Olopatadine Eye Drops and Allergy Skin Testing: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Olopatadine Eye Drops and Allergy Skin Testing (NCT00775658), a randomized na study. The primary outcome is The Difference in Area (cm2) of the Wheal Size in Histamine Skin Test Response During Treatment With Olopatadine Compared to Placebo..

2. Clinical Question

In practical statistical terms, the question is whether the recorded histamine wheal area differed between the two treatment conditions in this crossover study. A smaller wheal area would represent a smaller measured skin response, but the statistical result alone does not establish why a response changed or what it means for an individual. The registry brief summary states the study hypothesis that olopatadine eye drops would decrease allergy skin-test reactivity; that statement is the question being examined, not a conclusion about the result.

The crossover design means treatment conditions are intended to occur in sequence rather than being assigned as one permanent treatment per participant. Such a design can make within-participant comparisons relevant, because measurements from the same person may be related. The registry record does not explain how that relationship was handled in the reported statistical analysis. It also does not identify a more detailed primary analysis population or provide participant-level observations.

3. Trial Design

The registry identifies this as a randomized, double-blind crossover trial of olopatadine and placebo. It lists enrollment of 24 and a completed status. The recorded start is 2008-01, and primary completion is recorded as 2008-06. Phase is listed as NA. Vanderbilt University Medical Center is named as sponsor.

The treatment descriptions specify olopatadine 0.2% ophthalmic solution and placebo, described as normal saline ophthalmic solution. Each treatment description gives a daily schedule of 1 drop into each eye for 7 to 10 days, with a 7-10 day washout period between treatment periods. The registry lists the sequence labels olopatadine then placebo and placebo then olopatadine. However, both sequence descriptions contain the same wording about receiving olopatadine first and placebo afterward. This makes the written sequence details ambiguous; the actual order assigned to each sequence is not reported consistently in the registry record.

Randomization and double masking are recorded, but the method used to generate the randomization sequence, allocation concealment procedures, and who was masked are not reported in the registry record. The registry also does not report whether the washout duration varied between participants, how adherence was assessed, or whether deviations from assigned treatment occurred. These omissions limit how fully the operational conduct can be reconstructed from the record.

The primary outcome summaries list 21 observations for olopatadine and 21 for placebo. These are treatment-specific counts, not a reported count of distinct people completing both periods. The registry record does not explain how those counts relate to the enrollment of 24, whether any observations were excluded, or whether all participants contributed data to both treatment conditions. Those details matter in crossover studies because the unit of analysis and within-person pairing affect how treatment comparisons should be interpreted.

The record lists 2 outcome measures and 2 analyses, but it does not provide the original sample-size justification or a full statistical analysis plan. Specific procedures for handling period effects, sequence effects, carryover, missing measurements, or protocol deviations are not reported in the registry record.

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: Three participants withdrew prior to any study procedure: 1 had and upper respiratory infection, 1 required use of oral anti histamine and 1 was lost to follow up

First Intervention

Second Intervention

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 The Difference in Area (cm2) of the Wheal Size in Histamine Skin Test Response During Treatment With Olopatadine Compared to Placebo., measured over at baseline and at return visit (between study day 7-10).

The primary endpoint is the difference in area of the wheal produced by histamine skin testing during olopatadine treatment compared with placebo. The measurement is an area in cm2, a continuous outcome. The registry identifies the reported summary as a mean with standard deviation. The stated timeframe is at baseline and at a return visit between study day 7-10.

A wheal is the raised area assessed as part of a skin-prick response. Measuring its area turns the skin reaction into a quantitative outcome that can be summarized across participants and compared between treatment conditions. As a continuous measure, it can retain information about the size of individual responses rather than reducing responses to categories such as present or absent. The registry record does not give further detail about the measurement procedure, who measured the wheal, whether measurement was masked, or how baseline readings were incorporated into the primary comparison.

The endpoint is relevant to the trial’s stated question because the brief summary proposes that eye-drop treatment could change allergy skin-test reactivity. The endpoint is specifically a histamine response, not a general assessment of all allergy symptoms or all skin-test responses. The record also lists saline among the skin-testing substances in the official title, but the primary endpoint described in the results is the histamine wheal area. A numerical contrast in this measure therefore addresses the recorded histamine skin-test outcome; it does not by itself describe broader clinical effects.

