Evaluate the Safety and Efficacy of a New Artificial Tear for Use After LASIK Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluate the Safety and Efficacy of a New Artificial Tear for Use After LASIK Surgery (NCT00544713), a randomized na study. The primary outcome is Post LASIK Dry Eye Symptoms as Measured by Ocular Surface Disease Index (OSDI©) Score at Day 90.

2. Clinical Question

3. Trial Design

This was a randomized, double-masked, parallel-group trial listed as completed. The registry lists 228 participants enrolled. The primary outcome population is described as patients who started the study and were randomized, and the posted outcome summaries include 114 participants in each arm. These are reported analysis counts; the registry does not describe participant flow or explain whether any enrolled participants lacked a Day 90 outcome.

The intervention was a carboxymethylcellulose and glycerin based artificial tear, and the comparator was a carboxymethylcellulose based artificial tear. Both are listed as drug interventions. The study’s brief description identifies dry eye that may occur after LASIK surgery as the treatment context. More detailed eligibility criteria, study locations, dosing instructions, and treatment duration are not reported in the registry record.

The study start is listed as 2007-09, and primary completion is listed as 2008-06. The phase is recorded as NA. The sponsor is Allergan. The registry lists 13 outcome measures and 11 posted secondary outcomes. These counts describe the registered outcome information, but do not establish which outcomes were prespecified in a final analysis plan or which analyses were completed. Formal statistical methods, allocation concealment procedures, the specific roles covered by masking, and the original sample-size rationale are not reported in the registry record.

Randomization and masking are relevant design features, but their presence alone does not resolve every possible source of bias. A reader also needs information about conduct, follow-up, missing outcomes, and analysis choices. Those details are not reported in the registry record, so interpretation should remain focused on the posted summaries and their stated outcome definition.

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: Intent to Treat population defined as all patients who started the study and were randomized

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.

6. Primary Endpoint

The registry lists the primary outcome as Post LASIK Dry Eye Symptoms as Measured by Ocular Surface Disease Index (OSDI©) Score at Day 90, measured over Day 90.

The primary endpoint was Ocular Surface Disease Index (OSDI©) score at Day 90, described as a measure of post-LASIK dry eye symptoms. The registry says that the instrument covers 12 domains, with each domain rated on a 5-point scale. The domain scores are summed and normalized to a severity scale from 0 to 100. A score of 0 represents no dry eye symptoms, while 100 represents maximum severity.

OSDI is used to represent patient-reported symptom burden rather than a laboratory measurement or an eye examination finding alone. The specified direction matters for interpreting the group means: higher scores indicate worse reported symptoms, while lower scores indicate fewer symptoms on the scale. The registry reports the mean and standard deviation for each arm at Day 90. It does not report a baseline-adjusted estimate or describe how the endpoint was analyzed beyond identifying the summary measure.

The primary population is defined as patients who started the study and were randomized. This definition identifies the intended population for the primary outcome, but does not by itself explain how missing Day 90 scores, departures from assigned treatment, or other intercurrent events were handled. Such procedures are not reported in the registry record.

7. Results Posted in the Registry

• Carboxymethylcellulose Based Artificial Tear: mean 6.43 versus 8.32 on control; difference -1.89 (95% CI -4.41 to 0.63), standardized effect d = -0.19; z-test p = 0.141.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous: each participant’s OSDI score is expressed on a numeric severity scale, and the registry reports an average score for each arm. The posted arm summaries give the number of participants, mean, and standard deviation. A mean describes the group’s average score; a standard deviation describes how spread out participant scores are around that average. These quantities answer different questions and should not be interchanged.

What does the mean difference represent?

The posted comparison is a mean difference of -1.89, with the carboxymethylcellulose based artificial tear as the comparison arm and the carboxymethylcellulose and glycerin based artificial tear as the reference. In the reported summaries, the corresponding means are 6.43 and 8.32. A negative difference in this orientation is consistent with a lower average score in the carboxymethylcellulose-only group. Since higher OSDI scores indicate worse symptoms, the sign describes direction on the measured scale. It does not establish causation on its own, nor does it determine whether the difference is important to patients.

A mean difference is an absolute contrast in the measurement’s units. Here, those units are OSDI score points. It is not a relative percentage change, and no percentage transformation is needed to understand the reported estimate. The registry record does not report an adjusted model, such as one accounting for baseline scores or other covariates. The analysis details behind the sponsor’s result are not reported in the registry record.

