Evaluation of Astigmatic Contact Lenses: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Astigmatic Contact Lenses (NCT03086447), a randomized na study. The primary outcome is Monocular Distance Visual Acuity (VA).

2. Clinical Question

The registry describes the study as a randomized, double-masked, parallel-group comparison with bilateral dispensing. The primary outcome is framed as a count of eyes meeting a specified visual-acuity threshold, rather than as a mean score or a direct estimate of the difference between lens groups. The registry also lists overall comfort, overall handling, and monocular distance visual acuity among the secondary outcomes.

This clinical question is narrow by design: it concerns recorded study-lens performance during the stated follow-up, for the listed participant group and endpoint definition. It does not by itself answer broader questions about longer-term use, other contact lens wearers, or outcomes beyond those recorded in the registry. Details about the eligibility criteria beyond habitual soft toric lens use and age, and about any additional exclusions, are not reported in the registry record.

3. Trial Design

The registry describes a completed randomized, double-masked, parallel-group device study. Participants were assigned to an investigational lens or a marketed control lens. Both arms involved bilateral dispensing and a planned total of 5 visits; the stated wear period was 28-36 days. The study dates run from 2017-03-01 to primary completion on 2017-05-04. The registry lists enrollment as 267 and phase as NA.

The experimental arm is identified as Senofilcon C Toric (Test), and the control arm as Lotrafilcon B (Control). The arm descriptions specify habitual soft toric contact lens wearers aged 18 to 40 years. The registry lists 129 participants in the test arm and 132 in the control arm for the primary outcome display. Those figures describe the arm counts listed there; they do not, on their own, describe every step in participant flow or explain how enrollment relates to each analysis population.

Double masking and randomization are features intended to limit bias in treatment assignment and outcome assessment. The record does not describe who was masked, how the allocation sequence was generated or concealed, or whether masking was maintained throughout the study. Further details about site structure, recruitment, protocol deviations, and participant disposition are not reported in the registry record.

The study’s stated aim was to evaluate whether the investigational lens met design requirements related to corneal staining, lens fit, visual acuity, rotational performance, comfort, and handling. The posted outcomes include visual acuity, comfort, and handling, but the primary endpoint is specifically the count of eyes with acceptable monocular distance visual acuity. The registry does not report a protocol rationale for the enrollment target or a detailed statistical analysis plan.

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.

Before assignment: One subject was randomized to the Test lens but in actuality was dispensed the control lens, therefore his/her data was summarized with the control data, while this/her demographic information is apart of the Test summary.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects who administered the test article and have at least one observation after lens insertion.

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: All subjects that were randomized to a study lens.

6. Primary Endpoint

The registry lists the primary outcome as Monocular Distance Visual Acuity (VA), measured over Up to 4-Week Follow-up.

The primary endpoint is monocular distance visual acuity measured with the study lens. The registry says that acuity was assessed with a Snellen distance visual-acuity chart at an optical distance of 20 feet throughout the study. Observations from planned visits were classified as either 20/40 or better, called acceptable, or worse than 20/40, called unacceptable. A result of 20/40 with a negative modifier was classified as worse than 20/40. The reported endpoint is the number of eyes with acceptable acuity for each study lens, with a timeframe of up to 4-week follow-up.

As a general statistical concept, converting a measurement into an acceptable/unacceptable category can make a threshold-based outcome straightforward to interpret: it reports how many observations meet the defined criterion. The trade-off is that a binary classification does not retain all the detail of the underlying acuity measurements. Results near the cutoff may fall into different categories despite being close in value, while results within the same category can still differ from one another.

The endpoint is expressed in eyes, while the study arms are described in participants and the lenses were dispensed bilaterally. Eye-level outcomes from the same participant may be related. How that relationship was addressed in the sponsor’s primary analysis is not reported in the registry record. The registry’s primary population definition requires at least one observation after lens insertion, but further details about the timing or completeness of observations are not reported.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary analysis report?

The primary endpoint is a count of eyes classified as having acceptable monocular distance visual acuity. The registry reports the number of acceptable eyes by arm and a sponsor-reported binomial proportion with an Agresti-Coull confidence interval for the test group. A binomial proportion method is generally used for an outcome recorded in categories such as meeting or not meeting a criterion. The Agresti-Coull method is one way to estimate uncertainty around a proportion, including in settings where a simple normal approximation may be less suitable.