7. Results Posted in the Registry

• Olopatadine: mean 0.46 versus 0.46 on control; difference -0.00 (95% CI -0.16 to 0.16), standardized effect d = -0.01; z-test p = 0.971.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a Wilcoxon (Mann-Whitney) analysis for the comparison of olopatadine and placebo, with p = 0.57. This is a rank-based method: rather than relying directly on the difference between arithmetic means, it compares the ordering of observations across groups. Rank-based methods can be useful when distributional assumptions for a mean-based test are not suitable, although the exact interpretation depends on the design, the data distribution, and the implementation. The registry record does not describe the test’s detailed implementation or provide its test statistic, confidence interval, or participant-level data.

What does the summary-statistic recalculation estimate?

The recalculation uses the posted treatment means and standard deviations to estimate an unadjusted difference in means. The olopatadine mean is 0.461 cm2, and the placebo mean is 0.464 cm2. The reported recalculated mean difference is -0.003 cm2. A negative sign indicates that the olopatadine mean is lower than the placebo mean under the stated subtraction direction. The recalculated p-value is 0.971, and the recalculated confidence interval includes zero.

This calculation is based on group summaries rather than a reconstruction of the original participant-level analysis. It does not adjust for covariates or model the crossover sequence, period, or within-person correlation. The registry record does not report enough information to determine whether those features were addressed in the sponsor’s analysis. Because the design is crossover, the difference between an independent-group calculation and a paired analysis could matter: repeated measurements from the same participant may be correlated, and that correlation affects the uncertainty of a treatment contrast. No paired-difference summary is reported.

What does the p-value describe?

A p-value is a measure of compatibility between the observed data and a null hypothesis, conditional on the statistical model and assumptions used. It is not a measure of the size or clinical importance of an effect, and it is not the probability that the null hypothesis is true. The sponsor-reported p-value of 0.57 belongs to the Wilcoxon (Mann-Whitney) analysis. The recalculated p-value of 0.971 belongs to a separate unadjusted summary-statistic comparison. The two values differ because the methods and information used are not identical; the registry record does not provide enough detail to identify a single reason for the difference beyond this distinction.

In particular, a non-significant result does not establish equivalence. Equivalence requires a prespecified margin defining differences small enough to be considered practically equivalent and an analysis designed to evaluate that margin. Such a margin and an equivalence analysis are not reported in the registry record. Likewise, a superiority test and a failure to detect a difference cannot by themselves demonstrate that the treatments have the same effect.

Why does crossover structure matter?

In a crossover trial, participants receive more than one treatment condition over time. A comparison can potentially use each participant as their own reference, which may account for stable differences between people. But this depends on having measurements linked within participants and using an analysis that reflects that pairing. Treatment period, treatment order, and any lingering effect from an earlier period can also be relevant. The record identifies a randomized, double-blind crossover design and a washout period, but detailed analysis of pairing, period, sequence, or carryover is not reported in the registry record.

The registry’s reported method is named Wilcoxon (Mann-Whitney), a name commonly used for a rank-based comparison of independent groups. The record does not say whether the reported test was adapted for the crossover structure or how observations were arranged for it. It would therefore be inappropriate to assume that the analysis did or did not account for within-person pairing. The ambiguity in the sequence descriptions further limits interpretation of how treatment order was handled.

What do the standard deviations contribute?

The standard deviation describes how spread out observations are around a mean. The posted standard deviations are 0.279 cm2 for olopatadine and 0.254 cm2 for placebo. They describe variation in the outcome summaries, but they do not reveal the shape of the distributions, outliers, or the within-participant association between treatment responses. In a crossover study, that association is especially relevant to the precision of paired contrasts, and it cannot be recovered from the separate standard deviations alone.

Over-dispersion is a general concern when observed variability exceeds what a statistical model expects. In general, failing to account for excess variability can make uncertainty estimates too small and tests too optimistic. The registry record does not report a specific over-dispersion assessment or indicate that over-dispersion was identified in this trial. It should therefore be treated as a general statistical consideration, not a finding about these data.

