oral carotenoid supplementation, pooled data from
a lack of outcome-assessor masking. In contrast, the 2 studies63,69 involving 74 participants provided very low
double-masked RCT by Singh et al20 involving 120 certainty evidence for an improvement in CFF with the
participants found no significant between-group differ- carotenoid intervention relative to placebo. However, it is
ence (blue light blocking vs. no blue light blocking) in unclear if the observed between-group difference of 2 Hz
CFF, as a measurement of visual fatigue. is clinically meaningful if, indeed, CFF is a useful mea-
The other major intervention category was oral berry sure of visual fatigue at all.94e96 Further, the potential
extract supplementation. Pooled data from 7 mechanism of action of oral carotenoid supplementation
trials23,60,64e66,74,77 showed no reduction in visual fatigue in reducing eye strain is unknown, and the study popu-
with berry extract supplements compared with placebo in lation did not seem to have been assessed (at baseline)
study populations without established dietary deficiencies. with respect to a potential systemic deficiency that might
Use of this intervention also did not improve dry eye provide some justification for the benefit of oral
symptom scores. For both visual fatigue and subjective supplementation.
dry eye measurements, a range of questionnaires and Previous narrative reviews that focused on CVS man-
scales were used. Standardized MD was used to meta- agement4,15,21,105 have recommended optical
analyze relevant study data, but this parameter is not as interventions, artificial tears, ergonomic optimization,
readily interpretable with respect to clinical significance as adequate lighting, and frequent rest breaks to reduce
absolute mean differences. Unfortunately, absolute mean CVS. In contrast, the present systematic review found
differences can be derived only when consistent scales allow no benefit with multifocal lenses compared with a
data to be pooled directly. For future RCTs, it is recom- single-vision distance spectacles for managing CVS.
mended that an agreed set of validated questionnaires be Data to draw conclusions about the value of traditional
adopted by trialists to measure symptoms of CVS5 and dry medicines, environmental modifications, ergonomic
eye.97 adjustment, or artificial tears were limited. Although data
Although some controversy remains as to their benefit, available to assess the role of artificial tears in reducing
oral omega-3 fatty acid supplements also have gained CVS symptoms were limited, dry eye is associated
attention with respect to their potential usefulness in commonly with CVS,15 and artificial tears are a prevalent
improving ocular surface health via mechanisms that may dry eye management approach.106,107 Furthermore,
involve modulating tear homeostasis.98,99 Recent data although advising computer users to use the 20-20-20
suggest that omega-3 fatty acid supplements are a com- rule is a popular recommendation,21,105 only 1 RCT84
mon recommendation among eye care practitioners for evaluating this approach was identified. This study84
treating tear dysfunction.100 In the current review, pooled found no benefit in modifying visual fatigue symptoms,
data from 2 studies24,49 provided low certainty evidence and very low evidence for reduced dry eye symptoms,
for reduced dry eye symptoms in individuals with CVS using the 20-20-20 rule compared with a placebo
assigned to oral omega-3 supplements compared with pla- intervention involving as-required water intake.
cebo. Both studies used the Dry Eye Questionnaire and The current review included all interventions investi-
Scoring System score to quantify symptoms,101 which does gated for CVS to provide a comprehensive evidence syn-
not have a defined minimal clinically important difference. thesis that can assist practitioners with clinical
However, converting the observed between-group mean management decisions. It is acknowledged that only
difference in Dry Eye Questionnaire and Scoring System studies published in English were included, anddbecause
score to a percentage improvement (MD per maximum of an insufficient number of trials within each intervention
questionnaire score) indicates a 20% reduction in symp- categorydthe potential reason(s) for heterogeneity could
toms, which has been proposed to be clinically important for not be analyzed using meta-regression. Data extracted from
other validated dry eye symptom scales.102 clinical trial registry entries (n ¼ 2, both in the optical aids
19

