Detection and treatment of refractive errors were included as objectives of the Program Vision 2020: The Right to Sight (a global initiative of nongovernmental organizations and the WHO [World Health Organization]), which aimed to eliminate avoidable visual impairment (VI) worldwide by the year 2020.1 To provide data on the prevalence of uncorrected refractive error, Refractive Error Study in Children (RESC) surveys were performed in several populations worldwide with standardized protocols.2
After the initial RESC survey results were collected,3–8 minor changes were suggested to simplify execution of the protocols without compromising the quality of the results. Survey results demonstrated that children not attending school displayed variable rates of VI because of refractive errors.4,7,9 Thus, new RESC studies using a school-based protocol,10 including some in Brazil,11 were designed considering the high school attendance of many countries being a proxy for the population sampling and the methodological easiness of this protocol.
Furthermore, recent surveys have evaluated children screened by visual acuity (VA),11–13 applying this new approach not only with emphasis on the evaluation of the prevalence of refractive errors but on VI associated with refractive errors.
Although previous Brazilian surveys were reported using this new methodological approach in urban areas,8,11,12,14,15 they were performed using different school-based protocols after VA screening and may not represent most of the typical cities countrywide. In addition, the rates of Brazilian children attending school may vary between the beginning and the end of the school year. Therefore, we conducted a population-based survey in the middle of the school year using the RESC protocol in a city of the Central Region of Brazil (Gurupi) that represents the typical low-middle socioeconomic profile of the country.
A cross-sectional population survey was conducted among children aged 10 to 15 years (a priority age group proposed by the Refractive Error Working Group of the WHO [Refractive Error Working Group meeting draft, ICEE, Sydney, 2002; data not published]) who are residents of Gurupi city in Tocantins, a state in Brazil. The Brazilian Institute of Geography and Statistics16 census estimates 71,413 inhabitants in Gurupi, with 7098 (9.94%) of 71,413 individuals aged 10 to 15 years. Generally, Gurupi residents represent a young population (42% of age <25 years old), with a similar male-to-female ratio, a mixed racial-ethnic profile, and a gross domestic product (GDP) per capita of about US$ 6300.16
Cluster sampling was used to select the study population.2 For this estimated population of 7098 children, an 8% prevalence of VI was considered,11 with a confidence error bound of 20% and 95%, and adjustments to accommodate nonparticipation and the clustering design effect.
All 59 census districts evaluated geographically during the Brazilian national population census of 200016 were used as the sampling frame, each with an estimated 85 to 158 children. Twelve clusters were selected randomly, estimating 1512 children, exceeding the sample size calculation of 1452 individuals.
Institutional Review Board Approval
This study adhered to the tenets of the Declaration of Helsinki for human subjects research and was approved by the Ribeirão Preto Clinical Hospital Ethics Committee (HCRP; process number 1513/2007). Written informed consent was obtained from the parents or legal guardians of all participants before study enrollment.
All the fieldwork was carried out from June 2007 through August 2007. Six enumerators trained in assessing the population and in basic ophthalmic skills were sent to visit the 12 selected clusters.
The enumerators used a sequential approach in every house to recruit and examine children aged 10 to 15 years. The enumeration included those temporarily absent from the area if the time of absence was less than 6 months. Transient visitors were excluded from enumeration.
After parents or guardians of eligible children provided written informed consent, a form was completed with the children’s personal data, including name, age in years confirmed by birth certificates, sex, previous ophthalmic visits, and whether the children wore optical correction.
After acquisition of personal data, with the examiner positioned at a distance of 4 m, uncorrected VA was evaluated separately in each eye using a 5-line optotype “E,” representing 0.625 (20/32), adjusted to 4 m. Children who failed to identify correctly at least three of the five optotypes with one or both eyes were referred for evaluation by the ophthalmic consultant service.
After initial enumeration and tests, a card with the scheduled date of the eye examination was given to the parents or legal guardians of each child screened. Children who did not attend the specialized examination were contacted at least three times on separate occasions through phone calls and/or in-house visits. A second written informed consent was obtained from the person of the household accompanying the child to the ophthalmological visit. Children with spectacles were requested to bring them along on the day of the examination. Children who could not keep the scheduled date were given another date during the study period.
Two ophthalmologists (F.M.I., J.R.P.C.) from the reference ophthalmic service performed a retest of children’s uncorrected VA with a retroilluminated logarithm of the minimum angle of resolution (logMAR) chart with tumbling E optotypes and evaluated ocular motility. Cycloplegia was induced with two drops of 1% cyclopentolate for all referred children. Cyclopentolate drops were administered 5 minutes apart, with a third drop administered after 15 minutes. Cycloplegia and pupil dilation were evaluated after an additional 20 minutes. Complete cycloplegia was considered to be a pupillary dilation of 6 mm or more with absence of a pupil light reflex.
