Assessment of the Efficacy of an Experimental Occlusion Technology Dentifrice in Dentinal Hypersensitivity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assessment of the Efficacy of an Experimental Occlusion Technology Dentifrice in Dentinal Hypersensitivity (NCT02861664), a randomized na study. The primary outcome is Change From Baseline in Schiff Sensitivity Score at Week 8.

2. Clinical Question

This is a comparison of average changes between randomized groups, not a count of participants whose sensitivity improved. The posted primary results show a larger mean reduction in the test-dentifrice group than in the negative-control group. That pattern describes the trial’s measured outcome; it does not by itself establish how an individual participant would respond or how the result applies beyond the population studied.

3. Trial Design

The registry record describes a randomized, single-masked, parallel-group trial. Participants were assigned to either the stannous fluoride test dentifrice or the sodium monofluorophosphate negative-control dentifrice. The record lists an enrollment of 135 and a completed status. It gives a start date of 2016-09-21 and a primary completion date of 2016-12-01. The phase is listed as NA.

The study evaluated dentinal hypersensitivity over 8 weeks of twice-daily dentifrice use. Sensitivity assessments were scheduled at baseline and after 4 and 8 weeks, using evaporative air and tactile stimuli. The primary endpoint was change from baseline in Schiff Sensitivity Score at Week 8. Secondary outcomes included the Week 4 Schiff Sensitivity Score change and tactile-threshold changes at Week 4 and Week 8.

The primary endpoint population is described as the ITT population, with eligibility requiring randomization, at least one treatment exposure, and at least one post-baseline efficacy assessment. The posted primary arm summaries include 67 participants in each group. These endpoint counts are not necessarily a complete account of everyone enrolled or randomized, and no further participant-flow explanation is reported in the registry record.

The record names GlaxoSmithKline as sponsor. It identifies two interventions and gives arm descriptions for the dentifrices. Some control-arm instruction text in the registry record is incomplete, so details beyond the stated composition and brushing instructions are not reported in the registry record. The registry record also does not report the masking procedures, the method of sequence generation or concealment, or the original trial-specific sample-size assumptions.

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 273 participants were screened, out of which 135 participants were randomized to the study. 138 participants were not randomized because out of these, 135 participants did not meet the study criteria and 3 participants withdrew consent.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT population included all participants who were randomized, received the study treatment at least once and provided at least one post-baseline assessment of efficacy.

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: Intent-to-treat (ITT) population included all the participants who were randomized, received at least one dose of the study treatment and provided at least one post-baseline assessment of efficacy.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Schiff Sensitivity Score at Week 8, measured over At Baseline and Week 8.

The primary endpoint was change from baseline in Schiff Sensitivity Score at Week 8. An examiner assessed a participant’s response to an evaporative air stimulus after stimulating each individual tooth. The scale ranges conceptually from no response to a response considered painful that leads the participant to request that the stimulus stop. The registry record describes the scale’s categories and states that a reduction in score indicates improvement in sensitivity.

This endpoint turns a response to a controlled stimulus into an ordered score that can be summarized across participants. The posted analysis treats the outcome as a continuous measure and reports group means and standard deviations for change from baseline. That summary supports a comparison of average score changes; it does not show the distribution of individual responses or the proportion of participants who experienced a particular degree of change.

Using a baseline-to-follow-up change focuses the comparison on how scores shifted over the study period. The primary assessment at Week 8 corresponds to the end of the stated 8-week treatment period. The registry record does not report a threshold for a clinically meaningful change, so the statistical size of the mean difference cannot, on its own, determine how important that difference is to participants.

7. Results Posted in the Registry

• Test Dentifrice: Stannous Fluoride (SnF2): mean -0.94 versus -0.33 on control; difference -0.61 (95% CI -0.82 to -0.40), standardized effect d = -0.99; z-test p = <0.001.