category) and included in the quantitative syntheses should future studies, and thus enable enhanced data synthesis in
be interpreted with caution because, compared with results systematic reviews and meta-analyses. This would enable a
from published studies, registry results have not undergone clearer determination of the relative efficacy and safety of
a peer review process. Furthermore, among included trials, interventions to better inform clinical practice. The need for
participant eligibility criteria varied considerably. The a similar set of agreed core outcome measures for evaluation
breadth of populations evaluated in different studies in intervention trials has been identified in other areas of
included healthy individuals, computer users, self-reported ophthalmic research, including dry eye disease,108 cataract
symptomatic computer users, and participants who per- surgery,109 age-related macular degeneration,110 and
formed computer tasks for a specific duration per day. This uveitis.111
variability in study population may affect the generaliz- Further to the findings in this review, the following
ability of the findings. recommendations are pertinent to future CVS RCTs: (1) for
Another consideration is that, to optimize the evidence general considerations relating to the robust design, conduct,
capture, we did not exclude studies judged as having a high and reporting of RCTs, guidance outlined in the Consoli-
risk of bias based on the Cochrane risk-of-bias tool; dated Standards of Reporting Trials should be followed112;
however, risk of bias has been factored into the GRADE (2) symptomatic computer users should be enrolled
evidence certainty assessments for each outcome. There- because they constitute the relevant population for
fore, such certainty assessments apply only to the evidence evaluating CVS interventions; (3) the presence of CVS
in aggregate, and should not be applied to individual should be confirmed in the study population using an
studies making up this aggregate, which may differ sub- agreed validated questionnaire, such as the CVS
stantially in their individual quality). It should be noted questionnaire,7 to provide a consistent measurement of
that 2 studies70,71 from the same research group that baseline symptom severity; (4) when measuring visual
investigated yoga had multiple similarities, including fatigue symptoms (the most common outcome measure
publication year (2016) and recruited same sample size reported in studies included in this review), use of a
(n ¼ 291), and provided participants with similar standardized questionnaire5 is recommended; (5) outcome
interventions and comparators. Hence, the conclusions measure data should be reported in numeric form and
regarding yoga intervention need to be treated with should include all relevant details (e.g., means and
additional caution. In relation to the meta-analyses, the use standard deviations, or similar), either within the main
of SMD restricted our ability to combine both change from article itself or in the supplemental materials; and (6)
baseline and endpoint outcome data. The presentation of experimental studies measuring the change both before
single studies in the subgroup analysis of the forest plots and after computer use ideally should report the
(Figs 3 and 6) could be a potential limitation. In these experimental workstation, type of computer task, task
situations, interpretation of these subgroup analyses must duration, computer screen brightness and contrast, and
be treated with particular caution, and greater weight room temperature and humidity.
should be given to the total effect that represents an
analysis of multiple studies.
Ten of 45 studies included in this review did not Conclusions
report,25,51,56,57,60,68,83 or only partially reported,50,73,75 the
age of the included participants. Some of the outcome This systematic review finds low certainty evidence for a
measures we analyzed may be impacted by age (for possible role of omega-3 fatty acid supplementation in
example, CFF and amplitude of accommodation), and so a managing dry eyes symptoms associated with CVS. Low
limitation of our evaluation is that differences in outcomes certainty evidence was found for oral berry extract supple-
between studies could be the result of unreported mentation not to reduce visual fatigue or dry eye symptoms.
differences between the ages of the cohorts studied, rather Likewise, low certainty evidence for blue light-blocking
than a difference in the effectiveness of the interventions lenses being ineffective in reducing symptoms of visual
per se. Further, with inclusion of unadjusted estimates in fatigue was noted. For other interventions, insufficient evi-
the meta-analysis, any differences in confounding factors dence was found to establish their efficacy or safety with
(such as age) between intervention groups may be a po- any certainty. The review also highlighted a range of
tential source of confounding bias. Finally, a further po- frequent limitations in the design and reporting of RCTs
tential limitation is that quantitative data provided purely in investigating treatments for CVS and provides recommen-
graphical, rather than numeric, form was not considered in dations for how these shortcomings might be addressed in
this review. future studies.
Recommendations for Future Clinical Trials Acknowledgments
This review identified substantial inter-study variations in The authors thank Ji-Hyun Lee and Eve Makrai, Department of
methodology and outcome measure selection. These find- Optometry and Vision Sciences, The University of Melbourne, for
ings indicate that benefit would result from developing a their support in screening of articles and undertaking data extrac-
core outcome set for CVS trials to standardize reporting in tion for this review.
20

