Diffractive optical element (D.O.E) spectacle lenses for myopia prevention and control: a small-scale pilot study among younger Chinese children

Abstract

Aim

To assess the efficacy and safety of diffractive optical element (D.O.E) spectacle lenses in reducing the incidence and progression of myopia compared with currently available myopia control spectacle lenses in younger Chinese children.

Methods

This randomised, investigator-masked, parallel-controlled, small-scale pilot study included participants aged 6–8 years with a spherical equivalent refraction (SER) between 0.00 D and −1.75 D in either eye. Additionally, astigmatism and anisometropia did not exceed 1.50 D. Participants were randomly assigned to wear either the D.O.E (n=19) or highly aspherical lenslets ((H.A.L.T), n=17) spectacle lenses. Axial length (AL), choroidal thickness (ChT) and cycloplegic SER were measured at baseline and at 1, 2 and 3 months after wearing the spectacle lenses. Changes in AL, ChT and cycloplegic SER between the two groups were analysed using linear mixed models. Differences in compliance and subjective evaluations were also assessed.

Results

After 3 months, compared with the H.A.L.T group, the D.O.E group exhibited significantly smaller changes in cycloplegic SER (p<0.001), a significant reduction in AL (p=0.024) and a significantly greater increase in ChT (p=0.009). The estimated least-squares mean differences (95% CIs) compared with the H.A.L.T group were as follows: cycloplegic SER, 0.197 (0.068 to 0.326) D; AL, −0.027 (−0.072 to 0.018) mm; and ChT, 11.098 (0.067 to 22.130) μm. No significant differences were observed between the groups regarding compliance or subjective evaluation.

Conclusions

D.O.E spectacle lenses used in this pilot study demonstrated a statistically significant, though not clinically meaningful, effect in slowing myopia progression among younger Chinese children. The primary limitations were the small sample size and the short observation period, which restricted the ability to detect clinically significant changes. These findings are preliminary and require further validation before they can inform clinical decision-making.

Trial registration number

NCT06927414.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Novel spectacle lens technologies for myopia control demonstrate significant potential in children over extended periods.

WHAT THIS STUDY ADDS

• After 3 months, compared with currently available myopia control spectacle lenses, the diffractive optical element (D.O.E) spectacle lenses exhibited a statistically but not clinically significant effect in slowing myopia progression among younger children with high levels of compliance and favourable adaptability.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• D.O.E offers a new perspective on myopia prevention and control; however, these preliminary findings require further validation before impacting clinical decision-making.

Introduction

Myopia is a global public health concern, with its prevalence increasing and the age of onset becoming earlier in recent decades, particularly in East and Southeast Asia.^{1 2} High myopia, defined as a spherical equivalent refraction (SER) of less than −6.00 diopters (D), is associated with a heightened risk of vision-threatening ocular conditions.^{3 4} The risk of developing high myopia is notably higher in children whose myopia begins during early school age.⁵ Additionally, reducing myopic progression by 1.00 D during childhood could decrease the risk of myopic maculopathy by 40%.⁶ Therefore, slowing the progression of myopia and ultimately preventing its onset has become an essential component of public health strategies in many countries.

Current effective strategies for myopia control in children include optical devices, pharmacological interventions and behavioural modifications.⁷,⁹ Among the most commonly used optical devices with significant myopia control effects are orthokeratology, multifocal soft contact lenses, multiple-segment spectacle lenses and diffusion optics technology (D.O.T) spectacle lenses.¹⁰,¹³ However, the precise underlying mechanisms remain incompletely understood. Previous animal studies have demonstrated that during refractive development, hyperopic peripheral defocus induces posterior displacement of the retina, ultimately resulting in axial elongation. Conversely, myopic peripheral defocus has been shown to inhibit axial elongation.^{14 15} Certain peripheral designs, including defocus-incorporated multiple segment (D.I.M.S) and highly aspherical lenslets (H.A.L.T) spectacle lenses, have demonstrated reduced axial elongation and slower myopia progression compared with single-vision lenses.¹⁰¹⁶,¹⁹ Additionally, multicentre randomised controlled trials (RCTs) have reported that D.O.T spectacle lenses, which modulate retinal contrast by reducing signal differences between adjacent cones, effectively slow eye growth and myopia progression.^{11 20} These novel spectacle lens technologies show significant potential for myopia control in children over extended periods, despite the mechanisms of action not being fully elucidated.

