Introduction

The issue of myopia in children and adolescents has become a global public health concern. By 2050, the world’s myopic population is expected to reach 5 billion, accounting for approximately 50% of the total population1. China is among the countries with a high incidence of myopia, and in recent years it has gradually shown a trend towards higher prevalence and younger age2. According to the latest data released by the National Health Commission of the People’s Republic of China, the overall myopia rate among Chinese children and adolescents reached 52.7% in 2020, with rates of 35.6% in primary school students, 71.1% in middle school students, and 80.5% in high school students. Moreover, moderate to severe myopia accounted for half of these cases3. The development of myopia into high myopia can cause irreversible blinding eye diseases, such as glaucoma, retinal detachment, myopic macular degeneration and choroidal disorders. The risk of complications increases with the severity of myopia4,5. Myopia has become a major cause of visual impairment in Chinese children and adolescents, posing a significant threat to their physical and mental health, as well as to the socio-economic development. Therefore, the prevention and control of myopia are urgent.

While myopia is a polygenic disease, resulting from hereditary, environmental and lifestyle factors. A genome-wide association study identified only 336 genetic loci associated with ametropia, explaining a mere 18.4% of the heritability, thus implying a substantial contribution from environmental factors and lifestyle to the swift increase in myopia prevalence recently6. Conducting scientific and efficient epidemiological surveys of myopia is crucial for understanding the patterns of its occurrence and progression, exploring related influencing factors, and studying the mechanisms involved. It is also vital for early detection and the development of control strategies. Numerous studies7,8 have shown that the factors influencing the onset and progression of myopia are diverse, and many have been confirmed as risk factors. However, with the increasing prevalence and younger age trend of myopia, it is imperative to explore more risk factors. Previous studies9 has indicated that the prevalence of myopia is related to geographical environment, population dynamics and economic development. Most current studies on the visual changes in Chinese children and adolescents only describe these changes from a temporal perspective10. Given China’s vast territory and the significant regional differences in children’s and adolescents’ vision, temporal changes alone are insufficient to fully understand these variations.

Spatiotemporal characteristics refer to the characteristics and manifestations of a series of dynamic changes that occur over time in the two dimensions of time and space11. Panel data is a data type that contains both time series data and cross-sectional data characteristics. From a cross-sectional perspective, it involves data on multiple different individuals, units, etc. at the same point in time. From a time series perspective, it tracks and records the data changes of these same individuals, units, etc. at multiple different points in time12. The spatiotemporal evolution of the panel data of children and adolescents in China is helpful to deeply understand the dynamic process of myopia development in children and adolescents at the same time in different regions or at different times in the same region, as well as the interrelationship with the surrounding environment and other factors, and comprehensively explore the risk factors of myopia development in children and adolescents13. Geographic Information Systems (GIS) are technology systems based on geospatial databases that can link spatial data with various sources of data related to myopia in children and adolescents14. Through spatial analysis and visualization at a more refined geographical level, GIS can illuminate the levels of myopia in different provinces and cities, further defining the distribution dynamics and influencing factors of myopia15. This, in turn, provides reference information for effective prevention and control of myopia in children and adolescents.

In the previous study, we have found that the clustered areas shifted from northwest to southeast direction among Chinese students aged 7 to 18 from 1995 to 2014 and persisted over time13. However, with the increasing attention from the national and provincial governments in recent years towards the prevention and control of myopia among children and adolescents, a series of policies have been introduced to effectively mitigate the occurrence and development of myopia among them.

In August 2018, eight departments including the Ministry of Education jointly issued the Implementation Plan for the Comprehensive Prevention and Control of Myopia in Children and Adolescents, clarifying the responsibilities and tasks of families, schools, students, medical and health institutions and government departments at all levels, elevating the prevention and control of myopia in children and adolescents to a national strategy, and comprehensively implementing the comprehensive prevention and control of myopia16. Therefore, it is necessary to examine the spatiotemporal trends in the occurrence and development of myopia among children and adolescents at the national level, in order to provide reference for taking targeted measures to continuously and effectively reduce the myopia rates among children and adolescents in various provinces and cities. Consequently, building upon our previous research which the data analysis from 1995 to 2014, this study focuses on analyzing the distribution characteristics of myopia prevalence among Chinese students aged 7 to 18 from 2014 to 2019, aiming to provide theoretical support for the prevention and treatment of myopia among Chinese children and adolescents.

