Study of Benzoyl Peroxide Cream for Mild to Moderate Acne Vulgaris: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Benzoyl Peroxide Cream for Mild to Moderate Acne Vulgaris (NCT00558831), a randomized na study. The primary outcome is Subject Reported Change From Baseline Scale.

2. Clinical Question

The clinical question is framed as a comparison of two topical regimens within a pilot study. Its paired, split-face title indicates that the two approaches were intended to be compared on different sides of the face. That design can make within-person contrasts central: each participant may contribute information about both approaches. The registry record does not report the specific allocation procedure for sides, whether both sides were evaluable for every participant, or how the paired observations were analyzed.

The primary measure is a subject-reported change category, not a direct count of lesions or an independently assessed clinical rating. The categories range from worse to clear, with intermediate options for unchanged and degrees of improvement. The registry record does not report how many participants selected each category for either approach. Therefore, the stated question can be described, but the comparative answer cannot be recovered from the posted primary outcome information.

The population, treatment comparison, and timeframe should be read together. This is a small pilot record focused on a one-month assessment and a subjective change scale. It does not report the eligibility criteria, baseline severity distribution, or a detailed treatment protocol. Those details would help readers judge how the question was operationalized and which patients the findings might represent, but they are not reported in the registry record.

3. Trial Design

The study is titled as a split-face, paired-comparison pilot study of benzoyl peroxide 2.5% cream alone versus benzoyl peroxide 2.5% cream plus moisturizing lotion. It is listed as randomized, with no masking. The registry record classifies the model as single-group, while the title describes a paired comparison between two approaches. These registry descriptions are not identical, so the design should be understood from the available wording rather than by assuming details of the protocol or analysis.

The record lists 11 participants enrolled. It identifies acne vulgaris as the condition and lists Northwestern University as the sponsor. The study status is completed. Its listed phases are “NA.” The primary outcome timeframe is baseline and 1 month. The record does not give the treatment duration beyond that outcome timeframe, application frequency, adherence procedures, or the timing and methods of randomization.

In a split-face study, the design generally aims to compare approaches within the same participant, rather than treating each side as if it came from a different person. This is a general design explanation, not confirmation of the precise analysis used. Paired designs can account for characteristics shared by both sides of a participant, but they require analysis that preserves the pairing. The registry record does not report which side received which treatment, how side assignment was randomized, or whether the statistical analysis accounted for the paired structure.

The registry record lists one outcome measure and zero analyses. It does not report a statistical analysis plan, sample-size justification, interim analyses, or a protocol-level explanation of the design. The posted outcome is a multi-category change scale, and the registry information indicates that it was not reduced to one comparison because it has several categories or time points. This means the design can be described, but the intended primary statistical contrast and its implementation are not available in the record.

The study is explicitly described as a pilot. That label gives context for the study’s scale, but the registry record does not explain the pilot objectives or whether the study was intended to estimate feasibility, treatment effects, or both. The sample-size section accompanying this page is a planning illustration based on an observed effect; it is not the trial’s own design justification or evidence of observed power.

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

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.

6. Primary Endpoint

The registry lists the primary outcome as Subject Reported Change From Baseline Scale, measured over baseline and 1 month.

The primary endpoint is titled “Subject Reported Change From Baseline Scale.” It measures participants’ own assessment of change, with the timeframe recorded as baseline and 1 month. The response categories are -1 for worse, 0 for unchanged, 1 for mild improvement, 2 for moderate improvement, and 3 for clear. The unit is participants, and the parameter is listed as a number.

A patient-reported measure can capture how participants experience change, which may not be fully represented by a clinical count or observer rating. This is a general reason such endpoints are used. The scale is ordered: its categories move from worsening through no change to increasing degrees of reported improvement. However, the registry record does not define the instrument’s development, validation, administration, or whether participants reported separately for each side.