Singh et al
Footnotes and Disclosures

Originally received: October 12, 2021. No animal subjects were included in this study.
Final revision: April 26, 2022. Author Contributions:
Accepted: May 4, 2022. Conception and design: Singh, McGuinness, Anderson, Downie
Available online: ---. Manuscript no. OPHTHA-D-21-02036.
1 Analysis and interpretation: Singh, McGuinness, Anderson, Downie
Department of Optometry and Vision Sciences, The University of Mel-
bourne, Parkville, Australia. Data collection: Singh, McGuinness, Anderson, Downie
2
Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, Obtained funding: N/A; Study was performed as part of the authors’ regular
East Melbourne, Australia. employment duties. No additional funding was provided.
3
Melbourne School of Population and Global Health, The University of Overall responsibility: Singh, McGuinness, Anderson, Downie
Melbourne, Melbourne, Australia. Abbreviations and Acronyms:
*Both authors contributed equally as senior authors. CFF ¼ critical flicker-fusion frequency; CI ¼ confidence interval;
Disclosure(s): CVS ¼ computer vision syndrome; D ¼ diopter; GRADE ¼ Grading of
Recommendations, Assessment, Development, and Evaluation;
All authors have completed and submitted the ICMJE disclosures form. MD ¼ mean difference; RCT ¼ randomized controlled trial;
The author(s) have made the following disclosure(s): L.E.D.: Financial support SMD ¼ standardized mean difference; VDT ¼ visual display terminal.
e CooperVision, Azura Ophthalmics, Seqirus, Kedalion Therapeutics
Keywords:
Laura Downie, BOptom, PhD, an editorial board member of this journal, Computer vision syndrome, Digital eye strain, Visual fatigue, Blue light,
was recused from the peer-review process of this article and had no access
Eye, Vision, Digital, Eye strain.
to information regarding its peer-review.
HUMAN SUBJECTS: No human subjects were included in this study. Correspondence:
Laura E. Downie, PhD, Department of Optometry and Vision Sciences, The
Institutional Review Board approval was not required, given the retro-
spective nature of the study, and the requirement for informed consent was University of Melbourne, Parkville, Victoria, Australia 3010. E-mail:
waived. All research adhered to the tenets of the Declaration of Helsinki. ldownie@unimelb.edu.au.
References
1. American Optometric Association. Guide to the Clinical among Massachusetts clerical workers. J Occup Med.
Aspects of Computer Vision Syndrome. St Louis: American 1987;29(2):112e118.
Optometric Association; 1995:1. 12. Wolska A, Switula M. Luminance of the surround and visual
2. Hayes JR, Sheedy JE, Stelmack JA, Heaney CA. Computer fatigue of VDT operators. Int J Occup Saf Ergon. 1999;5(4):
use, symptoms, and quality of life. Optom Vis Sci. 553e581.
2007;84(8):738e744. 13. Wahlstrom J. Ergonomics, musculoskeletal disorders and
3. Ranasinghe P, Wathurapatha WS, Perera YS, et al. Computer computer work. Occup Med (Lond). 2005;55(3):168e176.
vision syndrome among computer office workers in a 14. Patel S, Henderson R, Bradley L, et al. Effect of visual
developing country: an evaluation of prevalence and risk display unit use on blink rate and tear stability. Optom Vis
factors. BMC Res Notes. 2016;9(1):150. Sci. 1991;68(11):888e892.
4. Blehm C, Vishnu S, Khattak A, et al. Computer vision syn- 15. Rosenfield M. Computer vision syndrome: a review of ocular
drome: a review. Surv Ophthalmol. 2005;50(3):253e262. causes and potential treatments. Ophthalmic Physiol Opt.
5. Sheedy JE, Hayes JN, Engle J. Is all asthenopia the same? 2011;31(5):502e515.
Optom Vis Sci. 2003;80(11):732e739. 16. Tosha C, Borsting E, Ridder 3rd WH, Chase C. Accommo-
6. Loh K, Redd S. Understanding and preventing computer dation response and visual discomfort. Ophthalmic Physiol
vision syndrome. Malays Fam Physician. 2008;3(3): Opt. 2009;29(6):625e633.
128e130. 17. Wiggins NP, Daum KM. Visual discomfort and astigmatic
7. Segui Mdel M, Cabrero-Garcia J, Crespo A, et al. A reliable refractive errors in VDT use. J Am Optom Assoc. 1991;62(9):
and valid questionnaire was developed to measure computer 680e684.
vision syndrome at the workplace. J Clin Epidemiol. 18. Thorud HM, Helland M, Aaras A, et al. Eye-related pain
2015;68(6):662e673. induced by visually demanding computer work. Optom Vis
8. Teo C, Giffard P, Johnston V, Treleaven J. Computer vision Sci. 2012;89(4):E452eE464.
symptoms in people with and without neck pain. Appl Ergon. 19. Lin JB, Gerratt BW, Bassi CJ, Apte RS. Short-wavelength
2019;80:50e56. light-blocking eyeglasses attenuate symptoms of eye fatigue.
9. Tauste A, Ronda E, Molina MJ, Segui M. Effect of contact Invest Ophthalmol Vis Sci. 2017;58(1):442e447.
lens use on computer vision syndrome. Ophthalmic Physiol 20. Singh S, Downie LE, Anderson AJ. Do blue-blocking lenses
Opt. 2016;36(2):112e119. reduce eye strain from extended screen time? A double-
10. Sanchez-Brau M, Domenech-Amigot B, Brocal-Fernandez F, masked randomized controlled trial. Am J Ophthalmol.
et al. Prevalence of computer vision syndrome and its rela- 2021;226:243e251.
tionship with ergonomic and individual factors in presbyopic 21. Coles-Brennan C, Sulley A, Young G. Management of digital
VDT workers using progressive addition lenses. Int J Environ eye strain. Clin Exp Optom. 2019;102(1):18e29.
Res Public Health. 2020;17(3):1003. 22. Kee CS, Leung TW, Kan KH, Lam CH. Effects of progres-
11. Rossignol AM, Morse EP, Summers VM, Pagnotto LD. sive addition lens wear on digital work in pre-presbyopes.
Video display terminal use and reported health symptoms Optom Vis Sci. 2018;95(5):457e467.
21