Materials and methods

Sources of materials

The prevalence of myopia among children and adolescents from 1995 to 2019 was sourced from the Report on Chinese Students’ Physique and Health17,18,19,20,21,22. The Report was compiled by the Chinese Students’ Physique and Health Research Group and has been reviewed by the Ministry of Education, the National Health Commission of the People’s Republic of China, and other departments, and it was open to the public. Our study focused on students aged 7–18 years from various provinces and cities across mainland China. The detailed sample size for each province has been described in the Report. And specific sampling methods and quality controls have been described in detail in previously published articles13. Myopia is defined as uncorrected distance visual acuity >0.00 log Mar and Sphere Equivalent (SE) ≤-0.5D with non-cycloplegic refraction in this study.

Spatial autocorrelation analysis

Spatial autocorrelation23,24 refers to the potential interdependence of variable observations within a certain study area. It includes two indices: global spatial autocorrelation and local spatial autocorrelation. The latter is a breakdown of the global index and measures the degree of similarity between a particular area and its surrounding regions. Local spatial autocorrelation can indicate the extent to which each local unit follows the overall global trend, including its direction and magnitude, and it highlights spatial heterogeneity, explaining how spatial dependence varies with location.

In this study, global Moran’s I index and local Moran’s I index were used to explore the spatial autocorrelation of myopia rates within the study area25. The value of Moran’s I ranges between [− 1, 1], I > 0 indicates a positive spatial correlation, while I < 0 indicates a negative spatial correlation, if the I value is close to 0, indicates that the cases are randomly distributed in space, no spatial correlation exists26,27. The larger the absolute value of I, the stronger the correlation. When Z > 1.96, P < 0.05 was considered to be statistically significant and a spatial autocorrelation existed28,29. Based on the geographical adjacency of national provincial administrative divisions, a spatial weight matrix was generated using an inverse distance-based spatial conceptualization method. Utilizing the defined weight matrix, Moran’s I index was calculated using ArcGIS 10.2 software30.

Hotspot analysis

The global G statistic was used to detect the presence of “positive hotspots” or “negative hotspots” of myopia rates across the entire study area. When G is greater than 0 and P is less than 0.05, it indicates the existence of high-value clusters in the study area. Conversely, when G is less than 0 and P is less than 0.05, it suggests the presence of low-value clusters in the study area31. The local G statistic was employed to determine whether each province (municipalities and autonomous regions) and its neighboring provinces are areas of high-value or low-value clusters. Z-score were used to measure the clustering patterns of myopia among children and adolescents in different provinces, and a Z-score greater than 1.96 indicates the presence of high-value clusters in the spatial unit and its surroundings, while a Z-score less than − 1.96 indicates the presence of low-value clusters in those areas32.

Spatio-temporal cluster analysis

To more accurately characterize the spatiotemporal clustering distribution of myopia rates, analyses were conducted beyond basic spatial autocorrelation. Specifically, SaTScan 9.5 and ArcGIS 10.2 (ESRI) software were utilized to perform cluster analysis and spatial distribution visualization on the myopia rate data from 1995 to 2019 across 31 provinces (municipalities and autonomous regions) in China. The detailed methodologies have been elaborated in other articles13.

Time evolution characteristic analysis

Stata 15.0 was used to analyze the time evolution patterns in the prevalence of myopia among adolescents aged 7–18 years across various provinces. This analysis33 involved conducting Beta convergence tests. If β is less than 0, it indicates that there is convergence in the changes of myopia rates among children and adolescents. This means that the occurrence and development of myopia are negatively correlated with its initial rates. In other words, provinces with initially lower (or higher) myopia rates experience faster increases compared to those with initially higher (or lower) rates. This suggests a trend where the disparity in myopia levels across provinces decreases over time, leading to a convergence trend. In the long run, the myopia rates among children and adolescents in all provinces are expected to become more uniform. Conversely, if β is greater than 0, it implies that there is no convergence trend in the changes of myopia rates. Instead, the trend is divergent, indicating that the rate of increase or decrease in myopia is positively correlated with its initial level, leading to a gradual widening of differences in myopia rates across provinces.

Results

Temporal characteristics

From 1995 to 2019, the average myopia rates among children and adolescents aged 7–18 years in China showed a increasing trend. The mean myopia rates for the years 1995, 2000, 2005, 2010, 2014, and 2019 were 35.9%, 41.5%, 48.7%, 57.3%, 57.1%, and 60.04%, respectively. In terms of different provinces, most followed this overall increasing trend. However, some provinces and cities, including Beijing, Shanghai, Tianjin and Jiangsu, saw their peak myopia rates among children and adolescents in 2014, followed by a decline in 2019 (Table 1).

