Myopia control in 2026: what the trial evidence supports

Childhood myopia is best understood as a growth problem: the eye elongates beyond its optical requirement, and each additional millimetre of axial length carries a measurable increase in the lifetime risk of retinal detachment, myopic maculopathy and glaucoma. Single-vision spectacles correct the refractive consequence of that growth; they do not alter its course. The clinically useful questions are which interventions do, by how much, in whom, and at what cost in side effects. Three have randomised-trial support worth examining closely. The figures below are quoted with their spread where the trials report it.

Orthokeratology: the ROMIO trial

The Retardation of Myopia in Orthokeratology (ROMIO) study randomised 102 Hong Kong Chinese children aged 6–10 years, with myopia of 0.50–4.00 D and astigmatism of 1.25 D or less, to overnight orthokeratology (a four-zone Menicon Z Night lens) or single-vision spectacles for two years (Cho & Cheung, Invest Ophthalmol Vis Sci 2012). Axial length was measured under cycloplegia every six months by a masked examiner; the children and their parents could not be masked. Seventy-eight children completed (37 ortho-k, 41 spectacles) and only completers were analysed.

Two-year axial elongation was 0.36 ± 0.24 mm with ortho-k against 0.63 ± 0.26 mm in controls (P < 0.01): a difference of 0.27 mm, or about 43% slower growth. Younger age at entry independently predicted faster elongation, and the treatment effect was largest in the youngest children — among 7–8-year-olds, 65% of spectacle wearers but 20% of ortho-k wearers were fast progressors (more than 0.36 mm a year). Six-monthly efficacy fluctuated (55%, 32%, 29%, 54%), which the authors attribute to natural slowing in the control group rather than loss of effect.

Discontinuation was substantial: 27% in the ortho-k arm and 20% in controls. Nine ortho-k children could not reach target correction with the single lens design, and five stopped for ocular-health reasons (three with rhinitis-related corneal staining, one with hyperaemia linked to poor compliance, one chalazion). There was no microbial keratitis and no serious adverse event, and ocular health was unaffected after cessation. Lenses and solutions were supplied under a university–Menicon agreement; the authors declared no personal conflicts.

Dual-focus soft contact lenses

Chamberlain and colleagues ran a three-year, double-masked randomised trial of the MiSight 1 day dual-focus lens against a single-vision daily disposable at four sites in four countries, enrolling children aged 8–12 with spherical equivalent (SE) refraction of −0.75 to −4.00 D (Optom Vis Sci 2019). Of 144 randomised, 109 (75.5%) completed — 53 test, 56 control. Cycloplegic SE progressed −0.51 ± 0.64 D with the dual-focus lens versus −1.24 ± 0.61 D with the control (difference 0.73 D, 59% less; P < .001), and axial length grew 0.30 ± 0.27 mm versus 0.62 ± 0.30 mm (difference 0.32 mm, 52% less). Refractive and axial change were tightly correlated (r = −0.90), which matters because axial length is the outcome that tracks later disease risk. No serious ocular adverse events occurred; four asymptomatic corneal infiltrative events were recorded, one in the test arm and three in controls.

A companion report by Lumb and colleagues covers wearer experience across six years (Cont Lens Anterior Eye 2023): after the masked phase, 52 continued the dual-focus lens (T6) and 56 switched to it from single vision (T3) without randomisation; 85 of 108 completed (79%; 77% T6, 80% T3). At the switch the ex-control children were more myopic (SE −3.45 versus −2.52 D; axial length 25.07 versus 24.76 mm). Ratings were high (94% or more barely noticed the lenses; over 90% saw well; weekday wear 13–14 hours), but annoying ghosting or haloes appeared in 8 of 1,034 dual-focus reports against 1 of 488 single-vision, and five children left for unacceptable vision, three at the switch itself. Entry required agreement to at least 10 hours’ daily wear, responses were self-reported and, the authors accept, parent and child answers may have mirrored each other. CooperVision sponsored the study; three of the five authors are its employees.

Low-concentration atropine: LAMP

The LAMP study (Chinese University of Hong Kong) randomised 438 children aged 4–12 with progressing myopia (at least 0.50 D in the prior year) to nightly 0.05% (n = 109), 0.025% (n = 108) or 0.01% atropine (n = 110) or placebo (n = 111) for one year (Yam et al., Ophthalmology 2019). Masking was double, though photophobia may have unmasked some. Fifty-five (12.6%) missed the one-year visit (6.4%, 15.7%, 11.8%, 16.2% by arm; P = 0.11); analysis was intention-to-treat.

SE change was −0.27 ± 0.61, −0.46 ± 0.45, −0.59 ± 0.61 and −0.81 ± 0.53 D (P < 0.001): 67%, 43% and 27% less than placebo (P < 0.001, < 0.001 and 0.006), with each concentration differing from the next (P = 0.01 and 0.05). Axial elongation was 0.20 ± 0.25, 0.29 ± 0.20, 0.36 ± 0.29 and 0.41 ± 0.22 mm (51%, 29% and 12% less). But while 0.05% and 0.025% separated from placebo (both P < 0.001), 0.01% did not (P = 0.18). Refraction is what the phoropter reports; axial length is the structural change carrying later disease risk, and the two correlated only moderately (r = 0.77). A dose that meets the refractive endpoint while leaving eye growth indistinguishable from placebo has not been shown to do what myopia control is for, as the authors concede.