This pilot study aimed to evaluate and compare the efficacy and safety of diffractive optical element (D.O.E) spectacle lenses with currently available myopia control spectacle lenses (H.A.L.T) in slowing myopia progression among school-aged children. These findings are intended to guide the selection of the most effective spectacle interventions for managing progressive myopia.

Methods

This 3-month, randomised, investigator-masked, parallel-controlled, small-scale pilot study, conducted between November 2024 and June 2025, evaluated D.O.E spectacle lenses for the prevention and control of myopia in children. The study was registered at ClinicalTrials.gov (ID: NCT06927414).

Participants

Healthy children aged 6–8 years with an SER between 0.00 D and −1.75 D in either eye, measured after cycloplegic refraction using 1% cyclopentolate eye drops, astigmatism ≤1.50 D, and anisometropia ≤1.50 D spherical equivalent were recruited. Additionally, participants were required to meet the following inclusion criteria: self-reported myopia progression of more than 0.5 D in the past year, based on school refractive screening or examination results from the previous academic year; best-corrected distance visual acuity (logMAR) of at least 0.1 in both eyes; near visual acuity of at least 0.0 in both eyes; and absence of ocular conditions that could affect refractive development (such as congenital or developmental ocular disorders). Participants were excluded if they had received any myopia management within 6 months prior to enrolment. Before obtaining consent and enrolment, participants were informed that they were required to wear the spectacle lenses for at least 12 hours each day throughout the study duration. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Randomisation

Eligible participants were consecutively enrolled and randomised in a 1:1 ratio into the D.O.E group (test) or the H.A.L.T group (control) using a computer-generated block randomisation sequence with a block size of four, created with Microsoft Excel 2019 by the study statistician. To ensure adequate allocation concealment, the sequence was stored centrally on a computer and could only be accessed when an eligible child was ready for randomisation. Following random assignment, participants were informed of their group allocation, while investigators and personnel involved in data collection and analysis remained blinded to the spectacle lens assignments.

Examination

From the baseline examination through the 1-, 2- and 3-month visits, participants first underwent axial length (AL) and choroidal thickness (ChT) measurements of both eyes without cycloplegia, followed by binocular automatic refractive SER measurements under cycloplegia. Cycloplegia was induced by administering two drops of 1% cyclopentolate, 5 min apart, with measurements taken 30 min after the second drop. At each visit, participants received ongoing education and compliance evaluations to enhance adherence. They retained the right to withdraw from the study at any time. Additionally, throughout the course of the study, participants had access to all relevant information related to the research.

D.O.E lens

The framework glasses operate based on the principle of light diffraction (figure 1). The lens centre contains a visual zone, which is surrounded by rings of optical element arrays. Although complete masking between the spectacle lenses was challenging due to the visibility of the optical elements under certain lighting conditions, the investigators conducting the measurements were instructed not to look directly at the participants’ spectacles to minimise potential bias.

Figure 1. (A) Diagram of the diffractive optical element (D.O.E) lens; (B) illustration of the imaging process of the D.O.E lens, where light transmitted through the lens enters the eye and creates diffraction patterns on the retina; (C) diffraction patterns produced when light passes through a slit; (D) photograph showing diffraction patterns generated by a single-wavelength laser pointer shining on the D.O.E lens.

