A Phase 2, Multicenter, Randomized, Double-Blind, Vehicle Controlled, Parallel-Group, Clinical Study in the Treatment of Acne Vulgaris: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2, Multicenter, Randomized, Double-Blind, Vehicle Controlled, Parallel-Group, Clinical Study in the Treatment of Acne Vulgaris (NCT03170388), a randomized phase 2 study. The primary outcome is Inflammatory Lesion Counts.

2. Clinical Question

3. Trial Design

The registry describes a phase 2, multicenter, randomized, parallel-group clinical study in acne vulgaris. Its title identifies the study as double-blind, while the registry’s masking classification is triple. The record does not explain which people were masked, so the difference in those descriptions cannot be resolved from the registry record. The sponsor was Bausch Health Americas, Inc.

Five groups are named: IDP-126 Gel, IDP-126 Component A, IDP-126 Component B, IDP-126 Component C, and IDP-126 Vehicle Gel. The posted primary outcome summaries include 146 participants for IDP-126 Gel, 150 for Component A, 146 for Component B, 150 for Component C, and 148 for vehicle. These are group-specific outcome counts; they need not describe every population used for every part of the study.

The registry lists enrollment of 741, a start date of 2017-10-05, primary completion on 2019-04-23, and status as completed. The registry record does not report the original design assumptions, the allocation sequence procedure, or the planned statistical analysis details. It also does not report the duration of treatment or follow-up in the registry record for this analysis.

The primary endpoint was absolute change in inflammatory lesion counts from baseline to Week 12. Six secondary outcomes are posted, including inflammatory and non-inflammatory lesion-count changes and global severity score success measures at Weeks 2, 4, and 8. The registry record lists no analyses for those outcomes in the information summarized for this analysis.

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: One participant in the IDP-126 Gel group did not receive study medication and thus is not included in the demography population.

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: One participant in the IDP-126 Gel group did not receive study medication and thus is not included in the demography population.

6. Primary Endpoint

The registry lists the primary outcome as Inflammatory Lesion Counts, measured over Baseline to Week 12.

The primary endpoint measures the change in inflammatory lesion count between baseline and Week 12. It is expressed as an absolute change in the mean number of lesions, rather than as a percentage change or a final count alone. A negative change indicates that the count at the later assessment is lower than at baseline on the endpoint’s scale.

Lesion counts give a numerical measure of one aspect of acne severity. Comparing mean changes between randomized groups can describe how the groups differed over the study period. The registry identifies the reported parameter as a least-squares mean and gives standard deviations as the dispersion measure. The displayed group estimates are -29.9 for IDP-126 Gel and -19.6 for IDP-126 Vehicle Gel.

For interpretation, the between-group estimate is more informative than looking only at whether each group’s count changed. A comparison accounts for the vehicle group’s observed change and estimates the difference between groups. That comparison remains specific to this endpoint, its Week 12 time point, and the study population. It does not, by itself, describe non-inflammatory lesions, global severity scores, longer-term outcomes, or individual responses. Further detail on the model used to obtain the least-squares means is not reported in the registry record.

7. Results Posted in the Registry

• IDP-126 Gel: mean -29.90 versus -19.60 on control; difference -10.30 (95% CI -13.04 to -7.56), standardized effect d = -0.86; z-test p = <0.001.
• IDP-126 Component A: mean -26.70 versus -19.60 on control; difference -7.10 (95% CI -9.81 to -4.39), standardized effect d = -0.60; z-test p = <0.001.
• IDP-126 Component B: mean -24.80 versus -19.60 on control; difference -5.20 (95% CI -7.92 to -2.48), standardized effect d = -0.44; z-test p = <0.001.
• IDP-126 Component C: mean -26.80 versus -19.60 on control; difference -7.20 (95% CI -9.90 to -4.50), standardized effect d = -0.60; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean change describe?

A mean change summarizes the average difference between baseline and a later measurement for a group. Here, the measurement is inflammatory lesion count and the later time point is Week 12. A negative group estimate means that the average count decreased on the change scale. The group mean does not imply that every participant had the same change, or even that every participant changed in the same direction.

The registry identifies the endpoint estimate as a least-squares mean. In general, a least-squares mean is a model-based group mean, often adjusted for terms included in a statistical model. It can differ from a simple arithmetic average if the model includes covariates or other structure. The model terms and the method used to estimate the least-squares means are not reported in the registry record, so the model’s adjustment cannot be specified.

