Twelve month refractive and axial length changes in the Israeli refractive error, activity, and devices (iREAD) study

Introduction

Myopia, or nearsightedness, is a highly prevalent refractive error worldwide¹. The progression of myopia, particularly in children, has been a subject of extensive research due to its implications for ocular health and quality of life. Numerous studies have identified various factors influencing myopia progression, including genetic predisposition, environmental influences, and lifestyle factors^2–11.

Parental myopia is a strong risk factor for the development of myopia, which may reflect both genetic predisposition and shared environmental influences^12,13. Linkage and GWAS studies have reported over 150 common variants for refractive error; however, these variants explain only approximately 8% of the phenotypic variance of this trait¹⁴. For those loci that have been identified, many are related to light-dependent pathways¹⁵. Despite recent success in identifying genetic variants associated with myopia, the rapid increase in prevalence is beyond what genetics alone could account for, which implicates environmental and behavioral factors⁸. According to an International Myopia Institute (IMI) white paper, education and time outdoors have emerged as risk factors with strong and consistent associations with myopia^8,16–18. However, the contributions and interactions between these factors are unknown.

In societies where formal education is limited and children do not regularly attend school, the incidence of myopia is notably low^19,20. However, as national education systems expand, a rise in myopia has been observed²¹, with evidence that myopia increases with exposure to schooling²². Children who are part of academically rigorous environments or who consistently achieve higher grades are more likely to develop myopia^23–25. This trend continues into adulthood, where those who have pursued more years of education or achieved higher educational qualifications exhibit greater myopia^26,27. On a global scale, countries experiencing a surge in myopia often stand out in terms of education. For example, East Asian countries, such as China²¹and Singapore²⁸, introduce educational demands at an early age, with academic expectations set as early as preschool^28,29. Young adults in these countries have a very high prevalence of myopia, estimated at over 80%³⁰. In contrast, regions with Western educational systems with less rigorous educational demands, such as in Europe³¹, have a myopia prevalence on the order of 27%. Mendelian randomization analysis supports a causal relationship between more years of education and increased myopia³².

Time outdoors has emerged as a major factor in protecting children from developing myopia, as shown in cross-sectional, ecological, longitudinal, and randomized controlled studies¹³. A large body of epidemiological evidence on the protective effects of time outdoors has been published³³, and a systematic review and meta-analysis has confirmed the association³⁴. Importantly, increased time outdoors can reduce the impact of parental myopia¹³and higher levels of near work⁹. The evidence for causality includes school-based intervention trials that have shown that increases in time outdoors of 40 to 80 min per day produced significant reductions in the incidence of myopia and its progression^16–18,35, consistent with the expectations from the epidemiological data. Other factors, such as near work and the use of electronic devices, have not consistently shown a strong association with myopia across studies, and causality has not yet been demonstrated^8,23,36–40. While seasonal variations in myopia progression suggest a potential influence of near work and time outdoors, the extent to which these factors directly impact progression remains uncertain.

