Comparative Study of the NIDEK TONOREF III and the Haag-Streit Goldmann Manual Tonometer and Comparison of Pachymetry Values for NIDEK TONOREF III With the NIDEK CEM-530: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparative Study of the NIDEK TONOREF III and the Haag-Streit Goldmann Manual Tonometer and Comparison of Pachymetry Values for NIDEK TONOREF III With the NIDEK CEM-530 (NCT03441477), a na phase not stated study. The primary outcome is Agreement of Intraocular Pressure.

2. Clinical Question

The central statistical question is whether the registry’s reported measurements establish agreement between the test device and its comparators. The endpoint description identifies the measurements and target population, but the numerical outcome results are not reported in the registry record. Therefore, the record does not allow readers to assess the size or precision of the observed differences. Nor does it establish from the information available whether the stated comparison objectives were met.

3. Trial Design

The registry describes a completed, single-group device study sponsored by Nidek Co. LTD. Allocation is listed as NA, masking as none, and the model as single group. The registry lists enrollment as 225. The recorded start date is 2017-12-17, the primary completion date is 2019-01-16, and the status is completed.

The devices named are NIDEK TONOREF III, the Haag-Streit Goldmann Manual Tonometer, and NIDEK CEM-530. The registry’s arm description lists adults 18 years old or older with NIDEK TONOREF III. The study summary describes the main objectives as comparing the TONOREF III tonometry values with those from a predicate device and assessing whether its pachymeter function is equivalent to a predicate device. The secondary objective concerns safety relative to the predicate devices.

This structure should not be read as a randomized comparison between separate treatment arms. The registry records no allocation scheme and identifies a single-group model; its own cautions also flag that this is not a randomized, like-for-like control comparison. The record does not describe the sequence of device measurements, whether measurements were taken from the same subjects, or how any order effects were handled. Those details are not reported in the registry record.

The primary endpoint has a one-day timeframe and is restricted to data from subjects with CCT from 500–600 µm and astigmatism up to 3D. The registry lists three outcome measures and no analyses. It does not report a protocol-level sample-size justification, a detailed statistical analysis plan, or the number of enrolled subjects contributing data to each endpoint. These omissions limit how fully the registered design can be linked to an analysis population.

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: Data of subjects with 500-600µm CCT and ≤3D Astigmatism.

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: A total of 225 subjects were enrolled in the study and comprised the safety population; 224 subjects completed the study. One subject was discontinued due to a protocol violation. 223 subjects were included in the CCT Efficacy Subpopulation. One subject was excluded from the CCT Efficacy Subpopulation due to unevaluable results.

6. Primary Endpoint

The registry lists the primary outcome as Agreement of Intraocular Pressure, measured over 1 day.

The primary endpoint is agreement of intraocular-pressure measurements, in mmHg, between NIDEK TONOREF III and the predicate tonometer. The registry describes the outcome as a mean difference between the devices and a standard deviation, with a one-day timeframe. The specified endpoint population consists of subjects with CCT from 500–600 µm and astigmatism up to 3D.

Agreement is relevant when two instruments are intended to measure the same quantity: the question is how closely their readings correspond, not whether a treatment changes a health outcome. A mean difference describes the average direction and size of the differences between readings. The standard deviation describes how much those differences vary around their mean. In general, a small average difference alone may not show whether individual readings are consistently close; the spread of differences is also important.

The endpoint description says the mean difference and standard deviation are shown in the outcome measures table, but their numerical values are not reported in the registry record. Without those values, the direction, magnitude, and variability of the recorded differences cannot be assessed. The record also does not state a prespecified acceptance margin for agreement. Its pachymetry objective is separate from this primary tonometry endpoint, and numerical pachymetry results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: intraocular pressure is measured in mmHg, and the registry specifies a mean and standard deviation for the difference between device measurements. For a continuous outcome, a mean provides a central summary of the measured values. A standard deviation describes their spread around that mean. Here, the endpoint description identifies the target summary as the difference between devices rather than a mean pressure value from one device alone.

In general, an estimated mean difference gives the average direction and magnitude of a comparison. Its sign depends on the order in which device readings are subtracted, so the order must be known to interpret which device tended to read higher. The registry description does not state that order. The standard deviation of differences is not the same as the standard deviation of readings from either device separately: it describes variation in the differences. A complete account of agreement often also considers the range of differences expected for individual measurements, but the registry record does not report such an analysis or its results.

A confidence interval, in general, describes uncertainty around an estimated quantity under the assumptions of the analysis. It helps distinguish an estimate that is precise from one that is compatible with a broad range of values. No confidence interval for the primary mean difference is reported in the registry record. Without an interval or other uncertainty measure, and without the mean difference and standard deviation themselves, the precision of the primary result cannot be evaluated from this record.

