Abstract
Citation information: Rosenfield M. Computer vision syndrome: a review of ocular causes and potential treatments. Ophthalmic Physiol Opt 2011, 31, 502–515. doi: 10.1111/j.1475-1313.2011.00834.x
Computer vision syndrome (CVS) is the combination of eye and vision problems associated with the use of computers. In modern western society the use of computers for both vocational and avocational activities is almost universal. However, CVS may have a significant impact not only on visual comfort but also occupational productivity since between 64% and 90% of computer users experience visual symptoms which may include eyestrain, headaches, ocular discomfort, dry eye, diplopia and blurred vision either at near or when looking into the distance after prolonged computer use. This paper reviews the principal ocular causes for this condition, namely oculomotor anomalies and dry eye. Accommodation and vergence responses to electronic screens appear to be similar to those found when viewing printed materials, whereas the prevalence of dry eye symptoms is greater during computer operation. The latter is probably due to a decrease in blink rate and blink amplitude, as well as increased corneal exposure resulting from the monitor frequently being positioned in primary gaze. However, the efficacy of proposed treatments to reduce symptoms of CVS is unproven. A better understanding of the physiology underlying CVS is critical to allow more accurate diagnosis and treatment. This will enable practitioners to optimize visual comfort and efficiency during computer operation.
Introduction
The use of computers and digital electronic devices for both vocational and non-vocational activities including e-mail, internet access and entertainment is almost universal in modern Western society. A recent estimate of internet usage by continent ranged from 77.4% of the population of North America to 10.9% of Africa, with an estimated 1 966 514 816 users worldwide (or 28.7% of the world’s population) (http://www.internetworldstats.com/stats.htm).
The viewing of digital electronic screens is no longer restricted to desktop computers located in the workplace. Today’s visual requirements may include viewing laptop and tablet computers, electronic book readers, smartphones and other electronic devices either in the workplace, at home or in the case of portable equipment, in any location. Furthermore, computer use is not restricted to adults. A recent investigation of over 2000 American children between 8 and 18 years of age reported that in an average day they spend approximately 7.5 h using entertainment media, 4.5 h watching TV, 1.5 h on a computer and over an hour playing video games.1 Some screen sizes may necessitate very small text which the observer frequently positions at a closer viewing distance than had previously been adopted for hard copy printed materials. These increased visual demands may give rise to a variety of symptoms which have been termed computer vision syndrome (CVS).
The American Optometric Association defines CVS as the combination of eye and vision problems associated with the use of computers. These symptoms result from the individual having insufficient visual capabilities to perform the computer task comfortably (http://www.aoa.org/x5374.xml). In a review of CVS, Thomson2 indicated that up to 90% of computer users may experience visual symptoms including eyestrain, headaches, ocular discomfort, dry eye, diplopia and blurred vision either at near or when looking into the distance after prolonged computer use. It is unclear whether this number has increased, given the increased use of electronic displays today. Further, Rossignol et al.3 reported that the prevalence of visual symptoms increased significantly in individuals who spent more than 4 h daily working on video display terminals (VDTs).
Asthenopia is a major complaint in subjects with CVS. The results of a 2008 questionnaire returned by over 400 computer operators in India revealed asthenopic symptoms in 46.3% of subjects.4 Similarly, a survey of 212 bank workers in Italy found asthenopic symptoms in 31.9% of the subjects, though it is worthwhile noting that this percentage was calculated after 87 subjects were excluded due to uncorrected hyperopia, undercorrected astigmatism, or overcorrected myopia, because the investigators wanted to investigate only subjects ‘without organic visual disturbances’.5 A higher prevalence was found in a study of 35 Mexican computer terminal operators where 68.5% of the subjects experienced symptoms.6 An Australian study of over 1000 computer workers found 63.4% reported symptoms with uncontrolled conditions; this number was reduced to 25.2% when an optimized, ergonomic desk and frequent work breaks were provided.7 It is unclear whether asthenopia during computer use is associated with age,4,7–9 although the prevalence does seem to be higher in females.10–14
In a review of asthenopia, Sheedy et al.15 noted that symptoms commonly associated with this diagnostic term included eyestrain, eye fatigue, discomfort, burning, irritation, pain, ache, sore eyes, diplopia, photophobia, blur, itching, tearing, dryness and foreign-body sensation. While investigating the effect of several symptom-inducing conditions on asthenopia, the authors determined that two broad categories of symptoms existed. The first group, termed external symptoms, included burning, irritation, ocular dryness and tearing, and was related to dry eye. The second group, termed internal symptoms, included eyestrain, headache, eye ache, diplopia and blur, and is generally caused by refractive, accommodative or vergence anomalies. Accordingly, the authors proposed that the underlying problem could be identified by the location and/or description of symptoms.