Outcome measures

The primary outcome measures were changes in cycloplegic autorefraction SER and AL from baseline to the 3-month visit. AL without cycloplegia was measured in both eyes using the IOL Master 500 (Carl Zeiss Meditec AG). Cycloplegic autorefraction was measured in both eyes using a Topcon KR-800 autorefractor (Topcon Co., Japan). Five AL and autorefraction measurements were obtained for each eye, and the average value was used in the statistical analyses. The SER was calculated as the spherical component plus half of the cylindrical component.

The ChT without cycloplegia was measured in both eyes using a spectral domain optical coherence tomography device (Velite C3000, Weiren Meditec, China) by the same experienced investigator at each visit. Subfoveal ChT was measured three times using a 9 mm line scan composed of 128 single B-scans. Additionally, six 6 mm radial scans centred on the fovea were obtained three times to calculate the average ChT across different regions. Each radial OCT image averaged 16 B-scans and contained 1024 axial scans. A 6×6 mm central circular region was automatically segmented according to the Early Treatment Diabetic Retinopathy Study (ETDRS) into three concentric circles: foveal (0–1 mm diameter), parafoveal (1–3 mm diameter) and perifoveal (3–6 mm diameter). The parafoveal and perifoveal areas were further divided into superior, inferior, temporal and nasal quadrants. The software automatically provided the average ChT from three images within each of the nine ETDRS locations. ChT was defined as the distance between Bruch’s membrane and the choroidal-scleral interface. Two masked, trained observers manually corrected boundaries on all images. Images were reanalysed if interobserver differences exceeded 10 µm. The averaged results from three measurements by both observers were used for analysis. Intraobserver and interobserver repeatability analyses demonstrated excellent consistency, with all intraclass correlation coefficients exceeding 0.993 across all choroidal regions.

Subjective experience data and self-reported wearing times (hours per day) were collected at the 1-month, 2-month and 3-month visits using a questionnaire. This questionnaire included three subscales—visual clarity, comfort and compatibility while wearing spectacles—with descriptors designed to help participants accurately rate their subjective experiences over the past week (see online supplemental material).

Sample size

This pilot study is designed to provide data to inform sample size calculations for a subsequent full RCT, provided the intervention appears feasible, safe and shows trends of efficacy.²¹ Sample size calculation requires four factors: significance level (α=0.05), power (80%), group difference and SD of the primary outcome.²² According to recent research,²³ the average 3-month SER progression difference between the H.A.L.T group and the single vision spectacle lens (SVL) group was 0.0525 D (with the SVL group at −0.0625 D and the H.A.L.T group at −0.01 D). Therefore, the group difference (SER) was set at −0.03 D, with an SD of 0.09 D. Randomisation was performed at a 1:1 ratio. Using PASS V.15.0 software, the minimum sample size needed per group was calculated to be 16 participants. Considering a 10% dropout rate during follow-up, each group will include 18 participants, making a total of 36 participants enrolled in the study.

Statistical analysis

All statistical analyses were conducted using IBM SPSS Statistics for Windows, V.21.0 (IBM Corporation, Armonk, NY, USA). Statistical significance was defined as p<0.05. Descriptive statistics summarised the baseline characteristics of the D.O.E and H.A.L.T groups. Fisher’s exact test and independent t-tests were employed to compare baseline features and changes in cycloplegic autorefraction, AL and ChT after treatment between the two groups. All post hoc multiple comparisons were adjusted using the Bonferroni correction. Between-group demographic differences were assessed using linear mixed models and χ² tests.

The linear mixed models included subject-specific random intercepts, with the eye (right or left) treated as a random factor. Primary outcomes (changes in cycloplegic autorefraction, AL and ChT) were analysed separately using linear mixed models that accounted for correlations between eyes and repeated measures. The models were adjusted for potential confounders, including baseline age, age at myopia onset, parental myopia, sex, baseline AL, baseline autorefraction and baseline ChT. Each model included treatment group (D.O.E group and H.A.L.T group), visit (baseline, 1-month, 2-month and 3-month visits) and the treatment group-by-visit interaction as fixed effects.