The reported test-group interval is 96.5 to 100 at the 95% confidence level, with a reported proportion of 100. This is an interval for the test-group proportion, not an interval for the difference between the test and control groups. The registry does not report the denominator associated with this proportion or a primary two-arm inferential analysis. Consequently, a direct comparative conclusion cannot be recovered from the displayed count and test-group interval alone.

Why are eyes and participants not interchangeable?

The endpoint is measured in eyes, and the registry describes bilateral lens dispensing. As a general statistical matter, two eyes belonging to one participant may share characteristics and outcomes. If observations are correlated within participants, treating every eye as statistically independent can make uncertainty appear smaller than it is. Analyses can account for this structure using methods that represent the participant-level clustering, but the primary analysis description does not say whether or how such dependence was addressed. The record therefore does not support an assumption about the method used.

What does a confidence interval communicate?

In general, a confidence interval describes the precision of an estimate under the model and assumptions used to calculate it. A narrower interval indicates greater precision within that framework; a wider interval indicates less precision. The interval does not measure bias, prove a treatment effect, or establish that all clinically important differences have been ruled out. For a comparative question, the interval should correspond to the contrast of interest. The reported primary interval is for the test-group proportion, so it does not directly quantify uncertainty about a difference between lenses.

How should the continuous secondary outcomes be read?

Overall comfort and overall handling are reported as means with standard deviations. A mean describes the center of the recorded scores, while a standard deviation describes spread among observations. An unadjusted difference between means is a simple comparison based on the posted group summaries. It does not account for design features or covariates unless those are explicitly included in the calculation. The recalculated estimates are therefore descriptive, unadjusted comparisons rather than replacements for the sponsor’s analysis.

The registry reports a linear mixed model for each of these secondary outcomes, with a mean difference in final values and a confidence interval. A mixed model can represent repeated measurements or other grouping structures, depending on its specification. The registry does not give enough detail about the model structure, included covariates, covariance assumptions, or handling of missing observations to determine how those features were implemented. The sponsor-reported model estimates and the simple unadjusted estimates need not match because they are based on different analytical approaches. The record does not explain the source of the differences.

For comfort, the posted summary statistics are a test-arm mean of 61.4 with standard deviation 23.45 among 108 observations, and a control-arm mean of 52.0 with standard deviation 24.45 among 120 observations. The recalculated unadjusted mean difference is 9.4, with a 95% confidence interval from 3.179 to 15.62 and an unadjusted p-value of 0.003. For handling, the test-arm mean is 63.5 with standard deviation 20.9 among 108 observations, and the control-arm mean is 65.7 with standard deviation 19.77 among 120 observations. The unadjusted mean difference is -2.2, with a 95% confidence interval from -7.496 to 3.096 and an unadjusted p-value of 0.416. These recalculations do not adjust for multiplicity.

What are the sample-size illustrations?

Sample-size illustrations based on an observed effect answer a planning question: how many participants per arm might be needed under specified assumptions to detect an effect of a stated size. They are not a measure of the probability that the completed study found a real effect, and they do not describe the trial’s actual design justification. The posted illustrations use a two-sided significance level of 0.05, equal allocation, and a normal approximation. They use fractions of the observed secondary effects as planning targets and report participants per arm for 80% and 90% power.

Because these illustrations begin with effects observed in the study, they are sensitive to uncertainty in those estimates. A planning scenario based on a smaller fraction of an observed effect requires more participants per arm under the stated assumptions. The illustrations should not be interpreted as observed power, as a revised account of the enrollment target, or as evidence that the trial was designed around those assumptions. A sample-size justification for the actual trial is not reported in the registry record.

What is not specified?

The registry does not report the full statistical analysis plan, the handling of missing data, a multiplicity adjustment, or details sufficient to reproduce the sponsor’s mixed models. It also does not report a primary between-group estimate for the count endpoint. These omissions limit how far the posted results can be taken beyond their direct descriptive meaning. General statistical interpretation depends on the estimand, analysis population, model, and assumptions; the record leaves some of those elements unspecified.

10. Analyses Reported by the Sponsor

The sponsor’s primary analysis reports a binomial proportion with an Agresti-Coull method for the test group: a value of 100 and a 95% confidence interval from 96.5 to 100. That result is distinct from a direct comparison of the test and control groups. The record also gives the number of acceptable eyes in each arm, but it does not report a sponsor-estimated between-arm contrast or its confidence interval for this endpoint.