Spatial characteristics

Between 1995 and 2019 in China, among the 30 provinces and cities, the top five regions with the highest myopia rates in children and adolescents were Jiangsu (62.86%), Shandong (62.53%), Zhejiang (61.38%), Shanghai (59.61%), and Shanxi (57.19%). On the other hand, the five regions with the lowest rates were Hainan (38.34%), Guizhou (40.00%), Xinjiang (42.91%), Guangxi (43.52%), and Jiangxi (46.44%). Overall, the eastern coastal areas of China had higher rates of myopia in children and adolescents, while the rates were relatively lower in the southwestern regions. The majority of other provinces and cities had myopia rates that were around the national average. (Fig. 1).

Table 1 The prevalence of myopia among students aged 7–18 years in China from 1995 to 2019.
Full size table
Fig. 1
figure 1

Spatial distribution of the prevalence of myopia among students aged 7–18 years in China from 2014 to 2019.

Full size image

Global spatial autocorrelation analysis

The global spatial autocorrelation analysis revealed that the Moran’s I values for the prevalence of myopia among Chinese children and adolescents in 2019 was 0.235665 (P < 0.05). The average Moran’s I value from 1995 to 2019 was 0.270920 (P < 0.05), indicating a statistically significant spatial autocorrelation in the prevalence of myopia among children and adolescents over this period at the regional level. Notably, the year 2000 exhibited the highest level of clustering (Moran’s I = 0.255), while 2005 showed the lowest (Moran’s I = 0.141). (Table 2).

Table 2 Global Spatial Autocorrelation Analysis among students aged 7–18 years in China from 1995 to 2019.
Full size table

Local spatial autocorrelation and hotspot analysis

The results of the local Moran’s I autocorrelation analysis and the local hotspot analysis were essentially consistent, indicating that Beijing, Shandong, Jiangsu, Anhui, Hebei, and Shanxi were high-high clustering areas, Yunnan was a high-low clustering area, Tianjin, Liaoning, Henan, and Hubei were low-high clustering areas; and Hainan, Guangdong, Guangxi, and Guizhou were low-low clustering areas. (Fig. 2).

Fig. 2
figure 2

the Local Spatial Autocorrelation and Hotspot Analysis among students aged 7–18 years in China from 1995 to 2019 (A: Local Spatial Autocorrelation Analysis; B: Hotspot Analysis).

Full size image

Spatio-temporal scanning analysis and temporal evolution characteristics

The spatiotemporal scanning analysis revealed that between 1995 and 2014, the concentrated areas of myopia among children and adolescents aged 7–18 years gradually shifted from west to east and from north to south across the country. Notably, from 2000 to 2010, Fujian Province consistently emerged as the primary cluster zone for myopia in this age group. By 2010, Zhejiang, Tianjin, and Hebei began to show increasing rates of concentration, culminating in Zhejiang Province becoming the leading cluster by 2019. Overall, from 1995 to 2019, the myopia clustering in Chinese children and adolescents aged 7–18 years initially moved from the northwestern regions towards the southeastern areas, then gradually shifted to focus around Tianjin, extending into Hebei, Shanxi, and other areas of North China, before eventually spreading throughout East China. (Fig. 3; Table 3).

The results of the Beta convergence test indicated that there was convergence in the changes of myopia rates among children and adolescents across various provinces (β < 0). Furthermore, from 1995 to 2019, the absolute value of the convergence coefficient β had shown an increasing trend, suggesting that the speed of convergence was accelerating. This means that provinces with initially lower rates of myopia are experiencing a faster increase in these rates compared to those with higher initial rates. Consequently, the disparity in myopia levels across provinces is gradually diminishing, eventually leading to a uniform level across all provinces. (Table 4).

Fig. 3
figure 3

The Spatio-temporal scanning analysis among students aged 7–18 years in China from 1995 to 2019.

Full size image
Table 3 The spatio-temporal scanning analysis among students aged 7–18 years in China from 1995 to 2019.
Full size table
Table 4 The Beta convergence test among students aged 7–18 years in China from 1995 to 2019.
Full size table

Discussion

At present, myopia, as a leading cause of poor vision, has emerged as a significant new health challenge worldwide34,35. According to the results of the National Survey on Myopia of Children and Adolescents conducted by the Ministry of Education and the National Health Commission, the prevalence of myopia with children and adolescents in 2018 was 53.6%, and it was 50.2% in 2019, which down 3.4% points from 201836. The ages 7–18 years are crucial for eye development, making the prevention and control of myopia especially vital. Our study showed that from 1995 to 2019, the prevalence of myopia among Chinese students aged 7–18 years has been rising annually and exhibits certain spatial clustering. Stage-wise spatiotemporal scan analysis revealed that the clustering of myopia among students initially shifted from the northwest to the southeast from 1995 to 2014, but gradually moved towards the North China regions centered around Tianjin and Shanxi from 2014, eventually affecting the entire East China region, where high-incidence areas continue to exist. And the differences in the degree of myopia among various regions tend to diminish and converge over time.