23. Ozawa Y, Kawashima M, Inoue S, et al. Bilberry extract 44. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
supplementation for preventing eye fatigue in video display inconsistency in meta-analyses. BMJ. 2003;327(7414):
terminal workers. J Nutr Health Aging. 2015;19(5):548e554. 557e560.
24. Bhargava R, Kumar P, Arora Y. Short-term omega 3 fatty acids 45. Higgins JP. Commentary: heterogeneity in meta-analysis
treatment for dry eye in young and middle-aged visual display should be expected and appropriately quantified. Int J Epi-
terminal users. Eye Contact Lens. 2016;42(4):231e236. demiol. 2008;37(5):1158e1160.
25. Gangamma MP, Poonam, Rajagopala M. A clinical study on 46. Schwarzer G, Carpenter JR, Rücker G. Small-study effects in
“computer vision syndrome” and its management with Triphala meta-analysis. In: Meta-analysis with R. Switzerland:
eye drops and Saptamrita Lauha. Ayu. 2010;31(2):236e239. Springer; 2015:107e141.
26. Koh S, Inoue R, Sato S, et al. Quantification of accommo- 47. Puhan MA, Schunemann HJ, Murad MH, et al. A GRADE
dative response and visual performance in non-presbyopes Working Group approach for rating the quality of treatment
wearing low-add contact lenses. Cont Lens Anterior Eye. effect estimates from network meta-analysis. BMJ. 2014;349:
2020;43(3):226e231. g5630.
27. Essilor. Crizal Prevencia. Available at: https://www.essilor. 48. Bhargava R, Kumar P, Phogat H, et al. Oral omega-3 fatty
com/en/brands/crizal-prevencia/. Accessed June 9, 2022. acids treatment in computer vision syndrome related dry eye.
28. Zeiss. Blue light protection for your eyes. Available at: https:// Cont Lens Anterior Eye. 2015;38(3):206e210.
www.zeiss.com.au/vision-care/spectacle-lenses-from-zeiss/dura 49. Biswas NR, Nainiwal SK, Das GK, et al. Comparative
vision-blueprotect.html. Accessed February 13, 2021. randomised controlled clinical trial of a herbal eye drop with
29. Bhargava R, Kumar P, Kumar M, et al. A randomized artificial tear and placebo in computer vision syndrome.
controlled trial of omega-3 fatty acids in dry eye syndrome. J Indian Med Assoc. 2003;101(3):208e209, 212.
Int J Ophthalmol. 2013;6(6):811e816. 50. Chatterjee PK, Bairagi D, Roy S, et al. Comparative rando-
30. Optimoz. Eye armourdvision & eye health support. Avail- mised active drug controlled clinical trial of a herbal eye drop
able at: https://www.optimoz.com.au/products/eye-armour. in computer vision syndrome. J Indian Med Assoc.
Accessed October 3, 2021. 2005;103(7):397e398.
31. iHerb. Whitaker Nutrition, Vision Essentials Gold. Available 51. Dhiman KS, Ahuja DK, Sharma SK. Clinical efficacy of
at: https://iherb.com/pr/whitaker-nutrition-vision-essentials- Ayurvedic management in computer vision syndrome: a pilot
gold-120-capsules/101140. Accessed October 10, 2021. study. Ayu. 2012;33(3):391e395.
32. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting 52. Galinsky T, Swanson N, Sauter S, et al. Supplementary
items for systematic reviews and meta-analyses: the PRISMA breaks and stretching exercises for data entry operators: a
statement. PLoS Med. 2009;6(7):e1000097. follow-up field study. Am J Ind Med. 2007;50(7):
33. Higgins JP, Savovic J, Page MJ, et al. Assessing risk of bias 519e527.
in a randomized trial. In: Cochrane Handbook for Systematic 53. Galinsky TL, Swanson NG, Sauter SL, et al. A field study of