Longitudinal studies have provided valuable insights into the progression of myopia over time and have shown that genetics², near work, and outdoor activity influence myopia progression^2,33,41. The Israel Refraction, Environment, and Devices (iREAD) Study¹⁰is a longitudinal study that takes advantage of this natural experiment to minimize the impact of genetics and to focus on the educational system of the three groups of boys with distinct schooling environments. Specifically, Ultra-Orthodox boys engage in the most rigorous educational system with intense near work beginning at an early age, followed by Religious and Secular boys. The three groups differ not only in their curricula, but also in the structure of their daily schedules and school calendars. Ultra-Orthodox boys begin formal education at age three and primarily study religious texts, often for long hours⁴². Religious boys follow a dual curriculum that integrates both religious and secular subjects, leading to a more balanced distribution of study time⁴³. Secular boys, who are enrolled in the standard Israeli education system, focus primarily on general education subjects and have structured school schedules that incorporate more extracurricular activities⁴⁴. Additionally, Ultra-Orthodox boys have significantly shorter school vacations than Religious and Secular boys^45–47, further increasing their cumulative academic workload over the years. By taking advantage of the population of Israeli Jews, the impact of genetics can be minimized, allowing the study to focus on behavioral differences in various subgroups. Baseline findings from the iREAD study revealed significant differences in behaviors and myopia prevalence between groups, with myopia prevalence being significantly greater in Ultra-Orthodox boys compared to Secular boys. Similarly, refraction of Ultra-Orthodox boys was significantly more myopic than Religious and Secular boys. Ultra-Orthodox boys spent significantly less time outdoors per day (as measured objectively) than Religious and Secular boys. Multivariate analyses, adjusted for age and parental myopia, showed that being Ultra-Orthodox along with daily near work and time in school increased the odds of myopia. Interestingly, Ultra-Orthodox boys used electronic devices significantly less than Religious and Secular boys. The similarity between the myopia epidemic in Ultra-Orthodox Jewish boys⁴⁸and the well-documented rise in myopia in East and Southeast Asian populations^21,28 suggests that fundamental factors, such as high educational demands, extensive near work, and limited outdoor exposure, play a critical role in myopia development and progression.

The aim of the 12-month follow-up of the iREAD study was to assess refractive change and axial elongation in these children and to determine associated factors. The current report represents the 12-month data from boys enrolled in the iREAD Study¹⁰. We hypothesized that the Ultra-Orthodox boys would have greater myopic progression and that less time outdoors and greater daily near work and time in school would be associated with this progression.

Results

Educational and near work characteristics

Results of the questionnaire revealed differing numbers of myopic parents, educational characteristics, and behaviors between the Ultra-Orthodox, Religious, and Secular groups (Table 2). The Ultra-Orthodox group was more likely to have myopic parents than the Religious (P = 0.003) and Secular (P < 0.003) groups. The Ultra-Orthodox group spent more hours in school from Sunday to Friday (6.9 ± 0.8 h) compared to the Religious (6.4 ± 0.6 h, P = 0.01) and Secular (5.9 ± 0.8 h, P < 0.001) groups. The Religious group also spent more time in school than the Secular group (P = 0.02). The Ultra-Orthodox group was significantly more likely to have learned to read before age 6 years compared to the Religious (P < 0.001) and Secular (P < 0.001) groups. Additionally, the Ultra-Orthodox group was less likely than secular children to have a cellphone (P < 0.001).

Significantly different patterns of behavior were observed in the use of electronic devices. The Secular group had significantly more handheld electronic device use (1.6 ± 1.4 h per day) compared to the Ultra-Orthodox (0.3 ± 0.3 h per day, P < 0.001) and Religious (0.8 ± 0.8 h per day, P = 0.002) groups. The Ultra-Orthodox group had significantly less overall screen time (0.8 ± 0.8 h per day) compared to the Religious (2.5 ± 2.3 h per day, P = 0.01) and Secular (4.7 ± 3.7 h per day, P < 0.001) groups, with the Religious group also having less screen time than the Secular group (P = 0.003). Additionally, the Ultra-Orthodox group spent significantly less time on intermediate near work (1.5 ± 0.7 h per day) compared to the Secular group (2.6 ± 2.0 h per day, P = 0.008). For far viewing activities, including watching television and playing video games, the Ultra-Orthodox group spent significantly less time (0.2 ± 0.3 h per day) compared to the Religious (0.8 ± 0.9 h per day, P = 0.007) and Secular (1.7 ± 1.2 h per day, P < 0.001) groups, with the Religious group also spending less time on these activities than the Secular group (P < 0.001). In contrast, there were no significant differences between groups for time spent on writing and reading printed material (P = 0.78), near work less than 40 cm (P = 0.28), or near and intermediate work less than 100 cm (P = 0.1).