From 1995 to 2019, the prevalence of myopia in children and adolescents in China has shown an upward trend, with the rate rising from 35.9 to 60.04%. This increase is not only due to the natural growth of myopia rates over time but also potentially linked to factors such as socio-economic development, advancements in medical technology, and increasing public awareness and detection of myopia37. It is worth noting that the current research conclusions on the relationship between socioeconomic development and myopia are inconsistent. Although most research results show that students with higher economic and social status will have higher education levels and greater learning pressure, which will lead to higher myopia rates38,39. However, there are also studies reporting contradictory results, that is, families with lower education level and lower income parents are more likely to suffer from myopia, which may be related to the poor hygiene conditions in their schools40.The results of this study indicated that in specific provinces like Beijing, Shanghai, Tianjin, and Jiangsu, the rate of myopia among children and adolescents gradually increased from 1995 to 2014, reaching a peak in 2014, and then began to decline. This trend might be attributed to these areas being economically more developed regions in China, where there is heightened sensitivity to the identification and control of myopia, positively impacting the efforts to curb the rise in myopia rates41. In summary, the current situation of myopia prevention and control in Chinese children and adolescents remains critical, necessitating concerted efforts from the entire society to safeguard the eye health of this young population.

In terms of spatial distribution, the prevalence of myopia among children and adolescents in the eastern coastal regions of China is relatively higher compared to less developed areas in the northwest and southwest. This finding aligns with previous research13 and is likely associated with factors such as higher levels of economic and social development and greater educational pressure in the eastern regions42. China’s vast expanse covers a wide array of provinces and cities with significant differences in economic and social development. This variation is reflected in the uneven levels of eye health among children and adolescents across different regions. Spatial analysis of the prevalence of myopia in this demographic can identify high-risk areas and specifically investigate the potential impact of social environmental factors on the development of myopia43. This study’s global spatial autocorrelation analysis indicated that from 1995 to 2019, there was a correlation in the prevalence of myopia among children and adolescents at the provincial and city levels in China, with the highest concentration occurring in the year 2000. This was followed by a trend of initial decline and then an increase, reaching another peak in 2019. The results of this study indicate a certain geographical correlation in the incidence of myopia among children and adolescents in China, with regions showing high similarities in myopia rates. Additionally, as socioeconomic development progresses, the degree of clustering also changes correspondingly. This characteristic provides convenience for regional governance or cross-regional collaborative management in the prevention and control of myopia among children and adolescents in China44,45.

Local spatial autocorrelation analysis revealed that Beijing, Shandong, Jiangsu, Anhui, Hebei, and Shanxi are high-high cluster, while Hainan, Guangdong, Guangxi, and Guizhou are low-low cluster. Studies have shown a significant correlation between the incidence and development of myopia and socio-economic growth38,39,40. Socio-economic development to some extent alters the environment and behavior of children and adolescents, such as increased educational pressure, more frequent use of electronic devices, longer screen time, and reduced outdoor activities, all of which have an impact on the vision of children and adolescents. Provinces like Beijing, Shandong, Jiangsu, Anhui, Hebei, and Shanxi are among the leading educational provinces in China. Research indicates that the rate of severe visual impairment in students from regions with higher education levels is more than twice that of students from areas with lower education levels46. Therefore, the prevalence of myopia among children and adolescents in these provinces is higher, showing a significant spatial clustering. In contrast, provinces such as Hainan, Guangdong, Guangxi, and Guizhou, besides having lower education levels, are located in the southern part of the country with lower latitudes, which means longer daylight hours. Previous studies47 have identified sunlight exposure as a significant protective factor against myopia, which may contribute to the lower incidence of myopia among children and adolescents in these provinces. Interestingly, research48 also suggests that peer effects significantly influence adolescents’ development of myopia. The peer effect refers to whether an individual adopts a certain behavior and the proportion of such behavior in its group, that is, the similar behavioral habits can lead to similar rates of myopia among students in the same region. By extension, this may also indirectly result in different spatial clustering patterns of myopia among children and adolescents in different provinces because of learn bad behavioral habits from each other. In light of this spatial clustering feature, the comprehensive myopia prevention and control efforts for children and adolescents in China, based on the provincial governance model where the central government signs myopia control responsibility agreements with each province. Additionally, cross-regional governance and multi-regional collaborative management approaches can be adopted for provinces with higher rates and rapid increases in the incidence of myopia among children and adolescents. It is essential to strengthen the incentive and constraint mechanisms of the central government towards local governments in these regions49. This would facilitate the implementation of cross-regional collaborative governance for the prevention and control of myopia among children and adolescents, effectively managing the widespread issue of myopia.

