Analysis of the Effect of Antimicrobial Soap on Bacterial Survival: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Analysis of the Effect of Antimicrobial Soap on Bacterial Survival (NCT02012348), a non-randomized na study. The primary outcome is Relative GAS Survival.

2. Clinical Question

The biological measurement is local and short-term: a skin swab is described as being collected 30 minutes after bacteria application and soap treatment. It is not a measure of clinical infection, illness, or longer-term effects. The registry labels the study nonrandomized and unmasked. It therefore does not describe a randomized, blinded comparison. The record does not report outcome values or an analysis comparing the soaps, so it is not possible to estimate the size or precision of any difference between them.

3. Trial Design

NCT02012348 is a completed, nonrandomized, parallel-model study sponsored by the University of California, San Diego. The registry records enrollment of 30 and lists three soap groups: control soap, antibacterial soap with triclocarban, and antibacterial soap with benzalkonium chloride. Masking is listed as none. The phase is listed as not applicable.

The brief summary describes applying an avirulent strain of S. pyogenes to forearms after washing them with different soaps, followed by bacterial sampling and cleansing. It says the investigator expected fewer bacteria to survive on skin washed with antimicrobial soap than on skin washed with control soap. This is a description of the proposed question and expectation, not evidence that the expected pattern was observed.

There is a design-description ambiguity. The registry lists a parallel model and group-specific arm descriptions, but the brief summary describes washing each of a subject’s forearms with different soap types. The record does not explain the allocation of soaps to people or forearms, whether observations were paired within a person, or how the analysis would account for such pairing. Allocation is explicitly nonrandomized, and the record flags that the arms are not a like-for-like control comparison. These features limit causal interpretation: differences could reflect the way people or forearms entered each group as well as the soap used.

The brief summary says the study would recruit up to 10 healthy volunteers, whereas enrollment is listed as 30. The record does not clarify whether the former is an earlier plan, applies to a subset, or has another explanation. Start and primary-completion dates are listed as 2013-12 and 2015-07, and the status is completed. Actual participant flow, the number contributing endpoint measurements, and exclusions 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.

Overall Study

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 Relative GAS Survival, measured over 30 minutes post-bacteria application.

The primary endpoint is Relative GAS Survival. It is described as the abundance of Group A Streptococcus relative to the abundance of other bacterial species isolated from a skin swab. Its unit is percentage overall bacterial abundance, and the summary measure is a mean with standard error. The stated timeframe is 30 minutes after bacteria application.

This endpoint expresses the target organism’s representation within the bacteria recovered from a swab, rather than simply reporting a clinical outcome. In general, a relative-abundance measure is compositional: its value can reflect changes in the target organism, in the other organisms used as the reference, or in both. It should therefore be interpreted as the defined relative measure, not automatically as an absolute count of viable GAS. The registry does not report an absolute bacterial count or additional details about how the relative percentage was calculated.

The endpoint is relevant to the study’s stated laboratory question because it measures bacterial survival on treated skin at a specified short-term sampling point. It does not, by itself, establish whether a person develops an infection or experiences a health benefit. Although the registry identifies this measure as primary, it also lists Relative GAS Survival among secondary outcomes, and records two outcome measures overall. The status and definition of the other outcome measure are not reported in the registry record.

No endpoint mean, standard error, group contrast, or confidence interval is reported in the registry record. Consequently, the posted information does not establish the observed relative survival in any soap group or the precision of a comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean describe?

The registry identifies the endpoint as continuous and specifies a mean as its summary measure. In general, a mean represents the average measured value in a group. For this study, that value would represent the group’s relative GAS survival as defined by the swab measurement. A group mean alone would not describe the spread of individual measurements, and it would not quantify a difference between soap groups.

What does the standard error describe?

The dispersion measure named for the endpoint is standard error. In general, a standard error describes the uncertainty of an estimated mean due to sampling variation, under the assumptions of the analysis. It is not the same as the spread of individual participant measurements. A standard error can help convey how precisely a mean is estimated, but it needs to be interpreted alongside the estimate and the analysis design. The registry does not post a standard error value for any group.

How would a comparison be interpreted?

In general, a comparison between two means can be expressed as a difference, with a confidence interval describing a range of values compatible with the data and analysis assumptions. A confidence interval communicates precision as well as the estimated direction and magnitude. A p-value addresses compatibility with a specified null model; it does not measure the size or importance of an effect. None of these comparison results is reported in the registry record, so no conclusion about statistical evidence or precision can be drawn.