Reviews of Interventions. Chichester: John Wiley & Sons; supplementary rest breaks for data-entry operators. Ergo-
2019:205e228. nomics. 2000;43(5):622e638.
34. Boutron I, Page MJ, Higgins JP, et al. Considering bias and 54. Hirayama M, Murat D, Liu Y, et al. Efficacy of a novel moist
conflicts of interest among the included studies. In: Cochrane cool air device in office workers with dry eye disease. Acta
Handbook for Systematic Reviews of Interventions. Chi- Ophthalmol. 2013;91(8):756e762.
chester: John Wiley & Sons; 2019:177e204. 55. Ho WY, Sung CY, Yu QH, Chan CC. Effectiveness of
35. Siegenthaler E, Bochud Y, Bergamin P, Wurtz P. Reading on computerized risk assessment system on enhancing workers’
LCD vs e-Ink displays: effects on fatigue and visual strain. occupational health and attitudes towards occupational
Ophthalmic Physiol Opt. 2012;32(5):367e374. health. Work. 2014;48(4):471e484.
36. Ide T, Toda I, Miki E, Tsubota K. Effect of blue light- 56. Horgen G, Aaras A, Thoresen M. Will visual discomfort
reducing eye glasses on critical flicker frequency. Asia Pac among visual display unit (VDU) users change in develop-
J Ophthalmol (Phila). 2015;4(2):80e85. ment when moving from single vision lenses to specially
37. Kang Y-Y, Wang M-JJ, Lin R. Usability evaluation of e- designed VDU progressive lenses? Optom Vis Sci.
books. Displays. 2009;30(2):49e52. 2004;81(5):341e349.
38. Murata K, Araki S, Yokoyama K, et al. Accumulation of 57. Joshi N, Ujwale R. A clinical study of the effect of tila taila
VDT work-related visual fatigue assessed by visual evoked padabhyanga on eye strain. Int J Res Ayurveda Pharm.
potential, near point distance and critical flicker fusion. Ind 2016;7(2):29e35.
Health. 1996;34(2):61e69. 58. Kawabata F, Tsuji T. Effects of dietary supplementation with
39. Deeks JJ, Higgins JP, Altman DG, Group CSM. Analysing a combination of fish oil, bilberry extract, and lutein on
data and undertaking meta-analyses. In: Cochrane Handbook subjective symptoms of asthenopia in humans. Biomed Res.
for Systematic Reviews of Interventions. Chichester: John 2011;32(6):387e393.
Wiley & Sons; 2019:241e284. 59. Kondo S, Sawano Y. Effects of a dietary supplement con-
40. Cohen J. Statistical Power Analysis for the Behavioral Sci- taining water chestnut extract and lutein on quantified VDT
ences. Burlington: Elsevier Science; 2013. workloaddaffected visual function in healthy middle-aged
41. Higgins J, Deeks J, Altman D. Special topics in statistics. In: adults. A randomized, double‒masked, placebo-controlled,
Cochrane Handbook for Systematic Reviews of Interventions parallel-group intervention study. Japanese Pharmacology
Version 5.1.0 [updated March 2011]. The Cochrane and Therapeutics. 2018;46(5):825e836.
Collaboration; 2011. www.handbook.cochrane.org. 60. Liang T, Yamashita S, Suzuki N, et al. Effect of a bilberry
42. Riley RD, Higgins JP, Deeks JJ. Interpretation of random extract (BILBERON)-containing diet on the improvement
effects meta-analyses. BMJ. 2011;342:d549. of eye fatigue-related symptoms (II)da randomized, double-
43. Borenstein M, Hedges LV, Higgins JP, Rothstein HR. A basic blind, placebo-controlled, parallel-group comparison study.
introduction to fixed-effect and random-effects models for meta- Japanese Pharmacology and Therapeutics. 2017;45(9):
analysis. Res Synth Methods. 2010;1(2):97e111. 1523e1534.
22