The endpoint is not a single binary response. Preserving its categories matters because reducing it to “improved” versus “not improved,” for example, would require a decision about which categories to combine. The registry record does not report such a decision, and the posted information does not give the number of participants in each category for each regimen. Without those results, the endpoint’s intended meaning is clear, but its observed distribution and comparative result cannot be described.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a randomized study, a split-face paired-comparison pilot, and a primary measure with several response categories. It also lists the overall model as single-group and lists zero analyses. These labels do not, by themselves, specify the statistical methods. The discussion below explains relevant concepts generally; it does not claim that the investigators used a particular test or model.

What does pairing mean?

In a paired design, two observations are linked because they come from the same participant. For a split-face comparison, the two sides of a participant can be exposed to different approaches. A within-person contrast can reduce the influence of characteristics shared across the participant’s face, because those shared characteristics do not vary between sides. This is a general rationale for pairing, not a finding about this study.

Pairing also changes the analysis. Treating the two sides as independent observations would ignore their connection and could misstate uncertainty. A paired analysis uses the relationship between the two observations. The registry record does not say how the side-specific responses were recorded or whether the analysis preserved the pairing. It also does not report a statistical method or analysis result, so the particular approach cannot be identified.

How should an ordered response scale be read?

The primary scale records a participant’s report of change using ordered categories: -1=worse, 0=unchanged, 1=mild improvement, 2=moderate improvement, and 3=clear. The order matters: the categories represent a progression in reported change. They are not necessarily equally spaced measurements. For example, moving from unchanged to mild improvement need not represent the same amount of change as moving from moderate improvement to clear. That distinction is general and cautions against automatically treating category labels as a continuous numeric measurement.

Analyses of an ordered outcome can use the full category distribution or can summarize it in a prespecified way. Each choice answers a somewhat different question. Combining categories may simplify a comparison, but it discards distinctions and depends on a chosen threshold. The registry record does not report a category-collapsing rule or a model for the ordinal responses. Its note that the outcome was not reduced to one comparison because it has several categories or time points reinforces that there is no single reported primary contrast to interpret.

What would an estimate and confidence interval communicate?

In general, an effect estimate summarizes the observed contrast under a defined analysis, while a confidence interval describes the uncertainty around that estimate under the method’s assumptions. A confidence interval is not a guarantee that the true effect lies within the interval, and its interpretation depends on the design, model, and data. For a paired outcome, the estimate and interval should reflect the paired structure. The registry record does not report an estimate or confidence interval for the primary endpoint, so there is no interval width or direction to evaluate.

A p-value, when reported, addresses how compatible data are with a specified null model under stated assumptions. It does not measure clinical importance, and it does not establish that an effect is absent when a result is not statistically significant. Here, no primary p-value or analysis is listed in the registry record. It would be inappropriate to infer one from the sample size or from the safety counts.

Why might the category distribution matter?

A single summary can conceal different response patterns. Two approaches might have similar numbers of participants with any improvement but differ in the degrees of improvement reported, or they might have similar averages while having different distributions. These are general possibilities, not claims about this trial. With the posted outcome counts absent, neither the overall distribution nor such patterns can be examined.

Small studies can yield imprecise estimates, and paired data can be especially sensitive to missing observations if one side or one response is unavailable. These are general statistical considerations. The registry record does not report missing outcome counts, an imputation method, or a sensitivity analysis. The enrolled count alone does not establish how many participants contributed to a specific analysis.

What does the planning illustration represent?

The sample-size section is a planning illustration based on an observed effect, with assumptions of a two-sided significance level of 0.05, equal allocation, and a normal approximation. It describes how a hypothetical study might be planned to detect a stated fraction of an observed effect. It is not the trial’s original design calculation, a retrospective measure of power, or evidence that an effect was observed in a particular direction. The registry record does not report the trial’s own sample-size justification.

Planning illustrations depend on assumptions about the outcome, analysis, effect size, and allocation. If the analysis is paired, the relation between paired observations can matter for planning; the record does not report assumptions for such a calculation. The illustration should therefore be read only as a planning aid and not as a claim about the adequacy of the completed pilot or the certainty of its primary findings.

