Introduction

Computer vision syndrome (CVS) is defined as “a complex of eye and vision problems related to near work experienced during computer use”1. Visual fatigue (VF) and digital eye strain (DES) terms are also used for CVS, reflecting the different digital devices related to potential health problems2. Symptoms related to CVS can be classified as visual, ocular, and extraocular symptoms3. Visual symptoms include blurred vision, visual fatigue or discomfort, and diplopia4,5,6,7. Ocular symptoms include dry eye disease, redness, eye strain, and irritation1,8,9. Extraocular symptoms include headache and shoulder, neck, and back pain3,4,10,11,12,13,14.

Individuals spend more time on electronic devices such as computers, laptops, smartphones, tablets, and e-readers, which contribute to CVS15. Children are also affected in CVS, as they spend many hours using electronic devices for schoolwork, playing video games, and sending and receiving text messages15. However, the use of these devices even for 3 h/day can lead to the development of CVS3.

The massive growth of digital devices has become an integral part of daily life, and millions of individuals of all ages are at risk of CVS16,17,18. In developed nations, engagement with digital devices has increased substantially in recent years across all age groups19,20,21,22. Moreover, digital device use has increased in developing countries, resulting in a high burden of CVS due to low accessibility, low utilization of personal protective equipment, and limited break time while using electronic devices. CVS is a major public health problem leading to occupational hazard, an increased error rate, impaired visual abilities, reduced productivity, and low job satisfaction23,24.

A review of the literature showed that factors associated with CVS can be classified as personal factors, which include poor sitting position, inappropriate eye-to-screen distance, insufficient working procedures, improper viewing angle and distances, age, medical diseases, and long duration of computer usage. The environment and computer factors such as improper workstations, poor lighting, contrast, and resolution rooms, slow refresh rate, glare of the display, excessive screen brightness, and imbalance of light between the computer screen and surrounding working room5,10,25,26,27,28.

Modern digital technology markedly influences the daily activities and lifestyles of people4,7. CVS has an effect on reduced productivity and visual and musculoskeletal impairment and a negative impact on cadiac rhythms and sleep patterns4,7,13,29,30. Although CVS is becoming a major public health problem, less emphasis is given, particularly in developing countries. There are primary studies on different continents; however, there are inconsistent findings in prevalence among the primary studies. Therefore, this systematic review aimed to estimate the pooled prevalence of computer vision syndrome.

Methods

Protocol and registration

This systematic review and meta-analysis was registered on PROSPERO with registration number CRD42022325167. Available at: https://www.crd.york.ac.uk/prospero/#myprospero.

Search strategies

The systematic review was developed using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines31, and the review procedure was reported using the PRISMA-P 2009 checklist32 (supplementary file 1). Published and unpublished studies were searched in databases such as Medline/PubMed, CINAHL, and Google Scholar from December 1 to April 9/2022. MeSH terms and entry terms were used to search studies from databases, and modifications were made based on the type of database (supplementary file 2).

Eligibility criteria

Inclusion criteria

  • The following criteria were considered to include studies:

Study area.

  • Anywhere

Study scope.

  • Studies that report the prevalence of CVS and its associated factors

  • Studies that report the prevalence of CVS

  • “Both community- and facility-based studies”

  • Quantitative results, if the study reported both qualitative and quantitative results

Study design.

  • Observational study designs, including cross-sectional and cohort study designs

Language.

  • English

Population.

  • All population groups

Publication year.

  • No restriction

Exclusion criteria

Studies were excluded if:

  • Other than English

  • Studies that did not report specific outcomes (prevalence) of CVS

  • No full-text article following email contact to the corresponding authors

  • Qualitative studies

  • Letters, conference abstracts, case reports, and reviews,

CoCoPop/PEO

Condition: computer vision syndrome.

Context: worldwide.

Population: All population groups.

Outcome/context: The primary outcome of the study was the pooled prevalence of CVS. The prevalence of CVS was considered when the studies reported the overall prevalence of CVS or either of CVS syndromes (blurred vision, eye strain/fatigue, discomfort, diplopia, dry eye disease, redness, irritation, headache, shoulder, neck, and back pain) in the primary studies.

Study selection

Endnote reference manager software33 was used to organize and remove duplicates, irrelevant titles, and abstracts. Duplicate studies were removed. An assessment of studies using the title and abstract was performed, and irrelevant titles and abstracts were removed. Study selection was performed independently by two reviewers (EW and AK). The selection procedures of the studies were presented using a PRISMA diagram.

Quality assessment

A full-text review of studies was performed before the inclusion of studies in the final meta-analysis using “The Joanna Briggs Institute Meta-Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI)”34 quality appraisal tool. The components of quality assessment include study setting, outcome and explanatory variable measurements, clear inclusion criteria, measurement criteria used, participants’ description, and valid statistical analysis. Independent quality assessment of the studies was reviewed by EW and AK, and studies with a quality score of 50% and above were included in the final systematic review and meta-analysis. Disagreement during quality assessment among reviewers was resolved with discussion. In addition, cross-referencing of the included articles was performed.

Data extraction

Independent data extraction was performed by the authors (EW and AK) using a pilot-tested data extraction Microsoft Office Excel sheet. The data extraction sheet elements included publication year, authors’ names, study design, country, sample size, response rate, prevalence and study subjects. Discrepancies were resolved by discussion between the authors (EW and AK). Contact with the corresponding authors of the studies was made for incomplete data, and the study was excluded if there was no response.

Data analysis

The extracted Excel data were imported into STATA version 14 for analysis. A narrative description and summary characteristics of the included studies were reported in tables and graphs. A random-effects model meta-analysis35 was used to estimate the overall effect size, and the results were presented using a forest plot.

