An Honest Read of the Ortho-K Myopia Control Evidence
By Dr. Mark Page6 min read
Myopia control marketing quotes percentages with a confidence the underlying literature does not support. That is unfortunate, because the actual evidence for Ortho-K is reasonably good and does not need embellishing. What it needs is accurate description: what kinds of studies exist, what they measured, how consistent the direction of effect is, how much the magnitude varies, and which questions remain genuinely open. This article gives that description without citing figures as though they were guarantees, so you can evaluate a claim the next time you encounter one.
On this page
What the Studies Actually Measured
What does the research say about Ortho-K and myopia progression?
Across randomised trials, prospective cohort studies, and pooled analyses conducted in several countries, children wearing Ortho-K showed less axial elongation than matched children in single-vision glasses or contact lenses. The direction of that finding is consistent across independent research groups, which is its main strength. The size of the reduction differs substantially between studies and between individuals, and most trials followed children for about two years rather than through their entire progression period.
The choice of outcome measure matters and is worth understanding. Nearly all credible myopia control research uses axial elongation as the primary endpoint rather than change in spectacle prescription, for two reasons. First, an eye reshaped by Ortho-K cannot be refracted meaningfully during treatment, so prescription change is not even available as a measure while lenses are being worn. Second, axial length is objective and directly tied to the long-term risks that motivate treatment.
The Kinds of Evidence and What Each Is Worth
| Design | Strength | Limitation |
|---|---|---|
| Randomised controlled trial | Allocation by chance removes selection bias. The strongest single design. | Cannot be masked, since children know whether they wear overnight lenses. Usually short and relatively small. |
| Prospective cohort with matched controls | Larger numbers and longer follow-up are practical. Reflects real clinical populations. | Families choosing Ortho-K may differ systematically from those who do not, which can bias results. |
| Retrospective chart review | Cheap, large, and reflects everyday practice. | Data quality varies and dropouts are often unaccounted for. Weakest of the three. |
| Meta-analysis | Pools results across studies, improving precision and revealing consistency. | Inherits the weaknesses of the studies it pools, and heterogeneity between them complicates interpretation. |
The practical value of having all four is that they fail in different ways. A finding that appears only in retrospective reviews would be weak. A finding that appears in randomised trials, is reproduced in cohorts, and survives pooling is considerably more robust, and that is roughly the situation for Ortho-K.
There is one structural problem that no Ortho-K trial has solved, and it is worth knowing about. These studies cannot be masked. A child either wears overnight lenses or does not, and both the child and the examiner know which. Masking usually protects against expectation influencing a subjective result, which is another reason the field settled on axial length as the endpoint. An instrument reading is far harder to nudge than a chart response, so the lack of masking damages these trials less than it would damage a study measuring symptoms or acuity.
How Big Is the Effect
Here is where care is required. Pooled estimates commonly describe Ortho-K as reducing axial elongation by something in the region of forty to fifty percent relative to single-vision correction over about two years. Several observations should accompany that framing.
- It is an average across groups, not a prediction for an individual. Some children in these studies showed very little elongation and some progressed nearly as fast as controls.
- It is relative, not absolute. A percentage reduction of a small underlying rate is a small absolute benefit. The same percentage applied to a fast progressor is a large one.
- It is time limited. Two-year figures should not be extrapolated linearly across a decade, and there is some evidence that relative benefit is larger early in treatment.
- Study populations differ. Much of the strongest work comes from East Asian populations where baseline progression rates are higher, which affects how the numbers transfer.
What Remains Genuinely Uncertain
Four open questions deserve to be stated rather than glossed over.
- Long-term outcomes. Most randomised data covers two years. What matters to a family is the prescription at age eighteen, and evidence covering that full arc is thinner and mostly observational.
- Rebound after discontinuation. Whether stopping treatment produces a period of accelerated growth that erodes the accumulated benefit is genuinely debated. See understanding rebound.
- Predicting individual response. No reliable pre-treatment test tells you which children will respond strongly. This is why measurement during treatment is not optional.
- Optimal design parameters. Whether treatment zone size, correction amount, and defocus ring position can be tuned to improve control is an active research area rather than settled practice.
How to Evaluate a Claim You Encounter
You will meet myopia control claims from practices, manufacturers, and other parents. Five questions separate a supported statement from a marketing one.
- What was measured? Axial length is the meaningful endpoint. A claim based on prescription change in Ortho-K patients should be treated sceptically.
- Compared with what? An effect size is meaningless without knowing the control group and its progression rate.
- Over what period? Two-year results are the norm. A claim implying lifetime effect from short-term data is overreaching.
- In whom? Age, ethnicity, and baseline progression rate all affect transferability.
- Was the outcome absolute or relative? These sound similar and mean very different things.
This practice publishes its own position on the evidence at clinical evidence and on realistic outcomes at realistic expectations. Both are written to the same standard as this article.
Frequently Asked Questions
Is Ortho-K the most effective myopia control option?
It is among the more effective optical options in published comparisons, but head-to-head trials against soft multifocals and modern defocus spectacles are limited and the differences are not large enough to declare a clear winner for every child.
The better question is which option a specific child will actually use consistently, since an unused treatment has no efficacy at all.
Why are there no long-term randomised trials?
Randomising a child to an untreated control group for a decade is difficult to justify ethically once effective treatments exist, and families in a control arm tend to leave.
The practical consequence is that long-term evidence is mostly observational, which is weaker but not worthless.
Does this research apply to my child?
Partly. Study populations differ from any individual patient in age distribution, ethnicity, baseline progression, and environment.
This is precisely why individual measurement matters. Your child's own axial length trend is better evidence about your child than any published average.
Should we wait for better evidence before starting?
Waiting is itself a decision with a cost, because progression continues during the wait and cannot be recovered afterward.
The current evidence is sufficient for most clinicians to recommend treatment for a progressing child. See why early intervention matters.
Keep reading
Last updated . Clinically reviewed by Dr. Mark Page.