Finally, multiple categories or time points can raise multiplicity questions if several contrasts are tested. Multiplicity refers generally to the increased chance of finding an apparently notable result when many hypotheses are examined without accounting for the number of tests. The registry record does not report an analysis plan, a hierarchy of comparisons, or multiplicity adjustment. Since the primary results and tests are not posted, the extent of any multiplicity issue cannot be assessed.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint includes multiple ordered response categories, and the registry note says it was not reduced to one comparison because it has several categories or time points. The record does not report which comparisons were planned, whether the categories were analyzed separately, or whether any multiplicity adjustment was used. In general, testing several outcomes or contrasts can increase the chance of a seemingly notable result; a prespecified analysis plan helps clarify how such testing is handled. No such plan is reported in the registry record.

Missing-data handling is also not reported. The record gives 11 participants enrolled, but it does not state how many contributed primary outcome responses, whether any responses were missing, or how missing values were treated. In a split-face design, it would also matter whether observations were available for both sides of each participant. These details are not reported in the registry record, so the enrolled count should not be read as the number analyzed.

The record does not specify an analysis population, such as whether the analysis followed randomized assignment or included only participants with complete observations. It lists zero analyses, and no statistical analysis plan is reported. Consequently, readers cannot determine which participants were included, how the paired structure was handled, or whether the analysis was selected before results were reviewed. These are reporting gaps, not evidence that a particular method was or was not used outside the registry record.

The absence of these details limits reproducibility and makes it difficult to distinguish planned analyses from post hoc choices. It also prevents assessment of whether the outcome categories, time points, and safety measures were treated as one primary family or as separate questions. No conclusions about the direction or statistical significance of the primary result should be drawn without the unreported analysis and outcome data.

11. 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 record lists serious adverse events separately from other adverse events. For benzoyl peroxide 2.5% cream, 0 participants were affected by a serious adverse event among 11 at risk; for benzoyl peroxide 2.5% cream plus moisturizing lotion, the serious-event count was also 0 among 11 at risk. The record lists other adverse events in 11 of 11 at risk for cream alone and 6 of 11 at risk for the cream-plus-lotion approach.

These figures are counts of participants, not event rates calculated for a broader population. The record does not identify the specific events in the “other” category, their severity, timing, duration, or relationship to treatment. It also does not explain the category definition. The death field is not reported, so it should not be interpreted as a reported count of zero deaths.

Zero serious events in these listed groups means none were recorded among the participants at risk in this registry entry; it does not establish that serious events cannot occur. The small pilot setting and the absence of detailed event descriptions limit what can be inferred. These numbers describe the posted record and are not a comprehensive safety assessment.

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

This record is useful for identifying the comparison the study set out to make: benzoyl peroxide 2.5% cream alone versus the same cream with moisturizing lotion, described as a randomized split-face paired-comparison pilot for mild to moderate acne vulgaris. It also identifies a patient-reported change scale and a baseline-to-1-month timeframe. Those details help define the clinical and statistical question.

The record does not provide the primary outcome-category counts, an effect estimate, a confidence interval, or a primary analysis. As a result, the posted information does not show how participants rated change under either approach or whether their reported changes differed. The study title and endpoint description cannot substitute for outcome results. The record’s safety counts are available, but they do not answer the primary efficacy question.

Nor does the record establish the size or precision of any treatment effect. A small pilot can be useful for generating information, but without the primary results and analysis details, the size and uncertainty of any comparison cannot be assessed. A sample-size planning illustration is not observed power and does not fill the result gap.

The design labels warrant careful reading. “Randomized” and “split-face, paired-comparison” describe important features, while the registry also labels the model as single-group. The record does not report how the pairing was analyzed or how the sides were assigned. Statistical conclusions require those details, particularly when outcomes from two sides of one participant are related.

