Effectiveness of Orthokeratology in Myopia Control: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effectiveness of Orthokeratology in Myopia Control (NCT03465748), a randomized na study. The primary outcome is Axial Length.

2. Clinical Question

The clinical rationale in the registry concerns myopia progression and axial elongation of the eye. It describes orthokeratology as a way to correct low to moderate amounts of myopia and frames the study as an investigation of whether it slows axial elongation and myopic progression. The stated interest in African American children reflects the investigators’ description of the prior evidence base; the posted outcome data do not separate results by race.

The registry identifies a randomized, parallel-group study and records the condition as myopia. The outcome is axial length in millimeters, measured at baseline and again at the 2-year mark. The posted results give the average axial length at the later assessment for each arm. They do not give baseline averages, within-person changes, or an adjusted comparison. As a result, the clinical question that can be answered from these results is narrower than the full question about slowing progression.

3. Trial Design

The registry describes a randomized, parallel-group study of orthokeratology and a control condition in children with myopia. The intervention is listed as a device called OrthoK. The study is marked completed, and enrollment is recorded as 45. The registry lists the phase as NA.

The arm descriptions refer to a planned 24-month comparison. The experimental arm is described as children wearing orthokeratology lenses; the control arm is described as children wearing single-vision spectacles. The brief study description, however, says the control group may use single-vision spectacles or soft contact lenses. This difference in descriptions is not resolved in the registry record, so the actual control treatment should not be assumed beyond what each description says.

There is also a design-description inconsistency concerning masking. The registry records masking as none, while the arm descriptions call the study single-masked. The record does not explain this difference or identify who, if anyone, was masked. Randomization is recorded, but details of sequence generation, allocation concealment, stratification, or implementation are not reported in the registry record.

The primary outcome is axial length at 2 years. The registry says it was measured at baseline and again at the 2-year mark, but the posted outcome values are later-assessment averages rather than changes from baseline. Endpoint counts are 15.0 for OrthoK and 7.0 for control, while enrollment is recorded as 45. The record notes that many subjects withdrew because of patient or parent scheduling. The relationship between enrollment, withdrawals, endpoint counts, and the populations used for other results is not fully described.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Axial length was measured in mm, many subjects withdrew from the study due to patient/parent scheduling

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Due to COVID, there were some limitations in recruiting for the study and finishing the enrollment.

6. Primary Endpoint

The registry lists the primary outcome as Axial Length, measured over 2 years.

Axial length is the measured distance along the eye from its front to its back. The registry reports it in millimeters and identifies it as the primary outcome, with an assessment at baseline and again at the 2-year mark. In the study rationale, progressive axial elongation is connected to myopic progression. Measuring axial length therefore provides an anatomical outcome relevant to the study’s question about progression.

The posted primary results give average axial length at the 2-year assessment: 25.32 mm for OrthoK and 26.08 mm for control. The registry also reports the full range for each arm. A mean describes the average in the participants included in that arm’s endpoint result; the range describes the span of reported values. Neither alone describes how each participant changed from baseline.

Although baseline measurement is described, baseline values and changes from baseline are not reported in the registry record. There is also no reported estimate of the between-arm difference, confidence interval, or statistical test for this endpoint. The result therefore supports a description of the endpoint values posted for the analyzed participants, not a quantified estimate of how much the groups changed differently over time.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted result is a mean axial length for each arm at the 2-year assessment. A mean is the arithmetic average of the measurements included in that arm’s endpoint summary. It gives a compact description of the group, but it does not show the distribution of measurements or the amount of variation around the average. The record reports a full range, which gives the low and high limits of the observed values, but does not report a standard deviation or another measure of spread around the mean.

As a general statistical explanation, a treatment comparison should reflect the randomized groups’ outcomes in a way that accounts for the study question and the timing of measurement. For an outcome measured at baseline and follow-up, analysts may compare follow-up values while adjusting for baseline, compare changes from baseline, or use a repeated-measures model. These approaches answer related but not identical questions and rely on different assumptions. The analysis method used for this trial is not reported in the registry record. The posted follow-up means should not be mistaken for a baseline-adjusted analysis or a comparison of change.

A confidence interval, as a general concept, describes a range of values compatible with an estimate under a specified statistical model and sampling assumptions. It communicates precision as well as direction. No confidence interval is reported for either mean or for a between-arm comparison. A p-value, also not reported, would address compatibility with a null hypothesis under a specified test; it would not measure the size or clinical importance of an effect. The registry provides neither a formal test nor enough summary information to reconstruct a reliable uncertainty interval for the group comparison.