Results

A total of 40 participants were screened and 38 eligible participants were randomly assigned to treatment groups (figure 2). Two participants dropped out: one withdrew without providing a reason and the other withdrew after learning their group assignment. Consequently, 36 participants were included in the study and 35 completed all visits, including 18 in the D.O.E group (one discontinued due to unsatisfactory vision at the 1-month visit) and 17 in the H.A.L.T group.

Figure 2. Flow chart showing the disposition of participants throughout the pilot study, including the reasons for discontinuation. D.O.E, diffractive optical element group; H.A.L.T, highly aspherical lenslets group.

Demographic and baseline characteristics

The demographic and baseline characteristics of the 36 randomised participants are presented in table 1. Although there were statistically significant differences in baseline cycloplegic SER and spherical power between the treatment groups (both p<0.001), baseline cycloplegic cylinder power, AL, ChT and other demographic characteristics did not differ significantly.

Table 1. Demographic and baseline characteristics of participants.

Variable	D.O.E group	H.A.L.T group	P value
Participants (n)	19	17
Age, years	7.21±0.71	7.47±0.63	0.255
Sex
Male, n (%)	10 (52.6)	5 (29.4)	0.192
Female, n (%)	9 (47.4)	12 (70.6)
Uncorrected visual acuity (logMAR)	0.28±0.13	0.33±0.12	0.075
Cycloplegic spherical equivalent refraction, D	−0.82±0.37	−1.16±0.43	0.001
Cycloplegic spherical power, D	−0.68±0.34	−1.03±0.48	0.001
Cycloplegic cylinder power, D	−0.28±0.31	−0.25±0.25	0.652
Axial length, mm	24.00±0.64	24.02±0.57	0.931
Central choroidal thickness, μm	278.21±44.29	278.35±33.38	0.988
Intraocular pressure, mm Hg	16.89±2.42	17.41±3.63	0.628
Age at onset of myopia, years	7.21±0.71	7.06±0.83	0.558
Parental myopia, %
None, n (%)	0	1 (5.8)	0.311
One, n (%)	6 (31.6)	8 (47.1)
Both, n (%)	13 (68.4)	8 (47.1)

Data are presented as mean±SD unless otherwise indicated.

D, diopter.AL, axial length; D.O.E, diffractive optical element; H.A.L.T, highly aspherical lenslets; SER, spherical equivalent refraction.

Efficacy

The changes in cycloplegic autorefraction, AL and ChT are shown in figure 3. The treatment effects, indicated by the least-squares mean differences at the 3-month visit for changes in autorefraction, AL and ChT between the treatment groups after adjusting for confounding factors, are shown in table 2.

Figure 3. Line graphs showing (A) observed mean change in cycloplegic spherical equivalent refraction, (B) observed mean change in cycloplegic spherical power, (C) observed mean change in cycloplegic cylinder power, (D) observed mean change in axial length, (E) observed mean change in choroidal thickness from baseline to the 3 months (3M) visit. The bar chart shows (F) the percentage of participants with an increase or no increase in cycloplegic spherical equivalent refraction, (G) the percentage of participants with an increase or no increase in axial length at the 3-month visit. Error bars represent SE. †Visits where significant differences (p<0.05) occurred between D.O.E group and the control H.A.L.T group. D.O.E, diffractive optical element group; H.A.L.T, highly aspherical lenslets group.

Table 2. Least square adjusted mean differences with adjusted 95% CIs with D.O.E group compared with the H.A.L.T group for the intention-to-treat participants for change in cycloplegic autorefraction, axial length and choroidal thickness from baseline to the 3-month visit.

Variable	D.O.E group vs H.A.L.T group
	Mean difference (95% CI)	P value
Change in cycloplegic autorefraction (D)
Spherical equivalent refraction, D	0.197 (0.068 to 0.326)	<0.001
Spherical power, D	0.158 (0.049 to 0.267)	<0.001
Cylinder power, D	0.034 (−0.036 to 0.104)	0.139
Change in AL (mm)	−0.027 (−0.072 to 0.018)	0.024
Change in choroidal thickness (μm)	11.098 (0.067 to 22.130)	0.009

D, diopter.AL, axial length; D.O.E, diffractive optical element; H.A.L.T, highly aspherical lenslets; SER, spherical equivalent refraction.