The study lists three soap groups, which could support more than one pairwise comparison. The record does not report which comparisons were planned, whether a single overall test was intended, or whether results would be adjusted for making multiple comparisons. It also does not report a statistical analysis plan. It would be inappropriate to assume a particular test or adjustment method.

Why does allocation matter?

In general, random allocation helps balance measured and unmeasured participant characteristics across groups on average. This study is explicitly listed as nonrandomized, and the registry notes that its arms are not a like-for-like control comparison. A difference between groups, if one were reported, could therefore reflect differences in the groups as well as a soap effect. The record does not describe a method for addressing confounding.

The description of using different soaps on each forearm also raises an analysis question. In general, paired observations from the same participant are correlated, because they share characteristics. Treating paired measurements as independent can misrepresent uncertainty. Yet the record lists a parallel model and does not say whether participants contributed one or multiple measurements, how the soap was assigned to each forearm, or whether within-person pairing was incorporated. No analysis method can be inferred from the design description alone.

What do the planning assumptions mean?

The registry states planning assumptions for a sample-size illustration: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also says the effect to detect is a stated fraction of the observed effect. These are assumptions for a planning illustration, not evidence about the study’s observed power. The observed effect is not reported in the registry record, and the record does not identify which two groups underpin the illustration. The registry does not report its own sample-size justification, so the illustration should not be mistaken for the original design rationale.

In general, a sample-size calculation depends on the outcome variability, the effect size considered relevant, the type of comparison, and assumptions about allocation and analysis. A planning illustration can show how required enrollment would change under different assumed effects, but it cannot show the probability that a completed study detected an effect. That probability cannot be reconstructed from the registry information, and no post hoc observed-power claim is warranted.

What is not established by the methods record?

The record does not report a statistical test, model, covariate adjustment, treatment of repeated measures, missing-data method, or analysis population. It also does not report interim analyses. In general, these details matter because they determine how estimates and uncertainty are calculated. Without them, readers cannot evaluate whether the method matched the apparent design, whether assumptions were checked, or whether results were sensitive to analytic choices. Most importantly, no outcome analysis is posted at all; descriptions of possible statistical concepts cannot substitute for the missing estimates.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Control Soap", "Antibacterial Soap With Triclocarban", "Antibacterial Soap + Benzalkonium Chloride")
n   <- c(10, 10, 10)
m   <- c(100, 75, 90)
sp  <- c(15, 20, 21.7)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Control Soap  10  100.000  47.434  66.068  133.932
CARM  Antibacterial Soap With Triclocarban  10  75.000  63.246  29.757  120.243
CARM  Antibacterial Soap + Benzalkonium Chloride  10  90.000  68.621  40.911  139.089

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures overall and names Relative GAS Survival as both the primary endpoint and a secondary outcome. The other outcome measure is not identified in the record. It also lists three soap groups, which makes the intended set of group comparisons relevant to interpretation. The registry does not report which comparisons were planned, whether an overall test would precede pairwise comparisons, or whether any adjustment for multiple testing was planned. No multiplicity procedure is reported, so the risk of drawing a chance-positive conclusion across several outcomes or comparisons cannot be assessed.

Missing-data handling is not reported in the registry record. There are no posted counts for participants with endpoint measurements, missing swabs, unusable measurements, or exclusions from analysis. The enrollment figure of 30 does not establish how many people contributed data to the primary endpoint. The brief summary’s statement that the study would recruit up to 10 healthy volunteers is not reconciled with the listed enrollment.

The analysis population is also not reported. The record does not say whether all enrolled participants, only participants with a measured swab, or another group would be included in the primary analysis. It does not describe how deviations from the planned soap treatment or incomplete sampling would be handled. These omissions matter because estimates based only on available measurements can be affected if the reasons for missingness relate to the outcome or treatment. No conclusions about completeness, exclusions, or bias from missing data can be drawn from the registry record.

12. 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 serious adverse event record lists 0 serious events among 10 participants at risk in each of the three soap groups: Control Soap, Antibacterial Soap With Triclocarban, and Antibacterial Soap + Benzalkonium Chloride. It also lists 0 deaths in each group. The other adverse-event counts are listed as 0 among 10 at risk in each group.

These are recorded counts, not comparative estimates of risk. The record does not provide definitions of the adverse-event collection period, event ascertainment, or the meaning of the category labeled “other.” It does not explain how the safety counts relate to the enrollment of 30 or to the brief summary’s description of up to 10 volunteers. The figures therefore should be read as the counts posted in the registry, not as proof that no unrecorded events occurred or as a basis for comparing safety between soaps. No rates, uncertainty intervals, or statistical safety comparisons are reported.