Agreement is not interchangeable with correlation. In general, two instruments can rank subjects similarly while producing systematically different readings; strong association alone does not establish that the readings are close enough to agree. The registry identifies an agreement endpoint and a mean difference with a standard deviation, but does not report correlation statistics or a fuller agreement analysis. It would be inappropriate to infer that any unreported method was used.

The study summary also describes a pachymeter comparison intended to assess equivalence. In general, an equivalence conclusion depends on a prespecified acceptable range for the difference and an analysis showing sufficient precision relative to that range. A nonsignificant test of difference is not, by itself, evidence of equivalence. The registry record does not report an equivalence margin, the statistical method, or numerical pachymetry results. Thus, the objective can be described, but its statistical resolution cannot be judged.

There is a separate planning illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers detecting a stated fraction of the observed effect. Such calculations are planning illustrations, not the trial’s own design justification and not observed power. They describe participants per arm for 80% power or other stated planning targets, rather than the number analyzed in this single-group registry design. The registry record does not report the study’s own sample-size rationale, assumptions, or analysis plan. The two-arm assumptions in a planning illustration should therefore not be mistaken for evidence that this study randomized participants into two arms.

In general, over-dispersion means that observed variability exceeds what a chosen statistical model assumes. For continuous measurement differences, an inaccurate variability assumption can affect standard errors and the coverage of confidence intervals. The registry record does not report a model, distributional checks, or an over-dispersion assessment, so there is no basis for saying whether it was relevant to the sponsor’s analysis. The described normal approximation belongs to the planning illustration and does not establish that the same approximation was used to analyze study measurements.

Several design details can affect measurement comparisons. In general, measurement order, repeated readings, subject-level characteristics, and missing paired observations may affect estimates of agreement. The record gives the endpoint-specific CCT and astigmatism criteria, but does not describe measurement order, repeatability procedures, handling of incomplete measurements, or adjustment for other factors. Those methods are not reported in the registry record, and should not be inferred from the study’s objective or device list.

The primary endpoint timeframe is one day. A timeframe identifies when an outcome is assessed in relation to the study record, but it does not by itself establish whether measurements were repeated, how much time separated devices, or how an endpoint value was chosen. Those procedural details are not reported in the registry record. The available description is therefore enough to identify the intended summary, but not enough to reproduce or fully assess the analysis.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("NIDEK TONOREF III Comparison")
n   <- c(120)
m   <- c(0.3)
sp  <- c(3.45)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  NIDEK TONOREF III Comparison  120  0.300  3.450  -0.324  0.924

11. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures, including the primary agreement outcome and a secondary outcome concerning adverse events. It does not report a statistical plan describing how multiple outcomes were handled. In particular, adjustment for multiple comparisons, a hierarchy of endpoints, and any distinction between confirmatory and exploratory analyses are not reported in the registry record. This matters because evaluating several outcomes can increase the chance of findings that appear noteworthy by chance when each is tested separately. No claim about such testing can be made from the record.

Missing-data methods and analysis populations are also not described. The registry identifies the primary endpoint population as subjects with CCT from 500–600 µm and astigmatism up to 3D, but it does not report how many subjects met those criteria, how many had usable measurements, or how incomplete measurements were handled. The registry lists enrollment as 225, but that does not establish the analyzed population for each outcome.

The record does not specify whether analyses used all available measurements, only complete device comparisons, or another population definition. It also does not report sensitivity analyses for missing observations. These details are important for judging whether the analyzed measurements represent the stated endpoint population and whether exclusions could affect the result. They are not reported in the registry record.

The registry lists zero analyses. This is a statement about what the registry record contains, not proof that no analyses were conducted elsewhere. It does mean the record does not document methods for multiplicity, missing data, or population selection, limiting independent assessment of those issues.

12. 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’s safety group is titled “NIDEK TONOREF III Comparison.” It lists 225 at risk for serious adverse events and zero affected, and 225 at risk for other adverse events and 1 affected. Deaths are listed as zero. These are counts in the registry record; no percentages or comparative effect estimates are reported.

The record does not list any serious adverse-event terms. It also does not describe the other adverse event, its severity, timing, relationship to a device, or how it was handled. The counts should therefore be read as a limited summary, not as a detailed account of event types or causes. In particular, the single other adverse event cannot be attributed to a device from the information in the record.

The listed figures do not provide a comparison between randomized groups: the study is registered as single group, with allocation NA and no masking. The record does not report a statistical comparison of safety outcomes. A zero count for serious events in the listed group is a descriptive observation, not evidence that serious events are impossible or that one device is safer than another. The safety data should be interpreted only within the scope of the counts and limited detail reported.

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

The study addresses a practical measurement question: how NIDEK TONOREF III readings compare with predicate devices for tonometry and pachymetry. Its registered primary endpoint focuses on mean difference and variability in intraocular-pressure measurements in a specified population. Those are relevant summaries for assessing whether two measurement devices give similar readings, but the registry record does not provide the numerical primary outcome values needed to judge their magnitude or precision.