What does the standard deviation tell the reader?

The standard deviation summarizes variability among individual scores within an arm. It is not a confidence interval and it is not a measure of how certain the estimated mean difference is. For example, the reported standard deviations are 11.12 and 8.036, while the reported arm means are 8.32 and 6.43. The spread indicates that individual scores vary around their respective group averages. The registry does not provide the score distributions, so skewness, floor effects, or unusual observations cannot be assessed from these summaries alone.

General statistical context: when individual observations vary more widely, the group mean can be estimated less precisely if other features of the study remain similar. The standard error of a mean difference reflects both variability and the number of observations contributing to the estimate. The recalculated comparison uses the posted summary statistics; it should not be mistaken for a reconstruction of an undocumented sponsor model.

How should the p-value be read?

The unadjusted recalculation gives p = 0.141. A p-value describes how unusual the observed estimate, or a more extreme one, would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, and it does not quantify the probability that a treatment works. The registry record does not report a prespecified significance threshold or the full analysis procedure. For that reason, the p-value is best read as a descriptive inferential result from the unadjusted recalculation rather than as a complete account of the trial’s planned analysis.

General statistical context: a p-value depends on the estimate, its uncertainty, the model, and the analysis choices. It does not capture the size of an effect or its clinical relevance. The mean difference and confidence interval should therefore be considered together with the p-value, outcome definition, study population, and limitations in reporting.

Why does the confidence interval matter?

The recalculated interval crosses zero, with its lower bound at -4.409 and its upper bound in positive values. Zero represents no mean difference on the scale. An interval crossing zero therefore includes no difference as one value compatible with the estimate and calculation assumptions, while also allowing for differences in either direction. A confidence interval is not a guarantee that the true value lies within the particular interval; under repeated use of the same procedure and assumptions, the method has a stated coverage interpretation. The confidence level used for this interval is not reported in the registry record.

General statistical context: confidence intervals help communicate precision because they show a range of estimates compatible with the observed data under a statistical procedure. A narrow interval indicates more precision than a wide one, but precision is not the same as clinical importance. The registry does not report a minimum clinically important OSDI difference, so the interval cannot be compared with a stated clinical threshold.

What does over-dispersion mean in general?

General statistical context: over-dispersion refers to data variability that exceeds what a chosen statistical model assumes. It matters when a model’s uncertainty calculation is too optimistic because it understates actual variation. The primary endpoint is a continuous score, and no over-dispersion assessment or model diagnostics are reported. It would therefore be inappropriate to assert that over-dispersion was present or that a particular correction was used.

The registry lists a randomized, double-masked, parallel design. These features describe how the study was organized, but they do not specify the statistical model, missing-data approach, multiplicity control, or the exact analysis population used in the sponsor’s analysis. Those elements are not reported in the registry record.

10. Confidence Intervals and Precision

The recalculated confidence interval for the mean difference extends from -4.409 to a positive upper bound and includes zero. This means the posted summary statistics do not pin down the direction of the difference with high precision: values favoring either arm, as well as no mean difference, remain compatible with the interval calculation. The exact upper bound is not repeated because it is reported with more than three decimal places.

The interval’s span is important because the point estimate alone can seem more definite than the available information warrants. The estimate of -1.89 is a single value, while the interval reflects uncertainty around that estimate under the recalculation’s assumptions. A confidence interval does not show every possible source of uncertainty. It cannot account for unreported missing-data decisions, outcome selection, analysis changes, or departures from model assumptions.

The arm-level standard deviations, 11.12 and 8.036, also help contextualize the uncertainty: participant scores vary within each group. The reported group sizes are 114 in each arm. These summaries support an unadjusted comparison but do not reveal the distributions of individual observations or the amount of missing endpoint data.

No clinically important difference threshold is reported in the registry record. Consequently, statistical compatibility with a range of mean differences cannot be translated into a judgment about clinical importance. Nor does the interval establish equivalence: an equivalence conclusion requires a prespecified margin and an analysis designed for that purpose, neither of which is reported. The most defensible reading is that the observed average scores differ numerically, but the recalculated interval leaves substantial uncertainty about the underlying difference.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total participants across both arms. They are illustrations based on the observed effect, not the original trial’s design rationale. The calculation assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

For planning for 100% of the observed effect, the illustration gives 414 participants per arm for 80% power and 555 participants per arm for 90% power. For planning for 75% of the observed effect, the corresponding figures are 736 and 985 participants per arm. For planning for 50% of the observed effect, they are 1,656 and 2,217 participants per arm.