The results of spatiotemporal scan analysis showed that during this period, the clustering areas of myopia among children and adolescents in China initially shifted from the northwest to the southeast regions, then gradually moved towards the North China region centered around Tianjin, and eventually spread throughout East China, with high-risk areas of myopia persistently present. Some of these research findings have been explained in previous articles and will not be elaborated here. Interestingly, starting in 2014, the clustering areas of myopia among children and adolescents began to shift towards Hebei with Tianjin at its core. A potential reason for this trend is that, in 2014, the national government proposed the Beijing-Tianjin-Hebei integration development strategy, which emphasized the collaborative development of education, healthcare, elderly care, and other social services across these three regions, with the goal of reducing disparities in basic public services. Consequently, this initiative led to changes in the levels of education and healthcare in the Tianjin and Hebei areas, resulting in an explosive increase in the myopia rate among children and adolescents50,51. The occurrence and development of myopia in children and adolescents are significantly correlated with the economic development level, education level, and urbanization level of each province. It is essential to focus on the occurrence and development of myopia among children and adolescents in economically developed regions such as East China and Central China, and to take timely intervention measures to protect the eye health of the youth.

To assess the temporal and spatial convergence of myopia in children and adolescents, which facilitates understanding the trends in changes of myopia and provides a basis for relevant departments to formulate effective measures to reduce the prevalence of myopia among children and adolescents. The Beta convergence test results indicated that, from 1995 to 2019, provinces with initially lower rates of myopia experienced a faster increase in prevalence compared to those with higher rates. The disparity in myopia levels among provinces was gradually diminishing, but the overall prevalence was increasing year by year. With the development of the socio-economy, regional economic conditions, social progress, education levels, and healthcare resources are becoming more equalized52,53. This leads to a faster growth rate of myopia in areas initially with lower rates of myopia in children and adolescents compared to areas with higher rates, ultimately converging42. Furthermore, another possible reason for this phenomenon is that provinces and cities with initially faster increases in children and adolescent myopia earlier attracted the attention of local governments and families, prompting timely measures to curb the onset and development of myopia54,55.

Myopia in children and adolescents is a societal issue derived from the coupling of multiple factors. The spatiotemporal evolutionary characteristics of myopia among Chinese children and adolescents is pronounced. Spatial analysis will assist policymakers in designing preventive interventions for areas at higher risk of myopia, thereby alleviating the burden on the health system.

Limitation

This study had several limitations. First, our study only included the data of Han students aged 7–18 years in the report on Chinese students’ Physique and Health, and did not include the visual acuity information of ethnic minority students, which may deviate the myopia distribution characteristics of 7–18 years -year-old students in China. Second, our study used screening visual acuity data. However, some studies have pointed out that noncycloplegic refractive errors is inappropriate in epidemiological studies of myopia56. Therefore, the results of this study may be biased. Finally, there may be other factors affect the results, such as socioeconomic development indicators, which were not included in the research, resulting in certain deviations in the results.

Conclusion

This study outlines the spatial distribution of myopia prevalence among students aged 7–18 years in China from 1995 to 2019. The research shows that from 1995 to 2019, the prevalence of myopia among Chinese children and adolescents has been rising annually and exhibits certain spatial clustering. The phased spatiotemporal scanning analysis indicated that the gathering area of myopia in students aged 7–18 years in China first shifted from northwest to southeast, and the gradually shifted to northern China regions with Tianjin as the core, and finally spread to the whole of East China, where high-risk areas regarding myopia continue to exist. At the same time, the difference in the level of myopia in each region will gradually shrink over time and will converge. There is a great deal of spatial variations in the pattern of the prevalence of myopia among children and adolescents in China. Spatial analysis will assist policymakers in designing preventive interventions for areas at higher risk of myopia, thereby alleviating the burden on the health system.