Objectively measured sleep, activity, and light exposure

Of the 125 children, 10 children did not have complete Actiwatch data: one watch was lost, six watches malfunctioned, and three watches lacked the minimum required four days of activity data. Therefore, analysis includes objective behavioral data for 115 children (Table 3). On average, children provided 9.5 ± 1.7 days and 10.8 ± 1.6 nights of valid Actiwatch data. The groups did not have a significant difference in number of valid days (Ultra-Orthodox: 9.2 ± 1.8 days; Religious: 9.6 ± 1.6 days; Secular: 9.8 ± 1.5 days, P = 0.09). However, there were differences in the number of valid nights, likely due to the Ultra-Orthodox children taking the watch off for Shabbat (Ultra-Orthodox: 10.2 ± 1.9 nights; Religious: 10.9 ± 1.4 nights; Secular: 11.3 ± 1.3 nights, P = 0.01), with a significant difference between number of nights for the Secular and Ultra-Orthodox groups (P = 0.01). No significant differences were observed between the groups for wake time (P = 0.28), bedtime (P = 0.56), sleep duration (P = 0.21), daily average white light exposure (P = 0.54), time spent outdoors (P = 0.1), daily average physical activity (P = 0.67), or time spent engaged in moderate to vigorous activity (P = 0.72).

Relationships of exposures with SER change and axial Elongation – Univariate analysis

Table 4 presents the predictor statistics for each exposure in the models assessing changes in SER and axial length. Baseline age and SER or axial length were control variables in each model. The sample size for each model varied depending on the availability of complete data for each child. All children had data for changes in axial length, while one child did not have data for changes in SER, resulting in a maximum sample size of 124 for the SER models. The only exposure with missing data was the number of myopic parents, with two children lacking this information. Consequently, the axial length model that included number of myopic parents had sample size of 123 subjects, while the SER model had sample size of 122.

In the univariate models of change in SER, significant associations were observed between group membership (Ultra-Orthodox, Religious, or Secular) and myopic shift (P = 0.005). Specifically, being in the Secular group was significantly associated with less myopic shift compared to being in the Ultra-Orthodox group, (beta value of 0.335, 95% CI: 0.137 to 0.533, P = 0.001). For axial length, the Secular group had significantly less axial elongation compared to the Ultra-Orthodox group (beta value of − 0.0973, 95% CI: − 0.1786 to − 0.016, P = 0.02). This suggests that group membership is a significant predictor of myopia shift in the univariate analyses.

The number of myopic parents was also a significant predictor of SER change (P = 0.019). In comparison to children with two myopic parents, children with no myopic parents showed significantly less myopia shift in SER (beta = 0.264, 95% CI: 0.029 to 0.499, P = 0.03), as did children with one myopic parent (beta = 0.212, 95% CI: 0.04 to 0.384, P = 0.02). For axial length, the difference was not statistically significant (P = 0.10), though there was a trend towards less axial elongation in children with no or one myopic parent.

None of the other behavioral characteristics examined showed significant associations with changes in SER or axial length. These findings suggest that these specific behavioral characteristics are not significant predictors of myopia progression or axial elongation over the one-year period studied.

Relationships of exposures with SER change and axial Elongation – Multivariate analysis

Table 5 summarizes the results for multivariate models of SER change and axial elongation fit to identify exposures that might be statistically significant when controlling for other exposures. The predictors, which are listed in the first column of Table 5, were selected because they were of theoretical interest. Two predictors emerged as being associated with greater SER change: group membership (P = 0.03) and number of myopic parents (P = 0.04). Being in the Secular group was associated with significantly less myopic shift compared to being in the Ultra-Orthodox group, with a beta value of 0.42 (95% CI: 0.10 to 0.74, P = 0.01). The Religious group showed a trend for less myopic shift, but it was not significant (beta = 0.18, 95% CI: − 0.07 to 0.44, P = 0.15). In addition, an increase of one myopic parent was associated with an average increase in myopic shift of − 0.13 D (beta = − 0.13, 95% CI: − 0.25 to − 0.007, P = 0.04). In this model, no other predictor (learning to read before age 6, all screen use, sleep duration, time outdoors, physical activity and base line data of SER or AL respectively) emerged as significant.