The 80% power column represents the estimated number of participants per arm under the stated assumptions for a study designed to detect the selected effect with 80% power. The 90% power column applies the higher power target, also under those assumptions. Power is a planning probability conditional on the chosen effect, model, and significance level; it is not a guarantee that a study will produce a statistically significant result.

Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants per arm in these illustrations. These figures are not observed power; they are planning illustrations derived from the observed effect and stated assumptions. They should not be interpreted as evidence about whether the completed trial was adequately powered, because its original design assumptions are not reported.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.890; pooled standard deviation σ = 9.704.
2. n = 2 × (1.96 + 0.842)² × 9.704² / -1.890² = 413.8, rounded up to 414 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -1.89                      # observed difference in means
sigma <- 9.70419808639539                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(544713)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 414 (80% power), 555 (90% power)
 75% of observed effect: n per arm = 736 (80% power), 985 (90% power)
 50% of observed effect: n per arm = 1,656 (80% power), 2,217 (90% power)
power.t.test: 414.8 
Simulated power at n = 414 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry identifies 13 outcome measures and 11 posted secondary outcomes, in addition to the primary OSDI endpoint. The secondary outcomes include patient acceptability, visual acuity, corneal topography, higher order aberration, Schirmer’s test, and corneal staining measures. Considering multiple outcomes can increase the chance of finding an apparently notable result by chance if each is tested separately. Whether that concern was addressed depends on the prespecified analysis plan and any adjustment or hierarchy used.

Multiplicity procedures, outcome hierarchies, and adjusted inferential results are not reported in the registry record. The primary comparison is an unadjusted recalculation from posted summary statistics. It should not be interpreted as accounting for all outcomes or analyses that may have been considered. The number of analyses is recorded as 0, but that registry entry does not establish that no analyses were conducted; it indicates that analyses are not posted in that record.

The endpoint population is described as Intent to Treat, including patients who started the study and were randomized. The registry reports 114 participants per arm for the primary outcome summaries, but does not report participant flow, the number with missing Day 90 scores, reasons for missingness, or how missing data were handled. It also does not state whether participants were analyzed according to random assignment if treatment use changed or stopped.

Missing-data handling can affect both the estimate and its precision, especially when the probability of missing an outcome is related to symptoms or treatment experience. General statistical context: complete-case analysis, imputation, and model-based approaches rely on different assumptions. No such method can be attributed to this trial from the registry record. The record also does not provide a protocol or statistical analysis plan with which to assess whether the posted outcome summaries and analyses match the original plan. These gaps limit assessment of selective outcome reporting and robustness.

14. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 114 participants at risk in each arm and records 0 serious adverse events in both groups. No serious adverse-event terms are listed. This is a report of the registry’s serious-event counts for the stated groups; it is not a guarantee that no event of any kind occurred or that no uncommon risk exists.

For other adverse events, the record lists 88 affected participants in the carboxymethylcellulose and glycerin based artificial tear arm and 82 in the carboxymethylcellulose based artificial tear arm, with 114 at risk in each group. These are group-level counts as recorded. The registry does not provide the event terms, severity breakdown, timing, or a statistical comparison for these other events, so the figures should not be used to infer which specific experiences occurred or whether the groups differed reliably.

Deaths are not reported in the registry record. The absence of a listed death count should not be interpreted as a confirmed zero. Overall, the posted safety information is limited to the summary counts and does not provide enough detail to characterize the types, duration, or clinical context of adverse events.

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

The posted results describe Day 90 OSDI averages in two randomized groups after LASIK surgery. The carboxymethylcellulose and glycerin based arm had a mean score of 8.32, and the carboxymethylcellulose based arm had a mean score of 6.43. The reported mean difference is -1.89, with the lower mean in the carboxymethylcellulose-only arm. Since higher OSDI scores represent worse symptoms, that direction is descriptive of the reported scale values.

The unadjusted recalculation gives p = 0.141, and the recalculated confidence interval crosses zero. These results do not establish a clear difference in average scores, but they also do not demonstrate that the treatments are equivalent or that there is no meaningful difference. Equivalence requires a defined margin and an appropriate design and analysis; no such information is reported in the registry record. The interval also does not establish whether any possible difference would be important to patients because a clinical-importance threshold is not reported.