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

The registry record describes a study intended to measure relative GAS survival on skin after washing with control soap or either of two antibacterial soaps. It specifies a swab-based endpoint and a stated timeframe of 30 minutes after bacteria application. It also records a nonrandomized allocation and no masking. These details identify the question and important features of the design.

The posted information does not show the result of that question. No group means, variability estimates, treatment contrasts, confidence intervals, or p-values are reported, and the registry lists no analyses. The expectation stated in the brief summary is not an observed result. Therefore, the record cannot establish whether relative GAS survival differed among the soap groups, how large a difference might have been, or how precisely it was estimated.

Even if a group difference were available, the nonrandomized design and the registry’s warning that groups are not like-for-like would require caution in attributing that difference to soap. The description of different soaps on each forearm also leaves open whether observations were paired within participants and how that dependence was handled.

The endpoint is a short-term relative bacterial abundance measurement. In general, it is not equivalent to an absolute bacterial count or a clinical outcome. The record does not report infection outcomes or longer-term follow-up. The results needed to assess the laboratory endpoint are absent, so the study description alone cannot support a conclusion about comparative performance or clinical effects.

14. Why This Trial Matters Statistically

This study illustrates why the design description and the outcome analysis must be read together. The registry identifies three soap conditions and a continuous, mean-based endpoint, but it also identifies nonrandomized allocation and warns that the arms are not a like-for-like control comparison. Those features are central to deciding whether a group contrast could be interpreted causally.

The forearm description creates a further statistical question: if different soaps were used on different forearms of the same participant, measurements might be paired. A parallel-model label does not resolve how the observations were allocated or analyzed. In general, the distinction between paired and independent observations affects the uncertainty assigned to a comparison. The registry does not report the details needed to determine which structure applies.

The record also demonstrates the importance of separating an investigator’s expectation from observed evidence. The stated expectation is that fewer bacteria would survive after antimicrobial soap, but no outcome estimates or analyses are posted. Without those, neither the direction of the observed data nor their precision can be evaluated. Planning assumptions cannot fill that gap; a sample-size illustration is not a result and is not observed power.

Finally, discrepancies and omissions deserve attention rather than silent resolution. The record lists enrollment of 30 while the brief summary says up to 10 volunteers, and it describes 30-minute measurement in the endpoint while the summary mentions several hours of incubation. Identifying these points helps readers understand what can and cannot be inferred from the registry record.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific limitations are apparent from the registry record. First, allocation is nonrandomized, and the record explicitly warns that the arms are not a like-for-like control comparison. No adjustment for differences between groups is reported. A causal interpretation of any contrast would therefore be limited even if estimates were available.

Second, the design descriptions do not fully align. The record calls the model parallel, while the brief summary describes washing each participant’s forearms with different soaps. It does not clarify whether comparisons were within-person, between groups, or both, nor how assignment to forearms was decided. The analysis structure and treatment of within-person dependence are not reported.

Third, the enrollment and timing descriptions are unresolved. Enrollment is listed as 30, while the brief summary says up to 10 healthy volunteers. The primary endpoint timeframe is 30 minutes after bacteria application, while the summary also mentions several hours of incubation. The registry does not explain these differences. Actual endpoint sample sizes and participant flow are not reported.

Fourth, the outcome reporting is incomplete for interpretation. There are two outcome measures listed, but only one is named; Relative GAS Survival appears as both primary and secondary. No outcome results or analyses are posted, and missing-data handling, analysis populations, and multiplicity procedures are not reported. The endpoint’s relative-abundance definition also does not supply an absolute bacterial count. These limitations prevent an assessment of effect size, precision, robustness, and the full analysis plan.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Analysis of the Effect of Antimicrobial Soap on Bacterial Survival — NCT02012348.
• Registry record last updated 2019-10-03; status last verified 2019-09.
• Results first posted 2019-10-03.
• 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.

19. Record Summary

NCT02012348 is a completed, nonrandomized, unmasked, parallel-model study sponsored by the University of California, San Diego. The registry lists enrollment of 30 and three soap groups: control soap, antibacterial soap with triclocarban, and antibacterial soap with benzalkonium chloride. The primary endpoint is Relative GAS Survival, defined as relative abundance of Group A Streptococcus in a skin swab, with a stated measurement time of 30 minutes after bacteria application.

The record reports no outcome analyses and no numerical endpoint results. It lists 0 serious adverse events and 0 deaths among 10 at risk in each group. Important details, including actual endpoint sample sizes, analysis methods, missing-data handling, and reconciliation of differing enrollment and timing descriptions, are not reported in the registry record. The posted information describes the study question but does not establish a comparative result.