The Effect of Periodontal Treatment and the Use of Dentifrice on Glycaemic Control in Diabetics: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Periodontal Treatment and the Use of Dentifrice on Glycaemic Control in Diabetics (NCT00763256), a randomized phase 3 study. The primary outcome is HbA1c Levels in Blood.

2. Clinical Question

This is a question about a between-group difference in a continuous measure at a stated follow-up time. The posted group means are 9.47 for the active comparator and 10.27 for the placebo comparator. The re-calculated mean difference is expressed as placebo comparator minus active comparator, giving 0.8 percentage points in the posted analysis. That direction describes the observed means; it does not establish why they differed or what would happen in another population.

The record does not report enough detail to reconstruct the full clinical question, including eligibility criteria, the specific periodontal procedures, or the intended relationship between periodontal treatment and dentifrice use. Those details are not reported in the registry record.

3. Trial Design

The registry identifies the study as a randomized, double-masked, parallel-group phase 3 trial. Its status is completed. The recorded enrollment is 33. The primary completion date is 2009-08, and the recorded start is 2006-05. Colgate Palmolive is listed as the sponsor.

The outcome data are presented for two groups: Active Comparator and B - Placebo Comparator. The arm counts reported with the primary outcome are 15 and 18, respectively. The record also lists Triclosan, Fluoride and Fluoride as interventions, both classified as drugs. Because the arm descriptions are blank, the registry record does not provide a detailed mapping of the listed interventions to the groups or a description of how treatment was delivered.

Randomization means that assignment to study groups was intended to use a chance-based process; the specific sequence-generation and allocation-concealment methods are not reported in the registry record. Double masking indicates that the registry classifies the trial as masked at two levels, but it does not identify which people were masked or how masking was maintained. The study model is parallel, so the listed groups are separate arms rather than a crossover comparison.

The primary outcome is HbA1c levels in blood at 12 months. The endpoint description says blood was taken from each subject and HbA1c measured by Queensland Health Pathology Scientific Services. Further details about visit schedules, adherence, co-interventions, or the treatment protocol are not reported in the registry record. The registry lists five outcome measures and five analyses, but does not provide a full account of how every outcome or analysis was defined. The brief summary states that the study tested dentifrice effectiveness in maintaining periodontal health; the primary endpoint listed for the posted results is glycaemic, rather than a periodontal measure.

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: All subjects passed a medical exam and met the inclusion/exclusion criteria.

Overall Study

Reasons for not completing this period:

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 HbA1c Levels in Blood, measured over 12 months.

The primary endpoint is HbA1c levels in blood, reported as a mean with standard deviation and measured in percentage units at 12 months. HbA1c is glycated hemoglobin. As the registry description explains, it is measured primarily to identify average plasma glucose concentration over prolonged periods. A blood measurement therefore provides an indicator that reflects a longer interval than a single point-in-time glucose reading.

For this trial, the posted endpoint is continuous: each participant has a measured level, and the group summaries describe the mean and the spread of those measurements. Comparing group means addresses whether the observed average levels differ between the active-comparator and placebo-comparator groups at the specified follow-up. It does not, by itself, describe every participant’s value or the distribution’s full shape.

HbA1c is relevant to the trial’s stated interest in glycaemic control, while the brief summary also names periodontal health as the focus of dentifrice testing. The registry record does not explain the hypothesized pathway connecting dentifrice use, periodontal health, and HbA1c, nor does it report a periodontal outcome among the posted primary results. The result should therefore be read as a comparison of the recorded HbA1c endpoint, not as a complete assessment of periodontal health or of all possible effects of the interventions.

7. Results Posted in the Registry

• B - Placebo Comparator: mean 10.27 versus 9.47 on control; difference 0.80 (95% CI -0.77 to 2.37), standardized effect d = 0.35; z-test p = 0.318.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous, and the registry reports a mean and standard deviation for each group. A mean gives the arithmetic average of the recorded values. A standard deviation describes how dispersed values are around that average. These summaries make it possible to describe the groups and to estimate a between-group mean difference, but they do not show the distribution of individual observations or reveal whether unusual values strongly influence the mean.

What does the mean difference represent?

The re-calculation defines the contrast as the B - Placebo Comparator mean minus the Active Comparator mean. With this ordering, a positive estimate means the placebo-comparator mean is higher. The reported estimate is 0.8 percentage points. The subtraction is an unadjusted comparison of the posted group means; it does not account for any baseline values or other variables. The record does not report an adjusted estimate.

A mean difference is expressed in the outcome’s units, here percentage units for HbA1c. It describes a contrast in group averages rather than a ratio, an individual response, or a probability of a particular outcome. Since the record provides only endpoint means in the posted comparison, it should not be interpreted as the average change in HbA1c from baseline.

Why use a confidence interval?