How should the group difference be read?

A mean difference compares the change estimate in one group with the change estimate in the comparator group. For IDP-126 Gel versus vehicle, the posted contrast is -10.3 lesions. Since both group estimates are changes, a negative contrast means that the estimated change was more negative in the IDP-126 Gel group. This interpretation depends on the direction of the endpoint scale and is not a separate measure of final lesion count.

Randomization is intended to make comparison groups comparable on average, including with respect to factors that are not measured. A randomized parallel-group comparison estimates a between-group contrast over the study period. The record identifies randomization and a parallel model, but does not give the allocation method or details of any adjustment used in the endpoint analysis.

What does a standard deviation tell the reader?

Standard deviation describes the spread of participant-level observations around a group mean. A larger standard deviation indicates more variation in those observations. It is not the same as the standard error, which describes how precisely a sample mean or estimate is measured, and it is not a confidence interval. The posted arm-level standard deviations are 11.86 for IDP-126 Gel, 11.74 for Component A, 11.71 for Component B, 11.69 for Component C, and 12.12 for vehicle.

In general, variation matters because greater spread makes a mean difference harder to estimate precisely when other features are unchanged. The displayed comparison calculations use an approximate pooled standard deviation, as indicated in their calculation notes. This is a summary-statistic approach and should not be mistaken for the full model used by the sponsor, which is not reported in the registry record.

What do confidence intervals and p-values add?

A confidence interval gives a range of values compatible with the data under the statistical method used. It conveys both the estimated direction and the precision of a comparison. A narrower interval indicates greater precision than a wider interval, for the same scale and context. It is not a range containing a fixed proportion of individual participant outcomes.

A p-value assesses how unusual the observed data, or data more extreme, would be under a specified null hypothesis and statistical model. It is not the probability that the null hypothesis is true, nor is it a measure of clinical importance. The posted comparison p-values are unadjusted. The registry record does not report whether the original sponsor analysis used a different model or a multiplicity adjustment.

Why distinguish the posted calculation from the original analysis?

The comparison statistics are presented as unadjusted summary-statistic comparisons using approximate standard deviations. They support a plain reading of the posted group summaries and are not a substitute for a complete trial analysis plan. The record lists zero analyses in the information summarized for this analysis, and does not provide a statistical analysis plan, model specification, missing-data procedure, or sponsor analysis output. Those omissions limit how specifically the inferential method can be reconstructed.

General statistical interpretation should also distinguish a result for one endpoint and time point from broader claims. The Week 12 inflammatory lesion count result does not establish what happened on the secondary measures at Weeks 2, 4, or 8. Although the registry lists those outcomes, their analysis estimates are not reported in the registry record.

10. Confidence Intervals and Precision

The IDP-126 Gel versus vehicle mean difference is -10.3 lesions, with a confidence interval from -13.04 to -7.559. The interval lies entirely below zero and is compatible with a negative difference across the range it covers. Its width communicates uncertainty in the estimate; it does not describe the variation in individual participant changes. The group standard deviations describe that participant-level spread separately.

The component comparisons with vehicle also have intervals entirely below zero: -9.81 to -4.39 for Component A, -7.924 to -2.476 for Component B, and -9.904 to -4.496 for Component C. These intervals quantify uncertainty for their respective group contrasts, not for comparisons between the active groups. The record does not report intervals for direct comparisons among IDP-126 Gel and the component groups.

All posted p-values for these comparisons are unadjusted and are <0.001. Statistical significance does not express the size of a difference or its practical importance. Precision is better assessed by considering the estimate and interval together, and by remembering that these intervals come from summary-statistic calculations rather than a documented full analysis model.

The interval’s exact coverage and method are not reported in the registry record. Nor are the assumptions underlying the original least-squares mean estimates. The numerical precision of the endpoints should therefore not be mistaken for certainty that the model fully captures all sources of uncertainty. The estimates describe the posted Week 12 endpoint and should not be extrapolated to other outcomes or time points.

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 the original trial’s documented design assumptions. For this continuous endpoint, the figures are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The registry record does not report the original sample-size justification or the assumptions used to plan the trial.

