Effective refractive error coverage and spectacle coverage among school children in Telangana, South India

Abstract

Background

Uncorrected refractive error (URE) is one of the leading causes of childhood vision impairment. Increasing effective refractive error coverage (e-REC) is one of the main indicators of WHO’s 2030 global eye health targets. The aim of this study is to estimate the e-REC and spectacle coverage among school children in Telangana, South India.

Methods

School children aged 4–15 years in the study locations underwent vision screening using 6/12 tumbling E optotype by trained community eye health workers in the schools. Those failing the initial vision screening and/or found to be having eye conditions were referred to a nearby referral centre appropriately, where they underwent detailed eye examination.

Results

A total of 774,184 children were screened in schools of which 51.49% were boys. The mean age was 9.40 ± 3.27 years. The prevalence of URE was 1.44% (95%CI:1.41–1.46) of which myopia was 1.38% (95%CI: 1.35–1.41). In multivariate analysis, the risk of myopia was higher among older children (Adj.OR: 17.04; 95%CI: 14.64–19.85), those residing in urban areas (Adj.OR:3.05, 95%CI:2.60–3.57), those with disabilities (Adj.OR:2.61, 95%CI:2.00–3.39) and among girls (Adj.OR:1.30, 95%CI:1.25–1.35) (P < 0.001). The overall e-REC was 56.97% and the spectacle coverage was 62.83%.

Conclusion

The need for interventions to improve e-REC to achieve 2030 global eye health target is eminent among children in this region. Improving refractive services through school eye health programs could aid in accelerating this process to achieve the target. Myopia being the most common type of RE, the risk factors included increasing age, urban location, and presence of disability.

Introduction

Globally, uncorrected refractive error (URE) is one of the leading causes of vision impairment in children [1]. Early detection and correction of refractive error (RE) will lead to a better visual outcome, improved quality of life and better future socio-economic prospects for children and their families [2]. The prevalence of RE is on rise globally, especially in Asia and Southeast-Asia due to increasing myopia prevalence in these regions [3]. In India, the prevalence of RE among school children was reported as 10.8% in a systematic review in 2018 [4]. With the changes in lifestyle choices such as excess digital gadget usage and less exposure to outdoor activities among school children, the risk of developing myopia is very eminent [4, 5]. Although the RE prevalence reported in the systematic review seems to be on the higher side, reports from south Indian states show relatively lower prevalence at 4.42% (2011–2015) and 2.28% (2014–2019) in Tamil Nadu and Andhra Pradesh respectively [6, 7].

On the other hand, World Health Organization’s (WHO) universal health coverage to achieve sustainable developmental goals (SDGs), focuses on combating vision impairment and its causes such as RE to improve the eye health among communities. One of the 2030 global eye health targets proposed during seventy-fourth world health assembly is to improve eye health needs by measuring and increasing effective refractive error coverage (e-REC) by 40% points [8]. Therefore, the aim of this study is to estimate the e-REC and spectacle coverage among school children in Telangana, India. This study is part of a large-scale school screening program named “Initiative for Screening Children for Refractive Errors and other Eye Health Needs (I-SCREEN)” which was conducted in Andhra Pradesh and Telangana states during 2015–2019. The data from Andhra Pradesh region was published earlier in July 2022 [7].

Methodology

The study protocol adhered to the tenets of declaration of Helsinki. Ethics approval was obtained from the local institutional review board of Hyderabad Eye Research Foundation, L V Prasad Eye Institute, India. The study methodology presented in this manuscript has also been described in our previous paper [7]. Prior to the study data collection, all the necessary approvals from the government authorities and a signed informed consent from the school managements were obtained. Prior to the screening, the parents were informed about the program in detail by the school managements.

Study area and population

The study included tribal, rural, and urban areas from Adilabad, Nirmal, Mahabubnagar and Hyderabad districts in Telangana state, India. (Fig. 1) Children aged 4–15 years and attending schools in the study area during June 2015 to April 2019, were included in the study.

Fig. 1.

Map of the study area.

