Clinical Study to Evaluate Dental Plaque: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Study to Evaluate Dental Plaque (NCT00758290), a randomized phase 3 study. The primary outcome is Dental Plaque Index.

2. Clinical Question

3. Trial Design

The registry classifies this as a randomized, double-masked, Phase 3 crossover study. It lists enrollment as 23 and the study status as completed. The start is recorded as 2008-05, with primary completion in 2008-07. The sponsor is Colgate Palmolive. These registry entries describe the study record; they do not establish how participants were randomized, how masking was maintained, or how the crossover periods were managed.

A crossover design means, in general, that participants receive more than one intervention in sequence, so comparisons may use measurements collected from the same people under different treatment conditions. That general feature can make within-person information relevant to the analysis. The registry record does not report the treatment sequence, period structure, washout arrangements, or whether the posted summary statistics preserve the paired nature of the crossover data. It also does not provide arm descriptions.

The study summary states that the research assessed anti-plaque efficacy of prototype toothpastes during a 3-week brushing regimen. The primary outcome is Dental Plaque Index, measured at a 4 Day timeframe. The record does not explain how that timeframe relates to the overall regimen. The posted outcome summaries list 12 observations for Fluoride/Triclosan and 12 for Triclosan/Fluoride, with means and standard deviations for each. Those group summaries do not reveal whether the same participants contributed measurements under both formulations.

The record lists one outcome measure and one analysis. It does not report the original sample-size justification, the planned statistical model, the randomization procedure, the analysis populations, or the handling of period and sequence effects. Those omissions matter when interpreting a crossover study because the design’s paired structure and period-related features may affect how an appropriate treatment comparison is estimated.

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 Dental Plaque Index, measured over 4 Day.

The primary endpoint is Dental Plaque Index, a continuous measure reported in units on a scale. The registry describes plaque units as measured on a scale between 0 to 5, with no plaque equal to 0 and 5 described as 2/3 of the tooth covered in plaque. The reported outcome timeframe is 4 Day. The measure therefore represents plaque coverage using a numerical score, rather than a count of events or a time until an event.

In general, a continuous endpoint is often summarized by a mean, which describes the average score in a group, and a standard deviation, which describes the spread of scores around that average. The registry identifies the summary parameter as a mean and the dispersion measure as a standard deviation. Those summaries give a compact description of the observed score distribution, though they do not show individual values or the shape of the distributions.

A plaque index is used in this study because the stated research question concerns anti-plaque efficacy. A numerical scale permits comparison of average plaque scores between the toothpaste formulations. Here, the posted means are 2.48 for Fluoride/Triclosan and 2.42 for Triclosan/Fluoride. A mean difference can describe the direction and size of the comparison, while an interval around that estimate indicates its precision. The record does not report further details about the scale’s measurement procedures, assessor training, or validation.

7. Results Posted in the Registry

• Triclosan/Fluoride: mean 2.42 versus 2.48 on control; difference -0.06 (95% CI -0.35 to 0.23), standardized effect d = -0.16; z-test p = 0.688.

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods help distinguish the observed group summaries from the uncertainty and assumptions involved in comparing them. The registry identifies a continuous primary outcome summarized by a mean and standard deviation, and it lists an ANOVA as the sponsor’s analysis method. A separate simple recalculation uses the posted group summaries to estimate an unadjusted mean difference and an interval. The record does not report the sponsor’s full statistical analysis plan, so these results should not be treated as though they came from the same fully specified model.

What does a mean difference describe?

A mean difference subtracts one group mean from another according to a chosen direction. In the posted comparison, the direction is Triclosan/Fluoride relative to Fluoride/Triclosan. The recalculated estimate is -0.06. A negative estimate means that the first group’s reported mean is lower than the comparator’s mean. The estimate is in the same units as Dental Plaque Index. It describes the observed contrast, not a percentage change and not a causal effect by itself.

The two posted means are 2.42 and 2.48. Each group summary also includes a standard deviation: 0.39 for Triclosan/Fluoride and 0.34 for Fluoride/Triclosan. A standard deviation characterizes variability among measurements around the group mean. It is not the uncertainty in the estimated mean difference. The standard errors listed for the group means, 0.113 and 0.098, address uncertainty in those means under the calculation used, but they do not give the complete crossover analysis.

What does an interval communicate?