For comfort and handling, the sponsor reports linear mixed-model estimates for final values. The comfort estimate is 6.8, with a 95% confidence interval from 2.1 to 11.5. The handling estimate is -3.1, with a 95% confidence interval from -7.2 to 0.9. These model-based estimates are not identical to the simple unadjusted recalculations from group means. The recalculations use posted summary statistics and do not reproduce the mixed model.

The record does not provide enough model detail to identify why the estimates differ. Potential analytical distinctions cannot be assigned as the cause without that information. The sponsor’s model-based results and the unadjusted recalculations should therefore be read as results from different approaches, not as interchangeable estimates. The registry labels the sponsor’s secondary analyses as non-inferiority analyses, but the non-inferiority margin and further hypothesis details are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and 8 analyses, and names overall comfort, overall handling, and monocular distance visual acuity among the secondary outcomes. It does not report a multiplicity adjustment or explain how inferential emphasis was assigned across the outcomes. When several outcomes are tested, general statistical reasoning cautions that unadjusted results can overstate the strength of evidence if readers focus on whichever result appears most notable. Whether that concern was addressed in the trial’s analysis plan is not reported in the registry record.

The record provides a population definition for the primary endpoint: subjects who administered the test article and had at least one observation after lens insertion. This defines a minimum observation requirement but does not establish whether the primary analysis followed intention-to-treat, per-protocol, or another convention. The analysis population for the secondary outcomes is not fully described. The enrollment and arm counts should not be treated as proof that every enrolled participant contributed to every endpoint.

Missing-data procedures are not reported in the registry record. The record does not state how observations unavailable after lens insertion were handled, whether incomplete visits led to exclusion, or whether any imputation was used. It also does not give reasons for missing observations or describe sensitivity analyses. These details matter because an analysis based only on observed data can be affected if missingness is related to outcome or group assignment. This is a general limitation; the record does not establish that such bias occurred.

For the primary outcome, the eye-level measurement and bilateral dispensing raise a further analytical question: whether the association between eyes from the same participant was represented in the analysis. That method is not described. For the secondary outcomes, linear mixed models are reported, but model structure and handling of repeated observations or missingness are not specified. The record therefore supports reporting the stated estimates and methods, while leaving important details about multiplicity, analysis populations, and incomplete data unresolved.

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 5% reporting threshold.

The registry reports zero serious adverse events and zero deaths in each safety group. It lists 132 participants at risk for the control group and 130 at risk for the test group. It also reports zero other adverse events in both groups, using those same at-risk counts. These are the safety figures posted for the study.

Zero recorded events means that no events in those categories are reported in the registry for the stated groups and counts. It does not prove that the risk of an event is absent, nor does it establish that the groups have identical safety profiles. With no events recorded, the figures do not provide a comparative estimate of the difference in event risk. The registry does not report additional detail about adverse-event definitions, ascertainment procedures, follow-up completeness, or event severity beyond the serious and other categories listed.

The safety-group counts are not identical to every arm count shown for the primary outcome. The registry does not explain the difference, so the safety figures should be described using their own denominators rather than assumed to represent the same analysis population as the primary endpoint. These results are a limited account of events recorded during the study period, not a complete characterization of safety in other settings or over longer use.

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

The posted primary result describes how many eyes were classified as having acceptable visual acuity under the registry’s definition, together with a sponsor-reported proportion and confidence interval for the test group. It does not report a direct, uncertainty-quantified comparison of acceptable-acuity outcomes between the investigational lens and the marketed control. The information supports describing the recorded results, but not a conclusion that one lens is superior, equivalent, or non-inferior on the primary endpoint.

The secondary results provide additional observations about comfort and handling, with sponsor-reported mixed-model estimates and simple unadjusted comparisons based on the posted summaries. These are different analyses and outcomes. The registry’s non-inferiority labeling for the sponsor’s secondary analyses cannot be interpreted fully without the margin and the prespecified decision criteria, which are not reported. A confidence interval that crosses zero in an unadjusted comparison is not, by itself, a non-inferiority test.

In general, statistical significance and clinical importance are different questions. A p-value or confidence interval describes evidence and precision under an analysis; it does not determine whether a difference matters to participants. The registry does not report thresholds for a clinically meaningful difference in comfort or handling, or further interpretation criteria for the visual-acuity endpoint. The score anchors for the scale-based secondary measures are also not described in the posted record.

