Efficacy of Three Experimental Toothpastes to Remove Plaque: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Three Experimental Toothpastes to Remove Plaque (NCT02992691), a randomized na study. The primary outcome is Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (TPI), Positive Control Versus [vs.] Negative Control.

2. Clinical Question

This question is narrow by design: it concerns a measured plaque-score change under the study’s brushing procedure. It does not establish how repeated use would compare, whether participants experienced a lasting difference, or whether the products would have the same performance under different brushing conditions. Those questions are not reported as primary questions in the registry record.

3. Trial Design

The registry describes a randomized, unmasked crossover trial sponsored by GlaxoSmithKline. Its status is completed. The study began on 2016-12-19, and primary completion is listed as 2017-02-02. Enrollment is recorded as 75. The trial is classified as phase NA, and the condition is dental plaque. These labels describe the registry record; they do not by themselves give details about how the trial was conducted.

The crossover model means that the study is organized around treatment comparisons in which participants may contribute observations under more than one treatment condition. However, the registry record does not report the treatment sequence, the number or timing of treatment periods, the washout procedure, or how period and sequence effects were handled. Those details are important for interpreting a crossover design because responses can depend on treatment order or carry over from an earlier treatment. Without the registered details, no specific assumptions about these features can be made.

Participants in each listed arm were instructed to put a full ribbon of dentifrice on a toothbrush and brush once for one timed minute. The study description focuses on plaque removal after a single brushing. Although the intervention list includes five sodium bicarbonate concentrations, the record does not identify which concentration corresponds to each product and control arm. The title refers to three experimental toothpastes, and the arm descriptions name Test Product 1, Test Product 2, Test Product 3, Positive Control, and Negative Control. The record does not report additional product composition details that would support a more specific mapping.

Masking is listed as none. The registry therefore does not describe blinding of participants or study personnel. The primary population was the intent-to-treat population, with eligibility for that analysis defined as randomized participants who received at least one dose and had at least one post-baseline efficacy measurement. Specific randomization procedures, the original design assumptions, the intended treatment sequence, and reasons for differences between enrollment and outcome counts 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.

Before assignment: A total of 100 participants were screened, out of which 25 participants were screening failure. Remaining 75 participants were enrolled in the study. Out of 75 enrolled participants, 6 did not meet eligibility criteria, 3 withdrew their consent, 10 participants, due to other reason (not specified). Remaining 56 were randomized in the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent to treat (ITT) population was the primary population of analysis. The ITT population was defined as those participants who were randomized, received at least one dose of study product and had at least one post-baseline efficacy measurement.

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 Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (TPI), Positive Control Versus [vs.] Negative Control, measured over Up to 5 weeks.

The primary endpoint is change from baseline, meaning the pre-brushing plaque score, to the post-brushing score after a single brushing treatment. A dental examiner used the Turesky Modification of the Quigley Hein Plaque Index to assess plaque on gradable teeth. The overall plaque score was calculated as an average over tooth sites for each participant. Each tooth was divided into six areas, and disclosed plaque on each surface was assigned a score on a scale from 0 to 5. A score of 0 means no plaque; the higher categories represent increasing plaque coverage, with a score of 5 for plaque covering 2/3 or more of the crown. Lower scores indicate less plaque area.

This endpoint is useful for a controlled brushing comparison because it converts visible plaque coverage into a continuous measure that can be summarized as a mean and compared between treatment conditions. Change from baseline focuses on how the measured score differs after treatment relative to the participant’s pre-brushing assessment. It does not directly measure every aspect of oral health, and a score change is not itself a measure of long-term benefit.

The registry lists the unit as a score on a scale, the summary statistic as a mean, the dispersion as a standard deviation, and the timeframe as up to 5 weeks. Those descriptions do not explain the exact timing of each assessment within that timeframe. The primary title specifies the positive-control versus negative-control comparison; the registered endpoint description does not define a separate clinically important difference for the score. Such a threshold is not reported in the registry record.