10. Confidence Intervals and Precision

The recalculated mean difference is -0.003 cm2, and its confidence interval extends across zero, from a negative value to a positive value. This indicates that the summary-statistic estimate is compatible with no mean difference and with differences in either direction under the assumptions of that calculation. The interval is more informative than the point estimate alone because it shows uncertainty around the estimated contrast. Its limits are not repeated numerically because the registry comparison reports them to more decimal places than the stated display limit.

A confidence interval is a range produced by a method that, under its assumptions, would cover the underlying parameter at the stated frequency across repeated applications. It is not a guarantee that the true effect lies within a particular interval, nor does it assign probabilities to individual parameter values after observing the data. The interval also depends on the analysis method and assumptions. The recalculated interval is based on posted summary statistics; it is not the confidence interval from the registry’s reported Wilcoxon (Mann-Whitney) analysis. The registry record does not report a confidence interval for that analysis.

The treatment means, 0.461 cm2 and 0.464 cm2, are close, but closeness of the point estimates does not by itself establish equivalence. A sufficiently informative equivalence assessment would require a clinically justified margin and an analysis planned for that purpose; these are not reported in the registry record. Nor can the precision be evaluated fully from separate group summaries in a crossover trial, because the within-participant variability and correlation are not reported.

The outcome summaries list 21 observations for each treatment. That sample size, the observed variability, and the crossover structure all bear on precision. Without participant-level data and a clearly reported paired analysis, the recalculated uncertainty should be understood as an unadjusted summary-based estimate rather than a complete account of uncertainty for the original design.

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 events. They are illustrations based on the observed effect rather than the trial’s original design justification. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect.

For planning at 100% of the observed effect, the figures are 124,150 participants per arm for 80% power and 166,201 participants per arm for 90% power. For planning at 75% of the observed effect, they are 220,710 participants per arm for 80% power and 295,468 participants per arm for 90% power. For planning at 50% of the observed effect, they are 496,597 participants per arm for 80% power and 664,803 participants per arm for 90% power.

The 80% power column describes the planned chance of detecting the chosen effect under the stated assumptions if that effect were present. The 90% power column uses a higher target chance of detection under those assumptions. These are design calculations, not results from the trial. A smaller effect is harder to distinguish from random variation, so planning to detect smaller effects requires more participants. Because the observed difference is very small, these illustrations become extremely large as the assumed target effect is reduced.

This is not observed power. The registry record does not report the trial’s own sample-size assumptions, original target enrollment, or justification for the enrolled number. The planning figures should not be interpreted as evidence about the probability that the trial detected an effect or as a retrospective measure of whether its result is valid.

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 = -0.003; pooled standard deviation σ = 0.267.
2. n = 2 × (1.96 + 0.842)² × 0.267² / -0.003² = 124,149.2, rounded up to 124,150 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 <- -0.003                      # observed difference in means
sigma <- 0.26679299091243                     # 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(775658)
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 = 124,150 (80% power), 166,201 (90% power)
 75% of observed effect: n per arm = 220,710 (80% power), 295,468 (90% power)
 50% of observed effect: n per arm = 496,597 (80% power), 664,803 (90% power)
power.t.test: 124150.2 
Simulated power at n = 124,150 per arm: 0.805 (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. Analyses Reported by the Sponsor

The registry’s own analysis is reported as a Wilcoxon (Mann-Whitney) test, with p = 0.57. It does not report a numerical treatment effect or confidence interval for that test. Separately, the summary-statistic recalculation estimates a mean difference of -0.003 cm2, with p = 0.971 and a confidence interval that crosses zero. The recalculation is an unadjusted comparison derived from the posted means and standard deviations, not a reproduction of the reported rank-based analysis.

The two p-values should not be treated as competing estimates from an identical procedure. The registry’s result comes from the named Wilcoxon (Mann-Whitney) method, while the recalculation targets a mean difference using group summaries. In addition, the registry describes a crossover design, but it does not explain whether the reported analysis accounted for pairing or other crossover features. The summary-statistic recalculation does not recover those features from the posted group-level information.