Side effects scaled with dose: accommodative amplitude fell 1.98, 1.61, 0.26 and 0.32 D (0.01% versus placebo, P = 0.89); photopic pupils widened 1.03, 0.76, 0.49 and 0.13 mm. Photophobia at two weeks affected 31.2%, 18.5%, 5.5% and 12.6% (P < 0.001), and 7.8%, 6.6%, 2.1% and 4.3% of completers at one year (P = 0.27); allergic conjunctivitis occurred in 3, 7, 7 and 7 children (P = 0.57). Acuity and NEI VFQ-25 scores (composite P = 0.12) were unaffected. Funding was public and charitable (Hong Kong Research Grants Council, CUHK, UBS Optimus Foundation, CUHK Jockey Club programme); no commercial interest was declared. No confidence intervals are reported, so the precision of the 0.21 mm axial difference for 0.05% cannot be judged. The authors’ caveats: one placebo-controlled year, an all-Chinese cohort, no rebound data yet.

Interpretation for New Zealand practice

None of these interventions prevents progression entirely, and trial effect sizes are group averages; individual children respond differently, and counselling should say so. But the direction is consistent: intervening during the progression window slowed axial growth by roughly 0.2–0.3 mm over one to three years in these cohorts, and slower growth is what is expected to reduce later disease risk — an inference from the epidemiology, not something these trials measured. Time outdoors remains the best-supported environmental measure for delaying onset. Treatment selection turns on age, refraction, progression rate, lifestyle and family logistics. Combining pharmacological and optical strategies is reasonable in selected cases, though it is less well tested than either alone.

Limitations of this evidence

• Completed-case analysis. ROMIO analysed 78 of 102 children and the dual-focus trial 109 of 144. If children who progressed faster or tolerated lenses poorly left preferentially, the effect estimates are biased upward.
• Masking. ROMIO could not mask participants; only the axial-length examiner was masked. LAMP and the dual-focus trial were double-masked; the six-year extension was open-label.
• Populations. ROMIO and LAMP enrolled Hong Kong Chinese children; the dual-focus trial was multinational but small per site. Transfer to New Zealand’s mixed population is plausible but untested.
• Duration and endpoints. LAMP is one-year data without rebound data; ROMIO ran two years and the dual-focus trial three. None reports adult outcomes; the link between slowed elongation and reduced pathology remains inferred.
• Industry involvement. Menicon supplied ROMIO’s lenses; the dual-focus trial and its wearer-experience report were sponsor-conducted. LAMP was publicly and charitably funded.
• Precision. The reports give means and standard deviations, not confidence intervals, for the between-group differences (0.27 mm, 0.32 mm, and 0.21 mm for 0.05% atropine); the true effects could be appreciably smaller or larger.

Clinically, we offer these interventions through the myopia control clinic at Rose Optometry, Hamilton, including orthokeratology (Shift lenses). Parents wanting an unhurried, no-cost discussion of the options can book a free Myopia Chat.

Jagrut Lallu — BOptom (Hons), MSc Specialty Lenses (Hons), FIAOMC. Therapeutic Optometrist & Contact Lens Specialist. Founder, New Zealand Eye Research Centre; Partner, Rose Optometry; WCO Asia-Pacific Myopia Ambassador; International Myopia Institute NZ Ambassador; Clinical Senior Lecturer, Deakin School of Optometry; Honorary Teaching Fellow, University of Auckland.

References

Sourced and verified via PubMed.

• Cho P, Cheung SW. Retardation of Myopia in Orthokeratology (ROMIO) study: a 2-year randomized clinical trial. Invest Ophthalmol Vis Sci. 2012;53(11):7077-85. PMID 22969068. https://doi.org/10.1167/iovs.12-10565
• Chamberlain P, Peixoto-de-Matos SC, Logan NS, et al. A 3-year randomized clinical trial of MiSight lenses for myopia control. Optom Vis Sci. 2019;96(8):556-567. PMID 31343513. https://doi.org/10.1097/OPX.0000000000001410
• Lumb E, Sulley A, Logan NS, et al. Six years of wearer experience in children participating in a myopia control study of MiSight 1 day. Cont Lens Anterior Eye. 2023;46(4):101849. PMID 37156658. https://doi.org/10.1016/j.clae.2023.101849
• Yam JC, Jiang Y, Tang SM, et al. Low-Concentration Atropine for Myopia Progression (LAMP) Study: a randomized, double-blinded, placebo-controlled trial of 0.05%, 0.025%, and 0.01% atropine eye drops in myopia control. Ophthalmology. 2019;126(1):113-124. PMID 30514630. https://doi.org/10.1016/j.ophtha.2018.05.029