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 in Schiff Sensitivity Score at Week 4

• 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 Dentifrice: Stannous Fluoride (SnF2): mean -0.63 versus -0.10 on control; difference -0.53 (95% CI -0.71 to -0.35), standardized effect d = -0.98; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Test Dentifrice: Stannous Fluoride (SnF2) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary outcomes follow the same broad direction as the primary result: the Week 4 Schiff Sensitivity Score change and the tactile-threshold outcomes at Week 4 and Week 8 are also framed as changes from baseline between the test and control groups. The available secondary summary indicates a larger mean reduction in Schiff score for the test group at Week 4. The registry record lists tactile-threshold changes as secondary outcomes, but does not include a sponsor-reported analysis for those outcomes in the posted details used for this interpretation.

These measures provide supporting context because they assess sensitivity at another time point or through a different stimulus. They do not replace the prespecified primary endpoint, and differences in measurement method mean that results need not be identical across endpoints. The primary result remains the main basis for interpreting the trial’s principal statistical comparison.

Secondary comparisons usually carry less inferential weight than the primary comparison when many outcomes or time points are examined. In general, making multiple statistical tests increases the chance of finding a nominally notable result even when there is no underlying effect. A multiplicity strategy can address that issue, but whether one was planned or used is not reported in the registry record. Accordingly, unadjusted secondary findings should be regarded as exploratory or supportive rather than as independent confirmation.

The record lists two secondary outcomes and the overall endpoint count, but does not describe a hierarchy among the secondary measures, a correction for multiple testing, or how missing observations were handled for each one. Those details would affect how strongly secondary results can be interpreted. The broad pattern is consistent with the primary comparison, but the available information does not justify treating every endpoint as equally conclusive.

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

• Change From Baseline in Schiff Sensitivity Score at Week 4
• Change From Baseline in Tactile Threshold at Week 4 and 8

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous in the posted analysis: each participant has a score change, and the group summaries describe the mean and standard deviation. A mean difference compares the average change in one group with the average change in the other. Since lower Schiff scores and reductions are described as favorable in the registry record, a negative difference for test minus control corresponds to a larger mean reduction in the test group.

What does the sponsor’s ANCOVA estimate represent?

The registry record reports an analysis of covariance, or ANCOVA, with a least-squares mean difference as its parameter. In general, ANCOVA compares group outcomes while accounting for one or more covariates in a statistical model. Adjusting for baseline outcome values is a common use of ANCOVA, but the specific covariates and model specification for this trial are not reported in the registry record. The least-squares mean difference is a model-based contrast, so it need not be identical to a simple subtraction of the posted group means.

The registry reports the sponsor’s estimate as -0.6, with a 95% confidence interval from -0.8 to -0.4 and p < .0001. The unadjusted recalculation uses posted group means and standard deviations to estimate the mean difference. It is not a reconstruction of the ANCOVA model and does not replace the sponsor’s reported analysis. The registry record does not report whether other covariates, stratification factors, or additional modeling choices were used.

What does the unadjusted comparison add?

A simple unadjusted comparison is useful as a check on the direction and approximate size of the posted result. It relies on summary statistics rather than participant-level data. It cannot account for any baseline adjustment or other features of the sponsor’s model. Agreement in direction and broad magnitude can help readers understand the summary outcome, but it does not verify every aspect of the original analysis.

The unadjusted p-value is a test of the null hypothesis of no average difference under the assumptions of that comparison. The reported p-value is not a measure of effect size. The estimated difference and its confidence interval convey magnitude and precision, while the p-value summarizes compatibility with a null contrast under a specified model. No multiplicity adjustment for the primary result is described in the registry record.

How should the confidence interval be understood?

In general, a confidence interval reflects sampling uncertainty under the analysis assumptions. A narrower interval indicates greater statistical precision than a wider interval on the same scale. It does not capture every source of uncertainty: it does not automatically account for missing-data bias, measurement limitations, model misspecification, or limits on generalizability. The sponsor’s reported interval and the recalculated interval arise from different methods, so they should not be treated as interchangeable estimates.