Singh et al
61. Maeda-Yamamoto M, Nishimura M, Kitaichi N, et al. 77. Sekikawa T, Kizawa Y, Takeoka A, et al. The effect of
A randomized, placebo-controlled study on the safety and consuming an anthocyanin-containing supplement derived from
efficacy of daily ingestion of green tea (Camellia sinensis L.) bilberry (Vaccinium myrtillus) on eye function: a randomized,
cv. “Yabukita” and “Sunrouge” on eyestrain and blood double-blind, placebo-controlled parallel study. Functional
pressure in healthy adults. Nutrients. 2018;10(5):569. Foods in Health and Disease. 2021;11(3):116e146.
62. Morita Y, Jounai K, Miyake M, et al. Effect of heat-killed 78. Yammouni R, Evans BJ. An investigation of low power
Lactobacillus paracasein KW3110 ingestion on ocular dis- convex lenses (adds) for eyestrain in the digital age
orders caused by visual display terminal (VDT) loads: a (CLEDA). J Optom. 2020;13(3):198e209.
randomized, double-blind, placebo-controlled parallel-group 79. Kosehira M, Machida N, Kitaichi N. A 12-week-long intake
study. Nutrients. 2018;10(8):1058. of bilberry extract (Vaccinium myrtillus L.) improved
63. Nagaki Y, Hayasaka S, Yamada T, et al. Effects of astax- objective findings of ciliary muscle contraction of the eye: a
anthin on accommodation, critical flicker fusion, and pattern randomized, double-blind, placebo-controlled, parallel-group
visual evoked potential in visual display terminal workers. comparison trial. Nutrients. 2020;12(3):600.
Journal of Traditional Medicines. 2002;19(5):170e173. 80. Kizawa Y, Sekikawa T, Kageyama M, et al. Effects of
64. Okamoto K, Munekata M, Ishii I, Najima M. A study for anthocyanin, astaxanthin, and lutein on eye functions: a
evaluating the effect of bilberry extract supplement on eye randomized, double-blind, placebo-controlled study. J Clin
conditions and functionsda randomized, placebo-controlled, Biochem Nutr. 2021;69(1):77e90.
double-blind study. Japanese Pharmacology and Therapeu- 81. Kan J, Wang M, Liu Y, et al. A novel botanical formula
tics. 2018;46(5):869e881. improves eye fatigue and dry eye: a randomized, double-
65. Okamoto K, Kushima M, Ishii I, Yamada T. Impacts of the blind, placebo-controlled study. Am J Clin Nutr.
intake of a dietary supplement containing bilberry extract on 2020;112(2):334e342.
improving eye functions and conditions caused by visual 82. Del Mar Segui-Crespo M, Ronda-Perez E, Yammouni R,
display terminal loadda randomized, double-blind, parallel- et al. Randomised controlled trial of an accommodative
group, placebo-controlled study. Japanese Pharmacology support lens designed for computer users. Ophthalmic
and Therapeutics. 2019;47(3):503e515. Physiol Opt. 2022;42(1):82e93.
66. Park CY, Gu N, Lim CY, et al. The effect of Vaccinium 83. Dabrowiecki A, Villalobos A, Krupinski EA. Impact of blue
uliginosum extract on tablet computer-induced asthenopia: light filtering glasses on computer vision syndrome in radi-
randomized placebo-controlled study. BMC Complement ology residents: a pilot study. J Med Imaging (Bellingham).
Altern Med. 2016;16:296. 2020;7(2):022402.
67. Robertson MM, Ciriello VM, Garabet AM. Office ergo- 84. Alrasheed SH, Alghamdi WM. Impact of an educational
nomics training and a sit-stand workstation: effects on intervention using the 20/20/20 rule on computer vision
musculoskeletal and visual symptoms and performance of syndrome. African Vision and Eye Health. 2020;79(1):
office workers. Appl Ergon. 2013;44(1):73e85. 1e6.
68. Sawant DP, Parlikar GR, Binorkar SV. Efficacy of Triphala 85. Kollbaum P. The impact of soft contact lens attributes on
Ghrita Netratarpan in computer vision syndrome. Interna- symptoms associated with digital eye strain in symptomatic
tional Journal of Research in Ayurveda and Pharmacy. soft contact lens wearers. Last updated January 7, 2020.
2013;4(2):244e248. Available at: https://clinicaltrials.gov/ct2/show/NCT03585
69. Stringham JM, Stringham NT, O’Brien KJ. Macular carot- 790. Accessed August 20, 2020.
enoid supplementation improves visual performance, sleep 86. Iacono D. Connecting contact lenses and digital technology.
quality, and adverse physical symptoms in those with high Last updated September 11, 2019. Available at: https://clin-
screen time exposure. Foods. 2017;6(7):47. icaltrials.gov/ct2/show/NCT02921087. Accessed August 20,
70. Telles S, Naveen KV. Effect of yoga on somatic indicators of 2020.
distress in professional computer users. Med Sci Monit. 87. Robertson J, Connor Jr C. The efficacy of computer eye drops
2006;12(10):LE21eLE22. compared to artificial tears. Optom Vis Sci. 2002;79(12):255.
71. Telles S, Naveen KV, Dash M, et al. Effect of yoga on self- 88. Deinema LA, Vingrys AJ, Wong CY, et al. A randomized,
rated visual discomfort in computer users. Head Face Med. double-masked, placebo-controlled clinical trial of two forms
2006;2:46. of omega-3 supplements for treating dry eye disease.
72. Wiggins NP, Daum KM, Snyder CA. Effects of residual Ophthalmology. 2017;124(1):43e52.
astigmatism in contact lens wear on visual discomfort in VDT 89. Saltaji H, Armijo-Olivo S, Cummings GG, et al. Influence of
use. J Am Optom Assoc. 1992;63(3):177e181. blinding on treatment effect size estimate in randomized
73. Yagi A, Fujimoto K, Michihiro K, et al. The effect of lutein controlled trials of oral health interventions. BMC Med Res
supplementation on visual fatigue: a psychophysiological Methodol. 2018;18(1):42.
analysis. Appl Ergon. 2009;40(6):1047e1054. 90. Lundh A, Lexchin J, Mintzes B, et al. Industry sponsorship
74. Yamashita SI, Suzuki N, Yamamoto K, et al. Effects of and research outcome. Cochrane Database Syst Rev. 2017;2:
MaquiBright() on improving eye dryness and fatigue in MR000033.
humans: a randomized, double-blind, placebo-controlled trial. 91. Singh S, Anderson AJ, Downie LE. Insights into Australian
J Tradit Complement Med. 2019;9(3):172e178. optometrists’ knowledge and attitude towards prescribing
75. Zhang M, Bi LF, Ai YD, et al. Effects of taurine supple- blue light-blocking ophthalmic devices. Ophthalmic Physiol
mentation on VDT work induced visual stress. Amino Acids. Opt. 2019;39(3):194e204.
2004;26(1):59e63. 92. Downie LE. Blue-light filtering ophthalmic lenses: to pre-
76. Guillon M, Maissa C, Pouliquen P, Delval L. Effect of scribe, or not to prescribe? Ophthalmic Physiol Opt.
povidone 2% preservative-free eyedrops on contact lens 2017;37(6):640e643.
wearers with computer visual syndrome: pilot study. Eye 93. Lawrenson JG, Hull CC, Downie LE. The effect of blue-light
Contact Lens. 2004;30(1):34e39. blocking spectacle lenses on visual performance, macular
23