7. Results Posted in the Registry

• Negative Control: mean -0.54 versus -0.57 on control; difference 0.03 (95% CI -0.06 to 0.12), standardized effect d = 0.12; z-test p = 0.529.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (Test Product 1, 2 and 3 vs Negative Control)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Test Product 1: mean -0.56 versus -0.54 on control; difference -0.02 (95% CI -0.11 to 0.07), standardized effect d = -0.08; z-test p = 0.677.
  • Test Product 2: mean -0.55 versus -0.54 on control; difference -0.01 (95% CI -0.10 to 0.08), standardized effect d = -0.04; z-test p = 0.831.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Test Product 1 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Test Product 2 and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (Test Product 1, 2 and 3 vs Positive Control)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Test Product 1: mean -0.56 versus -0.57 on control; difference 0.01 (95% CI -0.08 to 0.10), standardized effect d = 0.04; z-test p = 0.833.
  • Test Product 2: mean -0.55 versus -0.57 on control; difference 0.02 (95% CI -0.07 to 0.11), standardized effect d = 0.08; z-test p = 0.661.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Test Product 1 and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Test Product 2 and the control. The interval gives the range of effects that remain compatible with the data.

The secondary endpoints extend the plaque-score comparison to the experimental products against control dentifrices. Across those comparisons, the posted mean changes are close to one another, and the unadjusted estimates are small relative to the uncertainty described by their confidence intervals. The overall pattern is therefore broadly consistent with the primary result: the posted summaries do not show a clearly separated pattern of plaque-score changes. That is a description of the data, not a conclusion that the products have identical effects.

The secondary endpoints relate to the primary endpoint because they use the same general outcome framework: change in the Turesky Modification of the Quigley Hein Plaque Index after a single brushing. They add comparisons involving experimental products, whereas the primary comparison concerns positive control versus negative control. They do not answer a different long-term question, and they should not be interpreted as evidence about repeated use or outcomes beyond plaque-score change.

Secondary findings carry less weight when several comparisons are considered without a reported multiplicity strategy. Repeated testing makes it easier for a nominally notable result to arise by chance, so confirmatory interpretation normally depends on a prespecified testing hierarchy or another stated adjustment. Such a strategy is not reported in the registry record. In addition, the registry does not list sponsor-reported inferential analyses for the secondary endpoints. Their posted comparisons are therefore best treated as supporting descriptive evidence, rather than as independent confirmation of a treatment effect.

The registry record does not explain why some displayed secondary summaries and comparisons have different outcome counts, or provide enough detail to resolve all aspects of their analysis populations. The broad pattern should be interpreted alongside the limitations of the crossover analysis and missing-data reporting, not as a replacement for the sponsor’s primary analysis.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (Test Product 1, 2 and 3 vs Negative Control)
• Change From Baseline (Pre-brushing) After a Single Brushing Treatment (Post-brushing) Turesky Modification of Quigley Hein Plaque Index (Test Product 1, 2 and 3 vs Positive Control)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score summarized by a mean and standard deviation. A mean describes the center of the observed scores, while a standard deviation describes how spread out individual observations are around that mean. Neither measure, on its own, shows the uncertainty in a treatment comparison. The confidence interval for a difference is more directly relevant to precision because it gives a range of values compatible with the data and the analysis assumptions.

What does the unadjusted recalculation compare?

The recalculation uses the posted group means and standard deviations to form a simple, unadjusted comparison of the positive control and negative control. Its mean difference is 0.03, and its 95% confidence interval is -0.063 to 0.123. The associated unadjusted p-value is 0.529. “Unadjusted” means this calculation does not account for covariates or other model features beyond the posted summary statistics. It is not a substitute for the sponsor’s analysis, particularly when the original analysis uses a model such as ANCOVA.

The recalculation’s standard error is based on the amount of variation in the outcome and the group sizes used in the summary. A standard error reflects the uncertainty of an estimated mean or contrast, rather than the spread of individual scores. The group standard deviations are 0.247 for the positive control and 0.253 for the negative control. These summaries are useful for describing the data, but a simple calculation cannot reproduce a more detailed analysis if that analysis adjusted for baseline scores, periods, sequence, or other covariates. Which terms the sponsor included in the ANCOVA is not reported in the registry record.