In general, a confidence interval gives a range of values compatible with the data and a statistical model at a stated confidence level. Wider intervals indicate less precision, while narrower intervals indicate greater precision, conditional on the assumptions used to calculate them. The recalculated interval for the mean difference extends from -0.353 to 0.233. Because this range crosses zero, the summary-statistic recalculation allows both a lower and a higher mean for Triclosan/Fluoride relative to the comparator. The registry lists a 95% confidence level for the sponsor’s ANOVA analysis, but it does not provide that analysis’s interval endpoints.

Intervals are not guarantees that a true value lies within a particular range, and they do not directly describe the probability that a treatment has a particular effect. Their interpretation depends on the model and design assumptions. In a crossover study, within-participant correlation can influence uncertainty. The registry record does not include the individual paired measurements or enough model detail to determine how that correlation was handled by the sponsor analysis.

Why can a p-value differ between analyses?

A p-value summarizes how unusual the observed data, or data more extreme, would be under a specified null model and its assumptions. It is not an effect size, a measure of clinical importance, or the probability that the null hypothesis is true. The simple unadjusted recalculation gives p = 0.688, whereas the sponsor’s listed ANOVA gives p = 0.05. The discrepancy flag indicates that the values do not agree. Since the sponsor’s underlying model details and effect estimate are not reported in the registry record, the reason for the difference cannot be determined from the posted information.

ANOVA is a general method for comparing means through a model that partitions variation among terms. Which terms are included matters. In a crossover study, a model could potentially account for design features such as period or sequence, but the registry does not report whether these were included. It would be inappropriate to assume a particular model specification or to attribute the discrepancy to a specific feature.

Why can over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a statistical model assumes. If a model understates variability, uncertainty can be understated and inferential results can appear more decisive than warranted. The outcome in this study is continuous and the record reports standard deviations, but it does not report an over-dispersion assessment or a model diagnostic. Therefore, no conclusion about over-dispersion in this analysis is available from the registry record.

What is the role of the crossover structure?

In general, a crossover design can use each participant as their own comparison, potentially accounting for stable differences between participants. That benefit depends on the measurements and analysis retaining the pairing and on the design assumptions being suitable. Period effects, sequence effects, or carryover can also be relevant considerations in crossover analyses. The registry record does not report the sequence allocation, period-specific results, washout details, or the sponsor’s handling of these design features. The posted arm-level means and standard deviations cannot reconstruct a paired analysis because the association between measurements within participants is not reported.

These general principles explain why a summary-statistic recalculation and a sponsor analysis may differ, but they do not identify what happened in this study. The recalculation is explicitly unadjusted and uses the posted summaries. It should be read as a simple comparison, not as a replacement for an analysis that accounts for the crossover design. The sponsor analysis is the relevant sponsor-reported result, but its interpretation remains limited by missing model details and the absence of an effect estimate and interval.

10. Confidence Intervals and Precision

The recalculated mean difference is -0.06, with an interval from -0.353 to 0.233. The interval spans zero and includes values on either side of it. Thus, the simple comparison does not locate the difference precisely enough to settle its direction. The interval is also expressed in Dental Plaque Index units, so its limits should be understood on the scale described in the registry rather than as relative changes.

Precision concerns the range of effects compatible with the data and analysis, not whether a result is important. The group means are 2.48 for Fluoride/Triclosan and 2.42 for Triclosan/Fluoride; the reported standard deviations are 0.34 and 0.39. The posted summaries contain limited information for assessing the distribution of scores or the paired relationship in this crossover study. The registry record does not report the individual participant observations, so it is not possible to evaluate the shape of the score distributions or reconstruct within-participant variation.

The sponsor’s ANOVA result is listed with a 95% confidence level, but its confidence-interval endpoints and effect estimate are not reported in the registry record. This leaves the sponsor result without a corresponding displayed range for assessing precision. The sponsor’s p-value of 0.05 and the recalculated p-value of 0.688 should not be combined or treated as interchangeable: they come from results that do not agree, and the sponsor’s model details are unavailable. Because the discrepancy is flagged, the sponsor’s own analysis takes precedence for the sponsor-reported inference, while the recalculated interval remains a separate, unadjusted description of uncertainty from the posted summaries.

In general, greater precision would mean a narrower interval around an estimate. The available interval indicates uncertainty around the recalculated mean difference, but the registry does not state what size of difference would be meaningful. Statistical precision therefore cannot, on its own, establish practical importance.

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 display gives planning illustrations in participants per arm, not total events. Its assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record. These figures are planning illustrations, not observed power.

The row for planning for 100% of the observed effect gives 584 participants per arm for 80% power and 782 participants per arm for 90% power. The row for planning for 75% of the observed effect gives 1,038 participants per arm for 80% power and 1,390 participants per arm for 90% power. The row for planning for 50% of the observed effect gives 2,335 participants per arm for 80% power and 3,126 participants per arm for 90% power. These values are based on the observed effect used by the calculation and should not be read as the study’s enrollment target.

