Use of Topical NSAID to Reduce Pain in Oral Lichen Planus and Oral Lichenoid Lesions.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Use of Topical NSAID to Reduce Pain in Oral Lichen Planus and Oral Lichenoid Lesions. (NCT03509675), a randomized phase 2, phase 3 study. The primary outcome is Self-reported Pain Scores Using 100m Visual Analog Scale 0-100 Pain Score..

2. Clinical Question

3. Trial Design

The registry describes a randomized crossover trial of topical ibuprofen suspension and placebo for pain associated with oral lichen planus and oral lichenoid lesions. It lists the study as Phase 2 and Phase 3, with an enrollment of 23 and a completed status. The sponsor was the University of Washington. The start date was 2018-04-29, and the primary completion date was 2019-01-30.

In a crossover design, participants receive more than one treatment in sequence, so comparisons can draw on observations during different treatment periods. This general design can reduce the influence of stable differences between participants, but the analysis needs to account appropriately for the within-participant structure. The registry record does not report treatment sequence details, a period-specific analysis, or whether the analysis accounted for within-participant correlation. It describes a 7-day washout for participants already receiving active treatment at enrollment; the washout arrangements between study treatments are not reported in the registry record.

The record lists triple masking, while the brief summary describes the study as double-blinded. It does not identify which roles were masked. The active suspension contained ibuprofen at 100 mg per 5 ml; the placebo was compounded with a similar taste but without the active ingredient. The intervention listing also names an ibuprofen suspension at 5mg/100ml, which does not match the concentration in the arm description. The record does not explain this discrepancy.

Participants were asked to record spontaneous pain at day 0, day 4, and the end of day 7 for both treatment arms. The summary also describes initial contact after the first day of intervention and weekly reminder calls to support form completion and check for side effects. Further details about allocation procedures, treatment sequence, adherence, and the conduct of the trial 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.

First Intervention (7 Days)

Reasons for not completing this period:

Washout Period (7 Days)

Second Intervention )7 Days)

Reasons for not completing this period:

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 170 patients were assessed for eligibility through the Oral Medicine Clinical Services at the University of Washington before enrolling.

6. Primary Endpoint

The registry lists the primary outcome as Self-reported Pain Scores Using 100m Visual Analog Scale 0-100 Pain Score., measured over Total pain improvement over day 0 to day 7 (summation of improvement over 7 days).

The primary endpoint was self-reported pain improvement over day 0 to day 7, described as the summation of improvement over 7 days. Pain was recorded using a 100 mm visual analog scale, with 0 representing no pain and 100 the worst pain experienced. Participants recorded spontaneous pain at day 0, day 4, and the end of day 7 for each treatment. The registry states that lower pain scores mean a better outcome and reports the primary measure as a mean with standard deviation.

A patient-reported pain scale is relevant when the trial question concerns symptoms experienced by participants. It captures their own assessment rather than an observer's rating. As a general measurement principle, repeated assessments can describe how symptoms change during a treatment period, while a summary of improvement can condense those observations into a single endpoint. The registry does not give the exact formula used to sum improvement or explain how missing pain assessments were handled.

The posted results give an improvement mean of 12.3 for the ibuprofen rinse and 6.4 for placebo. These values are in the direction of greater recorded improvement with ibuprofen, but their interpretation depends on the definition and calculation of the summed endpoint. The record does not provide additional detail on that calculation. The primary outcome is therefore best read as the registry-defined summary of pain improvement, rather than as a direct final-day pain score.

7. Results Posted in the Registry

• Ibuprofen Rinse: mean 12.30 versus 6.40 on control; difference 5.90 (95% CI -5.53 to 17.33), standardized effect d = 0.30; z-test p = 0.312.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is reported as a mean and standard deviation rather than as a count of participants with an event. A mean describes the central value of the observations, while a standard deviation describes how spread out individual observations are around that mean. In the registry, the ibuprofen group's mean improvement was 12.3 with a standard deviation of 23.4; the placebo group's mean was 6.4 with a standard deviation of 15.3. These summaries describe the posted values but do not show the distribution of individual responses or how many participants improved.

The mean difference is the contrast between the group means, reported as 5.9. A difference is easier to interpret than two means alone because it expresses the comparison on the endpoint's scale. The sign depends on the subtraction order; the comparison lists ibuprofen against placebo. The endpoint is a summed improvement measure, rather than simply the pain score at the end of treatment. The exact summation method is not reported in the registry record, which limits assessment of how the contrast maps to the original pain assessments.