Corrective lenses were prescribed based on subjective refraction after cycloplegic autorefraction in children who demonstrated improved VA with refraction, considering medical decision. A spherical equivalent refractive error of at least −0.50 diopters (D) was defined as myopia and +2.00 D or more as hyperopia. Children were considered myopic if one or both eyes were myopic. Children were considered hyperopic if one or both eyes were hyperopic, as long as neither eye was myopic.
The examining ophthalmologist assigned a primary cause of VI for eyes with an uncorrected VA of 20/40 or worse. The causes of VI were classified as refractive, cataract, amblyopia, retinal diseases, trachoma, corneal opacities, or other. Moreover, refractive error was assigned as cause of VI for all eyes improved to 20/32 or better after refraction, even if other contributing diseases were present.
Completed household enumeration and clinical examination forms were reviewed for accuracy and missing data. The prevalence of different levels of VI (VA of 20/32 or worse) and blindness was calculated for uncorrected VA, presenting VA (representing VA with the participant’s habitual optical correction at the ophthalmic consultant’s office), and best-corrected VA. The latter measurement was based on subjective refraction obtained in children with reduced uncorrected VA.
Visual acuity categories were defined as normal/near vision (20/32 or better in both eyes), unilateral VI (20/32 or better in one eye only), mild impairment in the better eye (20/40 to 20/63 in the better eye), moderate impairment in the better eye (20/80 to 20/160 in the better eye), and blindness (20/200 or worse in both eyes).11
Visual impairment data were analyzed using descriptive statistical tools, including the mean with SD, proportions, percentages, and 95% confidence intervals (95% CIs). Comparisons of categorical variables were performed with contingency table tests. For the multivariate regression analysis, we used the following commands: PROC FREQ to calculate frequencies of the variables and PROC logistics for the calculation of odds ratios (crude and adjusted), using SAS 9 (SAS Institute Inc., Cary, NC). Statistical significance was assigned when p < 0.05.
From the 12 clusters, 1613 subjects were enumerated and 1590 subjects were examined, among whom 814 (51%) of 1590 were boys and 776 (49%) of 1590 were girls. Of all the children assessed, 246 (15%) of 1590 were 10 years old, 311 (20%) of 1590 were 11 years old, 340 (21%) of 1590 were 12 years old, 298 (19%) of 1590 were 13 years old, 220 (14%) of 1590 were 14 years old, and 175 (11%) of 1590 were 15 years old. The mean ± SD age was 12.4 ± 1.6 years for boys and 12.2 ± 1.6 years for girls.
Among those examined, 1152 (72%) of 1590 had not been examined by an ophthalmologist previously. One hundred eight children (108 [6.8%] of 1590) were wearing optical correction; most of them were girls (65 [60%] of 108). The extent of time for which children had worn glasses ranged from 1 to 120 months with an average of 28.6 ± 23.7 months for boys and 27.5 ± 24.5 months for girls.
Of the patients examined, 167 of 1590 were screened due to VI in at least one eye, and 126 (75%) of 167 attended office-based examinations. The prevalence of uncorrected, presenting, and best-corrected VA of 20/40 or worse in the better eye was 91 (5.72%) of 1590, 45 (2.83%) of 1590, and 13 (0.82%) of 1590, respectively (Table 1).
After subjective cycloplegic refraction, 65 (52%) of 126 children were given a prescription for optical correction. Among this group, 15 had their vision corrected with lenses with positive spherical equivalent, ranging from 0.25 to 7.00 spherical D. Fifty myopic children were prescribed negative spherical equivalent lenses (50 [39.7%] of 126; 95% CI, 31.1 to 48.2%) ranging from −0.50 to −9.00 spherical D; 23 (46%) of 50 were boys and 27 (54%) of 50 were girls. Thus, the frequency of myopia in the sample studied was 3.14% (50 of 1590; 95% CI, 2.28 to 4.00%), with no significant difference between boys and girls (3.07 vs. 3.61%; p = 0.578) (Table 2). Astigmatism of 1.0 D or more was found in 23 (55%) of 42 right eyes and 24 (51%) of 47 left eyes (Table 3).
Refractive error was the most common cause of VI in one or both eyes (65 [89.0%] of 73 children), as previously described. Other causes of VI in children were amblyopia (4 [5.5%] of 73), retinal disorders (mainly macular scars; 3 [4.1%] of 73), and congenital cataracts (1 [1.4%] of 73) (Table 4).
Given that myopia was the main cause of VI, a multivariate analysis of the following factors was performed: sex, age, previous use of optical correction, difference in income, and previous eye examination. No statistically significant associations were observed among any of the factors examined.