Definitions

RE were categorized based on Refractive Error Study in Children (RESC) study protocol as following; Spherical equivalent (SE) in the worse eye, of at least −0.50D or more as myopia and SE of ≥ + 2.00D as Hyperopia [9]. RE magnitude was categorised as follows; low myopia (SE: −0.50 to −5.75D), high myopia (SE: −6.00 and above) [10], low hyperopia (SE: +2.00 to +5.00D) and high hyperopia (SE: +5.25D and above) [11–13]. VI was defined as per the WHO ICD-11 classification which is presenting visual acuity (PVA) of worse than 6/12 in the better eye [14]. The e-REC was calculated for visual acuity of 6/12 or better as cut-off in the better eye and using the e-REC formula which included met need, under-met need, and unmet need as proposed by McCormick et al (e-REC = Met need/ (Met need+Under-met need+Unmet need)*100) [15]. The met need was defined as those wearing refractive correction and do not have VI (PVA of 6/12 or better), under-met need was those wearing refractive correction but cannot see 6/12, unmet need was those with PVA of worse than 6/12, improving with refraction and do not yet have refractive correction. The sum of met need, under met need and unmet need was considered as total refractive error burden.

Screening procedure and referral tracking

The initial vision screening was done by trained community eye health workers (CEHWs) who were trained as per the National program for prevention of blindness and vision impairment guidelines [16]. The CEHWs visited schools and screened children using 6/12 tumbling E optotypes kept at 3 metres distance. VA was measured using current spectacles if a child was already wearing a correction. A minimum of 3 out of 4 letters had to be identified correctly to pass the VA cut-off. Those who failed the VA cut-off or those who were found/reported to have any ocular symptoms/abnormalities were referred to a nearby vision centre (VC). At the VC an experienced vision technician (VT) performed basic dry refraction including objective retinoscopy and subjective acceptance. Additionally, a VT visited schools and performed dry refraction on those children who did not visit the VC as advised by the CEHWs. Children who required cycloplegic refraction (children under the age of 6 years, VA not improving beyond 6/9 with dry refractive correction, myopia of >6.00D, hyperopia of >2.00D, astigmatism of >2.00D) and those needing referral services for other ocular conditions to a nearby secondary eye care centre (SC). At the SC, children underwent complete eye examination including cycloplegic refraction using 1% cyclopentolate eye drops instilled one drop twice at 20 min interval following which wet retinoscopy was performed by a senior VT or an optometrist under the supervision of an ophthalmologist. Finally, post dilated fundus examination, final diagnosis and management was done by an ophthalmologist. Those requiring refractive correction or change of spectacles received spectacles at this stage for free of charge. Those requiring tertiary care services were referred to a higher centre at this stage.

Data analysis

Stata/SE 14.2, Special edition by Stata Corp LLC, TX, USA was used for the data analysis. Descriptive analysis to tabulate the summary statistics and univariate and multivariate logistic regression analysis for risk estimation using odds ratio were used. Chi-square test was performed to check significance within groups. P value of <0.05 was considered as significant.

Results

A total of 774 184 children were screened in 354 schools of which 51.49% (n = 398,596) were boys. The mean age was 9.40 ± 3.27 years (Range:4–15 years). Overall, 3.73% (n = 28 895 / 774,184) children were from schools in tribal areas, 55.45% (n = 429 277 / 774184) from schools in rural areas and 40.82% (n = 316 012 / 774 184) from school in the urban areas were screened. A total of 1.93% (N = 14,918 / 774,184) failed the VA cut-off (PVA < 6/12) in the initial vision screening test (PVA < 6/12) and were referred.

Prevalence of uncorrected refractive error

The prevalence of URE among those who had VI (<6/12) was 1.44% (n = 11,116 / 774,184; 95%CI:1.41–1.46) of which myopia was 1.38% (n = 10 686 / 774 184; 95%CI: 1.35–1.41); Hyperopia was 0.06% (n = 427 / 774 184; 95%CI:0.05–0.06). Based on the RE magnitude, low myopia was the most common error at 1.34% (n = 10,388 / 774 184; 95%CI: 1.28–1.33) followed by high myopia at 0.03% (n = 243 / 774 184; 95%CI: 0.02–0.04). RE prevalence was 0.58% (n = 169 / 28 895), 1.30% (n = 5599 / 429 277), 1.69% (n = 5348 / 316,012) in tribal, rural, and urban regions respectively which was statistically significant (P < 0.001) (Table 1).

