Refractive errors are common eye disorders and are leading causes of visual impairment in the general population. Children with uncorrected refractive error may experience reduced visual acuity, transient blurring, headache and persistent ocular discomforts particularly for close work which can impair reading efficiency and school performance.
This article documents the prevalence of refractive errors in school-age children of different ethnic origins. The goal is to identify possible variation in measuring techniques and diagnostic criteria, as well as limitations of studies, to provide a clear direction for future studies.
The review was undertaken through a detailed evaluation of peer-reviewed publications of primary research on this topic. The keywords for the search included ‘refractive error’, ‘hyperopia’, ‘myopia’, ‘astigmatism’ and ‘school children’. Only epidemiological studies with participants between 5 and 18 years of age were included.
Although several population and school-based studies have been conducted in various racial groups and populations, their findings were diverse owing to inconsistencies in the methods applied in identifying children in need of refraction, measurement techniques and diagnostic criteria for refractive errors. There are also some limitations associated with the sampling design and characteristics, which may have influenced the outcome measures.
Despite the problems inherent in the studies, the review indicates that refractive error in school-age children is a public health concern in those populations and warrants additional research that will provide reliable data for proper planning of intervention strategies.
Refractive errors (REs) including myopia, hyperopia and astigmatism are common eye disorders and are leading causes of visual impairment and treatable blindness in the general population.
This article presents a review of the prevalence of REs in school-age children, along with their association with age and gender. A discussion about variation in measuring techniques and diagnostic criteria, as well as limitations of studies, is provided to direct future studies. Considering the implications of uncorrected RE to academic achievement and overall well-being, this review could provide useful information for policymakers and can help in planning, provision and evaluation of child eye health services.
A literature search was conducted on the online databases of PubMed, Medline, OVID, Google Scholar, ScienceDirect and Embase from November 2016 to November 2017 using the following keywords: refractive error, hyperopia, myopia, astigmatism and school children. The review was restricted to primary research published in English and in peer-reviewed journals. Only epidemiological studies with stated measures of prevalence of corresponding RE among school-age children between 5 and 18 years of age were included.
In this narrative review, findings from studies that met the outlined criteria were reviewed. Variables of interests for review included the following: sample size and sampling method, participant characteristics including gender and age, prevalence rates of corresponding RE, information on diagnostic criteria and measurement techniques. A summary of each study was first presented and evaluated in relation to findings from other studies. Eligible studies on myopia, hyperopia and astigmatism were compared according to geographic regions or ethnicity.
Prevalence of hyperopia among school-age children in selected countries from various geographic regions.
Study | Country | Ethnicity | Age (years) | Sample size ( |
Definition criteria | Measurement technique | Prevalence (%) |
---|---|---|---|---|---|---|---|
Atowa et al. |
Nigeria | African | 8–15 | 1197 | SER ≥ 2.00 | Cycloplegic autorefraction | 0.9 |
Ovenseri-Ogbomo and Assien |
