Coronavirus disease 2019 outbreak and associated public health measures increase the progression of myopia among children and adolescents: Evidence synthesis

Authors: Watcharapalakorn A, Poyomtip T, Tawonkasiwattanakun P

Geographical coverage: China, Turkey and Argentina

Sector: Burden of disease

Sub-sector: Epidemiology

Equity focus: Children

Study population: Children aged 5-18 years

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: The COVID-19 outbreak led to changes in lifestyles and behaviour (less time outdoors; more screen time) that could conceivably have led to an increase in the progression of myopia among children. Although studies have suggested that the coronavirus disease 2019 (COVID-19) outbreak increased myopia progression, they had different settings and analysis methods. No meta-analysis to explore this hypothesis has been conducted to date.

Objectives: To compare myopia progression before and during the COVID-19 outbreak using meta-analysis.

Main findings:

Overall, study authors found the COVID-19 pandemic accelerated myopic progression compared to the past.

A total of 20 studies were selected according to the inclusion criteria, and eight were included in the meta-analysis. Participants’ ages ranged from 5 to 18 years, with more than 773,797 individuals. Three studies were conducted with relatively large sample sizes. Six studies were conducted in China, one in Turkey and one in Argentina. Three studies performed refractive error measurement using noncycloplegic refraction. There were two cross-sectional studies, one case-control study and five cohort studies. Based on The Newcastle-Ottawa Scale (NOS) included studies represented a median of good quality. Due to differing durations of refractive error measurement, the data was converted into annual myopia progression.

Based on the pooled analysis, authors indicated that the mean difference of annual myopia progression during the pandemic was 0.41 D higher (95% confidence interval [CI]: 0.35-0.48, p <0.01) than before the pandemic. Subgroup analysis using cycloplegic (mean difference, 0.30 D; 95% CI, 0.22-0.38; p <0.01) or noncycloplegic refraction (mean difference, 0.60 D; 95% CI, 0.27-0.93; p <0.01) indicated that the mean difference of annual myopia progression during COVID-19 significantly increased in both refractive measurements.

The authors identified two policy implications. Optometrists must provide myopia control education to patients as the standard of care. Building a myopia control practice can help provide the highest quality of care to patients. Parents should find ways to reduce the time spent performing near-work and promote outdoor activities for children.

A future study in which the inclusion and exclusion criteria are set based on younger age at baseline and age at myopic onset with respect to different ethnicities is required to validate the effect of public health measures on the progression of myopia.

Methodology:

Publications selected for analysis had the following characteristics: (1) case-control, cross-sectional, retrospective cohort and prospective cohort studies; (2) studies with mean spherical equivalent refraction (SER) in 2018, 2019 and 2020 or mean change in SER before and during the COVID-19 pandemic as primary or secondary outcomes; (3) studies with participants younger than 18 years; (4) articles published before the start search (8 October 2021); and (5) studies that provided informative data for calculating mean differences.

The exclusion criteria were as follows: (1) reviews, case reports, case series, conference abstracts and letters to the editor; (2) in vitro studies and animal experiments; (3) duplication; and (4) non-English articles.

A literature search was conducted in EMBASE, PubMed, ClinEpiDB and Web of Science databases to collect potentially eligible and relevant studies published until 8 October 2021. Two independent investigators performed the search and screened the titles and abstracts of identified potential studies to remove any irrelevant publications as per the inclusion and exclusion criteria. Additional publications were considered and registered in this study by screening review articles or relevant articles after searching for “myopia progression and COVID-19” in Google Scholar.

Two authors independently extracted data from selected studies. Discrepancies between the authors were discussed to resolve issues, and a third author was included to achieve the final consensus. The Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of the included studies, which was assessed by two authors independently, and the third author resolved any disagreement.

The heterogeneity of the included studies was evaluated using the I² value with Cochran’s Q test. Used a fixed-effect model to derive estimates due to the lack of evidence on heterogeneity (I² lower than 50%, p-value of Cochran’s Q test lower than 0.05). I² above 50%, the random-effects model with generic inverse variance method recommended by the Cochrane Handbook for Systematic Reviews, was used to analyse the difference in mean as a primary outcome. Subgroup analysis was classified into noncycloplegic and cycloplegic to explore the possible cause of heterogeneity. The radial plot was used to assess outlier influence when heterogeneity was detected. Publication bias identified by funnel plot with pseudo-95% CIs and Egger’s statistics. Sensitivity analysis was performed by subsequently omitting the studies to confirm the results. The meta-analysis was analysed using Review Manager. The significant value was p <0.05.

Applicability/external validity: Authors identify two limitations that may have affected the study’s applicability. The majority of data was from Chinese and Asian groups, and different durations were used to measure refraction progression.

Geographic focus: All included studies were undertaken in LMICs, with the majority in Asia. Given myopia is thought to be correlated with ethnicity, this affects the validity of the reported results.

Summary of quality assessment:

The methods employed for the identification, inclusion, and critical evaluation of the studies were largely solid, with all essential tasks being performed by the main authors. However, there is no indication of any efforts to incorporate unpublished works or works produced in languages other than English into the review. While the data analysis approach was generally robust, there was no effort to analyse the data based on the quality and risk of bias of individual studies. For these reasons, we assigned a medium level of confidence to the study’s results.