The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions. The 90% power column gives the corresponding planning count for a 90% chance. These are hypothetical design calculations. They are not observed power and do not describe the probability that the trial’s findings are true.

The rows represent planning for 100%, 75%, and 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so the planning calculation requires more participants per arm when the target effect is smaller. The figures differ across comparisons because the observed effects differ. For example, planning for Component B uses a smaller observed contrast than planning for IDP-126 Gel, and its associated participant requirements are consequently larger for the same stated power and fraction. These planning results depend on the listed assumptions and should not be read as a recommendation or as evidence that the study was designed around these particular targets.

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 = -10.300; pooled standard deviation σ = 11.992.
2. n = 2 × (1.96 + 0.842)² × 11.992² / -10.300² = 21.3, rounded up to 22 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 <- -10.3                      # observed difference in means
sigma <- 11.9915950588233                     # 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(3170388)
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 = 22 (80% power), 29 (90% power)
 75% of observed effect: n per arm = 38 (80% power), 51 (90% power)
 50% of observed effect: n per arm = 86 (80% power), 114 (90% power)
power.t.test: 22.3 
Simulated power at n = 22 per arm: 0.789 (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 a primary endpoint and six secondary outcomes. The secondary outcomes include changes in inflammatory and non-inflammatory lesion counts, percentage changes in those counts, and evaluator global severity score success measures at Weeks 2, 4, and 8. Several active groups are also compared with the same vehicle group. Testing multiple outcomes and contrasts can increase the chance that at least one result appears statistically notable by chance alone.

The posted p-values for the four comparisons are unadjusted. A multiplicity adjustment, when planned and used, can account for testing several hypotheses. Whether such a procedure was specified or applied in the sponsor’s original analysis is not reported in the registry record. The record also does not report a hierarchy of hypotheses, a gatekeeping strategy, or which outcomes were designated as confirmatory beyond identifying the primary endpoint.

Missing-data handling is not reported in the registry record. The record does not describe how missing Week 12 measurements were treated, whether any imputation was used, or what sensitivity analyses were conducted. Without those details, the implications of incomplete outcome data cannot be assessed from these summaries alone.

The primary endpoint population is also not fully defined in the information summarized for this analysis. The registry notes that one participant assigned to IDP-126 Gel did not receive study medication and was not included in the demography population. This statement does not establish whether that participant was included in, or excluded from, the primary endpoint analysis. The analysis population definitions and participant flow by outcome are not reported in the registry record. Therefore, the enrollment figure and the endpoint group counts should not be assumed to represent identical analysis sets.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: sickle cell anemia with crisis (0.7%), enteritis (0.7%), hyperbilirubinemia (0.7%), abortion induced (0.7%).

The registry reports serious adverse event figures by group using participants at risk and participants affected. The serious-event counts are 1 for IDP-126 Gel, 0 for Component A, 0 for Component B, 3 for Component C, and 0 for IDP-126 Vehicle Gel. The corresponding numbers at risk are 141, 146, 144, 148, and 146. No deaths are reported in any group.

These counts describe the registry’s serious adverse event summary. They do not establish that a study treatment caused an event, and they do not capture all aspects of safety. The record lists sickle cell anemia with crisis, enteritis, hyperbilirubinemia, and abortion induced as serious-event terms, each with a reported proportion of 0.7%. The record does not provide further context about timing, relatedness, or event narratives in the information.

The other adverse event counts are 30 for IDP-126 Gel, 29 for Component A, 6 for Component B, 19 for Component C, and 7 for vehicle, with each group’s reported number at risk listed alongside the safety results. These categories should not be conflated with serious adverse events. Counts alone do not establish comparative risk, particularly without fuller definitions, follow-up information, and an analysis of event severity and attribution.

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

The posted results describe group-average changes in inflammatory lesion counts through Week 12. The IDP-126 Gel group had a mean change of -29.9 lesions, compared with -19.6 for IDP-126 Vehicle Gel. Their estimated mean difference was -10.3 lesions, with a confidence interval from -13.04 to -7.559 and an unadjusted p-value of <0.001. These summaries indicate a more negative average change in the IDP-126 Gel group on this particular endpoint.

The estimate is not a measure of each participant’s response. Group averages can conceal differences in individual changes, and the confidence interval does not show the range of individual outcomes. The result also does not answer questions about non-inflammatory lesions, global severity scores, or the secondary outcomes at other time points. Those outcome results are not reported in the registry record summarized for this analysis.