The interval around the mean difference runs from -5.526 to 17.33. As a general statistical explanation, a confidence interval describes a range of values reasonably compatible with the data and the assumptions of a statistical procedure. Its width conveys uncertainty: a wider interval indicates less precision than a narrower interval on the same scale. This interval crosses zero, so the data do not narrowly pin down the direction of the difference. The registry record does not state the confidence level or identify a crossover-specific procedure used to produce the interval.

The unadjusted p-value is 0.312. As a general statistical explanation, a p-value measures how surprising data at least as extreme as those observed would be if a specified null model were true. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of the difference. “Unadjusted” indicates that the reported value has not been corrected for multiple statistical comparisons. The registry lists one primary outcome and no posted secondary outcomes, but the full analysis plan and any analyses not posted are not reported in the registry record.

The trial is identified as a crossover study. As a general statistical principle, measurements from the same participant across treatment periods are related, and an analysis should account for that within-person dependence. Period effects, treatment order, and possible carryover can also matter in a crossover comparison. The registry record does not report the statistical model, period or sequence terms, carryover assessment, or whether the posted comparison used paired observations. It is therefore not possible to verify from the record whether the analysis reflects the crossover design.

The standard deviations and comparison are presented with group counts of 23 for placebo and 23 for ibuprofen. The record lists an enrollment of 23, but it does not explain the relationship between enrollment and the group counts in the crossover analysis. It also does not report participant flow, withdrawals, the number of complete outcome records, or whether any imputation was used. This missing information matters because a mean comparison can be affected if outcome data are unavailable or incomplete in a way related to treatment or pain.

The registry record reports no analysis entries and no secondary outcomes. It does not report an original statistical analysis plan, the method used for the primary endpoint, assumptions used for the trial's original design, or any interim analyses. The comparison and planning illustrations should therefore not be mistaken for a complete account of the investigators' prespecified analysis. As a general principle, prespecification helps distinguish planned analyses from choices made after observing results; that distinction cannot be evaluated from the available record.

The posted summaries do not establish whether covariates were used, whether treatment sequence or period effects were adjusted, or whether a sensitivity analysis was conducted. They also do not provide participant-level data or distributions that could reveal outliers or departures from model assumptions. A standard deviation alone does not demonstrate that a normal model is appropriate. As a general statistical consideration, unusually skewed or over-dispersed observations can make a simple mean-based model less reliable, which is why distributional checks and design-appropriate analyses can be important. Such checks are not reported in the registry record.

10. Confidence Intervals and Precision

The mean difference is reported as 5.9, with an interval from -5.526 to 17.33. This interval is broad enough to include zero and values on either side of zero. It therefore does not identify a precise direction for the average difference. Its endpoints also represent contrasts that could be materially different in size, but the registry does not provide a threshold for judging clinical importance.

As a general statistical explanation, an interval expresses both an estimate and its uncertainty. A narrow interval can indicate that estimates are concentrated within a limited range, while a broad interval leaves more possible values compatible with the data and method. The interval is especially important because the point estimate alone, 5.9, can sound more definite than the uncertainty allows. The unadjusted p-value of 0.312 is consistent with the interval's inclusion of zero, but the interval makes the range of uncertainty more visible.

The trial enrolled 23, and the posted outcome summaries list 23 observations for each treatment group. Those figures do not establish that all relevant observations were complete or explain how a crossover analysis treated repeated measurements. The registry record does not report a confidence level for the interval, its exact calculation, or an analysis that explicitly accounts for treatment periods and within-participant correlation. Precision should therefore be interpreted in light of both the interval itself and the absence of detail about the method used to obtain it.

The interval is not evidence that the treatments are equivalent. Demonstrating equivalence or non-inferiority would require a prespecified margin and an analysis designed for that question; such a margin and analysis are not reported in the registry record. Nor does the interval by itself establish whether a possible difference would matter to participants. The posted estimate is best described as uncertain, with the observed data allowing a range that includes no average 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 sample-size figures are planning illustrations based on the observed effect, not a report of the trial's original design calculations. They give participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set as the stated fraction of the observed effect.

The 80% power column gives the planned participants per arm for a design that would have an 80% probability of detecting the target effect if that effect and the stated assumptions were correct. The 90% power column gives the corresponding per-arm planning figure for a design with a 90% probability. These are conditional planning probabilities, not guarantees about what a future study will find. This is not observed power.

The rows illustrate planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The listed per-arm figures are 177 and 236 for 100% of the observed effect, 314 and 420 for 75% of the observed effect, and 705 and 944 for 50% of the observed effect, for 80% and 90% power respectively. The pattern is expected: smaller effects are harder to distinguish from random variation, so detecting them requires more participants under the same assumptions.