Group and number of myopic parents did not emerge as significantly associated with axial elongation (P = 0.33 and P = 0.09, respectively). Baseline age was significantly associated with axial length elongation with beta value of − 0.06 (95% CI: − 0.08 to − 0.03, P < 0.001).

The original list of exposures for multivariate modeling (Table 5) included near work less than 100 cm, which was strongly correlated with screen use (Pearson correlation of 0.713). As both screen use and near work should not be included in the same model, a new model was created in which near work replaced screen use. Similar results were obtained, and two predictors emerged as being associated with greater SER change: group membership (P = 0.02) and number of myopic parents (P = 0.05). No predictors were significant for axial elongation (Supplemental data 02).

Discussion

The iREAD Study describes the impact of educational system and lifestyle on myopia progression. As hypothesized, Ultra-Orthodox boys had significantly more myopic shift than their Religious and Secular peers. This higher myopic shift was accompanied by significant differences in behaviors, including more time in school, learning to read at a younger age, and less use of electronic devices (including cell phones). In addition, Ultra-Orthodox boys had significantly more myopic parents. In both univariate and multivariate analyses, two factors emerged as being associated with a greater myopic shift: belonging to the Ultra-Orthodox group and having myopic parents. Despite results from the baseline iREAD study showing group differences¹⁰, time outdoors, near work at home, age learned to read, and hours in school did not emerge as being associated with a greater myopic shift.

The Jewish population in Israel originates from various European populations and consists of two distinct subcultures, Ashkenazi and Sephardic. Ashkenazi Jews originated in France, Germany, and Eastern Europe. Sephardic Jews originated in Spain, Portugal, North Africa and the Middle East. Genetic analyses ties Ashkenazi Jews to just 330 people from Middle Ages⁵⁰, contributing to relative genetic homogeneity in this population. Therefore, while not completely genetically homogenous, the effects of genetic and geographic variability are minimized. The different school systems in Israel for the Jewish population provide a natural experiment for the study of myopia. Natural experiments in healthcare research are observational study designs where interventions are determined by external variations beyond the researcher’s control, simulating randomization⁵¹. This approach utilizes quasi-random allocation methods to create comparable groups for study. While less definitive than randomized clinical trials (RCT) in establishing causality, natural experiments are invaluable in scenarios where RCTs are impractical or unethical, providing crucial insights into causal relationships between interventions and health outcomes. This method is particularly valuable for evaluating the effects of educational methods and other situational shifts on comparative groups. It would not be ethically acceptable to perform an RTC that randomly assigned some children to an intensive educational system and other children to a less rigorous system. However, in Israel, this occurs in a systematic manner because the Jewish population in Israel has three distinct educational systems, each associated with different lifestyles and prevalence of myopia^48,52. Despite different lifestyles and educations, the population in Israel is relatively genetically homogenous and geographically small^50,53, thereby minimizing confounding factors such as ethnicity and geographic location.