Ghana | African | 11–18 | 595 | SPH ≥ 0.75 | Non-cycloplegic retinoscopy | 5.0 |
Mehari and Yimer |
Ethiopia | African | 7–18 | 4238 | SER ≥ 2.00 | Non-cycloplegic retinoscopy | 0.3 |
Wajuihian an d Hansraj |
South Africa | African | 13–18 | 1586 | SER ≥ 0.50 | Non-cycloplegic autorefraction/Subjective refraction | 5.0 |
Aldebasi |
Saudi Arabia | Middle East | 6–13 | 5176 | SER ≥ 2.00 | Cycloplegic autorefraction | 0.9 |
Norouzirad et al. |
Iran | Middle East | 6–15 | 1130 | SER ≥ 2.00 | Non-cycloplegic retinoscopy | 12.9 |
He et al. |
Rural China | Asian/East | 13–17 | 2454 | SER ≥ 2.00 | Cycloplegic autorefraction | 1.2 |
He et al. |
Urban China | Asian/East | 5–15 | 4347 | SER ≥ 2.00 | Cycloplegic autorefraction | - |
Paudel et al. |
Vietnam | Asian/South East | 12–15 | 2238 | SER ≥ 2.00 | Cycloplegic autorefraction | 0.4 |
Goh et al. |
Malaysia | Asian/South East | 7–18 | 4634 | SER ≥ 2.00 | Cycloplegic autorefraction | 1.6 |
Dandona et al. |
Rural India | Asian/South | 7–15 | 3976 | SER ≥ 2.00 | Cycloplegic autorefraction | 0.4 |
Murthy et al. |
Urban India | Asian/South | 5–15 | 6447 | SER ≥ 2.00 | Cycloplegic autorefraction | 7.7 |
Zadnik et al. |
USA | Caucasian | 6–14 | 2583 | SER ≥ 1.25 | Cycloplegic autorefraction | 8.6 |
Kleinstein et al. |
USA | Caucasian | 5–17 | 2523 | SER ≥ 1.25 | Cycloplegic autorefraction | 12.6 |
O’Donoghue et al. |
United Kingdom | Caucasian | 6–7 | 392 | SER ≥ 2.00 | Cycloplegic autorefraction | 20.6 |
12–13 | 661 | 14.7 | |||||
Czepita et al. |
Poland | Caucasian | 6–18 | 5724 | SER ≥ 2.00 | Cycloplegic retinoscopy | 4.0 |
Ip et al. |
Australia | Caucasian | 11–14 | 2352 | SER ≥ 2.00 | Cycloplegic autorefraction | 5.0 |
Fotedar et al. |
Australia | Caucasian | 12 | 2233 | SER ≥ 2.00 | Cycloplegic autorefraction | 5.0 |
SER, spherical equivalent refraction.
As with studies on African populations, prevalence studies on children from other geographic locations also reported varied results. Although the studies in Asia utilised a logMAR protocol, common definition of SER 2 D or more, large sample sizes, differences in age group of the study participants and study locations (rural or urban) may have influenced the reported prevalence of hyperopia in the various studies reviewed. In rural China,
For studies conducted on caucasian populations, diagnostic criteria and age ranges of the study samples affected the reported prevalence of hyperopia (
Except for two studies in the United States,
Prevalence of myopia among school-age children in selected countries from various geographic regions.
Study | Country | Ethnicity | Age (years) | Sample size ( |
Definition criteria | Measurement technique | Prevalence (%) |
---|---|---|---|---|---|---|---|
Atowa et al. |
Nigeria | African | 8–15 | 1197 | SER ≤ −0.50 | Cycloplegic autorefraction | 2.7 |
Ovenseri-Ogbomo and Assien |
Ghana | African | 11–18 | 595 | SPH ≤ −0.50 | Non-cycloplegic retinoscopy | 2.6 |
Mehari and Yimer |
Ethiopia | African | 7–18 | 4238 | SER ≤ −0.50 | Non-cycloplegic retinoscopy | 6.0 |
Wajuihian and Hansraj |
South Africa | African | 13–18 | 1586 | SER ≤ −0.50 | Non-cycloplegic autorefraction | 7.1 |
Aldebasi |
Saudi Arabia | Middle East | 6–13 | 5176 | SER ≤ −0.50 | Cycloplegic autorefraction | 6.5 |
Norouzirad et al. |
Iran | Middle East | 6–15 | 1130 | SER ≤ −0.50 | Non-cycloplegic retinoscopy | 14.9 |
He et al. |
Rural China | Asian/East | 13–17 | 2454 | SER ≤ −0.50 | Cycloplegic autorefraction | 42.4 |
He et al. |
Urban China | Asian/East | 5–15 | 4347 | SER ≤ −0.50 | Cycloplegic autorefraction | 35.1 |
Paudel et al. |
Vietnam | Asian/South East | 12–15 | 2238 | SER ≤ −0.50 | Cycloplegic autorefraction | 20.4 |
Goh et al. |
Malaysia | Asian/South East | 7–18 | 4634 | SER ≤ −0.50 | Cycloplegic autorefraction | 20.7 |
Dandona et al. |
Rural India | Asian/South | 7–15 | 3976 | SER ≤ −0.50 | Cycloplegic autorefraction | 4.1 |
