The current research revealed that pseudomyopia was present in 8.29% of preschool children in the Tongzhou District of Beijing, China. The predominant refractive error among preschool children was mild hyperopia (72.67%), with a prevalence of myopia at 1.9%. Additionally, this study investigated the differences in refraction between noncycloplegic and cycloplegic conditions. The results indicated that noncycloplegic refraction was, on average, more myopic than cycloplegic refraction, with hyperopic participants experiencing the greatest deviation and myopic children showing the smallest deviation. Furthermore, the findings of this study indicate that assessing the refractive status of preschool children without the use of cycloplegia can lead to substantial inaccuracies, including underestimation of hyperopia prevalence and overestimation of myopia and emmetropia.

The median of cycloplegic SER in the present study was 1.25D (0.75-1.63D). These findings are in line with other published studies conducted on preschool-age participants. In the Guangzhou study, the mean cycloplegic SER was 1.42 ± 0.79D [14]. Two additional studies had similar results, with 1.20 ± 1.05D in the Shanghai study [22] and 1.38 ± 0.73 D in the Shenzhen study [23]. In a Singapore study, cycloplegic SER for participants aged 3–5 were1.14 ± 0.73 D, 1.38 ± 0.88 D, and 0.97 ± 0.90 D [24], respectively. Studies comparing noncycloplegic and cycloplegic refraction in children are listed in Table 3. We also compared the differences under cycloplegic and non-cycloplegic refraction conditions. After being divided into different groups according to their refractive states (myopia, emmetropia, and hyperopia), the DSE of hyperopic participants was greater than that of emmetropic and myopic participants, and comparable to previous results [2, 10, 12, 20]. In our study, the DSE of boys was lower than that of girls for all participants and consistent with prior findings [10, 11, 20]. Some studies, however, found no statistically significant association between the DSE and sex [2, 11]. Furthermore, an examination of the correlation between astigmatism and DSE revealed that WTR astigmatism exhibited an association with larger DSE, contrasting with the findings of the Lhasa Childhood Eye Study (LCES) [10]. Studies have also found that DSE decreases with age [3, 9, 11]. It is worth noting that our current study did not observe this pattern. Such inconsistencies could potentially be attributed to variations in environmental factors, age demographics, ethnic backgrounds, and lifestyle practices. Additional research endeavors are warranted to delve deeper into this relationship.

In the current study, low hyperopia was the most common refractive error(>70%), this finding was similar to that of previous studies [21, 23, 25] There was an overall prevalence of 1.9% of myopia in this study. In comparison with studies conducted in Shenzhen(1.3%) [23] and Guangzhou(1.0%) [14], a higher prevalence was found in the current study. As compared with that reported in studies conducted in Shanghai (3.7%) [22] and Singapore (> 6%) [24], the current study found a relatively lower prevalence. In summary, myopia was still very rare in this preschool population. It differed from several previous studies [13, 26, 27]. A majority of these published studies focused on school-aged children and adolescents. Due to cycloplegic methods, investigation time points, and home confinement regulations, the results of our study cannot be compared directly with those of previous studies. A Taiwanese study reported that the prevalence of myopia did not vary significantly under COVID-19 social restrictions in preschool-aged children [28], which is similar to the present study. There may be a couple of explanations for this. First, preschool-aged children have less dependence on computers and smartphones, and their screen time is also less than that of school-aged children [29]. Second, a few months of social restrictions might not be long enough to trigger myopia shift in preschoolers [28], as they often have a greater emmetropisation reserve [3]. The prevalence of refractive errors was similar between the sexes in our study, similar to the Singapore study [24] and Shenzhen study [23].The result was inconsistent with other studies; Shanghai [22] and Guangzhou [14] studies reported that boys had an ascending trend of myopia prevalence while girls had no this trend. Different studies have found a different gender effect on myopia prevalence in preschool children, possibly due to environmental factors such as near-work activities and education. Although the difference in the prevalence of myopia among different age groups was not statistically significant, 3-year-olds showed the highest prevalence in our study. Younger participants may be more sensitive to environmental change than older children [13]: more research is required.

The value of noncycloplegic assessments of the eye’s refractive state is limited. The noncycloplegic evaluation of refractive error tends to overestimate myopia and introduces a higher margin of error in the assessment of hyperopic and emmetropic refractive errors. Noncycloplegic measurements lead to incorrect refractive error classification. Thus, cycloplegia plays a crucial role in research on preschoolers’ refraction [30]. A survey in Anyang [3] found that many opticians still lack theoretical knowledge about the right age for cycloplegia, and there is very little recognition of the need for cycloplegia. Therefore, it is necessary to raise knowledge and awareness about the appropriate use of cycloplegia in children.

Pseudomyopia is a condition in which there is an increase in refractive power due to accommodation, it disappears with cycloplegia, and the affected eye is actually emmetropia or hyperopia. Its characteristics at different ages are not well understood. The hospital-based study tended to enroll more myopic children; thus, there is little information on the nature distribution of pseudomyopia [3]. In our study, the prevalence of pseudomyopia was 8.29%, which is consistent with the Beijing Myopia Progression Study [2]. Kang et al [3]investigated the prevalence of pseudomyopia in 6- and 13-year-old children and found that it was 18.9% and 24.1%, which was higher than that in our study. There might be an explanation for such inconsistency: the age of the subjects varied considerably among different studies, and the pseudomyopia rate may varied among different years and regions. Moreover, this study was conducted during the COVID-19 outbreak in China, while other studies reveal pre-COVID-19 data. Compared with non-pseudomyopic children, the pseudomyopic children had higher DSE. Kang et al. [3]. reported that the median DSE of 6-year-old pseudomyopia children was 1.13 D (0.63, 1.63). This result is significantly lower than that obtained in the present study, which is 2 D (1.63, 2.3). In light of the close relationship between pseudomyopia and accommodation [1], the possible reason is increased accommodation due to more intense near-work or screen time during the outbreak [31]. In addition, the prevalence of pseudomyopia was similar between the sexes, and the DSE was overall similar for males and females. These results might suggest that accommodation was not significantly different between sexes.

This study is subject to several limitations. Firstly, the auto-refractometer utilized is a commonly employed tool in epidemiological research due to its convenience and reliability. However, the accuracy of autorefraction may be influenced by children’s compliance during the examination and their visual behavior. To mitigate potential sources of interference, multiple measurements were taken and abnormal values were reevaluated. Secondly, the baseline data in this study is cross-sectional, precluding the ability to establish a causal relationship between pseudomyopia and the onset or progression of myopia. Therefore, longitudinal analysis should be conducted, with future research focusing on potential differences in myopia prevalence during and post-COVID. Additionally, it should be noted that the uneven distribution of participants across age groups, particularly with fewer cases in the 6-year-olds and 3-year-olds, is a common issue in longitudinal studies that may impact results and warrants further investigation. Furthermore, all study participants exhibited light to dark-brown iris colors, which is important to consider given the potential impact of iris color on the effectiveness of cycloplegic drugs. Caution should be exercised when extrapolating these findings to individuals with different iris colors. In conclusion, the present study provides definitive information about the prevalence of refractive error and pseudomyopia in preschool children in China during the epidemic period. Most of the preschool children in Beijing in this study have low hyperopia, the prevalence of myopia is relatively low, and pseudomyopia is more common. The lack of cycloplegia can lead to significant misclassifications of hyperopia, emmetropia, and myopia. Considering that cycloplegic agents can eliminate pseudomyopia, it is necessary to improve the knowledge and awareness about the proper use of cycloplegic agents. Cycloplegia refraction is essential in studies of refractive error in preschool children.