This study represents the first population-based survey of refractive error performed in Brazil using the RESC protocols. Previous Brazilian refractive error surveys were traditionally based on national or regional government and nongovernmental programs8,14,15 that were conducted with nonstandardized protocols. In one study, a modified RESC protocol was applied in a group of schools in the city of São Paulo, which is located in a circumscribed low- to middle-income area and has unique social characteristics.11
Data from population-based refractive error studies in Brazil should be analyzed in light of the fact that school dropout rates differ throughout the school calendar. Although most children attend school in the beginning of the year (97%), in some age groups, especially among 14- to 15-year-olds, the dropout rates can reach 25% during the last months of the school year17 probably because of socioeconomic issues. It is possible that timing of conducting school-based refractive error surveys is an important issue to be considered by investigators—at least in Brazil.
In addition, Gurupi is home to a young population (42% of age <25 years), with an equal male-to-female ratio, a mixed racial-ethnic profile, and a GDP per capita of about US$ 6,438. These data are similar to those of the entire country with respect to GDP per capita (average of US$ 8220) and the social characteristics of the population, including the adult education level and the age-sex pyramids.16 Similar to most Brazilian cities (96%), Gurupi has a population less than 100,000 individuals,16 with higher levels of poverty in its periphery. All these facts corroborate that Gurupi is a typical Brazilian city, and although our data should be considered as representative of most Brazilian cities, a direct generalization to the entire country may not be established.
In this study, refractive error was the leading cause of VI (80.0%), followed by amblyopia (5.5%) and retinal disorders (4.1%). Most of the refractive errors associated with VI were caused by myopia (3.14%) (Table 2) at levels close to those reported in previous Brazilian studies.11,12,14,15
The frequency of refractive error observed in this study is also quite similar to that reported in some South American and African studies.3,4,18 However, the frequency of myopia was lower than those reported in Europe and, particularly, in Asia.9,19–24 Interestingly, the prevalence of VI was also lower than those presented in recent studies.25,26 A definitive explanation for the ethnic differences related to refractive error prevalence, as cause of VI in children, is open to debate.
As in several Brazilian cities, there is a shortage of available eye care services in Gurupi City. There are only four ophthalmologists in this city, and many inhabitants may have limited access to eye care services. For example, 72% of the children in this study did not have a previous ophthalmologic evaluation. On the other hand, no trachoma case was detected, and this fact may suggest educational and sanitary enhancements in this region. However, our survey was not specifically designed to detect this disease.
Some modifications were made to facilitate data collection in the field. Visual acuity was measured with a line of optotypes on a portable chart, and only those children with an uncorrected VA worse than 20/32 were referred for ophthalmologic examination. Considering these modifications, we do not report the general prevalence of refractive errors but rather the frequency of VI attributable to refractive errors.11,15
Thus, in the context of refractive error prevalence and the variable rates of school dropout, results of this population-based study are consistent with previous school-based studies in Brazil.11 The results suggest that future school-based surveys of refractive error should be preferably performed during the first months of the school calendar.
Jayter Silva Paula
Departamento de Oftalmologia
Otorrinolaringologia e Cirurgia de Cabeça e Pescoço
Hospital das Clínicas de Ribeirão Preto
12° andar–Campus USP
Av. Bandeirantes 3900
14.049-900, Ribeirão Preto
São Paulo, Brazil
The authors thank Prof Solange Rios Salomão for her helpful discussions and suggestions.
The authors declare no conflicts of interest.
This submission has not been published anywhere previously, and it is not simultaneously being considered for any other publication.
Received June 1, 2012; accepted November 19, 2012.
1. Congdon NG, Friedman DS, Lietman T. Important causes of visual impairment in the world today. JAMA 2003; 290: 2057–60.
2. Negrel AD, Maul E, Pokharel GP, Zhao J, Ellwein LB. Refractive Error Study in Children: sampling and measurement methods for a multicountry survey. Am J Ophthalmol 2000; 129: 421–6.
3. Naidoo KS, Raghunandan A, Mashige KP, Govender P, Holden BA, Pokharel GP, Ellwein LB. Refractive error and visual impairment in African children in South Africa. Invest Ophthalmol Vis Sci 2003; 44: 3764–70.
4. Maul E, Barroso S, Munoz SR, Sperduto RD, Ellwein LB. Refractive Error Study in Children: results from La Florida, Chile. Am J Ophthalmol 2000; 129: 445–54.
5. Dandona R, Dandona L, Srinivas M, Sahare P, Narsaiah S, Munoz SR, Pokharel GP, Ellwein LB. Refractive error in children in a rural population in India. Invest Ophthalmol Vis Sci 2002; 43: 615–22.