Murthy et al. |
Urban India | Asian/South | 5–15 | 6447 | SER ≤ −0.50 | Cycloplegic autorefraction | 7.4 |
Zadnik et al. |
USA | Caucasian | 6–14 | 2583 | SER ≤ −0.75 | Cycloplegic autorefraction | 10.1 |
Kleinstein et al. |
USA | Caucasian | 5–17 | 2523 | SER ≤ −0.75 | Cycloplegic autorefraction | 9.2 |
O’Donoghue et al. |
United Kingdom | Caucasian | 6–7 | 392 | SER ≤ −0.50 | Cycloplegic autorefraction | 2.3 |
12–13 | 661 | 17.1 | |||||
Czepita et al. |
Poland | Caucasian | 6–18 | 5724 | SER ≤ −0.50 | Cycloplegic retinoscopy | 13.1 |
Fotedar et al. |
Australia | Caucasian | 12 | 2233 | SER ≤ −0.50 | Cycloplegic autorefraction | 9.8 |
SER, spherical equivalent refraction.
Variations in the prevalence of myopia in Asian children have also been widely reported, with considerable differences existing between various countries and study locations. Overall, the studies reviewed showed that myopia is more prevalent in East Asian and South-East Asian countries than in other parts of the world. For instance, studies by He et al. using cycloplegic autorefraction found that 35.1% and 42.4% of school-age children in rural
As with studies on African and Asian populations, the prevalence of myopia in Caucasian children was also influenced by the definition criteria and participants’ ages (
Prevalence of astigmatism among school-age children in selected countries from various geographic regions.
Study | Country | Ethnicity | Age (years) | Sample size ( |
Definition criteria | Measurement technique | Prevalence (%) |
---|---|---|---|---|---|---|---|
Atowa et al. |
Nigeria | African | 8–15 | 1197 | ≤ −0.75 | Cycloplegic autorefraction | 4.4 |
Ovenseri-Ogbomo and Assien |
Ghana | African | 11–18 | 595 | ≤ −0.75 | Non-cycloplegic retinoscopy | 6.5 |
Mehari and Yimer |
Ethiopia | African | 7–18 | 4238 | ≤ −0.75 | Non-cycloplegic retinoscopy | 2.0 |
Wajuihian and Hansraj |
South Africa | African | 1318 | 1586 | ≤ −0.75 | Non-cycloplegic autorefraction | 3.0 |
Aldebasi |
Saudi Arabia | Middle East | 6–13 | 5176 | ≤ −0.75 | Cycloplegic autorefraction | 11.2 |
He et al. |
Rural China | Asian/East | 13–17 | 2454 | ≤ −0.75 | Cycloplegic autorefraction | 25.3 |
He et al. |
Urban China | Asian/East | 5–15 | 4347 | ≤ −0.75 | Cycloplegic retinoscopy | 33.6 |
Cycloplegic autorefraction | 42.7 | ||||||
Paudel et al. |
Vietnam | Asian/South East | 12–15 | 2238 | ≤ −0.75 | Cycloplegic autorefraction | 20.4 |
Goh et al. |
Malaysia | Asian/South East | 7–18 | 4634 | ≤ −0.75 | Cycloplegic retinoscopy | 15.7 |
Cycloplegic autorefraction | 21.3 | ||||||
Dandona et al. |
Rural India | Asian/South | 7–15 | 3976 | ≤ −0.75 | Cycloplegic retinoscopy | 3.8 |
Cycloplegic autorefraction | 9.7 | ||||||
Murthy et al. |
Urban India | Asian/South | 5–15 | 6447 | ≤ −0.75 | Cycloplegic retinoscopy | 7.0 |
Cycloplegic autorefraction | 14.6 | ||||||
Kleinstein et al. |
USA | Caucasian | 5–17 | 2523 | ≤ −1.00 | Cycloplegic autorefraction | 28.4 |
Czepita et al. |
Poland | Caucasian | 6–18 | 5724 | ≤ −0.50 | Cycloplegic retinoscopy | 4.0 |
Robaei et al. |
Australia | Caucasian | 12 | 2353 | ≤ −1.00 | Cycloplegic autorefraction | 21.8 |
Previous studies exploring the prevalence of astigmatism in school-age children have also shown marked variations in prevalence levels (
The studies on Asian populations were consistent in the definition of astigmatism and the use of cycloplegic objective measurement methods. In most of the studies, both objective (retinoscopy and autorefraction) methods were applied and the results showed that autorefraction technique yielded higher values compared with the retinoscopic technique (
For studies on Caucasian children, the prevalence of astigmatism was also influenced by the definition criteria and measurement methods (
Most of the studies showed that the prevalence of hyperopia decreases significantly with age.