Findings of the current study suggest that something in the Ultra-Orthodox lifestyle is associated with more myopic refractive error and greater myopic shift than Religious and Secular lifestyles. However, the behavioral factors tested in this study did not account for the difference. Findings of the current study suggest that something in the Ultra-Orthodox lifestyle is associated with more myopic refractive error and greater myopic shift than Religious and Secular lifestyles. However, the behavioral factors tested in this study did not account for the difference. Notably, although myopic shift in the Ultra-Orthodox group was significantly greater than in the Religious group (P= 0.05), the difference was modest. This may be due to the fact that the Religious group exhibited behavioral patterns more similar to the Ultra-Orthodox than to the Secular group. Several hypotheses could explain this. First, near work at school was not examined in this study. It is known that boys in the Ultra-Orthodox school system spend a significant amount of time reading small text with font sizes as small as 1 mm in height, potentially leading to shorter working distances. Indeed, studies in adults show that Ultra-Orthodox men read at significantly shorter working distances than non-Ultra-Orthodox men⁵⁴. However, children in East and Southeast Asia did not show variation in print size, yet still experienced a significant myopia epidemic²¹. Second, it may be that differences in light exposure and near work between groups occur during school holidays and summer break, which were not assessed in the current study. Supporting this notion, Ultra-Orthodox schools in Israel typically have 50 more school days than Religious and Secular schools. For example, the Religious and Secular schools have summer break for the entire months of July and August. In contrast, Ultra-Orthodox schools only have three weeks of summer break, typically in July or August. Third, behavioral differences contributing to greater myopic shift could be exposures experienced at a younger age than included in this study. Ultra-Orthodox boys start formal schooling at the age of 3 years, which is likely associated with spending more time indoors engaged in near work at that age. Secular and Religious boys start formal schooling only at the age of 6 years, while generally spending much of their time playing before age 6 years. Finally, the difference could be some unknown aspect of the Ultra-Orthodox lifestyle not considered in this study.

In the multivariate analysis performed here, number of parents with myopia emerged as being independently associated with myopic shift. This might suggest a genetic basis in that the children inherit specific genes from their parents. This finding is consistent with previous studies, which have shown that parental myopia is a significant risk factor for myopia in children^{12,23,55–57}. However, differences between familiar risks and genetic risks should be considered, especially in the current study. It is possible that sex-linked genetic variants that are more prevalent in the Ultra-Orthodox group contribute to the higher prevalence of Myopia. However, as all three groups are of Jewish ancestry and the separation into different levels of religiosity occurred only a few generations ago^50,53we do not expect a significant difference between their genetic background. Thus, it is unlikely that genetics explains the difference between the Ultra-Orthodox, Religious, and Secular groups. Additionally, previous studies have shown that genetic variants explain only approximately 8% of the phenotypic variance in refraction¹⁴. Lastly, the rapid increase in prevalence of myopia in general is beyond what genetics alone could account for, which implicates environmental and behavioral factors^8,11. It is likely that Ultra-Orthodox parents and their children have similar behaviors that lead to greater myopia and progression, due to being in the same family.

At baseline, the iREAD Study¹⁰revealed significant differences in myopia prevalence among the groups. The Ultra-Orthodox group exhibited a higher prevalence of myopia (46%) compared to the Religious (25%) and Secular (20%) groups. This divergent rate of myopia in different populations of Israeli boys has been reported in several studies^48,52,58. On the other hand, at 12 months, this study did not find a significant difference in the prevalence of myopia between the groups^10,48,52. This may be due to the relatively small number of Secular children who complied with the 12-month visit. In general, Jerusalem has a much smaller Secular community than Religious or Ultra-Orthodox communities, posing challenges to recruitment. Despite this, refractive error became more myopic in all groups, indicating continued myopic shift. However, a trend of increasing myopia prevalence was observed across all groups, particularly in the Ultra-Orthodox group.

While no significant differences were observed in near work or time outdoors between groups, the observed progression may result from individual variations in susceptibility to myopia, such as baseline refractive error or genetic predisposition. Previous studies have demonstrated that children with a more myopic baseline refraction are at a higher risk for rapid progression^59,60, and those with myopic parents are more likely to develop and progress in myopia⁶¹. Some children may have already been on a trajectory for faster myopia progression independent of measurable lifestyle factors, as has been observed in other longitudinal studies^62,63.

Axial elongation over 12 months was similar between the groups of children examined here. The only factor associated with axial elongation in multivariate analysis was baseline age, highlighting that younger children are at a higher risk for more rapid myopia progression. This finding is consistent with existing literature, which indicates that children who develop myopia at an earlier age tend to experience more rapid axial elongation, leading to higher degrees of myopia later in life^59,64,65. In Taipei, children with more myopia at baseline experienced faster myopia progression over one year⁶⁰. A nationwide longitudinal study found that children aged 7–9 years with higher initial myopia (SE ≤ − 4.00D) had a higher proportion of myopia progression greater than − 0.50 D per year⁶⁶. These findings highlight the importance of closely monitoring and managing myopia progression from a young age.