The 95% interval reported by the sponsor is centered on its reported least-squares mean difference at the precision shown in the registry record. Its endpoints are both negative. The posted recalculation also has negative endpoints. These intervals support the conclusion that the estimated average contrast favors a larger reduction in the test group under the respective analyses, but they do not identify an individual response or establish clinical importance.

What is the role of the score’s scale?

The Schiff Sensitivity Scale is an ordered rating of response to evaporative air stimulation. The registry describes categories ranging from no response to a painful response prompting discontinuation. Treating score change as a continuous measure enables means and mean differences to be reported. As a general statistical consideration, this approach summarizes average changes in score units and may not describe how participants are distributed across the scale’s categories. The registry record does not report a category-level analysis for the primary endpoint.

The endpoint is a change from baseline, so the interpretation concerns average score movement over the stated period. The record does not provide individual-level data, baseline group means, or an estimate of variability in change beyond the posted summaries. This limits independent assessment of distributional assumptions and alternative analyses. The available information supports interpretation of the reported mean contrast, not a full reanalysis of participant-level responses.

11. Confidence Intervals and Precision

The sponsor’s primary analysis reports a least-squares mean difference of -0.6 with a 95% confidence interval from -0.8 to -0.4. The interval is entirely below zero and is reasonably concentrated around the reported estimate on the scale of the outcome. In general, a confidence interval communicates the precision of an estimate under the analysis assumptions: it shows the range of values compatible with the observed data and method at the stated confidence level.

The unadjusted recalculation reports a mean difference of -0.61 and a confidence interval from -0.820 to -0.401. Its interval is also entirely below zero. These limits belong to the summary-statistic recalculation, not the sponsor’s ANCOVA. Since the analyses differ, their intervals describe uncertainty under their respective methods and should not be combined.

The confidence intervals quantify sampling uncertainty, not every source of uncertainty. They do not account automatically for potential bias from missing outcome data, limitations of the score, the choice of analysis, or generalizability to populations not described in the registry record. The registry record does not report a threshold for a clinically meaningful difference. Thus the interval helps assess statistical precision and the direction of the estimated contrast, but it cannot alone determine the practical importance of the score difference.

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 sample-size figures are planning illustrations expressed as participants per arm, not total event counts. The calculations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set at a stated fraction of the observed effect. The trial’s original sample-size assumptions are not reported in the registry record.

The 80% power column indicates the planned number of participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the assumptions were correct. The 90% power column describes the corresponding planning target for a 90% chance. These are design concepts, not guarantees about what a future study will find.

The rows reduce the assumed effect from the full observed effect to 75% and 50% of that effect. Smaller effects require more participants for the same target power, all else equal. The listed counts therefore increase as the target effect becomes smaller. The curve should be read as a planning illustration of that relationship, not as a reconstruction of the original design or a claim about the original investigators’ calculations.

This is not observed power. It is not a measure of the probability that the completed trial’s result is correct, nor does it describe the chance that the trial detected an effect after observing its data. The figures depend on the stated planning assumptions and on using the observed effect as their reference, so they are illustrative rather than independent evidence about the treatment contrast.

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 = -0.610; pooled standard deviation σ = 0.619.
2. n = 2 × (1.96 + 0.842)² × 0.619² / -0.610² = 16.1, rounded up to 17 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 <- -0.61                      # observed difference in means
sigma <- 0.61857780432214                     # 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(2861664)
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 = 17 (80% power), 22 (90% power)
 75% of observed effect: n per arm = 29 (80% power), 39 (90% power)
 50% of observed effect: n per arm = 65 (80% power), 87 (90% power)
power.t.test: 17.2 
Simulated power at n = 17 per arm: 0.797 (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 sponsor’s reported primary analysis is an ANCOVA, with a least-squares mean difference of -0.6, a 95% confidence interval from -0.8 to -0.4, and p < .0001. The recalculated comparison is an unadjusted estimate based on the posted group means and standard deviations. It gives a mean difference of -0.61, a 95% confidence interval from -0.820 to -0.401, and p < 0.001.