The study summary states objectives to demonstrate comparability for tonometry, equivalence for pachymetry, and safety relative to predicate devices. Objectives describe what a study seeks to assess; they are not results. Because numerical tonometry and pachymetry outcomes, acceptance criteria, and analysis details are not reported in the registry record, the record does not establish that comparability or equivalence was demonstrated. Nor does it establish that either objective was not met.

The design information also limits the scope of interpretation. The registry identifies a completed single-group, unmasked study with allocation listed as NA, rather than a randomized comparison of treatment arms. Device measurement comparisons may still address agreement, but the record does not describe the measurement procedures or provide the statistical results needed to assess them. The listed safety counts likewise offer no comparative estimate.

Accordingly, the registry record supports describing the intended comparisons, the endpoint definition, the stated study population, and the recorded adverse-event counts. It does not support a conclusion about the extent of measurement agreement, the precision of any difference, the success of an equivalence assessment, or comparative safety. These are limits of what the record reports, not findings that the devices do or do not perform similarly.

14. Why This Trial Matters Statistically

Device-comparison studies raise a statistical question that differs from a treatment-effect question. When two instruments measure the same quantity, average difference is informative, but variability in differences matters as well. Reporting both allows readers to consider whether a small average offset conceals substantial disagreement for individual readings. The registry identifies these two summaries for intraocular pressure, making the intended statistical focus clear.

The record also illustrates the importance of separating study objectives from evidence of having achieved them. Terms such as comparable and equivalent require numerical results and, for an equivalence claim, an acceptance range and an appropriate analysis. Without those details, an objective in a study summary cannot be treated as a statistical conclusion. This distinction is useful when interpreting device studies and registry entries generally.

Design labels also shape interpretation. This study is registered as a single-group, unmasked study with allocation NA, so it should not be described as a randomized trial with a like-for-like control arm. A separate planning illustration based on equal allocation between two arms is not a description of the registered design. Keeping planning assumptions distinct from the actual design prevents readers from confusing hypothetical sample-size calculations with the study’s enrollment or analysis.

Finally, the registry’s limited outcome reporting demonstrates why effect estimates, uncertainty measures, endpoint definitions, analysis populations, and missing-data methods all matter. Without these elements, readers can identify the research question but cannot independently assess the strength or precision of the evidence. The study’s statistical importance lies in the measurement-agreement question it sets out to address, while the registry record leaves that question unresolved numerically.

15. 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.

A central limitation is that numerical primary outcome results are not reported in the registry record. Although the endpoint description specifies a mean difference and standard deviation, neither value is available in the record described. Numerical results for the stated pachymetry comparison are also not reported. This prevents assessment of the magnitude, direction, spread, and precision of the device differences.

The registry does not report a prespecified acceptance margin for agreement or equivalence. Without such a criterion, even an available estimate would be difficult to interpret against a stated threshold for acceptable measurement performance. Confidence intervals and inferential results are also not reported, and no analysis method is documented.

The endpoint population is restricted to subjects with CCT from 500–600 µm and astigmatism up to 3D, but the record does not state how many contributed measurements or how the subgroup criteria were applied. Enrollment is listed as 225, but endpoint-specific analysis counts are not reported. Measurement order, repeatability procedures, missing-data handling, and the analysis population are not reported either.

The design is registered as single group, unmasked, with allocation NA. The registry cautions that this is not a randomized comparison with a like-for-like control. The record does not explain measurement sequence or safeguards against possible order or observer effects. Safety reporting is limited to counts; the other adverse event is not described, and no comparative safety analysis is reported. These limitations concern the detail available in the registry record and do not establish what analyses may have been conducted outside it.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparative Study of the NIDEK TONOREF III and the Haag-Streit Goldmann Manual Tonometer and Comparison of Pachymetry Values for NIDEK TONOREF III With the NIDEK CEM-530 — NCT03441477.
• Registry record last updated 2023-07-03; status last verified 2022-01.
• Results first posted 2022-02-11.
• 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.

19. Record Summary

NCT03441477 is a completed single-group device study sponsored by Nidek Co. LTD. The registry lists enrollment as 225 and describes comparisons involving NIDEK TONOREF III, the Haag-Streit Goldmann Manual Tonometer, and NIDEK CEM-530. The primary endpoint is agreement of intraocular-pressure measurements, summarized by mean difference and standard deviation, with a one-day timeframe and a specified CCT and astigmatism population.

The study summary also states objectives for pachymetry equivalence and safety. The registry lists three outcome measures and zero analyses. Numerical primary and pachymetry results, analysis methods, missing-data procedures, and endpoint-specific analysis populations are not reported in the registry record. The safety group lists zero serious adverse events, 1 other adverse event, and zero deaths. These entries are descriptive and do not establish a comparative safety effect.