Regarding myopia, several studies reviewed were consistent in reporting a significant age increase in the prevalence of myopia.
It has been suggested that, on average, women have shorter axial length when compared with men.
There are some limitations associated with the studies reviewed, which may have influenced the interpretation of their findings and conclusions. All studies except Atowa et al.
This literature review has highlighted the prevalence of RE in school-age children in various countries. However, inconsistent methods were applied across studies in identifying children in need of refraction. Although a VA threshold of 6/9 or less can reliably detect myopia in school-age children, there is no reliable VA threshold for clinically significant hyperopia and astigmatism. High amounts of hyperopia (> 5 D) and astigmatism (> 1.5 D) have been reported in children who were able to read 6/6 (20/20) on the VA chart.
There is no consensus on the most appropriate method for the measurement of RE. Some studies reported myopia and hyperopia in terms of the spherical component, while others reported them based on the SER (sphere + ½ cylindrical components). Although an objective method (retinoscopy or autorefraction) was the preferred measuring technique, the use of cycloplegia was not a constant factor. Instead some studies utilised the plus lens test to screen for latent hyperopia because cycloplegia was contraindicated as accommodative tests were also included in their evaluations or for concerns of ethical issues.
A wide variety of criteria were applied in the diagnosis of individuals with different types of RE, with many studies focusing mainly on RE that significantly affects VA (
The studies
The disparity in the RE prevalence by regions and study locations can be explained by ethnicity and geographical factors. Hyperopia prevalence was low in African and East Asian populations compared with Caucasians. Similarly, myopia and astigmatism were higher in East and South-East Asian populations compared with other regions. Reports indicate that South-East Asian children are genetically predisposed to having myopia because of the influence of ethnicity, family history of myopia and schooling system.
This article indicates that the prevalence of RE in school-age children is a public health concern in the various study locations. The methodological differences, such as inappropriate study designs, variations in defining and quantifying the RE and improper measuring techniques, complicate the comparison of the corresponding findings. The article highlights the gaps in knowledge in this area of study, including the non-inclusion of low categories of RE, non-inclusion of all children for refraction within some studies, non-application of cycloplegia or the plus lens test, limited age range, small sample size and inappropriate sampling methods. The review of the literature also reveals regional variations in the prevalence of RE, which may be related to differences in socio-economic development, race, cultural factors as well as availability of interventions. Considering the implication of visual anomalies for academic achievement, as well as overall well-being, this review could provide useful information for policymakers and can help in planning, provision and evaluation of child health services. Future research should include near vision anomalies which are capable of affecting school performance even when VA is not affected. This would assist in developing broad interventions and management strategies targeting these conditions in school-age populations.
The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.
U.C.A. wrote the article. R.H. and S.O.W. provided feedback on the structure and content of the manuscript.