The figures depend on the observed effect estimate, which is uncertain in this trial, and on simplifying assumptions. The record's planning description specifies equal allocation and a normal approximation but does not describe an adjustment for the crossover structure, attrition, or missing observations. The original trial's own sample-size justification and design assumptions are not reported in the registry record. These planning values should therefore be read as illustrations rather than as a reconstruction of the investigators' original design process.

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 = 5.900; pooled standard deviation σ = 19.769.
2. n = 2 × (1.96 + 0.842)² × 19.769² / 5.900² = 176.2, rounded up to 177 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 <- 5.9                      # observed difference in means
sigma <- 19.7692943728399                     # 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(3509675)
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 = 177 (80% power), 236 (90% power)
 75% of observed effect: n per arm = 314 (80% power), 420 (90% power)
 50% of observed effect: n per arm = 705 (80% power), 944 (90% power)
power.t.test: 177.2 
Simulated power at n = 177 per arm: 0.804 (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 lists one primary outcome measure and no posted secondary outcomes. It does not report a family of additional planned endpoints or comparisons, nor does it describe any adjustment for multiple testing. The primary comparison is accompanied by an unadjusted p-value of 0.312. This means the displayed value should not be read as corrected for multiplicity. If other outcomes or analyses were examined but not posted, their number and role are not reported in the registry record.

As a general statistical principle, testing many outcomes or treatment contrasts increases the chance of finding an apparently notable result by chance. Prespecified primary outcomes and an explicit multiplicity plan help clarify which comparisons support confirmatory conclusions. The registry record does not report a complete statistical analysis plan, the planned hierarchy of tests, or whether any interim analyses took place. It is not possible to infer these details from the single posted outcome.

Missing-data handling is also not reported in the registry record. The posted outcome summaries list 23 for placebo and 23 for ibuprofen, but the record does not give a participant flow account, describe incomplete pain diaries, or say whether all recorded assessments contributed to the summed endpoint. It does not state whether observations were imputed, excluded, or analyzed using a model that accommodates missing data. A group count alone cannot show whether missingness affected the estimate.

The analysis population is not defined. The registry does not state whether the primary analysis included all randomized participants, only participants who completed both treatment periods, or another set. It also does not report adherence criteria or deviations from the assigned treatment sequence. These details are relevant in randomized crossover trials because incomplete participation in a treatment period can affect which within-participant comparisons are available.

Finally, the record does not describe how the crossover design was incorporated into analysis. Treatment order, period effects, within-participant correlation, and carryover assessment are not reported. These are not conclusions about a flaw in the analysis; they are details needed to judge it that the registry record does not provide.

14. 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.

The posted safety figures list zero serious adverse events in the placebo group and zero in the active ingredient group. They also list zero other adverse events and zero deaths in each group. The at-risk count is 23 for each group in both the serious and other adverse event summaries.

These figures are counts recorded for the trial, not estimates of how often an event might occur in a broader population. A zero count does not establish that an event is impossible or that a treatment has no safety risks. As a general statistical principle, uncommon events may not appear in a small trial, even when the trial reports none. The safety summaries do not provide event-specific descriptions, severity grades beyond the serious-event classification, or timing in relation to treatment.

The registry lists no serious adverse event terms. The brief summary says participants were contacted after the first intervention day and received weekly reminder calls to check for side effects. Further details about how adverse events were solicited, classified, or attributed to treatment are not reported in the registry record. The figures should therefore be interpreted as the recorded event counts for this trial, rather than as a comprehensive estimate of comparative safety.

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

The posted result describes a mean summed pain improvement of 12.3 with ibuprofen and 6.4 with placebo. The comparison reports a mean difference of 5.9, an interval from -5.526 to 17.33, and an unadjusted p-value of 0.312. The point estimate is in the direction of greater improvement with ibuprofen under the registry's endpoint description, but the interval includes zero and leaves substantial uncertainty about the direction and size of the average contrast.

This result does not establish that ibuprofen is better, worse, or equivalent to placebo. The interval is compatible with no average difference and with effects in either direction. The registry does not state a threshold for a clinically important difference, so the statistical estimate cannot establish whether any possible contrast would be meaningful to participants. The p-value also cannot answer that question; it is not a measure of effect size or clinical value.

The record documents a randomized, triple-masked crossover trial, but does not report the statistical model or how the crossover design was incorporated. It also omits details about treatment sequence, the handling of incomplete pain records, and the analysis population. These gaps limit assessment of the reliability and applicability of the reported comparison. The posted group counts and summary values alone cannot resolve them.