health and the sleep-wake cycle: a systematic review of the 103. Jones L, Downie LE, Korb D, et al. TFOS DEWS II man-
literature. Ophthalmic Physiol Opt. 2017;37(6):644e654. agement and therapy report. Ocul Surf. 2017;15(3):575e628.
94. Lee D-S, Ko Y-H, Shen I-H, Chao C-Y. Effect of light 104. Chang CH, Tseng PT, Chen NY, et al. Safety and tolerability
source, ambient illumination, character size and interline of prescription omega-3 fatty acids: a systematic review and
spacing on visual performance and visual fatigue with elec- meta-analysis of randomized controlled trials. Prostaglandins
tronic paper displays. Displays. 2011;32(1):1e7. Leukot Essent Fatty Acids. 2018;129:1e12.
95. Shen I-H, Shieh K-K, Chao C-Y, Lee D-S. Lighting, 105. Sheppard AL, Wolffsohn JS. Digital eye strain: prevalence,
font style, and polarity on visual performance and visual fa- measurement and amelioration. BMJ Open Ophthalmol.
tigue with electronic paper displays. Displays. 2009;30(2): 2018;3(1):e000146.
53e58. 106. Downie LE, Keller PR, Vingrys AJ. An evidence-based
96. Yan K, Rosenfield M. Digital eye strain and the critical flicker analysis of Australian optometrists’ dry eye practices.
fusion frequency. Optom Vis Sci. 2022;99(3):253e258. Optom Vis Sci. 2013;90(12):1385e1395.
97. Sakane Y, Yamaguchi M, Yokoi N, et al. Development and 107. Downie LE, Rumney N, Gad A, et al. Comparing self-reported
validation of the Dry Eye-Related Quality-of-Life Score optometric dry eye clinical practices in Australia and the
questionnaire. JAMA Ophthalmol. 2013;131(10):1331e1338. United Kingdom: is there scope for practice improvement?
98. Downie LE, Ng SM, Lindsley KB, Akpek EK. Omega-3 and Ophthalmic Physiol Opt. 2016;36(2):140e151.
omega-6 polyunsaturated fatty acids for dry eye disease. 108. Wang MTM, Craig JP. Core outcome sets for clinical trials in
Cochrane Database Syst Rev. 2019;12:CD011016. dry eye disease. JAMA Ophthalmol. 2018;136(10):
99. Britten-Jones AC, Kamel JT, Roberts LJ, et al. Investigating 1180e1181.
the neuroprotective effect of oral omega-3 fatty acid sup- 109. Mahmud I, Kelley T, Stowell C, et al. A proposed minimum
plementation in type 1 diabetes (nPROOFS1): a randomized standard set of outcome measures for cataract surgery. JAMA
placebo-controlled trial. Diabetes. 2021;70(8):1794e1806. Ophthalmol. 2015;133(11):1247e1252.
100. Zhang AC, Singh S, Craig JP, Downie LE. Omega-3 fatty 110. Krezel AK, Hogg R, Lohfeld L, et al. Core outcomes for
acids and eye health: opinions and self-reported practice be- geographic atrophy trials. Br J Ophthalmol. 2020;104(9):
haviors of optometrists in Australia and New Zealand. Nu- 1196e1202.
trients. 2020;12(4):1179. 111. Tallouzi MO, Mathers JM, Moore DJ, et al. COSUMO: study
101. Kumar P, Bhargava R, Kumar M, et al. The correlation of protocol for the development of a core outcome set for effi-
routine tear function tests and conjunctival impression cacy and effectiveness trials in posterior segment-involving
cytology in dry eye syndrome. Korean J Ophthalmol. uveitis. Trials. 2017;18(1):576.
2014;28(2):122e129. 112. Schulz KF, Altman DG, Moher D; CONSORT Group.
102. Miller KL, Walt JG, Mink DR, et al. Minimal clinically CONSORT 2010 statement: updated guidelines for
important difference for the ocular surface disease index. reporting parallel group randomised trials. BMJ. 2010;340:
Arch Ophthalmol. 2010;128(1):94e101. c332.
24