The results should not be extended automatically to people outside the described group of habitual soft toric lens wearers aged 18 to 40 years, to longer-term wear, or to outcomes that were not reported. The record describes a study with a wear period of 28-36 days and follow-up up to 4 weeks for the listed outcomes. It does not establish what would happen beyond that timeframe or in different populations. The findings are therefore best understood as endpoint-specific observations from this study, with comparative interpretation limited by the analyses and details reported.

14. Why This Trial Matters Statistically

This trial illustrates why the unit of measurement matters in clinical statistics. The primary outcome is counted in eyes, while randomization and arm descriptions refer to participants and the study dispensed lenses bilaterally. Eye-level measurements can be related within a participant. Statistical analyses need to distinguish the number of measured eyes from the number of independent participants when estimating uncertainty. The registry’s primary analysis description does not explain how that relationship was handled.

It also illustrates the difference between a descriptive count and a comparative effect estimate. The primary result gives counts by arm and a test-group proportion with an interval, but not a direct between-group estimate for the primary endpoint. A comparative question is most clearly answered by an estimate of the contrast and its uncertainty, using a defined analysis population and method. Without those elements, readers should avoid treating separate arm summaries as if they were a formal comparison.

The secondary outcomes highlight another common issue: different methods can yield different estimates from the same study. Simple calculations from group summaries are transparent but unadjusted; mixed models can incorporate structure specified by the analysis, though the relevant model details are not reported. Recognizing that distinction prevents an unadjusted recalculation from being mistaken for a reproduction of the sponsor’s analysis.

Finally, the registry’s secondary analyses are labeled non-inferiority, but the margin and supporting criteria are absent. Non-inferiority is not established merely by a nonsignificant difference; it depends on a prespecified margin and an appropriate analysis. This record therefore offers a useful example of why endpoint definitions, analysis methods, multiplicity, missing data, and estimands all shape what can be concluded from trial results.

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.

Several limitations specific to this record restrict interpretation. The primary endpoint is presented as a count of eyes, but the sponsor-reported primary analysis gives a test-group proportion only. The record does not report the denominator for that proportion, a corresponding control-group proportion analysis, or a direct between-group estimate with uncertainty. It is therefore not possible to evaluate a formal primary contrast from the posted primary analysis alone.

The endpoint is measured in eyes in a study with bilateral dispensing. The record does not describe whether the primary analysis accounted for within-participant association between eyes. This matters for how uncertainty is estimated. Similarly, the listed participant counts do not by themselves clarify how many observations contributed to each outcome or how participant flow affected each analysis.

Important design and analysis details are not reported in the registry record. These include the sample-size justification, primary hypothesis and decision criteria, non-inferiority margin for the secondary analyses, full model specifications, multiplicity procedures, missing-data handling, and detailed analysis populations. The record also does not explain the difference between the sponsor’s mixed-model estimates and simple unadjusted secondary calculations. These gaps limit reproducibility and make it difficult to assess the assumptions behind the estimates.

The follow-up period is short as described: lens wear was planned for 28-36 days, with outcome timeframes up to 4-week follow-up. The results do not address longer-term outcomes. The participant description is limited to habitual soft toric lens wearers aged 18 to 40 years, so generalization to other groups is uncertain. Scale interpretation is also constrained because score anchors and clinically meaningful thresholds for comfort and handling are not reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Evaluation of Astigmatic Contact Lenses — NCT03086447.
• Registry record last updated 2018-07-10; status last verified 2018-06.
• Results first posted 2018-07-10.
• 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

NCT03086447 is a completed randomized, double-masked, parallel-group study of Senofilcon C Toric (Test) versus Lotrafilcon B (Control). The registry lists enrollment as 267, with habitual soft toric contact lens wearers aged 18 to 40 years wearing their assigned lenses for 28-36 days and attending a total of 5 visits.

The primary outcome is the number of eyes with monocular distance visual acuity of 20/40 or better through up to 4-week follow-up. The registry reports 258 acceptable eyes in the test arm and 263 in the control arm, and a sponsor-reported test-group proportion of 100 with a 95% confidence interval from 96.5 to 100. It does not report a direct comparative analysis for this primary count endpoint.

Secondary outcomes include overall comfort and handling, with sponsor-reported linear mixed-model analyses. Multiplicity handling, missing-data procedures, full model details, and the sample-size justification are not reported in the registry record. The posted results should therefore be interpreted as endpoint-specific findings with important comparative and analytical details unresolved.