The estimates agree in direction and are close in magnitude, but they come from different analyses. ANCOVA is model-based and may adjust for covariates; the covariates and full model details are not reported in the registry record. The recalculation does not reproduce those adjustments, so it should be read as a supplementary check on the summary statistics, not as a replacement for the sponsor’s result.

The registry record does not flag a discrepancy between the reported and recalculated results. The sponsor’s own analysis takes precedence as the trial’s reported analysis. The recalculation can help readers relate the posted group summaries to the contrast, while the sponsor’s model remains the primary reported inferential result.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary endpoint and secondary outcomes, but does not report a multiplicity adjustment or a formal testing hierarchy. Multiplicity arises when several outcomes, time points, or comparisons are tested. In general, the more tests conducted, the greater the possibility of at least one apparently unusual result under a set of null effects. Adjustments or prespecified ordering can help control this issue. Without those details, primary and secondary findings should not be given the same inferential status.

The primary population is defined as participants who were randomized, received study treatment at least once, and provided at least one post-baseline efficacy assessment. This definition excludes randomized participants who did not meet those conditions from the primary endpoint population. The registry record does not report how many participants failed each criterion, why they were excluded, or whether any imputation or other missing-data method was used.

The posted primary arm summaries each include 67 participants. The safety figures use different at-risk counts, and the registry record does not explain the difference between the efficacy and safety populations. No participant-level flow, reasons for missing visits, or sensitivity analyses for missing outcomes are reported in the registry record. If missingness is related to outcome or treatment experience, estimates based only on observed assessments may be affected; this is a general concern, not evidence that such bias occurred in this trial.

The trial is described as randomized and single-masked, but the registry record does not report the sequence-generation or allocation-concealment procedures, who was masked, or the success of masking. It also does not give the complete ANCOVA specification. These omissions constrain assessment of the analysis and the protection against bias. The result can be interpreted from the reported estimates, while these unreported design and analysis details remain limitations of the registry record.

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 0% reporting threshold.

The registry record reports zero serious adverse events in each arm: the test-dentifrice group has 0 among 67 at risk, and the negative-control group has 0 among 68 at risk. It also reports zero for the other adverse-event category in each group, with the same at-risk counts. Deaths are reported as 0 in each group.

These are counts observed and posted for the trial, not proof that no adverse events of any kind occurred. A zero count means that no event in the relevant recorded category was reported among the listed at-risk participants. It does not establish that the underlying event risk is zero, particularly in a study of this size and duration.

The registry record does not provide serious adverse-event terms, since none are listed, or further detail about event definitions, collection procedures, severity, timing, or relatedness. The safety figures should therefore be read narrowly as the reported counts for the stated categories. They do not support broad conclusions about safety beyond the events and observation recorded.

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

The posted results show a difference in average change in Schiff Sensitivity Score at Week 8 between the randomized groups. The test-dentifrice group had a mean change of -0.94, compared with -0.33 for the negative-control group. The sponsor’s reported ANCOVA estimate was -0.6, with a 95% confidence interval from -0.8 to -0.4 and p < .0001. The unadjusted recalculation is similar in direction and approximate magnitude.

This describes a group-level average on the specified scale. It does not mean that every participant had a reduction, that every participant experienced the same change, or that an individual’s response can be predicted from the group mean. The score is an examiner-assessed response to an air stimulus, not a direct measurement of every aspect of a participant’s experience.

The statistical result does not establish how important the difference is to participants. The registry record does not report a minimum clinically meaningful change or another threshold for practical importance. Nor does a small p-value measure the size or value of an effect; it is a measure of compatibility with a null contrast under the analysis assumptions.