What does the sponsor’s ANCOVA add?

The reported analysis is an ANCOVA, or analysis of covariance, and its result is a net mean difference of -0.01, with a 95% confidence interval from -0.06 to 0.04 and p = 0.667. ANCOVA is a general method for comparing group outcomes while accounting for specified covariates. In a baseline-and-follow-up setting, such a model may be used to account for baseline outcome values, but the registry record does not identify the exact covariates or model specification. It is therefore appropriate to name the reported method without claiming which adjustments were made.

The sponsor’s reported estimate and the unadjusted recalculation are not identical. They are based on different analysis presentations: one is the registry’s reported ANCOVA result, and the other is a simple re-calculation from posted group summaries. A difference between them can arise when analysis methods or data handling differ, but the precise explanation is not reported in the registry record. The posted sponsor analysis should be distinguished from the recalculation rather than blended into one estimate.

How should the confidence interval and p-value be read?

A confidence interval describes uncertainty around an estimate under the assumptions of the method used. A narrower interval indicates greater precision; a wider interval indicates less precision. When an interval for a difference includes zero, the data are compatible with no mean difference as well as with values in either direction within the interval. The interval does not assign probabilities to particular true effects, and it does not establish that the groups are equivalent. The p-value is a measure of how unusual data at least as inconsistent with the null comparison would be under the specified model if the null were true. It is not the probability that the null is true or that a treatment has no effect.

Here, both the unadjusted and sponsor-reported intervals include zero. Their p-values are not small, and the estimates do not establish a clear difference between the positive and negative controls. That interpretation concerns statistical evidence for the comparison, not a decision about clinical importance. No clinically meaningful score difference or equivalence margin is reported in the registry record.

Why can over-dispersion matter in other outcome types?

Over-dispersion is a general issue for count outcomes: it occurs when observed variability exceeds what a simple count model expects. It can make standard errors too small if the model does not allow for the extra variability. It is not the main framework for this primary endpoint, which is a continuous plaque score summarized by means and standard deviations. The registry record does not report an over-dispersion model for this outcome.

What does the crossover label imply, and what remains unknown?

A crossover design can allow each participant to receive more than one treatment condition, so comparisons may need to consider within-participant correlation, treatment period, sequence, and possible carryover. These are general design considerations, not a description of the particular analysis. Although the registry labels the model as crossover, it does not report the treatment sequence, period structure, washout, or how those design features entered the analysis. The sponsor’s ANCOVA label alone does not supply those missing details.

The registry identifies the primary analysis population as intent to treat, defined as randomized participants who received at least one dose and had at least one post-baseline efficacy measurement. This is a registry-specific definition and does not mean that every randomized participant necessarily contributed an outcome to every comparison. Reasons for missing measurements and the handling of missing data are not reported in the registry record.

Why do secondary comparisons need caution?

Several treatment contrasts can increase the chance of finding an apparently unusual result simply through repeated testing. Multiplicity methods can control or otherwise address that concern, but the registry record does not describe a multiplicity adjustment. The posted secondary comparisons therefore should be interpreted as unadjusted evidence, not as confirmatory findings with a stated family-wise error strategy. The record also does not provide sponsor-reported inferential analyses for the secondary endpoints. These limitations do not change the arithmetic of their estimates, but they affect how much evidential weight to place on them.

11. Confidence Intervals and Precision

The primary comparison has two reported forms of uncertainty. The unadjusted recalculation estimates a mean difference of 0.03, with a 95% confidence interval from -0.063 to 0.123. The sponsor’s ANCOVA result is -0.01, with a 95% confidence interval from -0.06 to 0.04. Each interval includes zero and extends into both negative and positive values. The data therefore leave uncertainty about the direction and size of the difference.