Accordingly, the defensible interpretation is limited: the registry describes a completed pilot study with 11 participants enrolled, a one-month patient-reported endpoint, and posted adverse-event counts. It does not provide enough primary outcome information to judge the comparative result. This is not a recommendation about treatment and does not establish that either approach is better or worse.

13. Why This Trial Matters Statistically

This study illustrates a recurring issue in clinical trial interpretation: a clear research question and a named endpoint are not enough to evaluate a result if the outcome data and analysis are not posted. Here, the registry identifies a randomized split-face comparison and a patient-reported change scale, but it does not give the category counts or a primary analysis. Readers can understand what was intended to be measured without being able to assess the central comparison.

The paired design also highlights why the structure of the data matters. When two sides of the same participant receive different approaches, the observations are linked. Analyses that recognize this pairing answer a different statistical question from analyses that treat every side as independent. The record’s split-face title, randomized allocation, and single-group model label make clear why transparent reporting of the allocation and analysis would be especially useful.

The endpoint’s ordered categories matter as well. A scale from worse through clear preserves distinctions in participants’ reported experiences. Reducing an ordered outcome to a simpler summary can be reasonable when planned and justified, but it changes the question. The registry note indicates that the outcome was not reduced to one comparison, while the category results themselves are absent.

Finally, the record demonstrates the difference between safety counts, efficacy results, and planning calculations. The adverse-event figures are descriptive, and the sample-size section is a planning illustration. Neither answers the missing primary comparison. Keeping those types of information separate helps avoid overstating what a pilot record can establish and makes reporting gaps visible.

14. 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 most important limitation for interpreting this particular record is that primary outcome results are not reported in the registry record. Although the endpoint’s response categories and timeframe are described, the record does not give the number of participants in each category for either approach, an effect estimate, or an analysis result. The comparative primary question therefore remains unanswered by the posted information.

The sample is small: 11 participants are listed as enrolled. This limits the amount of information available, but the registry record does not report how many participants completed the primary measure or contributed paired observations. Enrollment should not be treated as the analyzed sample size. Eligibility criteria, baseline characteristics, and details of treatment exposure are also not reported in the registry record.

Several design and analysis details are missing. The record describes randomization and a split-face paired comparison, but does not state how sides were assigned, whether masking was used beyond the listed “none,” or how paired outcomes were analyzed. Its model is listed as single-group. The statistical analysis plan, sample-size justification, missing-data methods, and any interim analyses are not reported in the registry record.

The primary measure is subjective and uses ordered categories, but the record does not describe the instrument’s validation, administration, or whether participants assessed each side independently. The reported timeframe is baseline and 1 month; longer-term results are not described. The safety listing gives counts but not the specific other adverse events, their severity, timing, or attribution. These limitations constrain both statistical interpretation and generalizability, without establishing that unreported procedures or results did not exist elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study of Benzoyl Peroxide Cream for Mild to Moderate Acne Vulgaris — NCT00558831.
• Registry record last updated 2017-09-26; status last verified 2017-08.
• Results first posted 2010-11-22.
• 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.

18. Record Summary

The registry record describes a completed randomized pilot study titled as a split-face, paired comparison of benzoyl peroxide 2.5% cream alone versus the cream plus moisturizing lotion for mild to moderate acne vulgaris. It lists 11 participants enrolled, no masking, and an overall model classified as single-group.

The primary measure is a subject-reported change-from-baseline scale with categories from worse to clear, assessed at baseline and 1 month. The registry record does not report the category counts or a primary analysis. It lists zero analyses and does not report the statistical plan, missing-data approach, or sample-size justification.

Serious adverse events are listed as 0 among 11 at risk in each group. Other adverse events are listed as affecting 11 of 11 at risk with cream alone and 6 of 11 at risk with cream plus lotion. Specific event details and death counts are not reported in the registry record. The posted information describes the study question and selected safety counts, but does not establish the primary comparative result.