Participants randomised to the D.O.E group exhibited significantly less change in cycloplegic autorefraction—specifically, a less negative SER change than those in the H.A.L.T group (mean difference (95% CI) 0.197 (0.068 to 0.326) D, p<0.001). The D.O.E group showed significantly smaller changes in cycloplegic SER and spherical power compared with the H.A.L.T group at the 1-month (−0.016±0.220 D vs −0.204±0.343 D, p=0.009 and 0.027±0.259 D vs −0.147±0.326 D, p=0.016, respectively), 2-month (−0.029±0.262 D vs −0.261±0.327 D, p=0.002 and 0.014±0.264 D vs −0.168±0.168 D, p=0.012, respectively) and 3-month (−0.031±0.272 D vs −0.332±0.359 D, p<0.001 and 0.022±0.295 D vs −0.233±0.329 D, p=0.001, respectively) visits (figure 3A,B). Changes in cycloplegic cylinder power at the 3-month visit showed no significant difference between the two groups (figure 3C). The H.A.L.T group demonstrated a significant decrease in SER at the 3-month visit compared with baseline (−0.332±0.359 D, p<0.001), with no significant differences between the 3-month visit and the 1-month or 2-month visits. The D.O.E group did not exhibit any statistically significant variation in SER at the 3-month visit when compared with baseline, 1-month or 2-month visits.

Participants randomised to the D.O.E group exhibited a significantly greater change in AL than those in the H.A.L.T group (mean difference (95% CI) −0.027 (−0.072 to 0.018) mm, p=0.024). The D.O.E group showed a significant decrease in AL at the 2-month (−0.044±0.106 mm vs 0.010±0.010 mm, p=0.023) and 3-month (−0.042±0.113 mm vs 0.027±0.098 mm, p=0.008) visits compared with the H.A.L.T group (figure 3D). The H.A.L.T group showed no significant change in AL at the 3-month visit compared with the baseline, 1-month and 2-month visits. The D.O.E group demonstrated a significant decrease in AL at the 3-month visit compared with baseline (−0.042±0.113 mm, p=0.048), with no significant differences between the 3-month visit and the 1-month or 2-month visits.

Participants randomised to the D.O.E group exhibited a significantly greater increase in ChT than those in the H.A.L.T group (mean difference (95% CI) 11.098 (0.067 to 22.130) μm, p=0.009). The D.O.E group had a significant increase in ChT relative to the H.A.L.T group at both the 1-month (22.278±23.713 µm vs 12.912±13.219 µm, p=0.045) and 3-month (30.417±23.896 µm vs 18.500±16.864 µm, p=0.019) visits (figure 3E). Both groups demonstrated a significant increase in ChT at the 3-month visit compared with baseline (both p<0.001), with no significant differences between the 3-month visit and the 1-month or 2-month visits.

Neither the D.O.E group nor the H.A.L.T group showed any significant interaction with age, age at myopia onset, sex, parental myopia, autorefraction, AL or ChT at baseline in relation to changes in autorefraction, AL and ChT.

Distribution of participants experiencing myopia progression

Among all participants, the 3-month changes in SER significantly correlated with the changes in AL (r=−0.738, p<0.001) and ChT (r=−0.572, p<0.001).

Change in the SER

The distribution of changes in SER showed a statistically significant difference between the D.O.E group and the H.A.L.T group (χ²=13.688, p=0.002). A greater proportion of participants in the H.A.L.T group experienced a power increase of ≥−0.50 D compared with the D.O.E group (7/17, 41.2% vs 1/18, 5.6%) (figure 3F).