A confidence interval conveys uncertainty around an estimated difference. In general, it is a range produced by a statistical procedure that is designed to have a stated coverage rate over repeated use under its assumptions. It is not a guarantee that the true difference lies in the particular interval, nor does it assign a probability to a fixed parameter after the data have been observed. The re-calculated interval is described as 95% and spans values on both sides of zero. That means the posted summary statistics are compatible with no mean difference as well as with differences in either direction within the interval.

Intervals are useful because they show the scale of uncertainty, while a p-value alone does not. A confidence interval that includes zero does not prove equality. It indicates that zero is among the values not excluded by that interval procedure. The breadth of the interval also matters: a wide interval leaves a broad range of differences compatible with the data and signals limited precision.

What does the p-value say?

The re-calculated p-value is unadjusted and is based on a two-sided comparison. In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions held. It is not the probability that the null hypothesis is true, nor is it a direct measure of clinical importance. The re-calculated value, 0.318, should be read together with the estimate and confidence interval, rather than as a stand-alone verdict.

The registry’s reported analysis is also described as a two-sided t-test and gives p=0.05, without a reported estimate or interval for that test. The registry record does not specify additional details of that test, such as whether it was adjusted or which analysis population was used. The re-calculation from the group summaries is therefore not a substitute for a complete account of the sponsor’s analysis.

Why is the standard deviation relevant?

The standard deviations are 2.2 in the active-comparator group and 2.4 in the placebo-comparator group. Variation among participants affects how precisely group means and their difference can be estimated. Greater dispersion generally makes a mean difference harder to estimate precisely when other design features are held constant. The posted re-calculation uses a pooled standard deviation of 2.312. Pooling is a method that combines information about variability across groups under an assumption that a common variance is a suitable representation. The registry record does not state that this assumption was part of the trial’s own analysis plan.

Over-dispersion is a general concern when outcome variability exceeds what a statistical model assumes. If a method understates variability, standard errors may be too small and confidence intervals may be too narrow. The posted summaries do not establish whether over-dispersion occurred in this trial, and the record does not report diagnostics for distributional assumptions. These are general methodological considerations, not reported findings about this dataset.

What does an unadjusted re-calculation mean?

The re-calculation uses the group means, standard deviations, and group counts posted for the endpoint. It is called unadjusted because it does not include covariates or other factors. Such a calculation helps readers understand the numerical comparison implied by the summaries, but it cannot reproduce analyses that use individual-level records, covariate adjustment, a different analysis set, or additional modeling choices. The registry does not report whether any such choices were used for its own reported p-value.

Because this is a randomized trial, the group comparison is the central design-based contrast. However, randomization alone does not answer questions about missing observations, adherence, protocol deviations, or whether the analysis included every randomized participant. Those details are not reported in the registry record. The estimates should be interpreted as results reported for the analyzed group summaries, with those limitations in mind.

10. Confidence Intervals and Precision

The re-calculated mean difference is 0.8, with a 95% confidence interval that runs from below zero to above zero. This interval includes the value representing no difference and allows for compatible values in either direction. Its span indicates that the point estimate should not be treated as a precise determination of the size or direction of the underlying group difference.

Precision concerns how narrowly the data locate an estimate, not whether a result is important. The group standard deviations, 2.2 and 2.4, are large in relation to the observed mean difference, and the reported arm counts are 15 and 18. The interval’s breadth reflects uncertainty in the estimated contrast. It does not mean that all values within the interval are equally plausible, and it does not guarantee that values outside it are impossible.

The interval pertains to the mean difference as calculated from the posted summaries. It is not an interval for individual HbA1c measurements, a change from baseline, or a measure of periodontal health. The registry’s own reported t-test gives p=0.05 but does not include an associated confidence interval or effect estimate, so its precision cannot be assessed from that reported analysis alone.

In general, larger samples can improve precision when other conditions are similar, while greater outcome variability can reduce it. This trial’s posted group sizes and variability help contextualize the uncertainty but do not justify assuming what a larger or differently designed study would find. The interval should be considered alongside the endpoint definition, follow-up, and the missing information about analysis populations and missing data.

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 planning figures are participants per arm, not total event counts. The event-count convention applies to time-to-event calculations, whereas this endpoint is continuous. The figures are illustrations based on the observed mean difference and are not the trial’s original design assumptions. The original sample-size assumptions are not reported in the registry record.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. For planning at 100% of the observed effect, the figures are 132 participants per arm for 80% power and 176 participants per arm for 90% power. At 75% of the observed effect, they are 234 per arm for 80% power and 312 per arm for 90% power. At 50% of the observed effect, they are 525 per arm for 80% power and 702 per arm for 90% power.