It is important to identify whether symptoms (both internal and external, see above) are specific to computer operation, or are simply a manifestation of performing a sustained near-vision task for an extended period of time. If no physiological or subjective differences exist when patients view materials either on electronic screens or in printed form, then there would be little justification for special attention being paid to the visual demands encountered during computer operation. The electronic screen would simply represent another visual target. However, there is evidence that the two forms of target presentation are not equivalent. For example, Sheedy et al.16 compared the performance of an editing task when the material was either presented on a VDT or in hard copy form. They observed that subjects made fewer errors and performed the task quicker with the hard copy presentation. Similar findings of fewer errors when viewing printed materials have also been reported in other studies.17–19 More recently, Chu et al.20 compared ocular symptoms immediately following a sustained near-task viewed either on a computer monitor or in hard copy format. Identical text was used in the two sessions, which was matched for size and contrast. In addition, target viewing angle and luminance were similar for the two conditions. Significant differences in median symptom scores were found with regard to blurred vision during the task and the mean symptom score. In both cases, symptoms were higher during computer use. Accordingly, it appears that the symptoms associated with CVS do not result from simply performing a near-vision task for a prolonged period of time. Even when viewing a modern flat panel monitor, subjects reported significantly greater blur during the computer task, when compared with a hard-copy printout of the same material.
In addition to the discomfort experienced during computer operation, symptoms of CVS may also have a significant economic impact. As noted above, symptoms can increase the number of errors made during a computer task as well as necessitating more frequent breaks. Musculoskeletal injuries associated with computer use may account for at least half of all reported work-related injuries in the USA.21 Indeed, Spekléet al.22 noted that conservative estimates of the cost of musculoskeletal disorders to the United States economy as reported in 2001, when measured by compensation costs, lost wages and reduced productivity were between 45 and 54 billion dollars annually or 0.8% of gross domestic product. Further, the prevalence of neck, shoulder and arm symptoms in computer workers may be as high as 62%.23 In addition to productivity costs, it was estimated in 2002 that employers in the USA paid approximately $20 billion annually in workers compensation resulting from work-related musculoskeletal disorders.24 When considering CVS specifically, Daum et al.25 estimated that provision of an appropriate refractive correction alone could produce at least a 2.5% increase in productivity. This would result in a highly favourable cost-benefit ratio to an employer who provided computer-specific eyewear to their employees. Accordingly, it is clear that the economic impact of CVS is extremely high, and minimizing symptoms that reduce occupational efficiency will result in substantial financial benefit. It should also be noted that both national and international regulations have been issued with regard to health and safety requirements for workers using VDTs to minimize these disorders [e.g. European Council directive 90/270/EEC, the United Kingdom Health and Safety (Display Screen Equipment) (DSE) regulations and the Australian Occupational Health & Safety Act of 2000].
Effect of uncorrected refractive error
Given the need to achieve and maintain clear and single vision of relatively small targets throughout the computer task, it is important that the retinal image be focused appropriately. Thus, spherical hyperopia and high myopia should be corrected to reduce the ocular stimulus to accommodation and minimize blur. Additionally, the correction of small astigmatic errors may also be important to reduce symptoms of CVS. In two similar experiments, Wiggins and Daum26 and Wiggins et al.27 examined the effects of uncorrected astigmatism while reading material from a computer screen. In both studies the authors observed that the presence of 0.50–1.00 D of uncorrected astigmatism produced a significant increase in symptoms. Interestingly, Wiggins et al.27 tested subjects with up to 1 D of residual astigmatism who were corrected with spherical soft contact lenses. This is a common clinical practice. The residual uncorrected astigmatism produced a significant increase in symptoms during the computer task. Accordingly, the authors suggested that symptoms could be reduced either by fitting these individuals with toric contact lenses, or alternatively by using a spectacle overcorrection to correct the residual astigmatism during computer operation.
A recent study in our laboratory (paper in preparation) recorded ocular symptoms (both internal and external) using a written questionnaire immediately after a sustained period of reading from a computer monitor either through the habitual distance refractive correction or with a supplementary −1.00 or −2.00 D oblique cylinder added over these lenses. Additionally, the distance correction condition was repeated on two occasions in 12 subjects to assess the repeatability of the symptom questionnaire. The results showed no significant difference between the habitual correction conditions, but the change from 1 to 2 D of induced astigmatism produced a significant increase in post-task symptoms. These results are shown in Figure 1. The presence of uncorrected oblique astigmatism will reduce visual acuity significantly. The increase in target blur will make performing the task more difficult, thereby leading to an increase in symptoms such as eyestrain and headache. Therefore, the correction of astigmatic refractive errors may be important in minimizing symptoms associated with CVS...
