Prevalence of refractive errors among children in Saudi Arabia: a systemic review

Author: Alghamdi W.

Geographical coverage: Saudi Arabia

Sector: Burden of disease

Sub-sector: Epidemiology

Equity focus: Children aged 4-14

Study population: Children aged 4-14 in Saudi Arabia.

Review type: Other review

Quantitative synthesis method: Narrative review and meta-analysis

Qualitative synthesis method: Not applicable

Background:

Refractive error (RE) is a major public health problem that affects a large proportion of children. The World Health Organization identified uncorrected RE (URE) as the leading cause of vision impairment globally. In 2016, more than 6.1 million (30%) of the Saudi population were estimated to be below the age of 14 years. However, a limited number of studies have investigated the prevalence of RE and URE in children in Saudi Arabia, and their results have been variable.

Objectives: The aim of this review was to estimate the prevalence of RE and URE in school-aged children of 4 to 14 years of age in Saudi Arabia.

Main findings:

The review included a total of eight school-based studies. The studies varied in sample size, age group, refraction method used and the definition of refractive errors. The definition of uncorrected refractive error was similar across all studies that reported it. Authors reported high heterogeneity between the included studies reporting the prevalence of all refractive errors in children (Cochran’s Q-test, p <0.001; I² = 98%). Although the funnel plot analysis showed no presence of publication bias in the review, the author did not exclude its presence due to the small number of studies included in the analysis.

Among the overall pooled population of 12,247 children, the estimated prevalence of refractive error reported by the author was 17.5% (95%CI: 11.1- 25). In the five studies that reported the URE (n = 10198), the pooled prevalence was 16.8% (95% CI: 11.4-21.3). The overall prevalence of refractive error in boys (n = 6114) and girls (n = 5563) were 16.8% (95% CI: 10.8-24.1) and 17.7% (95% CI: 10.2-25.9), respectively. The author noted no significant difference in the prevalence of RE between boys and girls (p = 0.1984). The review found myopia as the most prevalent refractive error 40.8% (95% CI: 16.1-69.9) followed by astigmatism 29.7% (95% CI: 6.1-61.7 and hyperopia 28.3% (95% CI: 16.9-41.2). Overall, according to the author, this meta-analysis indicated a high burden of refractive errors in Saudi children affecting nearly one in every five children, and both females and males appear to be equally affected. The author notes that these results highlight the high demand for a new national prevention strategy to reduce the numbers of uncorrected RE in the country effectively.

The author noted a relatively small number of studies and the fact that few were conducted in densely populated areas of Saudi Arabia.

Methodology:

In this review, data from Saudi Arabia published between 1 January 2000 and 1 October 2020 was included. Other inclusion criteria included any quantitative study that reported the prevalence of any refractive errors from a Saudi population. Then, only data related to children between 4 and 14 years was then selected. Two independent reviewers performed the following review process: (A) searching and selecting studies; (B) critical review and methodological assessment of the studies; (C) extraction of relevant data; and the author then performed statistical analysis and interpretation.

The author conducted a search on the following databases: Cochrane Library, PubMed, OVIS and EMBASE. The methodological quality of the eligible studies was assessed using the 10- item checklist developed by Munn et al.

MedCalc Statistical Software was used to calculate the pooled estimate of refractive errors and uncorrected refractive errors in children. MedCalc uses the Freeman-Tukey double arcsine transformation with a random-effects model. Additionally, the pooled prevalence of refractive errors by gender and types among children and for gender and types were performed. Chi-square-based Q test was used to assess the heterogeneity of reported prevalence among studies, while publication bias was assessed through funnel plots.

Applicability/external validity:

The author does not specifically address generalisability of the findings. However, it is noted that no data was available from large and highly populated regions in Saudi Arabia, including Mekka, Assir and northern provinces in which each province has significant differences in demographic and socioeconomic conditions.

Geographic focus:

The review focuses on Saudi Arabia only.

Summary of quality assessment:

This systematic review has a clear inclusion criteria and the search covered a sensible period of time given the objectives of this study. Two reviewers independently screened studies for inclusion in the review, extracted data and conducted the quality assessment of included studies. The author also used appropriate methods to pool data of cross-sectional studies. However, important limitations were identified. The database search was not comprehensive enough to ensure all relevant studies were included in the review. Furthermore, it is not clear if: 1) the author considered the inclusion of unpublished studies; 2) language bias was avoided by searching for studies written different languages; and 3) references of included studies were reviewed for additional studies. This means some important literature may have been missed, impacting on the reliability of the review findings. Where studies of differing risk of bias are included, results are not analysed separately by risk of bias status (which we understand may not have been possible, due to the small number of studies involved). Therefore, “low confidence” was attributed to this review’s findings.