A systematic review of amblyopia prevalence among the children of the world

Author: Ali M, Morteza G, Hossein H, Hesam M, Fereshteh F, Nazli T, Fariba P.

Geographical coverage: Worldwide

Sector: Burden of disease

Sub-sector: Epidemiology

Equity focus: Children

Study population: Children with amblyopia

Review type: Other review

Quantitative synthesis method; Meta-analysis

Qualitative synthesis method: Not applicable

Background: Amblyopia is one of the most prominent causes of visual impairment in children and adult populations. Several studies have evaluated the prevalence of amblyopia, reporting the disorder to be present in two to three per cent of the population.

Objectives: To assess the global prevalence of amblyopia disease in the children.

Main findings:

Overall findings of this review showed that the total prevalence of amblyopia is 4.3%, but this estimate varies in all continents, especially in Africa.

A total, the authors included a total of 131 studies the review. This study includes data from different countries across the world, such as USA, China, Iran, Turkey, Nepal, Germany, Saudi Arabia, Hong Kong, Sweden, Brazil, India, Ireland, Denmark, Netherlands, Taiwan, Japan, Pakistan, Bangladesh, Cameroon, Tanzania, Bulgaria, Palestine, and Croatia. However, the authors did not specify the exact number of studies or participants from each country.

Overall, the quality of these studies was reported to be high, so all studies were included in data extraction and synthesis. Out of 57 studies, five cohort studies, one clinical trial, one case series and 50 cross-sectional studies were included.

The global prevalence of amblyopia in children was determined to be 4.3% (pooled prevalence: 4.3%, 95% CI: 2.6%-7.00%, p-value 0.0001). The studies showed a high level of heterogeneity (Q: 48281.18, df: 56, p-value 0.001, I²: 99.88). Subgroup analysis revealed that the prevalence of amblyopia in cross-sectional studies was 4.31% (pooled prevalence = 4.31%, 95% CI:2.53%-7.37%, p-value 0.0001), while in cohort studies it was 2.07% (pooled prevalence = 2.07%, 95% CI:1%-4.31%, p-value 0.0001). When analysed by continent, America had the highest prevalence at 5.58% (95% CI: 2.23%-13.94%, p-value 0.0001), while Europe (4.58%, 95% CI:2.57%-8.14%, p-value 0.0001), Asia (3.83%, 95% CI:1.85%-7.92%, p-value 0.0001), and Africa (0.71%, 95% CI:0.002%-172.46%, p-value 0.05) had the lowest prevalence rates.

Authors note that further worldwide high-quality and valid studies should be carried out to allow the calculation of the real prevalence of amblyopia among children of the world. In Africa, because of the lack of well-designed research studies, poor standards of assessing amblyopia, and insufficient treatment evaluation, the continued scarcity of robust evidence leaves gaps in knowledge that need to be addressed.

Methodology:

The study analysed articles on children diagnosed with amblyopia, primarily caused by strabismus or refractive disorder1. It excluded cases arising from eye trauma or congenital cataract. The research incorporated articles evaluating global amblyopia prevalence, conducted via pre-school screening programs or referrals to medical centres and clinics. It considered observational studies, including prospective and retrospective cohort reviews, and various analytical and descriptive cross-sectional studies. The study also included experimental and quasi-experimental studies, such as randomised and non-randomised controlled trials, assessing amblyopia prevalence at baseline4. The analysis covered articles published in any language from 1963 to June 2019.

The study employed a three-stage search strategy. Initially, a search was performed on the Medline (PubMed) database, followed by an analysis of the words in the title and abstract sections, and the index terms of the articles. A second search was then conducted using all identified keywords and index terms across several databases in June 2019. This included Medline (via PubMed), EMBASE, Scopus, Web of Science, ProQuest (for unpublished studies and dissertations), and Google Scholar. The reference lists of all selected reports and articles were also reviewed for additional data.

Post-search, all citations were imported into EndNote software and duplicates were removed. Two independent researchers reviewed the titles and abstracts against the inclusion criteria. Full-texts of potentially eligible studies were thoroughly evaluated by two independent reviewers, with any disagreements resolved through discussion or a third reviewer’s input. Studies not meeting the inclusion criteria were excluded.

Eligible studies underwent critical appraisal by two independent reviewers using a standardized tool from the Joanna Briggs Institut1. Any disagreements were resolved through discussion or a third reviewer. Data extraction was performed by two independent reviewers using a modified JBI tool. Discrepancies were resolved through discussion or a third reviewer. Authors were contacted for any missing or additional data.

Data was pooled using a statistical meta-analysis with comprehensive meta-analysis software (CMA) v5.2 Software. For analysis, effect sizes were detailed in proportion with a confidence interval of 95% and were calculated. Heterogeneity was assessed using the standard chi-squared and I² tests. Statistical analyses were performed using the random-effects method. Analyses of subgroups were performed according to their continents. As there were 10 or more studies included in the meta-analysis, a funnel plot was generated in CMA v5.2 to assess the bias of publication. Statistical tests for the asymmetry of the funnel plot such as the Egger test, Begg test and Harbord test were conducted as needed.

Applicability/external validity: Authors note high heterogeneity between studies, which may make their pooled findings less applicable.

Geographic focus: The review offers distinct prevalence estimates for each continent, except for Africa due to the availability of only two studies. The search was conducted without geographical restrictions and included studies from all continents. This encompassed several LMICs such as China, Turkey, Nepal, Iran, Tanzania, Southern India, and Bangladesh.

    Summary of quality assessment:

The methods for selecting, including, and critically appraising the studies were generally rigorous, with two researchers performing all essential tasks. The search was highly inclusive and did not have any language or publication limitations. However, it’s not clear whether there were any date restrictions on the search. The data analysis approach was thorough in all respects, leading us to have medium confidence in the review’s findings.