In plain terms, the 80% power column gives the planned participants per arm under the stated assumptions for a design intended to have an 80% chance of detecting the selected effect if that effect is real and the assumptions hold. The 90% power column describes a higher target chance under those same assumptions, which requires more participants per arm. The display also illustrates that smaller effects need more participants: planning for a smaller fraction of the observed effect produces larger sample-size estimates. The figures do not establish that the completed study had adequate power or that its result is conclusive.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.060; pooled standard deviation σ = 0.366.
2. n = 2 × (1.96 + 0.842)² × 0.366² / -0.060² = 583.7, rounded up to 584 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.0600000000000001                      # observed difference in means
sigma <- 0.365855162598534                     # 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(758290)
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 = 584 (80% power), 782 (90% power)
 75% of observed effect: n per arm = 1,038 (80% power), 1,390 (90% power)
 50% of observed effect: n per arm = 2,335 (80% power), 3,126 (90% power)
power.t.test: 584.6 
Simulated power at n = 584 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The registry lists the sponsor’s analysis as ANOVA, with a p-value of 0.05 and a stated confidence level of 95%. It does not report an effect estimate, confidence-interval endpoints, or details of the model used. In particular, the registry record does not describe how the analysis handled the crossover structure. The p-value therefore gives only a partial account of the sponsor’s result.

The simple recalculation from the posted group summaries gives a mean difference of -0.06, an interval from -0.353 to 0.233, and an unadjusted p-value of 0.688. These are not the same result as the sponsor’s listed ANOVA p-value. The discrepancy flag is true, so the sponsor’s own analysis takes precedence when reporting the sponsor’s inference. The recalculation should be presented separately as an unadjusted re-analysis, not as a correction or replacement for the sponsor’s result.

The registry does not provide sufficient detail to explain why the results differ. It would be speculative to assign the difference to a particular model term, crossover feature, or data-handling decision. Without the sponsor’s estimate and interval, readers cannot compare the size and precision of the sponsor’s estimated effect directly with the recalculated one. The posted results support reporting both analyses with their limitations, while keeping their distinct methods and conclusions clear.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and one analysis, with Dental Plaque Index identified as the primary outcome. It also records zero secondary outcomes posted. These entries describe what appears in the registry record; they do not establish that no other outcomes were measured during the study or that no additional analyses were conducted. Any broader outcome inventory or analysis history is not reported in the registry record.

Multiplicity arises, in general, when several outcomes, comparisons, time points, or analytical choices are considered. Testing multiple questions can increase the chance of finding an apparently notable result by chance unless the analysis plan accounts for that. The record does not report a multiplicity strategy, a prespecified hierarchy of tests, or whether any adjustment was planned. Since the registry lists only one outcome measure and one analysis, the information available does not support a fuller assessment of multiplicity.

The record also does not report missing-data counts, reasons for missingness, imputation methods, or sensitivity analyses. Although enrollment is listed as 23 and the outcome summaries list 12 observations in each group, the record does not explain how these counts relate, whether the same participants contributed under both formulations, or whether any observations were excluded. It is therefore not possible to determine the amount or handling of missing data from these summaries.

The analysis population is not defined in the registry record. It does not say whether the analysis included all randomized participants, only participants who completed the study, or another set. Nor does it report a per-protocol or intention-to-treat analysis. These omissions constrain interpretation because population selection and missing-data handling can affect the estimated comparison. They also limit the ability to assess how closely the posted results reflect the randomized crossover study as a whole.

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

The posted data describe average Dental Plaque Index scores for two toothpaste formulations at the recorded 4 Day timeframe. The means are 2.48 for Fluoride/Triclosan and 2.42 for Triclosan/Fluoride. The simple unadjusted recalculation estimates a mean difference of -0.06 for Triclosan/Fluoride relative to Fluoride/Triclosan. This is a description of the posted comparison, not a statement that either formulation works or does not work.

The recalculated interval, -0.353 to 0.233, includes zero and covers both negative and positive differences. This indicates uncertainty about the direction and size of the mean difference under the recalculation’s assumptions. It does not prove there is no difference, and it does not establish equivalence. Equivalence would require an appropriately designed and analyzed assessment against a prespecified margin, which is not reported in the registry record.