The safety summaries report no serious adverse events, other adverse events, or deaths in either group. Those observations describe what was recorded in this trial, not the full range of possible harms. The enrollment was 23, so the results offer limited information about uncommon events or how effects might vary across people with different characteristics.

In short, the registry provides a small, uncertain comparison of a patient-reported pain-improvement measure. It does not support a definitive treatment conclusion, and it cannot replace a complete analysis report with details of the crossover model, missing data, and prespecified analysis plan.

16. Why This Trial Matters Statistically

This trial illustrates why a treatment contrast should be read together with its uncertainty. The mean difference of 5.9 appears to favor greater summed improvement with ibuprofen under the endpoint description, but the interval from -5.526 to 17.33 includes zero and extends in both directions. The estimate alone is therefore not enough to characterize how firmly the data distinguish the treatments.

It also demonstrates the importance of matching analysis to design. A crossover design can use within-participant comparisons, but it creates analytical requirements involving repeated observations, treatment periods, and order. The registry identifies the crossover model but does not report how those features entered the analysis. Statistical interpretation depends not only on the reported means and p-value but also on whether the analysis reflects the study's structure.

The outcome is patient-reported pain, a directly relevant measure for a question about symptomatic relief. At the same time, the endpoint is described as a summed improvement over day 0 to day 7, and the exact summation method is not reported. Clear endpoint construction is essential for interpreting what a mean difference represents and for comparing results across studies.

The safety figures offer another statistical lesson: no recorded events are not the same as proof of no risk. Small studies can provide useful observations while remaining unable to characterize uncommon outcomes precisely. The sample-size illustrations likewise show how design targets depend on the effect researchers aim to detect; smaller target effects require larger planned samples under the stated assumptions.

Finally, this record shows the value of transparent reporting. Details on missing data, analysis populations, multiplicity, and crossover-specific modeling would help readers judge the strength and limits of the posted result. Without those details, careful interpretation means describing the estimate and its uncertainty rather than making a categorical treatment claim.

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 enrollment was 23, and the posted outcome counts list 23 in each treatment group. This is a limited evidence base for estimating an average treatment contrast, particularly when the interval around the difference is broad. The registry does not report whether all participants completed all treatment periods or whether all planned pain assessments were available.

The endpoint is described as total pain improvement over day 0 to day 7, but the exact calculation for summing improvement is not reported in the registry record. Pain was recorded at day 0, day 4, and the end of day 7. Without the calculation details, the relationship between these observations and the final summary measure is not fully transparent.

Although the study is identified as a crossover trial, the record does not report treatment sequences, washout arrangements between the study treatments, or the analysis method used to account for repeated observations. The brief summary mentions a 7-day washout for participants already on active treatment at enrollment, but does not explain whether or how washout was handled between treatment periods. Period effects and carryover assessment are not reported.

The registry describes masking differently in two places: it lists triple masking, while the brief summary calls the trial double-blinded. The record does not identify the masked roles. The intervention listing also gives a concentration of 5mg/100ml, while the active arm description specifies 100 mg per 5 ml. The reason for the difference is not reported.

Further limitations include the absence of a reported analysis plan, missing-data procedures, participant flow, analysis population, and clinical importance threshold. The record reports no secondary outcomes or analysis entries. These omissions limit assessment of analytic choices, robustness, and the broader clinical meaning of the primary estimate.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Use of Topical NSAID to Reduce Pain in Oral Lichen Planus and Oral Lichenoid Lesions. — NCT03509675.
• Registry record last updated 2020-05-08; status last verified 2020-04.
• Results first posted 2020-05-08.
• 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

The registry describes NCT03509675 as a completed randomized, triple-masked crossover study sponsored by the University of Washington. It enrolled 23 people with oral lichen planus or oral lichenoid lesions and compared topical ibuprofen suspension with placebo. The registered phase is Phase 2 and Phase 3.

The primary outcome was a summed measure of self-reported pain improvement over day 0 to day 7, based on a 100 mm visual analog scale. The posted means were 12.3 for ibuprofen and 6.4 for placebo. The mean difference was 5.9, with an interval from -5.526 to 17.33 and an unadjusted p-value of 0.312. The interval includes zero, so the direction and size of the average contrast remain uncertain.

The safety figures report zero serious adverse events, zero other adverse events, and zero deaths in each group. Details of the crossover analysis, missing-data handling, analysis population, and original design assumptions are not reported in the registry record.