Refractive and accommodative anomalies
As noted earlier, the presence of uncorrected ametropia may lead to an increase in symptoms. Given that individuals may spend many hours (often continuously) viewing electronic screens, it is important that they are able to maintain a clear image of the target over time. There is little evidence to support the proposal that the accommodative demands of the VDT differ from viewing printed materials at the same distance and gaze angle. However, the presence of any refractive or accommodative anomaly (e.g. accommodative infacility or insufficiency) could impact upon the patient’s level of visual comfort during the task. In examining patients with CVS, the following clinical parameters should be assessed [with all near testing being performed at the distance(s) at which the electronic screen(s) are positioned]:
- Best corrected visual acuity
- Refractive error (including binocular balancing)
- Accommodative error (lag) at the appropriate working distance
- Monocular and binocular amplitude of accommodation
- Monocular and binocular accommodative facility
- Negative and positive relative accommodation
Patients with accommodative anomalies may benefit from measures to improve the accuracy and dynamics of their accommodative response, including vision therapy and/or the provision of lenses to provide a clear image of the target at the required viewing distance and gaze angle. If adjustments can be made to optimize the design of the workstation, these should also be discussed with the patient.
In addition, patients should be advised regarding their working times. Fixation on any near object for a sustained period of time, whether a computer screen or printed material may lead to asthenopia. Indeed, Henning et al. compared computer workers typing performance at baseline, when they were allowed three 30 s breaks plus a 3 min break each hour, and a rest break plus exercise condition where stretching exercises were introduced during the breaks. A 5% and 15% improvement in productivity was observed for the breaks and breaks plus exercise conditions, respectively. Accordingly, it seems reasonable that any patient should be advised to take breaks and to look into the distance periodically in order to reduce the accommodation and vergence responses.
Vergence anomalies
It has been demonstrated that subjects reading text on a computer were most symptomatic when they converged accurately on the screen, i.e. having ortho associated phoria (AP), when compared with individuals having exo AP who were less symptomatic (see Figure 3). Given this finding, one might conjecture whether intentional correction of a subject’s AP to an exo posture would reduce asthenopic symptoms after a computer task. Accordingly, our laboratory compared post-task symptoms immediately following a continuous 20 min reading task from a desktop computer monitor at a viewing distance of 50 cm. Each subject (n = 40) attended for two trials. In the first session, subjects wore the amount of prism corresponding to their AP at the computer test distance over their refractive correction. In the second trial, subjects were given 3Δ more base-out prism than their measured AP in order to induce an AP of 3Δ exo. No significant difference in total ocular symptom score between the two groups was observed following these two conditions. However, five individuals showed a significant preference for the ortho condition, whereas nine subjects showed a marked preference for the 3Δ exo condition. Further analysis indicated differences in binocular accommodative parameters between these two subgroups. The group whose symptoms were reduced with exo AP had significantly lower negative relative accommodation (NRA) and a higher accommodative response (i.e. smaller lag) when compared with the ortho AP preferred group. No significant difference in either distance or near heterophoria was observed between the two groups. Thus a subgroup of patients may exist whose symptoms of CVS can be alleviated by creating exo AP. This proposal should be examined further in a larger population.
Any vergence anomaly which would cause difficulty with maintaining clear and single vision of printed text at near (e.g. uncompensated heterophoria, convergence excess or insufficiency or vergence infacility) is likely to give rise to symptoms during sustained viewing of an electronic screen at near. Accordingly, when examining patients with CVS, the following clinical vergence parameters should be assessed [with all near testing being performed at the distance(s) at which the electronic screen(s) are positioned]:
- Near point of convergence
- Near heterophoria
- Horizontal and vertical fixation disparity and/or associated phoria.
- Vergence facility
- Vergence ranges (negative and positive relative vergence)
- Stereopsis
- AC/A and CA/C ratios.
Dry eye
As noted previously, the presence of dry eye may play a significant role in the aetiology of CVS. Computer use has been associated with both a reduced rate of blinking and a high number of incomplete blinks when compared with viewing hard copy materials. Additionally, the environments where computers are located often have low ambient humidity and forced-air heating or air conditioning which may exacerbate symptoms of dry eye. Accordingly, practitioners should consider both patient education and a range of therapies available to attenuate this condition.