The findings do not support medical advice or a conclusion that either artificial tear works or does not work. They are a statistical description of the posted outcome data, subject to uncertainty and incomplete reporting. The registry does not describe how missing outcomes were handled, how multiple outcomes were addressed, or the sponsor’s detailed primary analysis method. These omissions make it difficult to evaluate the robustness of the result beyond the posted summary statistics.

The safety summary likewise requires restraint. Serious adverse events are recorded as 0 in both groups, while other adverse events are summarized as 88 and 82 affected participants. The registry does not list serious-event terms or detail the other events, and deaths are not reported. These counts describe what is recorded, not the full safety profile or the comparative likelihood of particular events.

16. Why This Trial Matters Statistically

This trial illustrates why a clinical result should be read through both its estimate and its uncertainty. The two reported Day 90 means are close on a scale ranging from 0 to 100, but the mean difference and confidence interval are more informative than the direction of the means alone. The interval crosses zero, so the posted summary statistics do not establish a precise direction for the underlying mean difference.

It also highlights the distinction between statistical and clinical interpretation. A p-value does not measure the size or importance of a result, and the OSDI scale’s range does not supply a threshold for a meaningful change. Without a reported clinical-importance threshold, readers cannot use these summaries to decide whether a difference of the observed size would matter to patients. The estimate, interval, endpoint meaning, and study design all need to be considered together.

The registry’s limited analysis detail is itself statistically important. The outcome population is identified, but missing-data procedures, multiplicity methods, and the sponsor’s analysis plan are not reported. These details can influence how confidently readers interpret an estimate, especially when the record lists multiple outcome measures. The recalculation is useful as an unadjusted check of the posted summaries, but it cannot replace an undocumented trial analysis.

Finally, the planning illustration demonstrates how sample-size needs depend on the effect a future study aims to detect. Smaller target effects require more participants per arm under the stated assumptions. Those planning figures do not reveal the original design rationale or the power of the completed trial. Together, these points make the trial a useful example of separating observed summaries, inferential uncertainty, planning assumptions, and claims that the registry record cannot substantiate.

17. 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 are specific to the available record. The registry reports the primary outcome means, standard deviations, and group sizes, but does not report individual-level data or score distributions. It is therefore not possible to assess skewness, unusual observations, or how scores were distributed across the OSDI scale. The record also does not state whether the comparison was adjusted for baseline OSDI score or other variables.

Although the primary population is described as patients who started the study and were randomized, participant flow and missing Day 90 outcome counts are not reported. The registry does not explain how missing scores were handled, whether all randomized participants contributed to the posted means, or whether treatment discontinuations affected the analysis population.

The record lists multiple outcomes but does not report a multiplicity strategy, outcome hierarchy, or full statistical analysis plan. The sponsor’s formal analysis details are not reported, and the unadjusted recalculation from summaries cannot address choices that may have been made in the original analysis. No clinically important OSDI threshold is reported, limiting interpretation of the numerical magnitude.

The intervention and comparator are named, but dosing, treatment duration, adherence, eligibility criteria, and detailed masking procedures are not reported in the registry record. Safety reporting is also brief: serious-event counts are given without event terms, other adverse events are reported as aggregate affected-participant counts, and deaths are not reported. These gaps limit assessment of both the outcome analysis and the broader trial context.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Evaluate the Safety and Efficacy of a New Artificial Tear for Use After LASIK Surgery — NCT00544713.
• Registry record last updated 2011-12-14; status last verified 2011-11.
• Results first posted 2011-12-14.
• 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.

21. Record Summary

NCT00544713 is listed as a completed randomized, double-masked, parallel-group study sponsored by Allergan. It enrolled 228 participants and compared a carboxymethylcellulose and glycerin based artificial tear with a carboxymethylcellulose based artificial tear in the context of dry eye after LASIK surgery. The primary outcome was OSDI score at Day 90 in an Intent to Treat population defined as patients who started the study and were randomized.

The posted means were 8.32 and 6.43, respectively, with a mean difference of -1.89. An unadjusted recalculation gives p = 0.141, and its confidence interval crosses zero. The registry also reports 0 serious adverse events in each group and aggregate other adverse-event counts of 88 and 82. Detailed analysis methods, missing-data handling, multiplicity procedures, and deaths are not reported in the registry record.