Table 1.

Prevalence of uncorrected refractive error among school children in Telangana.

	N (%)
Total	N = 774,184
No RE	763 068 (98.56)
Overall RE prevalence	11 116 (1.44)
Regional specific RE prevalence
Tribal (n = 28,726)	169 (0.59)*
Rural (n = 423,678)	5 599 (1.32)*
Urban (n = 310,664)	5 348 (1.72)*
*P < 0.001
RE magnitude based on spherical equivalent
Low hyperopia (+2.00 to +5.00D)	387 (0.05)
High hyperopia (+5.25D and over)	40 (0.01)
Low myopia (−0.50 to −5.75D)	10 388 (1.34)
High myopia (−6.00D and over)	243 (0.03)

Risk factors for myopia

Multivariable analysis showed an increased risk of myopia with increasing age; children who are in 11–15 years age group showed higher odds of developing myopia (Adj.OR: 17.04; 95%CI: 14.64–19.85) followed by those in 6–10 years age group (Adj.OR: 5.37; 95%CI: 4.60–6.27) as compared to 4–5 years age group (Fig. 2). Children in the rural and urban locations had 2.40 times (95%CI:2.05–2.80) and 3.05 times (95%CI:2.60–3.57) the odds of developing myopia respectively as compared to those studying in schools in the tribal locations. Disability increased the risk of developing myopia by 2.61 times (95%CI:2.00–3.39) and girls had 1.30 times (95%CI:1.25–1.35) the odds of developing myopia as compared to boys (p < 0.001) (Table 2).

Fig. 2.

Increasing trend of uncorrected refractive error prevalence with age.

Table 2.

Risk factors for myopia among school children in Telangana (n = 10,631).

	Univariate		P value	Multivariate		P value
Age	OR	95% CI		OR	95% CI
4–5 years	Reference			reference
6–10 years	7.76	6.77 - 8.88	<0.001	5.37	4.60–6.27	<0.001
11–15 years	23.34	20.41 - 26.68	<0.001	17.04	14.64–19.85	<0.001
Gender
Male	Reference			Reference		<0.001
Female	1.33	1.28–1.38	<0.001	1.3	1.25–1.35	<0.001
School location
Tribal	Reference			Reference
Rural	2.6	2.22 - 3.04	<0.001	2.4	2.05–2.80	<0.001
Urban	7.11	6.09 - 8.31	<0.001	3.05	2.60- 3.57	<0.001
Disability
No disability	Reference			Reference
Disability present	2.66	2.05 - 3.45	<0.001	2.61	2.00–3.39	<0.001

^*P value not statistically significant (P > 0.05).

P < 0.05 was considered as significant.

Regional variations in prevalence of uncorrected refractive error

While comparing different regions, the URE prevalence was higher in Hyderabad (urban) at 2.89% (n = 1121/38,735; 95% CI: 2.73–3.07;) followed by Nirmal district (including Mudhole and Nizamabad regions) at 1.54% (n = 4531/293,687; 95% CI:1.50–1.59;), Mahabubnagar 1.26% (n = 3175/252,968; 95% CI: 1.21–1.30;) and Adilabad 1.21% (n = 2289/188,794; 95% CI: 1.16–1.26). (Fig. 3) The URE prevalence increased with age; Age groups 8 years and older showed a steady increase in prevalence (Fig. 2).

Fig. 3.

Regional variations in uncorrected refractive error prevalence in Telangana.