The heterogeneity of studies was assessed by the I2 statistic36. I2 statistics of 25, 50 and 75% showed low, moderate and high heterogeneity, respectively, with p < 0.05. Publication bias was assessed using visual observation of the funnel plot37 and Egger’s test at p < 0.0538. To identify the sources of heterogeneity among the studies, subgroup analysis and meta-regression39 were performed based on country and sample size. Moreover, sensitivity analysis was performed to assess the effect of the study on the overall effect size.

Results

A total of 735 articles were retrieved using electronic database searches: PubMed, Google Scholar, and CINHAL. Seventy-seven articles were excluded due to duplication, and 559 articles were excluded because they were not related to the title and abstract. Ninety-nine full-text articles were assessed for quality eligibility, and 57 articles were excluded based on the quality appraisal tool because they were irrelevant, had no full text available, or were duplicates. Three articles were identified through a cross-reference search of the included studies. Finally, 45 articles were included in the systematic review and meta-analysis (Fig. 1).

Figure 1
figure 1

PRISMA flow diagram studies screening, and selection on computer vision syndrome, 2022.

Full size image

Characteristics of the included studies

A total of 45 cross-sectional studies with 17,526 sample sizes were included in this systematic review and meta-analysis: four studies in Saudi Arabia40,41,42,43, two studies in Nigeria44,45, three studies in Ghana46,47,48, four studies in Pakistan49,50,51,52, three studies in Spain53,54,55, seven studies in Ethiopia56,57,58,59,60,61,62, one study in Jordan63, two studies in China64,65, one study in Iran66, three studies in Egypt67,68,69, eight studies in India18,70,71,72,73,74,75,76, one study in Nepal77, one study in Sri Lanka29, two studies in Brazil78,79, one study in Beirut80, one study in Japan81, and one study in Thailand82. The sample size ranged from 74 in China64 to 2210 in Sri Lanka29 (Table 1).

Table 1 Characteristics of included studies in the meta-analysis of computer vision syndrome, 2022.
Full size table

Pooled prevalence of computer vision syndrome

The pooled prevalence of computer vision syndrome was 66% (95% CI: 59, 74). The lowest proportion included study was in Japan, 12% (95% CI: 9, 15)81, and the highest was in Pakistan, 99% (95% CI: 97, 100) 52. The I2 test showed that there was heterogeneity among the included studies (I2 = 99.42%, p value < 0.001) (Fig. 2).

Figure 2
figure 2

Forest plot showing the pooled prevalence of computer vision syndrome, 2022.

Full size image

Subgroup analysis by country

Subgroup analysis was performed based on country, and the prevalence of computer vision syndrome was highest in Pakistan (97%, 95% CI: 96, 98) and lowest in Japan (12%, 95% CI: 9, 15). The studies that showed significant heterogeneity were studies in Saudi Arabia (I2 = 99.41%, p value < 0.001), Ethiopia (I2 = 72.6%, p value < 0.001), Egypt (I2 = 80.06%, p value < 0.001), and India (I2 = 98.04%, p value < 0.001) (Table 2).

Table 2 Subgroup analysis by country on computer vision syndrome, 2022.
Full size table

Meta regression

Meta-regression was performed to identify the source of heterogeneity across the studies by country and sample size. Meta-regression indicated that heterogeneity was not associated with country or sample size (p value > 0.05) (Supplementary file 3 Table S1).

Publication biases

Publication bias was checked using dot plots, and visual inspection suggested asymmetry (Supplementary file 4: Figure S1).Moreover, publication bias was not shown by Egger’s test (p = 0.21) (Supplementary file 5 Table S2).

Sensitivity analysis

The sensitivity analysis was performed, and no single study unduly influenced the overall effect estimate of CVS (Supplementary file 6 Table S3).

Discussion

This systematic review and meta-analysis aimed to assess the pooled prevalence of computer vision syndrome. Although there are primary studies conducted on CVS, there are inconsistent findings on prevalence results. Moreover, there are no systematic reviews and meta-analyses on the pooled prevalence of computer vision syndrome. Therefore, findings from this systematic review and meta-analysis will help policy-makers design appropriate strategies to reduce computer vision syndrome-related public health concerns.

The pooled prevalence of computer vision syndrome was 66% (95% CI: 59, 73). The pooled prevalence was in line with the study done in India COVID-19 pre lockdown, 64.3%86. However, the pooled prevalence was lower than that in studies performed in India during the COVID-19 lockdown, 87.3%86, Europe, 90%87, and Ethiopia, 73.21%88. The difference might be due to differences in study period, study setting, socioeconomic differences, awareness and behavioral change in the prevention of computer vision syndrome. Moreover, the precision of the diagnostic instruments used to record the prevalence of CVS may be the cause of a wide range of variations. Whether through direct or online surveys, the majority of papers used purely subjective questions. As most surveys rely solely on the existence of one or more CVS complaints to diagnose CVS without connecting these complaints to the time of screen use and the long-term frequency of these complaints for months, studies may exaggerate the true prevalence of CVS11,89. Additionally, the disparity may be caused by how people use screens, particularly smartphones, or screen abuse, such as poor lighting, uncomfortable seating positions, close eye-screen distance, improper visualization gaze, uncorrected refractive errors, prolonged continuous screen hours, a lack of breaks, viewing screens in the dark, and poor screen design.

This study has the following limitations: articles published only in English were included, and it was difficult to determine the cause-effect relationship, as all the studies were cross-sectional designs. Additional database searches, such as Science Direct, Web of Science, ProQuest, Scopus, EMBASE, etc., we’re not performed due to the lack of free access and we recommend funding to expand database searches. Moreover, this study was reported from 20 countries, which might lack representativeness.

Conclusion

Nearly two in three participants had computer vision syndrome. Thus, preventive practice strategic activities for computer vision syndrome are important interventions.