A statistical comparison is not, by itself, a judgment about clinical importance. The registry information does not state a threshold for a clinically meaningful difference, and this analysis does not assign one. The p-value does not measure the probability that the intervention is effective or that the result will recur in another study. Interpretation also depends on analysis details that are not reported, including the model specification, missing-data approach, and any multiplicity strategy.

The safety counts likewise have a limited meaning: they report events and participants affected as recorded, but do not establish causation or provide a complete assessment of the balance of benefits and risks. Taken together, the figures are a statistical account of the posted trial outcomes, not personal medical guidance or a basis for an individual treatment decision.

16. Why This Trial Matters Statistically

This study illustrates how a randomized vehicle-controlled design can estimate a difference in outcome change, rather than relying only on the change observed within an intervention group. The vehicle group also showed a negative mean change in inflammatory lesion counts. Comparing the active group with that control makes the contrast, not either group’s change in isolation, central to interpretation.

The study includes the gel, three component groups, and vehicle. That structure creates several useful comparisons but also raises an important statistical issue: multiple comparisons need clear interpretation. The posted contrasts each compare an active group with vehicle, and their p-values are unadjusted. Without the original multiplicity plan or full analysis output, readers cannot determine whether these comparisons were treated as confirmatory or exploratory in the sponsor’s analysis.

The study also shows why estimates and precision belong together. The mean difference communicates the scale and direction of the contrast, while its confidence interval conveys uncertainty. Standard deviations, in turn, describe variation among observations. Keeping these concepts separate helps prevent a group average, a p-value, and an interval from being treated as interchangeable evidence.

Finally, the registry offers planning illustrations derived from observed contrasts, distinct from the trial’s original design calculations. Those illustrations are useful for understanding how smaller target effects require larger per-arm samples under stated assumptions, but they do not explain how the completed study was planned. The record’s limits on analysis populations, missing data, and statistical methods are therefore part of the statistical interpretation, not minor details.

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 does not report the original statistical analysis plan, the model specification for the least-squares means, the method used to construct the confidence intervals, or the method for handling missing outcome data. The comparisons available in this analysis are unadjusted summary-statistic calculations, so they cannot substitute for a documented sponsor analysis.

Multiplicity is an important limitation. The record lists multiple groups and secondary outcomes, but it does not report a correction or testing hierarchy. The posted p-values are unadjusted, and the registry record does not make clear how the sponsor interpreted the full family of comparisons.

The analysis population is not fully described. A note says that one participant in the IDP-126 Gel group did not receive study medication and was excluded from the demography population, but the record does not say how that participant was handled for the primary endpoint. Enrollment and group-specific endpoint counts are not interchangeable definitions of the analysis set.

Further limitations include the absence of posted secondary-outcome estimates in the information summarized for this analysis and the lack of direct comparisons among IDP-126 Gel and its component groups. The masking description also differs between the study title, which says double-blind, and the registry classification, which says triple; who was masked is not reported. Finally, adverse-event counts alone do not provide timing, relatedness, event duration, or a complete comparative safety analysis.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Phase 2, Multicenter, Randomized, Double-Blind, Vehicle Controlled, Parallel-Group, Clinical Study in the Treatment of Acne Vulgaris — NCT03170388.
• Registry record last updated 2021-11-10; status last verified 2021-10.
• Results first posted 2021-11-10.
• 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.

21. Record Summary

The registry record describes NCT03170388, a completed phase 2 randomized, multicenter, parallel-group study of IDP-126 Gel, three component groups, and IDP-126 Vehicle Gel for acne vulgaris. The sponsor was Bausch Health Americas, Inc. The registry lists enrollment of 741, a start date of 2017-10-05, and primary completion on 2019-04-23.

The primary endpoint was absolute change in mean inflammatory lesion counts from baseline to Week 12. IDP-126 Gel had a posted group estimate of -29.9 lesions, compared with -19.6 for vehicle; the mean difference was -10.3, with a confidence interval from -13.04 to -7.559 and an unadjusted p-value of <0.001. The record also lists six secondary outcomes and group-level safety counts.

Important analysis details—including the original statistical plan, multiplicity handling, missing-data methods, and full analysis population definitions—are not reported in the registry record.