Random allocation is intended, in general, to make treatment groups comparable in expectation at the start of a trial. It does not guarantee that the observed groups are identical at baseline, particularly when the endpoint counts are limited. Baseline axial-length results and baseline balance information are not reported in the registry record. Consequently, the later means cannot show whether the groups began at similar levels.

Incomplete outcome data matter because the posted endpoint counts differ from the recorded enrollment, and the registry notes withdrawals related to patient or parent scheduling. General methods for missing data include complete-case analyses, imputation, and likelihood-based approaches; each has assumptions about why measurements are missing and how missing outcomes relate to observed information. The registry record does not report which approach was used, whether missingness was assessed, or whether sensitivity analyses were performed. It is therefore not possible to determine from the record how the withdrawals were handled analytically.

The record lists one outcome measure and zero analyses. No statistical analysis plan, model specification, covariate adjustment, or interim analysis information is reported in the registry record. Multiplicity procedures are likewise not reported. With no secondary outcomes posted, the available evidence is limited to the stated primary endpoint result and adverse-event counts; absence of posted analyses does not establish what analyses may have been conducted outside the registry.

General interpretation of a range should remain distinct from interpretation of inferential uncertainty. A full range describes the observed extremes in a group. It is sensitive to the values at the edges and does not directly quantify uncertainty in the mean or comparison. A confidence interval for a mean or treatment contrast would be needed for a direct account of precision, but none is reported. The reported ranges also cannot substitute for individual participant trajectories or change-from-baseline summaries.

Planning calculations are a separate statistical use of observed results. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set as a stated fraction of the observed effect. Such calculations are planning illustrations rather than evidence that a particular effect was detected with a specified probability in this completed study. The actual planning outputs and the trial’s original sample-size justification are not reported in the registry record. No observed-power claim can be inferred from the posted endpoint summaries.

Finally, the record has an inconsistency in its description of masking: the registry entry says none, while the arm descriptions say single-masked. Masking can reduce some forms of measurement or assessment bias when implemented, but the record does not clarify who was masked or how the inconsistency should be interpreted. The design details that would support evaluation of those procedures are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies axial length as the primary outcome and records one outcome measure. It lists zero analyses and no secondary results as posted. No multiplicity adjustment or testing hierarchy is reported in the registry record. With no formal endpoint test reported, there is no stated family of hypothesis tests to interpret. The absence of a reported adjustment does not show whether one was or was not used in analyses beyond the registry entry.

Missing outcome data are a notable limitation. The recorded enrollment is 45, while the axial-length endpoint counts are 15.0 in OrthoK and 7.0 in control. The registry states that many subjects withdrew because of patient or parent scheduling. It does not provide a participant flow account, group-specific reasons for missing endpoint measurements, or the timing of withdrawals.

The record does not report whether the primary analysis included all randomized participants, only participants with follow-up measurements, or another analysis population. It also does not describe imputation, a missing-data model, sensitivity analyses, or assumptions about the outcomes of participants who withdrew. General statistical practice treats these details as important because conclusions may depend on how missing measurements are handled.

The safety counts use different at-risk denominators from the primary endpoint counts. The registry does not explain the relationship between those populations. These denominators should therefore be interpreted within their respective outcome summaries rather than combined or treated as interchangeable. Overall, the registry does not report enough detail to assess the extent or consequences of missing data for the primary comparison.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for participants at risk in each arm. In OrthoK, 27 were at risk and 0 were affected; in control, 14 were at risk and 0 were affected. Deaths are recorded as 0 in both groups. No top serious-event terms are listed.

For other adverse events, 27 were at risk in OrthoK and 14 in control. The record reports 1 affected in OrthoK and 0 in control. The registry’s categories separate serious adverse events from other adverse events; the single event in the latter category should not be described as serious based on these counts.

These are event summaries for the recorded safety populations. The registry does not give event descriptions for the other adverse event, its timing, duration, outcome, or an assessment of relatedness to the intervention. It also does not explain how the safety populations relate to the participants with primary endpoint measurements. Zero recorded serious events does not establish that no events occurred outside the reporting captured in the registry, and the counts alone do not compare risk in a way that accounts for follow-up time or missing information.

12. What the Result Does, and Does Not, Mean

The posted results describe average axial length at the 2-year assessment among participants included in the endpoint summaries. The OrthoK average is 25.32 mm, with a full range of 25.07 to 26.23 mm. The control average is 26.08 mm, with a full range of 25.59 to 26.74 mm. These values are descriptive observations, not a complete estimate of how treatment changed axial length relative to control.