Refractive error burden and effective coverage

A total of 2.1% (n = 16,234/774,184) children had refractive correction at the time of screening. The overall RE burden (Met need+Under-met need+Unmet need) was 3.34% (N = 25,836 / 774,184). Of the overall burden, the met need was 1.90% (n = 14 720 / 774 184), under-met need was 0.20% (n = 1514 / 774,184), unmet need was 1.24% (n = 9602 / 774,184). The e-REC was 56.97%. The e-REC also increased with age, and it ranged between 31.58% among the 4 years olds to 68.57% among the 15 years old children. The e-REC among boys was 47.04% and 39.06% among girls. The e-REC in schools in the tribal regions was as low as 40.49% followed by 59.55% in rural areas and 58.75% in urban areas. The overall spectacle coverage was 62.83%. The spectacle coverage also increased with age with 38.38% among 4 years olds to 72.63% among 15 years old children. (Fig. 4) The spectacle coverage was marginally lower among boys at 57.61% as compared to 61.64% among girls. The spectacle coverage was 51.41%, 70.89%, 69.91% in the tribal, rural, and urban regions respectively.

Fig. 4.

Effective refractive error coverage and spectacle coverage age-wise distribution.

Discussion

Refractive error and myopia

URE prevalence in Telangana regions ranged between 1.21 and 2.89% in the current study. The estimates from schools in tribal and rural areas show a lower prevalence as compared to what was reported in Andhra Pradesh (2.38%) [7], however the prevalence from urban locations is comparable (2.89%). The increased risk of RE in the urban areas was also reported in the systematic review and a recent multicentric study from India as well as reports from outside India [4, 17, 18]. The overall URE prevalence was 1.44% in this study which is lower as compared to other studies from south as well as north India [4, 7, 19]. The reason for the lower prevalence in this study could be due to the fact that many refractive error studies used ≤6/9.5 as the screening cut-off whereas in this study it was <6/12 [18]. Additionally, it could also be due to the underestimation of RE owing to the methodology of this study which did not include children who were already wearing spectacles and did not fall into the VI category (<6/12) at the initial screening. This would have underestimated the low myopic as well as latent hyperopic refractive error among these children. Therefore, these aspects should be kept in mind while interpreting and comparing the RE prevalence estimates from this study.

Myopia was the most common type of RE among school children reported in this study accounting for 95.83% of all URE, which is in line with other studies from within as well as outside India [6, 7, 18, 20]. Although, a systematic review reports an exponential increase in myopia prevalence in Asian context, in India which represents south Asia, it seems to be still on the lower side compared to other East Asian regions, North America and Europe [21, 22]. However, a recent report predicting the time trends on the myopia prevalence points to an eminent possibility of myopia epidemic in urban India in the near future [23]. The same report also shows a 4.76 times (4.44% to 21.15%) increase in myopia prevalence between 1999 to 2019 among 5–15 years age groups in the urban context. Another study reporting myopia prevalence among adolescents and young adults (6–22 years) in Hyderabad which is an urban setting, during 2013–2015, shows an increased prevalence of myopia at 29.8% [24]. Inclusion of age groups over 15 years also could be one of the reasons for the increased RE prevalence reported in the above study compared to the current study.

The North Indian Myopia study (NIM) (5–15 years age group) and The Sankara Nethralaya Tamil Nadu Essilor Myopia study (5–16 years age group) from south India also reported an increased myopia prevalence in the recent times that included rural, suburban and urban areas [5, 25, 26]. The prevalence reported in these studies ranged from 6% in rural schools to 17% in urban settings. However, the RE prevalence found in Telangana regions is much lower. One reason could be due to the RE definition wherein, both these studies considered PVA of <6/9 or <6/9.5 as the cut-off for refraction whereas this study considered PVA < 6/12. Secondly, the NIM study all children with refractive error underwent cycloplegic refraction. Although the current study shows much lower prevalence (2.80%) in the same region as compared to what has been reported by the above studies, there is 1.51 times more prevalence in the urban areas as compared to rural settings. Hence, the possibility of increasing prevalence of myopia in urban settings seem to be realistic and of a public health concern. A recent study also reported increasing myopia prevalence among rural school children in north India [26]. However, the rural prevalence still seems to be lower as compared to the urban settings. This study findings are also consistent with similar findings.