The sponsor’s ANOVA result is listed as p = 0.05, while the recalculated unadjusted p-value is 0.688. Because the results do not agree and the sponsor’s analysis details are absent, the two values should be described as distinct analyses rather than reconciled by assumption. The sponsor’s own analysis takes precedence for the sponsor-reported inference. The absence of the sponsor’s effect estimate and interval means its result cannot be interpreted for magnitude and precision from the registry entry alone.

The outcome scale provides a measure of plaque coverage, but the record does not report a threshold for a clinically meaningful difference. The statistical results therefore do not establish whether the observed contrast is important in practical terms. The crossover classification also matters: without details on the treatment sequences, periods, paired measurements, and model specification, the posted group summaries cannot fully show how the design was incorporated into the analysis.

In short, the record supports a description of the observed means and two nonmatching inferential results, each with limitations. It does not establish treatment equivalence, clinical importance, or a definitive treatment effect. Further details about the sponsor model, participant-level crossover data, outcome handling, and analysis population are not reported in the registry record.

16. Why This Trial Matters Statistically

This study illustrates why an estimated difference, its uncertainty, and the analysis method all matter. The group means are close, but an observed contrast alone does not show how precisely the underlying difference has been estimated. The recalculated interval crosses zero, while the sponsor’s ANOVA is reported with a different p-value. Reading only one number would conceal the disagreement and the incomplete information about the sponsor’s analysis.

The crossover classification makes the distinction between a group-level summary comparison and a design-aware analysis especially relevant. In general, a crossover can contain within-participant information that is not represented by separate means and standard deviations alone. The posted summaries do not report the pairing or the period-specific data. The registry also does not describe the ANOVA model, so readers cannot determine how the design was accounted for.

The record is also useful for distinguishing statistical evidence from practical importance. The mean difference is expressed on the plaque-index scale, but no meaningful-difference threshold is reported. A p-value cannot supply that missing clinical context, and a confidence interval cannot determine importance without a relevant benchmark.

Finally, the planning illustration shows how sample-size requirements depend on the effect chosen for planning and the desired power. It is separate from evidence about whether this completed trial was adequately planned or powered, because the original design assumptions are not reported. Together, these issues demonstrate why clinical-trial interpretation requires attention to design, estimation, uncertainty, and reporting completeness rather than reliance on a single result.

17. Limitations

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

The registry record leaves important design details unspecified. Although the study is classified as randomized, double-masked, and crossover, it does not report the randomization procedure, treatment sequences, period structure, washout arrangements, or how masking was maintained. Arm descriptions are blank. These gaps limit assessment of whether the study’s design features were handled appropriately in the analysis.

The participant and outcome counts are not reconciled. Enrollment is listed as 23, while the outcome summaries report 12 observations for each group. The record does not state whether these are paired measurements from the same participants, explain any exclusions, or describe missing observations. It also does not define the analysis population.

The primary outcome is described as Dental Plaque Index on a scale between 0 to 5, with 5 described as 2/3 of the tooth covered in plaque. Further measurement details, including assessor procedures and scale validation, are not reported in the registry record. The timeframe is listed as 4 Day, while the study summary describes a 3-week brushing regimen; the relationship between those time descriptions is not explained.

The sponsor’s ANOVA is reported with p = 0.05 but without an effect estimate, interval endpoints, or model details. The unadjusted recalculation differs, and the registry does not explain the discrepancy. Information on the original sample-size rationale, interim analyses, missing-data handling, multiplicity, and sensitivity analyses is not reported in the registry record. These limitations restrict assessment of the robustness, precision, and practical importance of the findings.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Clinical Study to Evaluate Dental Plaque — NCT00758290.
• Registry record last updated 2008-11-25; status last verified 2008-11.
• Results first posted 2008-11-25.
• 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 completed Phase 3 study is recorded as randomized, double-masked, and crossover, with Colgate Palmolive listed as sponsor. The registry dates are 2008-05 for study start and 2008-07 for primary completion. Its summary describes a 3-week brushing regimen using prototype toothpastes.

The primary endpoint is Dental Plaque Index at a 4 Day timeframe. Posted means are 2.48 for Fluoride/Triclosan and 2.42 for Triclosan/Fluoride. The simple recalculation gives a mean difference of -0.06, an interval from -0.353 to 0.233, and an unadjusted p-value of 0.688. The sponsor’s ANOVA is listed with p = 0.05, but its effect estimate and interval endpoints are not reported. The discrepancy is flagged, so the sponsor’s own analysis takes precedence for its reported inference.

Missing design and analysis details—including crossover-model specification, analysis population, and missing-data handling—limit further interpretation. The posted summaries do not establish equivalence or clinical importance.