The result does not report baseline values or change from baseline, so it cannot show how far either group’s axial length changed during follow-up. Nor does the registry report a formal between-group estimate, confidence interval, or statistical test. Without an estimate of the contrast and its precision, readers cannot assess how compatible the observed difference is with chance variation or quantify the uncertainty around it.

The record also notes substantial withdrawal in general terms and provides endpoint counts below recorded enrollment. The reasons are attributed to patient or parent scheduling, but the participant flow and handling of missing outcomes are not described in detail. The endpoint results may consequently reflect a subset of those enrolled, and the registry does not establish how this affected the group summaries.

These results do not establish whether orthokeratology slows myopia progression, whether any difference is clinically important, or whether the outcomes apply to particular racial groups. The investigators’ interest in African American children is stated, but race-specific results are not reported. The results also do not resolve the control-treatment description or the masking inconsistency. The appropriate reading is limited to the axial-length values and safety counts recorded, with substantial uncertainty about comparative inference and follow-up completeness.

13. Why This Trial Matters Statistically

This trial is statistically notable because its stated clinical question concerns progression over time, while the posted primary result is an average at the 2-year assessment. For a longitudinal question, the distinction between a follow-up value and a change or adjusted comparison is central. The registry reports baseline measurement in the outcome description, but does not post baseline values or a change-based analysis. That limits how directly the posted numbers address slowing of progression.

The difference between recorded enrollment and endpoint counts is also important for interpretation. The registry attributes many withdrawals to scheduling, but does not report a detailed flow or missing-data analysis. In a randomized study, incomplete follow-up can weaken the comparability created by randomization if the participants with observed outcomes differ from those without them. The record does not show whether that occurred, so it remains an uncertainty rather than a demonstrated bias.

The masking descriptions are inconsistent: the registry records none, while the arm descriptions say single-masked. This matters when assessing how outcome measurement and study conduct may have been influenced, yet the record does not clarify the procedures. The control description likewise differs between registry text, leaving some uncertainty about the comparator actually received.

Finally, the results are not accompanied by a formal comparison or estimate of precision. The means and ranges offer a useful description of the observed endpoint data, but they cannot replace an analysis that addresses baseline, missingness, and uncertainty. The trial illustrates why outcome summaries need to be read alongside design and analysis details before drawing conclusions about a randomized comparison.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary endpoint results are limited by the lack of baseline values, changes from baseline, and a reported between-arm estimate. Although the registry says axial length was measured at baseline and at the 2-year mark, only the later averages and full ranges are posted. No standard deviations, confidence intervals, or p-values are reported, so precision and formal comparative evidence cannot be assessed.

Follow-up completeness is another concern. Enrollment is recorded as 45, but endpoint counts are 15.0 for OrthoK and 7.0 for control. The registry notes that many subjects withdrew because of patient or parent scheduling, without a detailed flow or group-specific accounting. The analysis population and handling of missing measurements are not reported in the registry record.

Descriptions of study procedures do not fully agree. The registry records masking as none, while the arm descriptions refer to single-masking. The control arm is described as single-vision spectacles in one place and as single-vision spectacles or soft contact lenses in another. These inconsistencies constrain interpretation of the trial procedures.

The project description emphasizes African American children and comparison with other races, but the posted results do not include race-specific counts or outcomes. The record also does not report the statistical analysis plan, adjustment strategy, interim analyses, sample-size justification, or details of adverse-event ascertainment. These gaps limit the ability to evaluate the strength, applicability, and robustness of the reported findings.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effectiveness of Orthokeratology in Myopia Control — NCT03465748.
• Registry record last updated 2023-09-28; status last verified 2023-09.
• Results first posted 2023-09-28.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry describes a completed, randomized, parallel-group study of orthokeratology for myopia control in children. Enrollment is recorded as 45. The primary outcome is axial length in millimeters at the 2-year assessment. The registry reports a mean of 25.32 mm and a full range of 25.07 to 26.23 mm for OrthoK, and a mean of 26.08 mm and a full range of 25.59 to 26.74 mm for control.

Endpoint counts are 15.0 and 7.0, respectively. The record says many subjects withdrew because of patient or parent scheduling, but does not report a detailed missing-data analysis. No formal between-arm estimate, confidence interval, or p-value is reported. Serious adverse events and deaths are recorded as 0 in both groups; other adverse events are recorded as 1 in OrthoK and 0 in control. The record contains unresolved differences in its descriptions of masking and the control condition.