AL growth

Although there was no statistically significant difference in the distribution of AL changes between the D.O.E group and the H.A.L.T group (χ²=3.821, p=0.319), we observed that more than half of the participants in the D.O.E group (12/18, 66.7%) and nearly half in the H.A.L.T group (7/17, 41.2%) exhibited AL growth of ≤0 mm (figure 3G).

Adaptation, wearing time and compliance

The self-reported adaptation and wearing times for the D.O.E and H.A.L.T group across three follow-up visits are presented in online supplemental table 1. Although there was no statistically significant difference in self-reported adaptation time between the two groups, a greater proportion of participants in the D.O.E group required a longer adaptation period than those in the H.A.L.T group (adaptation time ≥1 week: 4/18, 22.3% vs 0/17, 0%, respectively). Participants demonstrated progressive improvement in compliance with wearing spectacle lenses throughout the follow-up period. Overall compliance was relatively high at the 3-month visit, with over 60% of participants reporting adherence to the prescribed wearing schedule (22/35, 62.9%, ≥12 hours). No statistically significant differences were observed between the two groups at any of the three visits. In addition, there was no significant correlation between adaptation time and overall compliance. Moreover, adaptation and wearing time had no statistically significant effect on changes in cycloplegic SER and AL.

Subjective data

There were no statistically significant differences between the D.O.E and H.A.L.T groups in self-reported visual clarity (eg, when looking at close or distant objects), night vision, colour vision or visual comfort (eg, walking, going up and down stairs) (online supplemental table 2). Additionally, no significant correlation was found between adaptation time and overall clarity or comfort. No serious adverse events or device-related adverse events were reported during the study period.

Discussion

This 3 months, small-scale pilot study aimed to evaluate the effectiveness and safety of functional spectacle lenses designed based on the principle of light diffraction for myopia prevention and control.

The research results indicated that, compared with H.A.L.T spectacle lenses, D.O.E spectacle lenses slowed the progression of myopia, as measured by changes of 0.197 D (95% CI 0.068 to 0.326 D) in cycloplegic autorefraction, −0.027 mm (95% CI −0.072 to 0.018 mm) in AL and 11.098 µm (95% CI 0.067 to 22.130 µm) in ChT over a 3-month period. Although the differences observed between the two groups reached statistical significance, the effect size was small and fell within the instrument’s noise threshold, potentially limiting its clinical relevance.²⁴ These findings are preliminary and require further validation before they can inform clinical decisions.

The study population comprised children aged 6–8 years, a critical period for refractive development and myopia prevention and control in Chinese children, to investigate the effects of D.O.E spectacle lenses. The rationale is as follows: First, prior epidemiological data indicate that home confinement during the COVID-19 pandemic was associated with a significant myopic shift in children aged 6–8 years (eg, a sharp increase from 5.7% at age 6 to 21.5%), whereas rates among 9- to 13-year-olds changed less significantly, suggesting heightened sensitivity to environmental factors in younger children.²⁵ Second, a longitudinal study revealed that over 50% of children who develop myopia between ages 7 and 8 progress to high myopia in adulthood, while delaying onset reduces this risk, underscoring the importance of early intervention.⁵ Additionally, at this developmental stage, children’s behavioural patterns are not yet firmly established, and compliance with interventions by both children and their caregivers tends to be high, enhancing the effectiveness of myopia control. Therefore, this study focused on Chinese children aged 6–8 years.