In plain language, the 80% power column gives the planned number per arm for a design with an 80% chance of detecting the specified effect under the stated assumptions, if that effect and model assumptions are correct. The 90% column describes the corresponding planning target for a 90% chance. A higher power target requires more participants under otherwise unchanged assumptions. Smaller effects also require more participants to detect, as the increasing figures for smaller fractions illustrate.

These are planning illustrations, not observed power. They do not show the probability that the trial’s result is true or the probability that a future study will succeed. They also do not establish that the trial was designed using these assumptions.

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 = 0.800; pooled standard deviation σ = 2.312.
2. n = 2 × (1.96 + 0.842)² × 2.312² / 0.800² = 131.1, rounded up to 132 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 <- 0.799999999999999                      # observed difference in means
sigma <- 2.31182095522821                     # 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(763256)
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 = 132 (80% power), 176 (90% power)
 75% of observed effect: n per arm = 234 (80% power), 312 (90% power)
 50% of observed effect: n per arm = 525 (80% power), 702 (90% power)
power.t.test: 132.1 
Simulated power at n = 132 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. Analyses Reported by the Sponsor

The registry’s own analysis is listed as a two-sided t-test and reports p=0.05. It does not give an estimate or confidence interval for that analysis. A separate, simple unadjusted re-calculation from the posted endpoint summaries gives a mean difference of 0.8 and an unadjusted p-value of 0.318, with a 95% confidence interval crossing zero.

These results should not be conflated. The re-calculation is derived from the posted group means, standard deviations, and counts; it may not reproduce an analysis that used different data or methods. The registry record does not explain how its reported p-value was obtained or why it differs from the re-calculated value. The record does not flag a discrepancy, and there is not enough information to determine whether the analyses used different populations, adjustments, or other assumptions.

The registry’s reported result remains the sponsor’s recorded analysis, while the re-calculation is a transparent unadjusted check of the posted summaries. Neither the p-value alone nor the difference between these values establishes a clinical interpretation. Readers should consider the reported estimate and its uncertainty together with the limitations in the analysis details.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists five outcome measures and five analyses, while the posted record indicates zero secondary outcomes posted. That information does not identify which outcomes were designated primary before the results were known, how the analyses were ordered, or whether any adjustment for multiple testing was planned or used. The multiplicity strategy is not reported in the registry record.

Multiplicity is a general statistical issue that arises when several outcomes, time points, groups, or tests are examined. With more opportunities to find an unusual result, the chance of at least one apparently notable result can increase. Methods that control this issue may adjust thresholds or estimates, but no such method can be inferred from the number of registered outcomes alone. The record does not state whether the reported p=0.05 was adjusted for multiple comparisons.

The registry does not report the analysis population for the primary endpoint. It is therefore not possible to tell from the record whether the analysis included all randomized participants, only participants with observed 12-month values, or another defined group. The posted arm counts for the endpoint are 15 and 18, but those counts alone do not establish the flow of participants from randomization through follow-up or explain exclusions.

Missing-data handling is also not reported in the registry record. The record does not say whether any HbA1c measurements were missing, how missing values were treated, or whether sensitivity analyses were conducted. No assumption about the amount or pattern of missing data should be made from the endpoint summaries.

Details of the statistical analysis plan, covariate adjustment, protocol deviations, interim analyses, and outcome-selection procedures are not reported in the registry record. Without those details, the posted results cannot establish whether the reported analysis was prespecified or whether alternative analyses were considered. These gaps limit assessment of the analysis process, rather than changing the numerical summaries that are posted.

15. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The serious adverse event figures report zero affected participants in each group: 0 among 15 at risk in the Active Comparator group and 0 among 18 at risk in the B - Placebo Comparator group. The figures for other adverse events also report zero affected participants in both groups, with the same respective at-risk counts. No serious adverse event terms are listed.

These entries describe what is recorded in the registry; they do not establish that no adverse events occurred outside the recorded categories or reporting process. Death counts are null for both groups, which means a count is not reported in the registry record. A missing count should not be interpreted as a reported zero.

Adverse-event summaries are most informative when read with definitions, observation periods, collection procedures, and information about how many participants were monitored. The registry record does not provide additional details about event definitions or the safety-monitoring process. The figures should therefore be read as the recorded counts, not as a complete characterization of safety.

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

The posted primary endpoint results show mean HbA1c levels of 9.47 for the Active Comparator group and 10.27 for the B - Placebo Comparator group at 12 months. The unadjusted re-calculated mean difference is 0.8, with the placebo-comparator mean higher. The 95% confidence interval crosses zero, and the unadjusted p-value is 0.318. These data leave substantial uncertainty about the size and direction of the underlying group difference.

The registry also reports p=0.05 from a two-sided t-test, without an accompanying estimate or interval. The record does not explain how this reported analysis relates to the re-calculation. The two results should be read as distinct analyses rather than combined into a single conclusion. The registry record does not flag a discrepancy, and the information available does not support a definitive explanation for the different p-values.