Dry eye therapies which have been proposed to minimize symptoms of CVS include the use of lubricating drops, ointments and topical medications for blepharitis or allergic conditions. Additionally, blink training to increase the blink rate during computer use, as well as changes in ambient humidity, hydration (drinking more water) and redirection of heating and air conditioning vents have all been proposed.
The benefit of many of these therapies for minimizing CVS symptoms is unproven. For example, Acosta et al. observed that topical instillation of an elastoviscous solution did not produce a significant change in the computer-induced reduction in blink rate. Mean blink rates for the computer condition with and without instillation of the elastoviscous solution were 6.4 and 6.1 blinks min−1, respectively. Accordingly, it is uncertain whether the use of lubricating or rewetting drops will indeed reduced CVS symptoms. With regard to increasing the blink rate, Portello and Rosenfield compared post-task symptoms when subjects were either allowed to blink voluntarily or when the blink rate was consciously increased using a metronome during computer use. Although a significantly increased blink rate was recorded in the metronome condition (23.5 vs 11.3 blinks min−1 in the control session), no significant difference in post-task CVS symptoms was found either in the entire population tested (n = 23) or in those subjects reporting the highest symptom scores in the control condition. Furthermore, several subjects stated that increased conscious blinking interfered with their ability to perform the task satisfactorily, which may limit the practicality of this advice.
Furthermore, Acosta et al. noted that blowing an air stream onto the face while subjects were playing a computer game did not produce a significant change in blink rate. Accordingly, to date there appears to be little experimental evidence to support many of the therapeutic interventions that have been proposed. Further work is required to determine what aspects of dry eye treatments will indeed reduce CVS symptoms.
Conclusions
As noted above, the use of electronic devices to view small type for many hours, frequently at close working distances, has become commonplace in modern society in patients of all ages. Many individuals use multiple devices such as a desktop and laptop computer as well as one or more hand-held devices. These present a variety of visual demands that are significantly different from those of printed materials in terms of working distances, gaze angles and text sizes. It is no longer reasonable to assume that a patient will read text at a viewing distance of approximately 40cm with their eyes depressed. Accordingly, a significant change in both optometric testing methods and the design of ophthalmic lenses (particularly for the correction of presbyopia) will probably be required.
Given that the prevalence of symptoms (including eyestrain, headaches, ocular discomfort, dry eye, diplopia and blurred vision) may be as high as 90%, it is likely that an increasing number of patients will present for eye examinations due to symptoms associated with CVS. Practitioners need to consider what are appropriate examination procedures and treatment regimens for these individuals. Near testing at a single distance and gaze angle such as is commonly employed when a nearpoint card is positioned in the primary position at a viewing distance of 40cm is not adequate. The assessment of oculomotor functions at multiple viewing distances and gaze angles may be required. One should note that this cannot be achieved when the patient is viewing through a phoropter, and nearpoint testing in free space, with the patient wearing a correction mounted in a trial frame is required.
In addition, prescribing routines may need to be reconsidered. For example, small refractive errors (such as astigmatism between 0.50 and 1.00D), which might have been left uncorrected in the past (particularly in contact lens wearers), should be corrected in a patient who is viewing an electronic screen for an extensive period of time. Similarly, instances of low to moderate oculomotor anomalies or cases of dry eye, which might previously have been left uncorrected may be of sufficient magnitude to cause significant symptoms when combined with prolonged computer use.
It is worth noting that the symptoms of CVS associated with accommodation and vergence disorders do seem, in most cases, to be a result of viewing a visually demanding near target for an extended period of time and not specific to the electronic monitor. In contrast, symptoms of dry eye do appear to be directly related to computer use due to the position of the monitor (producing increased corneal exposure), reduced blink rate, increased partial blinking and other environmental factors. Further research is required to determine the efficacy of dry eye treatments in reducing symptoms of CVS.
Given the remarkably high number of hours per day that many (or perhaps most) individuals now spend viewing small text on electronic screens at close working distances and varying gaze angles, it is incumbent upon all eye care practitioners to have a good understanding of the symptoms associated with, and the physiology underlying CVS. As modern society continues to move towards greater use of electronic devices for both work and leisure activities, it seems likely that the visual demands that these place upon our patients will only continue to increase. An inability to satisfy these visual requirements could present significant lifestyle difficulties for patients.
Acknowledgement
I would like to acknowledge the assistance of the following colleagues in this research: Yuliya Bababekova, Jaclyn Benzoni, Christina Chu, Juanita Collier, Rae Huang, Jennifer Hue, Marc Lay, Joan Portello, Daniel Recko, Elizabeth Wickware and Abraham Zuckerbrod.