In contrast to our hypothesis, time outdoors and daily light exposure did not emerge as significant predictors of myopic shift. Children in the current study spent 1.0–1.3 h per day outdoors, as measured objectively. This is similar to time spent outdoors in children during school sessions in the United States⁶⁷, when measured using the same methods. At baseline, Ultra-Orthodox boys had more outdoor time than Religious and Secular boys. However, the differences at 12 months were not significant. These findings contrast with multiple large-scale studies demonstrating a protective effect of outdoor time against myopia development and progression^34,68. One possible explanation is that other factors, such as near work intensity or indoor lighting conditions, may have outweighed the protective benefits of outdoor exposure. Additionally, it is possible that the relationship between time outdoors and myopia progression is more complex and influenced by factors such as age, genetic predisposition, or cumulative lifetime exposure rather than short-term changes over a 12-month period. Future studies should investigate whether earlier or more prolonged outdoor exposure has a stronger protective effect.

Our study found that while early reading significantly differed between groups, it was not an independent risk factor for myopia progression in either univariate or multivariate analyses. It is possible that the observed relationship between early reading and myopia may be confounded by other factors, such as group-specific behaviors or cultural influences. In particular, early reading may be strongly associated with the ultra-Orthodox Jewish population in our sample, where intensive near-work activities from a young age are common. Alternatively, it may reflect an early-life exposure effect, wherein the cumulative burden of near work over time contributes to myopia development^22,69–73. This interpretation aligns with a recent review from Brennan et al.⁷⁴, who provided robust evidence that education plays a causal role in myopia development. Their review synthesized large-scale studies from China, where school entry age is strictly regulated, and demonstrated that cumulative years of schooling, rather than chronological age, are the primary driver of myopia onset and progression. The natural experiment created by school enrollment policies showed that children exposed to an additional year of education had significantly higher rates of myopia, independent of their birth month^22,69–74. These findings reinforce the idea that prolonged exposure to near work and educational demands during early childhood is a key factor in myopia development. Thus, while our study did not identify early reading as an independent risk factor, it is possible that its impact is embedded within broader educational and environmental influences.

A recent systematic review of the association between screen time and myopia reported that findings across studies are conflicting, with no clear association between screen time and myopia prevalence, incidence, or progression in children⁷⁵. On the other hand, other systematic reviews have revealed that children who spend more time using screens have a higher risk of developing myopia^76,77. Our findings support that, in Israel, the use of electronic devices is not driving the myopia epidemic. Ultra-Orthodox boys, who exhibited higher rates of myopia progression than Religious and Secular boys, generally have limited access to and use of screens. The discrepancy between our findings and those of others may be attributed to the specific characteristics of the Ultra-Orthodox population, where screen use is minimal and other factors, such as early schooling, are more prominent. Interestingly, Ultra-Orthodox boys spend much less time using hand-held devices and have lower total screen time, in contrast to widespread speculation about the role of screen exposure in myopia development. This observation aligns with historical evidence from East Asia, where the myopia epidemic emerged before the advent of smartphones or mass access to the internet²¹.

The questionnaire used in the current study asked parents to estimate time engaged in near work while children were at home. Findings showed that near work performed at home was not a significant predictor of myopia. Previous studies have shown mixed results regarding the impact of near work on myopia. While some studies report associations between increased near work with myopia^9,78,79, others report no or limited correlations^39,80,81. Given that Ultra-Orthodox boys spend a significant portion of their day engaged in near work activities at school, it is possible that near work during school or other unmeasured factors related to near viewing behaviors, such as the specific types of near work, near viewing distance, and near viewing breaks, play a significant role in myopia progression. Near work performed during school hours was not measured in this study, which could be a critical factor. The cumulative effect of these educational practices over time could contribute to the higher rates of myopia progression observed in this group. Future studies should adopt a more nuanced approach by considering the specific educational practices and near work environments of different cultural groups to better understand their impact on myopia progression. Additionally, use of wearable sensors for continuous and objective measurement of near viewing behaviors would offer more precise quantification^54,82.