The result does not show that the groups were equivalent: an interval spanning zero is not evidence that all meaningful differences have been ruled out. Nor does the observed difference alone establish that one intervention caused the other group’s mean to be higher. The record supplies no baseline outcome comparison or change-from-baseline estimate, so the result concerns endpoint means at 12 months, not change over time.

The findings do not answer every question raised by the trial title or brief summary. The primary result is about HbA1c, and no periodontal outcome result is described in the posted primary comparison. The available information also does not establish how broadly results might apply beyond the trial’s participants and setting. Missing analysis details, including the population analyzed and missing-data procedures, further limit interpretation.

In short, the posted summaries describe a numerical difference between group means, but the re-calculated interval reflects uncertainty that includes no difference. The record’s own p-value is reported separately and without supporting estimate details. Neither result should be treated as a complete account of clinical benefit, harm, or periodontal outcomes.

17. Why This Trial Matters Statistically

This trial illustrates why a reported test result and an estimated effect answer different statistical questions. A p-value describes the compatibility of data with a specified statistical model under its assumptions. An estimate gives the direction and scale of the group contrast, while a confidence interval communicates uncertainty around that estimate. For the posted HbA1c summaries, these elements do not all point to a single unambiguous interpretation: the re-calculated mean difference is 0.8, its interval crosses zero, and the registry reports a separate p-value of 0.05.

The comparison also highlights the value of making analysis details available. A reader can reproduce a simple unadjusted comparison from the group summaries, but cannot determine why the registry’s reported test yields a different p-value without information about the exact data and methods used. The record does not report an estimate or interval for the registry analysis, making it harder to assess its magnitude and precision.

Clinical research often places several questions side by side. Here, the brief summary refers to maintaining periodontal health, whereas the primary outcome listed in the results is HbA1c at 12 months. Distinguishing the stated study focus from the specific endpoint prevents an outcome result from being generalized to questions it does not directly measure.

The record also demonstrates that a completed randomized design does not eliminate uncertainty. The endpoint group counts are modest, and the recorded variability is substantial relative to the mean difference. The planning illustrations make clear that detecting smaller effects generally requires larger per-arm samples under the stated assumptions. Those illustrations are not evidence about the trial’s original planning or observed power.

Careful interpretation therefore depends on the endpoint, effect estimate, uncertainty, reported analysis, and what remains unreported. The result is most useful when these elements are considered together rather than reduced to a single p-value.

18. Limitations

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

The registry record leaves important clinical and operational details unspecified. Arm descriptions are blank, so the listed Triclosan, Fluoride and Fluoride interventions cannot be mapped in detail to the study groups from this record alone. The procedures used for periodontal treatment, the dentifrice regimen, adherence, and any other care are not reported in the registry record.

The primary outcome description identifies HbA1c measurement at 12 months and names the testing service, but does not provide baseline HbA1c summaries or a change-from-baseline analysis. The endpoint data therefore support a comparison of reported group means at follow-up, not an assessment of individual change. The record does not describe the timing of any other measurements or explain whether all outcome measures were assessed on the same schedule.

The analysis population, missing-data handling, covariate adjustment, and sensitivity analyses are not reported in the registry record. The record also does not explain the relationship between the registry’s reported p=0.05 and the unadjusted re-calculation from the group summaries. These omissions limit independent assessment of the analysis and its robustness.

The record does not report detailed eligibility criteria, recruitment setting, participant characteristics, allocation concealment, or which people were masked. It identifies the trial as randomized and double-masked, but the labels alone do not resolve those design details. The results should be understood in the context of this incomplete record, without assuming unreported protocol or analysis features.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: The Effect of Periodontal Treatment and the Use of Dentifrice on Glycaemic Control in Diabetics — NCT00763256.
• Registry record last updated 2015-07-28; status last verified 2015-07.
• Results first posted 2012-04-02.
• 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.

22. Record Summary

The registry describes a completed randomized, double-masked, parallel-group phase 3 trial sponsored by Colgate Palmolive. Recorded enrollment is 33, with primary completion in 2009-08. The study concerns periodontal disease and lists Triclosan, Fluoride and Fluoride interventions.

The primary endpoint is mean HbA1c in blood, measured in percentage units at 12 months. The recorded means are 9.47 for the Active Comparator and 10.27 for B - Placebo Comparator. The unadjusted re-calculated mean difference is 0.8, with a 95% confidence interval crossing zero and an unadjusted p-value of 0.318. The registry separately reports a two-sided t-test with p=0.05 but no associated estimate or interval.

The endpoint summaries describe a numerical group difference with uncertainty. Further details about analysis populations, missing data, intervention mapping, and the relation between the reported analysis and re-calculation are not reported in the registry record.