In a recent study published in the BMC Public Health, researchers systematically reviewed and quantitatively evaluated the epidemiological evidence on the association between screen time exposure and myopia in children and adolescents.
Study: The association between screen time exposure and myopia in children and adolescents: a meta-analysis. Image Credit: Inna Kot/Shutterstock.com
Background
Myopia, or near-sightedness, is a refractive error characterized by the excessive elongation of the ocular globe, leading to an increased risk of pathological eye changes such as cataracts, glaucoma, retinal detachment, and macular degeneration, which can cause irreversible vision loss.
Environmental factors like education, near-work, and outdoor activities significantly influence myopia prevalence. Screen time, including the use of computers, televisions, video games, and mobile devices, is now pervasive in children’s and adolescents’ lives, with increased exposure from a younger age.
Recent epidemiological studies show conflicting results regarding the correlation between screen time and myopia.
Further research is needed to clarify the conflicting findings on the relationship between screen time exposure and myopia in children and adolescents and to inform effective prevention and control strategies.
About the study
In this systematic review and meta-analysis, researchers conducted literature screening, data extraction, risk of bias assessment, and analysis.
Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) 2020 statement, eligibility criteria included studies on children and adolescents examining screen time exposure (categorical or continuous) and reporting adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for myopia.
Observational studies using cross-sectional, cohort, or case-control designs were included, selecting the most recent and comprehensive publication when multiple studies reported on the same population.
Literature searches in PubMed, Embase, and Web of Science up to June 1, 2023, used screen-related and myopia-related terms. Exclusions included reviews, letters, commentaries, occupational exposure studies, non-human studies, ecological studies, and studies lacking effect estimates.
Data extraction included author details, publication year, country, study design, sample size, screen device type, myopia definition, results (ORs and 95%CIs), and confounder adjustments. The quality assessment used the Newcastle Ottawa Scale (NOS), scoring high, moderate, or low-quality studies.
Statistical analysis employed R software, using fixed-effect or random-effect models based on heterogeneity. Subgroup and sensitivity analyses were conducted, and publication bias was evaluated using funnel plots, Egger’s test, and the trim and fill method.
Study results
The study, 6,493 articles were identified from PubMed, Embase, and Web of Science. After excluding 1,159 duplicated studies, 5,295 unrelated studies were removed during the title and abstract screening.
Thirty-nine articles were assessed for eligibility, but 20 were excluded due to univariate analysis, no available data, or lack of myopia prevalence.
Ultimately, 19 studies involving 102,360 participants were included, with 91,282 in cross-sectional studies (N=15) and 11,078 in cohort studies (N=4).
Thirteen studies (68%) used cycloplegic refraction, three (16%) used self-reported myopia, and three (16%) performed optometry without cycloplegia. The studies originated from nine countries: two from North America, seven from Europe, six from East Asia, two from South Asia, and two from Southeast Asia.
According to the NOS checklist, 14 studies (74%) were considered high quality (score ≥ seven stars), while the remaining five studies (26%) were of moderate quality (score 5 or 6 stars).
Possible sources of bias included small sample sizes in six studies, insufficient strategies to deal with confounding factors in five studies, lack of adjustment for key confounders in five studies, and not using cycloplegic refraction to confirm myopia cases in five studies.
Eleven studies involving 90,415 participants examined the relationship between categorical screen time exposure (high vs. low) and myopia in children and adolescents. The highest category of screen time exposure was significantly associated with myopia in cross-sectional studies (OR=2.24, 95%CI: 1.47–3.42) and cohort studies (OR=2.39, 95%CI: 2.07–2.79).
Subgroup analysis by screen device type showed significant associations with myopia for computers and televisions but not smartphones. Subgroup analysis by study quality, geographic region, and research period revealed significant associations in high-quality studies, East and South Asia, and research conducted after 2008.
Eight studies involving 11,925 participants analyzed continuous exposure (per 1 h/d increase) to screen time and myopia. There was no association in cross-sectional studies (OR=1.15, 95%CI: 0.97–1.37), but a significant association was found in cohort studies (OR=1.07, 95%CI: 1.01–1.13).
Significant associations were observed for computer screen time in cross-sectional studies and in East Asia. Due to the limited number of studies, further subgroup analyses were not performed for cohort studies.
Publication bias was detected in cross-sectional studies for both high vs. low screen time and per 1 h/d increase screen time groups, as indicated by Egger’s test. After trim and fill analysis, the pooled ORs remained significant.
Sensitivity analysis showed robust results for the high vs. low screen time group but not for the per 1 h/d increase screen time group.
Conclusions
To summarize, this comprehensive meta-analysis found significant associations between screen time from computers and televisions and myopia, but not smartphones. Regional differences were noted, with significant associations in East and South Asia.
The study emphasized the need for targeted prevention strategies, including reducing near-work activities and promoting outdoor time.
Artemisinin resistance detected in children with severe malaria in Uganda