6. Murthy GV, Gupta SK, Ellwein LB, Munoz SR, Pokharel GP, Sanga L, Bachani D. Refractive error in children in an urban population in New Delhi. Invest Ophthalmol Vis Sci 2002; 43: 623–31.
7. Pokharel GP, Negrel AD, Munoz SR, Ellwein LB. Refractive Error Study in Children: results from Mechi Zone, Nepal. Am J Ophthalmol 2000; 129: 436–44.
8. Kara-José N, Holzchuh N, Temporini ER. [Refractive errors in school children in the city of Sao Paulo, Brazil]. Bol Oficina Sanit Panam 1984; 96: 326–33.
9. Zhao J, Pan X, Sui R, Munoz SR, Sperduto RD, Ellwein LB. Refractive Error Study in Children: results from Shunyi District, China. Am J Ophthalmol 2000; 129: 427–35.
10. He M, Huang W, Zheng Y, Huang L, Ellwein LB. Refractive error and visual impairment in school children in rural southern China. Ophthalmology 2007; 114: 374–82.
11. Salomão SR, Cinoto RW, Berezovsky A, Mendieta L, Nakanami CR, Lipener C, Munoz Ede H, Ejzenbaum F, Belfort R Jr., Pokharel GP, Ellwein LB. Prevalence and causes of visual impairment in low-middle income school children in Sao Paulo, Brazil. Invest Ophthalmol Vis Sci 2008; 49: 4308–13.
12. Thorn F, Cruz AA, Machado AJ, Carvalho RA. Refractive status of indigenous people in the northwestern Amazon region of Brazil. Optom Vis Sci 2005; 82: 267–72.
13. Sapkota YD, Adhikari BN, Pokharel GP, Poudyal BK, Ellwein LB. The prevalence of visual impairment in school children of upper-middle socioeconomic status in Kathmandu. Ophthalmic Epidemiol 2008; 15: 17–23.
14. Garcia CA, Oréfice F, Nobre GF, Souza Dde B, Rocha ML, Vianna RN. [Prevalence of refractive errors in students in Northeastern Brazil]. Arq Bras Oftalmol 2005; 68: 321–5.
15. Bechara SJ, Kara-Jose N. [Detection and treatment of patients with amblyopia in the city of Sao Paulo, SP, Brazil]. Rev Saude Publica 1987; 21: 326–30.
17. DRE. Relatório de evasão escolar na cidade de Gurupi-TO. Brazil: Delegacia Estadual da Educação-Gurupi; 2007.
18. Jiménez R, Soler M, Anera RG, Castro JJ, Perez MA, Salas C. Ametropias in school-age children in Fada N’Gourma (Burkina Faso, Africa). Optom Vis Sci 2012; 89: 33–7.
19. Quek TP, Chua CG, Chong CS, Chong JH, Hey HW, Lee J, Lim YF, Saw SM. Prevalence of refractive errors in teenage high school students in Singapore. Ophthalmic Physiol Opt 2004; 24: 47–55.
20. Fan DS, Lam DS, Lam RF, Lau JT, Chong KS, Cheung EY, Lai RY, Chew SJ. Prevalence, incidence, and progression of myopia of school children in Hong Kong. Invest Ophthalmol Vis Sci 2004; 45: 1071–5.
21. Edwards MH, Lam CS. The epidemiology of myopia in Hong Kong. Ann Acad Med Singapore 2004; 33: 34–8.
22. Lam CS, Lam CH, Cheng SC, Chan LY. Prevalence of myopia among Hong Kong Chinese schoolchildren: changes over two decades. Ophthalmic Physiol Opt 2012; 32: 17–24.
23. Rezvan F, Khabazkhoob M, Fotouhi A, Hashemi H, Ostadimoghaddam H, Heravian J, Azizi E, Khorasani AA, Yekta AA. Prevalence of refractive errors among school children in Northeastern Iran. Ophthalmic Physiol Opt 2012; 32: 25–30.
24. Logan NS, Shah P, Rudnicka AR, Gilmartin B, Owen CG. Childhood ethnic differences in ametropia and ocular biometry: the Aston Eye Study. Ophthalmic Physiol Opt 2011; 31: 550–8.
25. Mehari ZA, Yimer AW. Prevalence of refractive errors among schoolchildren in rural central Ethiopia. Clin Exp Optom 2013; 96: 65–9.
26. Cheng HM, Sun HY, Lin DP, Chang HH, Chen ST, Yeh SM, Peng ML, Tseng JK, Su KC, Tseng KW, Chen BY, Hsiao CJ, Huang SY, Cheng CY. Characterising visual deficits in children of an urban elementary school in Taiwan. Clin Exp Optom 2012; 95: 531–7.