In our study, nearly all participants had at least one parent with myopia. Changes in cycloplegic SER, AL and ChT in both groups showed no significant correlation with age, age at myopia onset, sex or parental myopia status. Previous studies have demonstrated that a younger baseline age and an earlier onset of myopia are associated with faster myopia progression.⁵²⁵,²⁷ Additionally, younger age was significantly correlated with greater AL elongation in patients treated with D.I.M.S spectacles.²⁸ However, another study found no correlation between changes in SER and AL with age in patients treated with H.A.L.T spectacles.¹⁸ These discrepancies may be related to differences in the design of the spectacle lenses used in various clinical trials and the age of the participants. Previous research has identified sex-based differences in refractive responses to various lenses, suggesting a potential interaction between sex and specific optical designs.²⁹ However, our study did not observe any statistically significant differences. This finding warrants further investigation in future research, as it may impact personalised myopia management strategies. Previous studies have also found that parental myopia significantly influences changes in AL and autorefraction, highlighting a complex relationship between genetic susceptibility and various measures of myopia progression.³⁰ However, in our study, the effects of D.O.E and H.A.L.T spectacle lenses were not influenced by a family history of myopia.

These newly developed D.O.E lenses incorporate high-precision optical diffraction elements into the lens surface. Unlike single-vision lenses, D.O.E lenses modulate the physical properties of incoming light and induce a light-attraction effect. This effect produces a coloured spectral strip, a relatively dark region and another coloured spectral strip on the peripheral retina under white light conditions. Therefore, we speculate that D.O.E lenses may influence the periodic excitation-inhibition cycles and the on-off pathways of ganglion cells through alternating light and dark signals, thereby affecting eyeball development.^{31 32}

The subjective experience was positive when wearing both the D.O.E and H.A.L.T spectacle lenses, with no significant differences observed between the two groups. Previous studies have shown that visual discomfort can significantly reduce compliance in paediatric populations.³³ Although participants in the D.O.E group reported a longer adaptation period to spectacle lenses than those in the H.A.L.T group at the 1-month visit, there was no significant difference in wearing time between the two groups during the 3-month follow-up period. Furthermore, our study found no correlation between adaptation time and subjective experience. These results suggest that any initial adaptation challenges associated with certain optical configurations are resolved with continued wear, indicating adaptation mechanisms similar to those observed with aspherical lenslet designs.³⁴ While previous studies have identified a complex interaction between compliance and indicators of myopia progression, our study did not observe a similar association among autorefraction, AL changes, and compliance.

This study has several limitations. First, as a pilot study, it involved a small sample size and a short observation period, which may have limited the ability to detect clinically significant changes in AL or SER. Regardless of the evaluation technique used, a change in refractive error of at least ±0.50 D should be considered the minimum significant change in refractive status.²⁴ We observed that the difference in SER between the two groups was small; although statistically significant, it was within the instrument’s noise threshold, which may limit its clinical relevance. Additionally, statistically significant differences in baseline cycloplegic SER and spherical power were found between the two groups. Although the linear mixed model adjusted for baseline autorefraction to reduce confounding, these baseline differences may still introduce residual bias, potentially affecting the validity of causal inferences regarding myopia progression differences. Furthermore, we excluded individuals who had received myopia treatment within 6 months prior to enrolment. For participants who received treatment more than 6 months earlier, we did not record the type of intervention or the duration of the treatment. Therefore, it is essential to include a larger sample size and conduct long-term data collection to validate the effectiveness of myopia prevention and control during childhood development. Second, although the investigators were blinded to the spectacle lens assignments, we could not guarantee that they did not see the participants’ spectacle lenses during the study. To minimise bias, the personnel responsible for data collection and analysis were also blinded to the spectacle lens assignments. Third, participants’ self-reported myopia progression over the past year and daily spectacle-wearing hours in the past week may be subject to recall bias. Furthermore, this study included only Chinese children; therefore, the results may not be generalisable to other ethnic groups. Finally, this study did not quantify the potential impact of near-work and outdoor activities on myopia progression.

Conclusions

This small-scale pilot study demonstrated that D.O.E spectacle lenses had a statistically significant, though not clinically meaningful, effect in slowing myopia progression in younger Chinese children compared with H.A.L.T spectacle lenses. High levels of compliance, favourable adaptability and the absence of treatment-related adverse events were observed, indicating that both lens types have a satisfactory safety profile. These findings are preliminary and require further validation before they can inform clinical decision-making.

Data availability statement

Data are available upon reasonable request.