In this study older age, female gender, presence of disability, and urban settings were found to be associated with higher risk of myopia. These findings are consistent with the findings reported in a systematic review, meta-analysis and other studies [22, 25, 27]. Apart from the risk factors mentioned above, there are also other risk factors which are associated with myopia such as parental myopia, history of premature birth, time spent outdoors, socio-economic status, amount of near work, hours of sleep, digital gadget usage and population density, which have been reported [5, 21, 22, 26, 28]. However, these risk factors were not included in this study and hence cannot be commented upon.

Effective refractive coverage

The overall e-REC among school children in Telangana region was 56.97% during 2015–2019. At present there are no e-REC data in school children to compare from India or even other parts of the world. This is the first study reporting e-REC among school children. In children the e-REC increased with age which, seemed to range between 31% (younger children)-69% (older children), and it was about 8% higher in boys as compared to girls among school going children in this study. This is an important finding suggesting that there is a gender inequality in terms of availability of refractive services for girls. However, addressing this area further by identifying the reasons/barriers for such disparity will add more value for planning interventions in the future. Additionally, the e-REC among school children in the tribal locations where 10% lower as compared to rural as well as urban locations. This disparity could be due to the lack of availability and accessibility of refractive services in tribal regions as compared to rural and urban settings. The spectacle coverage among school children was found to be close to 63% in this study which is comparable to 66% which was reported during 2019 in schools in New Delhi [29], and comparable to what was reported in China [30, 31], However, it seems to be much higher than what was reported in Chile (10%) [32].

Strengths and limitations

This was a large-scale school screening study and hence included close to 0.8 million children. The data from this study also represents a wider geographical reach and hence providing a more generalizable estimates for the state of Telangana. This study included school going children who are ≥ 4 years of age and hence might have missed out information on children who are not attending schools and those who are less than 4 years old. However, as the school enrolment in these regions ranged between 83–96% in Telangana during the study period [33], this data is more likely to represent data of children from population-based studies in this region. Secondly, as the study considered only those who failed visual acuity cut-off of worse than 6/12 for refraction, there is a possibility of underestimation of lower magnitude RE, especially myopia. Details on risk factor for myopia such as time spent outdoors, digital gadget usage, time spent on near activities and parental myopia were not collected, hence, this should be kept in mind while interpreting the data on myopia risk factors. Finally, in this study children who were prescribed spectacles at vision centres did not undergo cycloplegic refraction, hence there is a possibility of missing out on significant hyperopic refractive errors especially among younger children.

Conclusion

It is evident that URE is one of the growing problems among school children in these regions of Telangana. Implementation of public health intervention strategies such as school eye health programs with the provision for good quality affordable spectacles, in addition to bridging the gender inequality and increasing refractive services to more neglected regions such as tribal areas, would aid in increasing the e-REC among school children in this region. These efforts would lead to achieving the 2030 global eye health targets towards attaining universal eye health for all without causing financial impoverishment for those who need these services.

Summary

What was known before

• One of the two major indicators for 2030 global eye health targets set by World Health Organization (WHO) include increasing effective refractive error coverage (e-REC) by 40%.

• However, currently there is no evidence reporting on e-REC among children in the literature both within and outside India.

What this study adds

• This study reports on both e-REC and spectacle coverage among school children covering a larger geographical area as part of a large-scale school vision screening program.

• The findings from this study would serve as a baseline data for future school eye health programs and research planning in the topic.

Author contributions

First author: Winston D Prakash. Data cleaning and analysis, interpretation, and manuscript writing. Second Author: Srinivas Marmamula: Study design, data collection and manuscript review. Third author: Jill Keeffe: Study design and manuscript review. Fourth & corresponding author: Rohit Khanna: Study concept, design, data interpretation and manuscript review.

Funding

This study was funded by Lavelle Fund for the Blind, USA, Sun Pharma Corporate Social Responsibility grant and Hyderabad Eye Research Foundation, India.

Data availability

The dataset can be made available on request.

Competing interests

The authors declare no competing interests.

Ethics approval

This study was approved by Institutional Review Board of Hyderabad Eye Research Foundation prior to the screening. Ethics reference number: LEC 01-15-011.