The results apply to the endpoint population and study conditions described in the registry record. The record does not report detailed eligibility criteria, participant characteristics, or enough information to judge how closely the trial population resembles other populations. The finding therefore should not automatically be generalized to people or settings not characterized in the record.

The reported zero serious adverse-event counts are limited to recorded events during the trial and do not establish an absence of risk. In combination, the efficacy and safety summaries describe what was measured and recorded; they do not provide medical advice or settle questions beyond the trial’s outcomes and available reporting.

18. Why This Trial Matters Statistically

This trial illustrates several statistical distinctions that matter in clinical research. First, a change-from-baseline endpoint and a between-group contrast answer different questions from a within-group change alone. Both groups had negative mean changes, but the primary comparison concerns the difference between their average changes. A change in one arm by itself cannot establish a treatment contrast.

Second, the registry presents a sponsor-reported ANCOVA alongside an unadjusted recalculation from group summaries. Those estimates are close in direction and magnitude, yet they arise from different methods. Keeping the analyses distinct prevents a summary-statistic check from being mistaken for a reproduction of the trial’s full model.

Third, uncertainty and importance are not the same. The confidence intervals describe statistical precision, while the registry record does not report a threshold for a clinically meaningful score difference. A statistically clear average contrast therefore does not, by itself, answer how important that contrast is to participants.

Finally, endpoint hierarchy, missing-data handling, and multiplicity affect interpretation. The primary outcome is identified, but the record does not report a testing hierarchy for secondary outcomes or methods for missing observations. Such information is necessary to judge the strength of secondary evidence and the robustness of the estimates. These considerations show why reading an effect estimate together with its analysis method, uncertainty, endpoint definition, and reporting limits is essential.

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 trial-specific details needed for a fuller assessment are not reported in the registry record. These include detailed eligibility criteria and participant characteristics, the allocation-sequence and concealment methods, who was masked, and whether masking was maintained. The record describes the study as randomized and single-masked but does not explain those procedures.

The primary population definition requires treatment exposure and a post-baseline efficacy assessment. The registry record does not provide a participant-flow account, explain missing assessments or exclusions, or describe imputation and sensitivity analyses. The primary endpoint summaries therefore cannot be assessed against a complete account of all randomized participants.

The sponsor reports an ANCOVA, but the covariates and full model specification are not reported in the registry record. The rationale for the original sample size and the trial’s own design assumptions are also not reported. The planning illustration based on the observed effect should not be mistaken for that original design rationale.

The record does not report a clinically meaningful threshold for Schiff score change, a multiplicity procedure for secondary outcomes, or detailed category-level primary results. Adverse-event reporting is limited to the posted counts; event definitions and collection details are not reported in the registry record. These limitations constrain conclusions about clinical importance, missing-data robustness, secondary inference, and the completeness of safety reporting.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Assessment of the Efficacy of an Experimental Occlusion Technology Dentifrice in Dentinal Hypersensitivity — NCT02861664.
• Registry record last updated 2018-08-10; status last verified 2018-06.
• Results first posted 2018-08-10.
• 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 record describes a completed, randomized, single-masked, parallel-group study of dentinal hypersensitivity, sponsored by GlaxoSmithKline. The trial compared a stannous fluoride test dentifrice with a sodium monofluorophosphate negative-control dentifrice during 8 weeks of twice-daily use. Enrollment was 135; the primary endpoint population summaries include 67 participants in each arm.

The primary outcome was change from baseline in Schiff Sensitivity Score at Week 8. The sponsor reported an ANCOVA least-squares mean difference of -0.6, with a 95% confidence interval from -0.8 to -0.4 and p < .0001. The unadjusted recalculation from posted summaries gives a similar direction and magnitude. The record also lists secondary sensitivity outcomes and reports zero serious adverse events in each arm. Important details about the original sample-size rationale, missing-data methods, multiplicity, and full model specification are not reported in the registry record.