View publication stats

OPHTHA12065S200

Uploaded by

Copyright:

Available Formats

OPHTHA12065S200

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

OPHTHA12065S200

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Interventions for the Management of Computer Vision Syndrome: A

Article in Ophthalmology · May 2022

Sumeer Singh Laura Downie

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Supplemental material available at www.aaojournal.org.

ª 2022 by the American Academy of Ophthalmology https://doi.org/10.1016/j.ophtha.2022.05.009 1

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

Thorud et al18 studied the association between extraocular Eligibility Criteria

Methods Data Extraction and Management

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

0.11 unit higher (0.14 unit lower to NCT0292108786

Ophthalmology Volume -, Number -, Month 2022

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

difference between intervention

Interventions for Computer Vision Syndrome

with the greatest number of studies considered to have a high

progressive addition (multifocal) spectacle lenses,22,56,82 single-

1. Primary outcome. Eight studies measured subjective visual

single-vision distance correction (Fig 3; 3 studies, 262

multifocal compared with single-vision lenses (mean difference

In addition, Wiggins et al72 provided very low certainty evi-

blocking lenses (Table S6).

lenses compared with both low blue-light blocking and

2. Secondary outcomes. Three studies considered dry eye

vision contact lenses after 1 week (44 participants; MD, 0.70

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

Interventions for Computer Vision Syndrome

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

Ophthalmology Volume -, Number -, Month 2022

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

arms: MD relative to placebo, 0.40 unit

Interventions for Computer Vision Syndrome

Certainty of the Evidence Estimate (95% Conﬁdence Interval)

vs. placebo Sunrouge, Yabukita, and placebo green

Ophthalmology Volume -, Number -, Month 2022

adverse events in 2 contact lens-wearing groups after 1 week, in

difference in the risk of an adverse event between multifocal and

single-vision lenses (Fig S1, available at www.aaojournal.org; 3

n ¼ 166; and maqui berry, n ¼ 174) relative to a placebo.

1. Primary outcomes. Seven studies evaluated subjective vi-

certainty evidence for no improvement in visual fatigue score

studies; 322 participants; SMD, e0.27 units; 95% CI, e0.70 to

with oral berry extract supplementation relative to placebo at the

CI, e1.39 to 0.69 Hz; I2 ¼ 29%; P ¼ 0.51).

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:32 am ce

Very low certainty evidence was found for no difference in

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:33 am ce

(F: 8:00 AMe5:00 PM/day) SVL (F: 8:00 AMe5:00 PM/

Ophthalmology Volume -, Number -, Month 2022

Intervention(s) (Dose; Comparator (Dose; Recruited Intervention

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:33 am ce

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:33 am ce

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:33 am ce

REV 5.6.0 DTD OPHTHA12065_proof 2 July 2022 12:33 am ce