This study has limitations that should be considered. Sample size, particularly within each subgroup (Ultra-Orthodox, Religious, and Secular), was relatively small, which may limit the generalizability of the findings. The lack of statistical significance for changes in prevalence and axial length between groups underlines the challenges associated with sample size constraints. While consistent trends were observed, significant associations between near work, time outdoors, and myopia progression, reported in previous studies^{9,34,68,78,79}, were not detected in our analysis. However, this does not necessarily indicate the absence of an effect but rather suggests that a larger sample may be required to detect these associations with greater confidence. Alternatively, it is possible that key exposures occurred during periods not measured in this study. Notably, near work during school hours was not assessed, which may be a crucial factor given that Ultra-Orthodox boys engage in extensive near work throughout the day. Additionally, this group has shorter school vacations compared to religious and secular children^45–47, potentially leading to reduced outdoor light exposure. Another consideration is that early reading and light exposure at a younger age may be contributing factors, as Ultra-Orthodox boys begin formal schooling at age 3. This aligns with recent findings^22,69–73that demonstrated that additional years of education are strongly associated with increased myopia progression, reinforcing the importance of cumulative educational exposure. Despite these limitations, the study’s tightly controlled methodology and use of wearable sensors to quantify behaviors enhance the robustness of the findings. However, the study relied on self-reported estimates for near work and screen use, which is known to be subject to reporting bias. Additionally, parents only reported on after-school behaviors. Detailed information about children’s activities during school hours, which may significantly influence myopia progression, were lacking. Future studies should consider integrating both objective measurements and real-time self-reported surveys to capture a more comprehensive picture of near work behavior during school hours. In particular, wearable sensors capable of tracking near viewing distances and durations, could provide more accurate insights into how these behaviors contribute to myopia progression. The study’s follow-up period was limited to 12 months; longer-term studies are needed to better understand the progression of myopia over time. Finally, the study focused exclusively on boys, raising the question of whether similar results would be observed in girls from these populations. Gender differences in myopia progression have been observed in some studies, with girls generally showing more myopia and faster progression^13,83. Investigating whether the same patterns hold true for girls in similar cultural contexts would be valuable, particularly given that Ultra-Orthodox girls attend different types of schools which are more similar to Religious and Secular schools.

In conclusion, this study highlights the significant impact of lifestyle and educational environment on myopia progression. The significant differences in myopic shift among Ultra-Orthodox, Religious, and Secular children highlight the need for culturally tailored interventions. Although concerns about screen use and its association with myopia are widespread, our study did not find a significant link between screen use and myopia progression in the Ultra-Orthodox population, likely due to their low levels of screen exposure. This finding reinforces that extensive screen use is not required for myopia progression, as Ultra-Orthodox boys exhibited high progression despite minimal screen exposure. Similar trends have been observed in East Asia and Singapore, where the myopia epidemic emerged before widespread access to digital devices, indicating that intensive near work, regardless of medium, is a key factor. Additionally, findings emphasize the importance of considering baseline age and parental myopia when assessing myopia risk, as younger children and those with myopic parents are at higher risk for rapid progression. These associations should inform the development of personalized strategies to manage and prevent myopia, particularly in high-risk populations.

Methods

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

E.S., A.G.S., L.O. and R.D. initiated the idea and made substantial contributions to the conception and design of the work, the acquisition, global analysis, and interpretation of data. They wrote the main manuscript text and prepared the figures, and the tables. K.D. performed data analysis. L.T.S and L.A.J performed statistical analysis. J.L. worked on data acquisition. All authors reviewed and revised the manuscript and commented on all parts and have approved the submitted version.

Funding

United States-Israel Binational Science Foundation 2019053.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations