Exploratory Clinical Study to Investigate the Ability of an Experimental Combination Dentifrice to Protect Sensitive Teeth: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Exploratory Clinical Study to Investigate the Ability of an Experimental Combination Dentifrice to Protect Sensitive Teeth (NCT05243745), a randomized na study. The primary outcome is Mean Schiff Sensitivity Score (Average of the Two Selected Test Teeth) at Baseline (Day 0).

2. Clinical Question

3. Trial Design

This was an 8-week, randomized, controlled, examiner-blind, proof-of-principle study. The registry classifies it as a parallel-group trial with triple masking and random allocation. It lists 120 participants enrolled and the study status as completed. The recorded start date is 2022-02-28, and the primary completion date is 2022-10-03. The registry identifies HALEON as the sponsor and lists the study phase as not applicable.

Four arms were described: the test product containing 3% PVM/MA plus 5% KNO3; Comparator 1 containing 3% PVM/MA; Comparator 2 containing 5% KNO3; and a negative control of regular fluoride toothpaste. The arm descriptions specify brushing two selected test teeth first, followed by the whole mouth for one timed minute, twice daily. The particular individuals or roles covered by the triple-masking label, and the procedures used to maintain masking, are not reported in the registry record.

The posted baseline endpoint analysis includes 29 participants in the test-product arm, 31 in Comparator 1, 31 in Comparator 2, and 29 in the negative-control arm. These are the group sizes associated with the reported baseline scores, not a description of every participant’s progress through the trial. For the primary endpoint, the analysis population was modified intent-to-treat: randomized participants had to use study product at least once and have at least one post-baseline clinical assessment of Schiff or tactile sensitivity.

The registry also lists follow-up outcomes at Days 3, 14, 28, and 56, including Schiff sensitivity, tactile threshold, and dentine hypersensitivity experience questionnaire measures. Detailed results for those follow-up outcomes are not reported in the registry record. The original sample-size justification, recruitment details, and reasons for any differences between enrollment and endpoint group sizes 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 133 participants were screened of which 120 participants were randomized to a treatment with either one of the 4 products: test product (combination toothpaste of 3 percent \[%\] methyl vinyl ether/maleic anhydride co-polymer \[PVM/MA\] + 5% Potassium Nitrate \[KNO3\]), comparator 1 (toothpaste containing 3% PVM/MA), comparator 2 (toothpaste containing 5% KNO3), or negative control (a regular fluoride toothpaste).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The modified intent-to-treat (mITT) population included all randomized participants who completed at least one use of study product and had at least one post-baseline clinical performance assessment (Schiff sensitivity assessment or Tactile sensitivity assessment).

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Safety population included all randomized participants who completed at least one use of study product.

6. Primary Endpoint

The registry lists the primary outcome as Mean Schiff Sensitivity Score (Average of the Two Selected Test Teeth) at Baseline (Day 0), measured over Baseline (Day 0).

The primary endpoint is the mean Schiff sensitivity score at baseline (Day 0), averaged across two selected test teeth. An examiner assessed response to an air stimulus directed at the tooth surface. The Schiff scale ranges from 0 to 3: 0 indicates no response, while 3 indicates a response considered painful with a request to stop the stimulus. A lower score indicates less observed sensitivity on this scale. The assessment used a one-second air application, and the mean was calculated from two test teeth in different quadrants of the mouth.

As a general statistical explanation, an examiner-based ordinal score can provide a structured way to record responses to a standardized stimulus. Averaging measurements across two selected teeth gives a participant-level summary for the endpoint, though it does not make the scale a direct measurement of every aspect of sensitivity. The registry identifies the population for this endpoint as modified intent-to-treat: randomized participants who used study product at least once and had at least one post-baseline Schiff or tactile assessment.

The timing is essential to interpretation. This primary measure is explicitly at baseline, before the follow-up assessments described in the registry. It can show whether the groups began with similar or different mean scores, but it cannot by itself show change during treatment. Follow-up change in Schiff score is listed as a secondary outcome, but detailed results for that measure are not reported in the registry record.

7. Results Posted in the Registry

• Test Product: mean 2.43 versus 2.44 on control; difference -0.01 (95% CI -0.23 to 0.21), standardized effect d = -0.02; z-test p = 0.930.
• Comparator 2: mean 2.44 versus 2.44 on control; difference 0.00 (95% CI -0.22 to 0.22), standardized effect d = 0.00; z-test p = 1.000.
• Negative Control: mean 2.34 versus 2.44 on control; difference -0.10 (95% CI -0.30 to 0.10), standardized effect d = -0.25; z-test p = 0.333.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a continuous outcome summarized by an arm mean and standard deviation. The mean is the arithmetic average of the participant-level Schiff scores used in the analysis. The standard deviation describes how spread out those participant scores were around the mean. As a general explanation, these summaries provide different information: the mean locates the group’s average score, while the standard deviation describes participant-to-participant variability. Neither alone conveys how precisely the group mean or a difference between means has been estimated.

What does a mean difference represent?

The posted comparisons express each arm’s baseline mean relative to Comparator 1. A negative mean difference means the first named arm’s mean is lower than the reference mean; a positive difference means it is higher. The difference of -0.01 for the test product, 0.0 for Comparator 2, and -0.1 for the negative control are calculated from the posted baseline summary statistics. They do not represent changes from baseline, because the primary endpoint is measured at baseline. The registry gives no follow-up treatment-effect estimates in the results described by these figures.

As a general statistical explanation, a mean-difference analysis is most directly interpretable when the outcome is measured on the same scale in all groups. The Schiff scale is bounded from 0 to 3 according to the endpoint description. A mean difference summarizes the average separation, but it does not show the full distribution of responses, the proportion at each score, or how many individuals improved. Those details are not reported in the registry record.

How should the confidence interval be read?

A confidence interval is an uncertainty range around an estimated difference, constructed by a statistical procedure with a stated coverage level. The posted comparisons give 95% confidence intervals. As a general explanation, an interval that includes zero does not rule out no mean difference under the calculation used, and its width indicates the range of differences compatible with the observed data and method. The test-product and negative-control intervals include zero. Comparator 2’s interval extends from -0.22 to 0.22 and also includes zero. These intervals concern baseline comparisons, not treatment effects over follow-up.

The recalculated intervals use unadjusted comparisons from summary statistics. The registry’s reported comparison summaries include standard errors and pooled standard deviations, but no detailed account of the trial’s own inferential procedure is reported in the registry record. The exact analysis model, any covariate adjustment, and any approach to repeated measurements are not reported in the registry record. Therefore, the recalculated comparisons should not be mistaken for a documented sponsor analysis plan.

What do p-values add?

The unadjusted p-values are 0.93 for the test product, 1.0 for Comparator 2, and 0.333 for the negative control, each compared with Comparator 1 at baseline. As a general explanation, a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be under the model and assumptions used. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. These p-values do not establish equivalence and do not answer whether scores changed during treatment.

“Unadjusted” means that the reported p-values do not account for a multiplicity adjustment across a family of comparisons. The registry record does not report whether the sponsor used another adjustment or a prespecified hierarchy for analyses. Because the endpoint is baseline and the calculations are unadjusted, these p-values should be read narrowly as part of the descriptive comparison of starting group means.

What is the role of the analysis population?

The modified intent-to-treat population for the primary endpoint includes randomized participants who used study product at least once and had at least one post-baseline clinical performance assessment, either Schiff or tactile sensitivity. As a general explanation, defining an analysis population makes clear which randomized participants contribute to an estimate. A modified intent-to-treat definition may exclude some randomized participants; the registry record does not report the number excluded from this population or the reasons. The group sizes attached to the baseline endpoint are 29, 31, 31, and 29.

The registry’s outcome list includes several repeated follow-up assessments and questionnaire domains. Repeated measures raise general analytic questions, such as how to account for measurements from the same participant and how to handle incomplete visits. The model for those assessments, missing-data assumptions, and any imputation procedures are not reported in the registry record. No specific method should be inferred from the list of assessment days alone.

The trial record classifies the primary measure as a mean of a continuous outcome and supplies standard deviations for the arms. This supports a straightforward unadjusted mean comparison as a recalculation, but does not reveal whether the trial’s planned analysis used that approach. The registry record does not report the original statistical analysis plan, sample-size justification, interim analyses, or a sponsor-reported inferential result for the baseline endpoint. The calculation method for the planning illustrations is reported separately and should not be confused with the original trial design.

10. Confidence Intervals and Precision

The 95% confidence intervals quantify uncertainty around the recalculated baseline mean differences under the unadjusted comparison method. For the test-product comparison, the interval includes zero; the interval for the negative-control comparison also includes zero. Comparator 2’s interval runs from -0.22 to 0.22, likewise spanning zero. These are intervals for baseline differences relative to Comparator 1, not intervals for change from baseline or for an 8-week treatment effect.

As a general statistical explanation, interval width depends in part on the amount of information in the data and the variability of participant scores. The endpoint groups contain 29 or 31 participants, and their reported standard deviations range from 0.356 to 0.442. The confidence intervals therefore describe the precision of comparisons based on these group summaries, not the certainty of every individual participant’s response. The record does not report precision estimates for the secondary follow-up outcomes in the results.

All three recalculated intervals include zero, and the point estimates are -0.01, 0.0, and -0.1. This means the posted baseline summaries do not distinguish the corresponding mean differences from zero using these intervals. It does not demonstrate that the arms are equivalent or that any differences are clinically unimportant; an equivalence margin and the relevant follow-up estimates are not reported in the registry record. The test-product and negative-control interval limits are retained in the calculations but are not reproduced numerically because their recorded precision exceeds three decimal places.

Confidence intervals express statistical precision under the calculation’s assumptions. They do not address possible selection, measurement, or analysis limitations, and they cannot compensate for the fact that the primary endpoint is at baseline. Follow-up confidence intervals for change scores are not reported in the registry record.

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 illustrations are planning examples based on observed comparison effects, not the trial’s own design justification. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is a stated fraction of the observed effect. The figures give participants per arm, not total participants, for these continuous-outcome comparisons. The planning rows cover 100%, 75%, and 50% of the observed effect for comparisons where an illustration is listed.

For the test-product comparison, the illustration lists 30,402 participants per arm for 80% power and 40,699 participants per arm for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, it lists 54,047 per arm for 80% power and 72,353 per arm for 90% power. For 50% of the observed effect, it lists 121,605 per arm for 80% power and 162,794 per arm for 90% power. For the negative-control comparison, the corresponding rows list 255 and 341 per arm for 100% of the observed effect; 453 and 607 per arm for 75%; and 1,019 and 1,364 per arm for 50%. No planning illustration is listed for Comparator 2.

In plain language, the 80% column is the planned number per arm under the stated assumptions for a design with an 80% chance of detecting the chosen effect if that effect is present. The 90% column describes a design targeting a higher chance of detection and therefore requires more participants under the same assumptions. A smaller target effect is harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants. The sample-size curve can be read as illustrating this relationship between target effect and required enrollment.

These figures are not observed power. They are planning illustrations calculated from the observed effect, and they do not establish the trial’s original sample-size target or justification. The original design assumptions are not reported in the registry record.

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 = -0.010; pooled standard deviation σ = 0.440.
2. n = 2 × (1.96 + 0.842)² × 0.440² / -0.010² = 30,401.0, rounded up to 30,402 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 <- -0.00999999999999979                      # observed difference in means
sigma <- 0.440073504832059                     # 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(5243745)
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 = 30,402 (80% power), 40,699 (90% power)
 75% of observed effect: n per arm = 54,047 (80% power), 72,353 (90% power)
 50% of observed effect: n per arm = 121,605 (80% power), 162,794 (90% power)
power.t.test: 30402 
Simulated power at n = 30,402 per arm: 0.798 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The registry lists a broad set of outcomes, including Schiff sensitivity, tactile threshold, and several dentine hypersensitivity experience questionnaire measures. Follow-up time points include Days 3, 14, 28, and 56. Multiple outcomes, time points, arms, and pairwise comparisons create a general multiplicity concern: the more statistical questions examined, the greater the chance that some results appear unusual by chance if all are treated as independent confirmatory tests.

The recalculated p-values for the baseline comparisons are explicitly unadjusted. The registry record does not report a multiplicity adjustment, a hierarchy of hypotheses, or which outcomes were designated confirmatory beyond identifying the primary measure. The original statistical analysis plan and any prespecified approach to multiplicity are not reported in the registry record. It is therefore not possible to determine from these details whether or how the trial’s own analyses controlled the overall false-positive risk.

The primary endpoint analysis population is defined as modified intent-to-treat, requiring at least one use of study product and at least one post-baseline Schiff or tactile assessment. The registry record does not report the number of randomized participants who failed to meet this definition, the reasons for non-inclusion, or the detailed analysis populations used for secondary outcomes. The record lists 120 participants enrolled, while the baseline endpoint groups include 29, 31, 31, and 29 participants; reasons for the difference are not reported in the registry record.

Missing follow-up data are especially relevant when outcomes are measured repeatedly. As a general statistical explanation, results can depend on whether missingness is related to participants’ observed or unobserved outcomes and on the assumptions used to address it. The registry record does not report missing-data rates by visit, imputation procedures, sensitivity analyses, or the assumptions used in any repeated-measures analysis. These gaps limit evaluation of the follow-up outcomes, whose detailed results are also not reported in the registry record.

The registry does not report interim analyses, stopping rules, or the original sample-size justification. Their absence from the record should not be interpreted as evidence that a particular procedure was or was not used.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The posted adverse-event figures report zero participants with serious adverse events in each of the four arms: Test Product, Comparator 1, Comparator 2, and Negative Control. The numbers at risk for the serious-event summaries are 29, 31, 31, and 29, respectively. The reported death count is zero in every arm. These are counts during the trial record’s reported observation, not estimates of the frequency of uncommon events outside the study.

As a general statistical explanation, a zero count means no event was recorded in the reported group; it does not establish that the underlying risk is zero. Small groups provide limited information about uncommon outcomes. The registry record does not report serious-event terms, because none are listed, and the posted results do not describe the full range, timing, or severity of non-serious events. The adverse-event figures should therefore be read as a narrow account of reported serious events and deaths, not as a comprehensive assessment of every possible safety outcome.

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

The posted primary results describe baseline Schiff sensitivity scores in the four randomized groups. Their means were 2.43 for the combination test product, 2.44 for 3% PVM/MA, 2.44 for 5% KNO3, and 2.34 for regular fluoride toothpaste. Recalculated unadjusted comparisons with Comparator 1 are small in the displayed mean differences, and each reported 95% confidence interval includes zero. These observations concern starting scores only.

The study’s stated aim was to investigate protection from dentine hypersensitivity during an 8-week proof-of-principle study. However, a baseline primary measure cannot answer whether sensitivity changed with product use. The registry lists follow-up changes in Schiff scores, tactile thresholds, and questionnaire outcomes, but detailed follow-up results are not reported in the registry record. No conclusion about improvement, lack of improvement, or comparative protection over time follows from the baseline comparisons alone.

As a general statistical explanation, lack of a statistically distinguishable baseline difference is not evidence that groups are equivalent, and it does not establish that a product has or lacks an effect. Equivalence requires a suitable design and a margin defining differences small enough to count as equivalent; such a margin is not reported in the registry record. Likewise, p-values do not measure clinical importance, and the baseline mean differences do not show the distribution of individual responses.

The safety record reports zero serious adverse events and zero deaths in each arm among the participants included in the relevant summaries. This describes the recorded counts in this study and does not rule out uncommon events. Together, the available information supports a limited statistical reading: baseline group scores can be compared descriptively, and serious-event counts can be stated, but follow-up efficacy and a broad safety profile cannot be assessed from these posted results.

16. Why This Trial Matters Statistically

This trial illustrates why the timing and definition of an endpoint matter as much as its name. Schiff sensitivity is clinically relevant to the study question, but the listed primary measure is explicitly at baseline. It describes the starting score and cannot establish a change during treatment. A reader who notices only that the study aimed to investigate protection could otherwise mistake the baseline comparisons for follow-up treatment estimates.

The design also makes clear that one combination product was considered alongside its individual ingredients and a regular-fluoride negative control. That structure can support distinct comparisons, but interpretation still depends on the outcomes and analysis plan actually reported. The posted recalculations compare baseline group means, while the registry’s follow-up outcome list does not include detailed follow-up estimates in the results. Statistical design alone cannot substitute for reporting those estimates.

Precision is another important lesson. The recalculated confidence intervals include zero, but this does not demonstrate equivalence. The planning illustrations show how strongly required enrollment depends on the effect a future study aims to detect: smaller target effects require larger participant counts under the stated assumptions. Those illustrations are based on observed effects and are not the original trial’s rationale or evidence of its power.

Finally, the registry’s analysis-population definition, outcome list, and adverse-event counts provide useful context while leaving important questions unanswered. The original analysis plan, missing-data approach, multiplicity strategy, and detailed secondary results are not reported in the registry record. Recognizing what a result can and cannot answer is central to sound biostatistical interpretation.

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 primary endpoint is a baseline mean rather than a change from baseline or a follow-up comparison. This is the central limitation for interpreting the posted primary values against the study’s stated aim of investigating protection from sensitivity. Detailed results for the listed follow-up outcomes are not reported in the registry record, so change over time and comparative follow-up effects cannot be assessed from these results.

The primary endpoint results are available as arm summaries and recalculated unadjusted comparisons. The registry record does not report the original statistical analysis plan, any covariate-adjusted primary analysis, or the method planned for repeated measurements. The recalculated contrasts should therefore be distinguished from a documented sponsor analysis. In addition, the recalculated p-values are unadjusted, and the registry record does not describe a prespecified approach to multiplicity across the many outcomes and assessment times.

The analysis population is defined, but the number of randomized participants who did not meet that definition and the reasons are not reported in the registry record. The record reports 120 enrolled and baseline endpoint group sizes of 29, 31, 31, and 29; it does not explain the relationship between enrollment and those analysis counts. Follow-up completeness, missing-data handling, and sensitivity analyses are not reported in the registry record.

Masking is labeled triple, but the roles masked and the procedures used are not reported. Eligibility details, the original sample-size justification, and interim analyses are also not reported in the registry record. The serious-event counts are zero in all arms, but the group sizes are limited and the record does not give detail on non-serious events. These limitations constrain both the assessment of the follow-up question and the breadth of conclusions possible from the registry results.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Exploratory Clinical Study to Investigate the Ability of an Experimental Combination Dentifrice to Protect Sensitive Teeth — NCT05243745.
• Registry record last updated 2024-09-20; status last verified 2024-05.
• Results first posted 2024-09-20.
• 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

NCT05243745 is a completed, randomized, parallel-group, triple-masked, 8-week proof-of-principle study sponsored by HALEON. It enrolled 120 participants and compared a toothpaste containing 3% PVM/MA plus 5% KNO3 with 3% PVM/MA alone, 5% KNO3 alone, and regular fluoride toothpaste. The registry lists the study start as 2022-02-28 and primary completion as 2022-10-03.

The primary measure is the mean Schiff sensitivity score at baseline, averaged across two selected test teeth. Recalculated unadjusted baseline comparisons with Comparator 1 have confidence intervals that include zero. They do not establish follow-up effects. Change-from-baseline outcomes are listed, but detailed results are not reported in the registry record. The posted safety counts show zero serious adverse events and zero deaths in each arm.