The registry record does not provide the test statistic, participant-level data, or details needed to reproduce the sponsor’s test. The posted results therefore support reporting the sponsor’s stated method and p-value alongside, rather than replacing it with, the separate mean-based recalculation. Neither result establishes equivalence or resolves the design and reporting uncertainties.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 2 outcome measures and 2 analyses, while the number of posted secondary outcomes is 0. It identifies the histamine wheal area comparison as the primary endpoint. The record does not describe whether the trial prespecified multiple hypotheses, whether any multiplicity adjustment was planned, or how the two listed outcome measures and analyses relate to one another. A multiplicity strategy is therefore not reported in the registry record.

Multiplicity is a general statistical issue that arises when several outcomes, time points, or comparisons are tested. The more opportunities there are to find a nominally notable result, the more important it can be to state in advance which comparisons are primary and how interpretation will account for the full set. That general principle does not establish that multiplicity affected this trial’s reported result; the record does not give enough detail to assess that question.

The primary outcome timeframe mentions baseline and a return visit between study day 7-10, but the registry does not explain the exact analysis of baseline values or whether change from baseline was used. It also does not report a prespecified missing-data method, imputation procedure, or sensitivity analysis. The treatment-specific outcome summaries list 21 observations each, while enrollment is 24. Reasons for the difference and the handling of absent measurements are not reported in the registry record. It is therefore not possible to determine whether missingness was related to treatment, outcome, or participant characteristics.

The analysis population is not described in detail. The record does not state whether the comparison included all randomized participants, only participants who completed both treatment periods, or another defined group. It also does not report how protocol deviations, adherence, or incomplete crossover sequences were handled. These details are important because exclusions after randomization can affect the comparability created by random assignment.

For a crossover design, the analysis plan may also need to address sequence, period, within-participant pairing, and carryover. The record reports a washout period but does not report how these design features were assessed or incorporated statistically. The inconsistent written sequence descriptions add uncertainty about how treatment order was recorded. These are reporting gaps; they do not establish that a particular analytical choice was made or that the result was biased.

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

The posted results describe mean histamine wheal areas of 0.461 cm2 during olopatadine treatment and 0.464 cm2 during placebo treatment. The unadjusted recalculated difference is -0.003 cm2, and the recalculated confidence interval includes zero. The sponsor-reported rank-based analysis has p = 0.57. Together, these results do not show a clear separation between the reported treatment summaries in this dataset, but they do not prove that the treatments are equivalent.

A statistical comparison of this endpoint is not a direct measure of allergy symptoms, daily functioning, or other clinical outcomes. The record identifies allergic rhinitis and allergic conjunctivitis as conditions, but the primary endpoint is a histamine skin-test response. The result should therefore be interpreted as a finding about the recorded skin-test measure, not as a general conclusion about clinical effects.

It is also important not to read a non-significant p-value as evidence that no effect exists. A study can fail to distinguish an effect from random variation when its estimate is uncertain, and an equivalence conclusion requires a defined margin and appropriate analysis. Neither an equivalence margin nor an equivalence analysis is reported in the registry record. The confidence interval’s crossing of zero communicates uncertainty in the recalculated mean contrast; it does not demonstrate that all values within the interval are clinically unimportant.

The crossover design can be informative when within-participant comparisons are analyzed appropriately, but the record does not describe how pairing, period, or sequence effects were handled. Outcome counts are listed as 21 per treatment, and the record does not explain their relation to enrollment of 24. These uncertainties limit how confidently the posted summaries can be generalized beyond the specific observations and analysis reported.

The adverse-event listing records 0 at risk and 0 affected for both olopatadine and placebo, while deaths are null. Since the at-risk counts are recorded as zero, these entries do not establish that no adverse events occurred among participants; they do not provide usable denominators for interpreting event frequency. The registry record does not provide further adverse-event detail in the listed terms.

Accordingly, the result is best read as a close pair of observed means accompanied by uncertainty and incomplete reporting about analysis and participant accounting. It does not establish a treatment recommendation, demonstrate equivalence, or determine what would occur in a different population or study setting.