A confidence interval is not a range in which the true value is guaranteed to lie, nor does it mean that values inside the interval are equally likely. In general, it represents a range generated by a method that, under repeated use and its assumptions, has the stated coverage. For an individual trial, it is best read as a measure of the precision of the estimate. The sponsor’s interval and the recalculated interval should be kept separate because they arise from different analysis approaches.

The intervals are consistent with a small difference in either direction, but the registry record does not define a threshold for a clinically important plaque-score change. Statistical uncertainty therefore cannot be translated into a judgment about whether any difference would matter clinically. The record also does not report a formal equivalence or non-inferiority margin. An interval containing zero is not proof of equivalence, and a non-significant p-value does not show that two products have identical effects.

Precision depends on the amount of variation, the number of usable observations, and the analysis method. The reported standard deviations and outcome counts help describe the observed data, but they do not resolve uncertainty about crossover handling, missingness, or model specification. The primary record’s confidence intervals are consequently the clearest available summary of uncertainty, subject to the limitations of the reported analysis details.

12. 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 planning figures are illustrations based on the observed effect, not the original trial’s design assumptions. The record specifies a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. Because these are continuous outcomes, the figures are participants per arm, not total events. The original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm that, under those assumptions and for the specified effect, would yield an 80% chance of detecting a difference using the stated significance threshold if that effect were present. The 90% power column uses a higher chance of detection, so its planning requirement is larger. These are planning concepts, not guarantees: actual performance depends on whether the assumptions about effect, variation, allocation, analysis, and missing data are appropriate.

The planning illustrations show estimates for detecting 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller effects require more participants because they are harder to distinguish from random variation. The estimates are therefore sensitive to the observed effect used as their starting point. Where an observed effect is small, planning from that estimate can produce very large participant requirements; such figures should not be read as evidence that the original trial was designed to detect that effect.

The plotted curve is another way to read the same planning relationship: as the target effect becomes smaller, the required participants per arm increase. These calculations are not observed power. They do not describe the probability that the completed trial detected its actual effect, and they do not replace a prospective design calculation based on a justified clinically meaningful target and an appropriate model.

13. 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 = 0.030; pooled standard deviation σ = 0.250.
2. n = 2 × (1.96 + 0.842)² × 0.250² / 0.030² = 1,090.3, rounded up to 1,091 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 <- 0.0299999999999999                      # observed difference in means
sigma <- 0.250017999352047                     # 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(2992691)
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 = 1,091 (80% power), 1,460 (90% power)
 75% of observed effect: n per arm = 1,939 (80% power), 2,595 (90% power)
 50% of observed effect: n per arm = 4,362 (80% power), 5,839 (90% power)
power.t.test: 1091.2 
Simulated power at n = 1,091 per arm: 0.805 (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.

14. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary analysis as an ANCOVA with a net mean difference of -0.01, a 95% confidence interval from -0.06 to 0.04, and p = 0.667. This is the trial’s reported inferential result. It differs from the unadjusted recalculation from posted group summaries, which estimates a mean difference of 0.03, with a 95% confidence interval from -0.063 to 0.123 and p = 0.529.

The recalculation is a simple comparison of summary statistics and does not reproduce the sponsor’s ANCOVA. ANCOVA can account for covariates specified in the model, but the registry record does not state which covariates were included or provide the full model specification. The record also does not explain the numerical difference between the two estimates. It would therefore be inappropriate to claim that the difference results from a particular adjustment, crossover feature, or data-handling choice.

Both estimates have confidence intervals that include zero, and each is an uncertain estimate of the control-group contrast. The sponsor’s result should be read as the primary analysis reported by the registry; the recalculation is supplementary context for understanding the posted summaries. Neither should be presented as a direct comparison of an experimental toothpaste against no brushing. The sponsor’s analysis does not establish equivalence, and the unadjusted calculation does not supersede the reported ANCOVA.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when an analysis makes several statistical comparisons. Even if each comparison uses the same nominal significance threshold, examining several results increases the chance that at least one will appear unusual by chance. A prespecified testing hierarchy, family-wise error procedure, or other multiplicity strategy can clarify how such results should be interpreted. The registry record does not report a multiplicity adjustment or a hierarchy for the primary and secondary comparisons. It is therefore not possible to determine from the record whether the posted secondary p-values, which are unadjusted, were intended as confirmatory tests or as exploratory comparisons.

The registry identifies an intent-to-treat population for the primary analysis and defines it as randomized participants who received at least one dose of study product and had at least one post-baseline efficacy measurement. This definition is narrower than a general statement that every randomized participant is included regardless of outcome availability. The record does not give a full participant flow for each outcome, explain why the analysis counts differ from enrollment, or state how missing baseline or post-baseline measurements were handled. Reasons for missingness, imputation procedures, and sensitivity analyses are not reported in the registry record.

These omissions matter because missing outcomes can affect both the mean comparison and its uncertainty, especially if the likelihood of missingness is related to treatment or outcome. That is a general statistical concern; it does not establish that missing data biased this trial’s result. The record does not provide enough information to assess the direction or magnitude of any such effect.

The crossover designation also raises general questions about period, sequence, and carryover effects, as well as correlation between measurements from the same participant. The registry does not report treatment order, washout details, or how these features were modeled. Nor does it provide a full statistical analysis plan, the ANCOVA covariates, or the original sample-size justification. These elements are not reported in the registry record, so no assumptions about their use should be made.

16. 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

Serious events with the highest arm-level rate: gingivitis (3.8%), oral herpes (3.8%), gingival erythema (3.8%), nasopharyngitis (3.6%), tongue coated (1.9%).

The registry lists serious adverse events by study group, alongside the number considered at risk. Serious adverse events were reported for 5 participants in Test Product 1, 2 in Test Product 2, 8 in Test Product 3, 6 in the Positive Control group, and 6 in the Negative Control group. The corresponding at-risk figures are 55, 55, 52, 55, and 55. No deaths are listed in any group. For other adverse events, the reported affected count is 0 in each group, with at-risk figures matching the listed group counts.

These are descriptive safety counts from the registry record. They do not establish that a study product caused an event, nor do they establish that an event was unrelated. The record also lists serious-event terms with percentages, including Gingivitis, Oral Herpes, Gingival Erythema, Nasopharyngitis, Tongue Coated, and Angular Cheilitis. Those percentages should not be interpreted as comparative risks without clear group-specific attribution and fuller context.

The posted figures do not give detailed event narratives, timing, duration, severity beyond the serious-event classification, or a causal assessment for each event. Such information is not reported in the registry record. Counts from this study should therefore be read as a description of events recorded during the trial, not as a comprehensive estimate of safety in other populations or with repeated use.

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

The primary result describes a comparison of plaque-score changes between a positive control and a negative control after a single brushing. The posted means are -0.57 and -0.54, respectively. The sponsor’s ANCOVA estimate is -0.01, with a 95% confidence interval from -0.06 to 0.04; the unadjusted recalculation is 0.03, with a 95% confidence interval from -0.063 to 0.123. Both intervals include zero, so the available estimates do not establish a clear difference between the controls.

This does not mean that the two controls are proven equivalent. Establishing equivalence requires a prespecified margin defining how large a difference could be considered acceptably small, together with an analysis designed to assess that margin. No such margin is reported in the registry record. The same caution applies to the experimental products: a result that does not clearly distinguish two conditions is not proof that they have exactly the same effect.

The result also does not compare brushing with no brushing. Both control groups underwent a brushing procedure, and the endpoint is change in a plaque score after a single brushing. The registry record does not establish how repeated use would affect plaque, whether any observed change persists, or whether these score changes translate into other oral-health outcomes. A plaque index is a measurement of disclosed plaque coverage; it is not a direct measure of every outcome that might matter to participants.

Interpretation is also limited by incomplete reporting of the crossover procedures, the ANCOVA model, treatment sequence, missing-data handling, and multiplicity. The trial provides a specific controlled comparison under its recorded single-brushing procedure. It should not be extended beyond that setting or used to make claims about longer-term effectiveness, broader clinical outcomes, or safety outside the observations recorded for this trial.

18. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the distinction between a point estimate and the uncertainty around it. A mean difference alone can appear directional, but its confidence interval shows whether the data locate that difference precisely. In the primary comparison, both the unadjusted recalculation and sponsor-reported ANCOVA have intervals that include zero. The central estimates therefore need to be read alongside their uncertainty rather than as stand-alone evidence.

It also highlights why analysis methods matter. The registry reports an ANCOVA result, while the posted group summaries permit a simple unadjusted recalculation. The estimates differ, and the record does not give enough model detail to explain exactly why. This is a useful reminder that summary statistics cannot always reconstruct a trial’s primary analysis, especially in designs that may involve repeated treatment conditions or covariate adjustment.

The crossover label adds another statistical consideration: within-participant correlation and treatment order can affect how comparisons should be estimated. The record does not report enough detail to assess those issues. Clear reporting of sequence, periods, washout, analysis populations, missingness, and model specification would make the result easier to evaluate.

Finally, the planning illustrations show how estimates based on a small observed effect can imply large sample requirements, and why a prospective target effect is preferable to relying only on an observed estimate. The illustrations are not observed power and do not describe the original design. Together, these points make the record a useful example of how to separate descriptive outcomes, sponsor analysis, recalculation, planning assumptions, and unanswered methodological questions.

19. 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 limitations are specific to the information recorded for this trial. Although the model is labeled crossover, the registry does not report the treatment sequences, period structure, washout, or the method used to account for within-participant correlation, period effects, or carryover. These are central details for understanding a crossover comparison, so the design label alone is not enough to assess how well the analysis reflects the study structure.

The intervention list names five sodium bicarbonate concentrations, but the record does not map those formulations to the three test products and two controls. The positive and negative controls are not otherwise described in enough detail to characterize their formulations. This limits interpretation of what chemical contrast the primary comparison represents.

The sponsor’s primary result is labeled ANCOVA, but the covariates and full model specification are not reported. The primary population is defined, yet the record does not explain outcome counts, differences in counts across the posted summaries, missing-data reasons, or missing-data methods. It also does not provide a sample-size justification for the original design, a multiplicity strategy, or a stated clinically important plaque-score difference. These items are not reported in the registry record.

The study evaluates plaque-score change after a single brushing, and the endpoint timeframe is listed as up to 5 weeks. The record does not give enough timing detail to determine the exact relationship between the assessment schedule and that timeframe. It also does not support conclusions about repeated use, persistence of change, other oral-health outcomes, or longer-term safety. The unmasked design is recorded, but the registry does not describe further safeguards against assessment or performance bias. These limitations qualify the scope of interpretation without changing the reported numerical results.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy of Three Experimental Toothpastes to Remove Plaque — NCT02992691.
• Registry record last updated 2019-05-29; status last verified 2019-05.
• Results first posted 2018-10-29.
• 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.

23. Record Summary

The registry describes a completed, randomized, unmasked crossover trial of three experimental toothpastes, a positive control, and a negative control. Enrollment is listed as 75. The study assessed change in plaque score after a single brushing using the Turesky Modification of the Quigley Hein Plaque Index, for which lower scores indicate less plaque area.

The primary comparison was positive control versus negative control. The sponsor reported an ANCOVA net mean difference of -0.01, with a 95% confidence interval from -0.06 to 0.04 and p = 0.667. An unadjusted recalculation from posted summaries estimated a mean difference of 0.03, with a 95% confidence interval from -0.063 to 0.123 and p = 0.529. Both intervals include zero, and the two estimates should remain distinct because they come from different analysis approaches.

Two secondary endpoints address experimental-product comparisons. Their posted results are unadjusted, and a multiplicity approach is not reported in the registry record. The record also omits important details about crossover procedures, the ANCOVA specification, missing-data handling, and the original sample-size assumptions. The findings should be understood as a limited comparison of plaque-score changes under the recorded single-brushing conditions.