16. Why This Trial Matters Statistically

This trial raises a useful statistical issue: the design and the analysis need to match. In a crossover study, the same participant can contribute observations under more than one condition. The association between those observations can influence the estimated uncertainty of the treatment contrast. Separate group means and standard deviations do not reveal that association, and the registry record does not report a paired-difference summary or explain how the analysis handled the crossover structure.

The record also illustrates why an analysis label and a numerical result are not enough to reconstruct a statistical analysis. It names a Wilcoxon (Mann-Whitney) method and reports p = 0.57, but does not provide the test statistic, detailed implementation, confidence interval, or participant-level data. A separate recalculation from the posted summary statistics estimates a mean difference of -0.003 cm2 and gives p = 0.971. Those estimates answer related but not identical statistical questions, so readers need to distinguish method from result rather than selecting a p-value in isolation.

Another important issue is the difference between statistical uncertainty and equivalence. Similar treatment means and a confidence interval that includes zero may be compatible with little difference, but they do not establish that any difference is small enough to be unimportant. That conclusion would require a prespecified margin and an analysis designed to assess it; neither is reported in the registry record.

Finally, enrollment and analyzed counts are not interchangeable. Enrollment is 24, while each treatment summary lists 21 observations. The record does not explain how participants contributed data across the crossover periods. Transparent reporting of participant flow, missingness, treatment order, and analysis population would help readers understand what the estimate represents. These points make the trial a useful example of how statistical interpretation depends on design details and reporting, not only on the size of a point estimate.

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.

The registry record leaves several trial-specific questions unresolved. The crossover sequence descriptions are inconsistent: both named sequences describe olopatadine followed by placebo. The record therefore does not clearly establish the actual treatment order for each sequence. Although a washout period is described, the record does not report whether carryover was assessed or how period and sequence effects were handled.

The statistical reporting is limited. The sponsor’s analysis is identified as Wilcoxon (Mann-Whitney), with p = 0.57, but no test statistic, confidence interval, detailed implementation, or participant-level data are reported. The summary-statistic recalculation uses treatment means and standard deviations and cannot reconstruct a paired analysis. The record does not state whether the sponsor’s method accounted for the crossover structure.

Enrollment is 24, whereas the primary outcome summaries list 21 observations for each treatment. The reasons for this difference, participant flow through both periods, and methods for handling missing outcome measurements are not reported in the registry record. The analysis population and the treatment of protocol deviations are also not described.

The record does not report the original sample-size justification, the detailed statistical analysis plan, or a rationale for the chosen outcome and analysis. It also does not report an equivalence margin or an equivalence analysis. Consequently, the result cannot be interpreted as evidence of equivalence merely because the observed means are close.

The adverse-event entries list 0 at risk and 0 affected for each group, with deaths recorded as null. The zero at-risk entries do not offer denominators from which adverse-event occurrence could be evaluated. More complete adverse-event reporting is not present in the registry record. These limitations concern what can be established from the record; they do not demonstrate that any particular procedure was omitted during the trial.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Olopatadine Eye Drops and Allergy Skin Testing — NCT00775658.
• Registry record last updated 2017-07-12; status last verified 2017-06.
• Results first posted 2017-07-12.
• 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

NCT00775658 is a completed randomized, double-blind crossover trial sponsored by Vanderbilt University Medical Center. It enrolled 24 and compared olopatadine 0.2% ophthalmic solution with placebo for a histamine skin-prick response. Treatment descriptions specify daily eye drops for 7 to 10 days and a 7-10 day washout period, although the written sequence descriptions are inconsistent.

The primary outcome is mean wheal area in cm2. The posted means are 0.461 for olopatadine and 0.464 for placebo, with 21 observations listed for each treatment. The registry reports a Wilcoxon (Mann-Whitney) analysis with p = 0.57. A separate unadjusted recalculation gives a mean difference of -0.003 cm2 and p = 0.971; its confidence interval includes zero.

The record does not explain the outcome counts, provide detailed crossover-analysis methods, or report the original sample-size justification. The findings describe the posted histamine skin-test summaries and should not be read as establishing equivalence or broader clinical effects.