Anti-inflammatory Steroids in the Prevention of Tooth Sensitivity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Anti-inflammatory Steroids in the Prevention of Tooth Sensitivity (NCT02956070), a randomized na study. The primary outcome is Visual Analogic Scale (0-100) for Tooth Sensibility..

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial. The listed masking status is quadruple masking, while the brief summary calls the study triple-blind. Those descriptions differ, and the record does not explain the discrepancy or identify which roles were masked. The listed phase is NA, and the condition is inflammation.

The registry lists an enrollment of 70. The arm descriptions list 35 participants in the experimental group and 35 in the placebo group. The brief summary refers to selecting and recruiting 66 volunteers, but that wording does not match the enrollment and arm counts recorded elsewhere. The record does not explain the difference. It also does not describe how the random allocation sequence was generated or concealed.

The experimental intervention is named dexamethasone acetate. The brief summary describes dexamethasone capsules, 8mg each, and a placebo gel. The placebo arm is described as receiving placebo capsules and a desensitizing gel containing 6% potassium nitrate and 0.10% fluoride. The planned treatment involved two office whitening sessions separated by 7 days. The summary describes administering the experimental capsules before each session.

The record lists Universidade Federal Fluminense as sponsor. It records the study as completed, with a start of 2015-10 and primary completion of 2016-05. These design and timing details describe the registry entry; they do not establish whether every planned procedure was completed exactly as described.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Visual Analogic Scale (0-100) for Tooth Sensibility., measured over One hour.

The primary endpoint is the mean score on a visual analog scale for tooth sensibility, assessed one hour after bleaching. The scale runs from 0 to 100: 0 means no sensibility and 100 means maximum sensibility. The endpoint is continuous, so the posted results describe each group's average score and standard deviation, then compare the group means.

A numerical scale allows the trial to represent reported sensitivity as a graded measure rather than simply classifying each participant as having or not having sensitivity. Comparing group means estimates the average difference between the assigned regimens at the stated assessment time. The standard deviation describes how dispersed individual scores were within each group; it is not the uncertainty interval for the difference between group means.

The one-hour timing defines the assessment window for this outcome. The result therefore addresses scores at that time point, not sensitivity at other times or across the entire period after whitening. The registry lists three outcome measures, but only the primary endpoint is described in the posted results summarized for this comparison. Details of additional outcome measures are not reported in the registry record.

7. Results Posted in the Registry

• Experimental: mean 6.31 versus 11.14 on control; difference -4.83 (95% CI -12.72 to 3.06), standardized effect d = -0.29; z-test p = 0.230.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous score, and the registry identifies the summary statistic as a mean with standard deviation. A mean describes the center of the scores within a group. A standard deviation describes how spread out the individual scores are around that mean. These answer different questions: the group standard deviations describe participant-level variation, while the confidence interval describes uncertainty in the estimated difference between the group means.

What does the mean difference represent?

The posted comparison is the experimental-group mean minus the placebo-group mean. The experimental mean is 6.31 and the placebo mean is 11.14, and the posted difference is -4.83. Because the scale assigns lower values to less reported sensibility, the negative sign indicates a lower observed average score in the experimental arm. It does not by itself establish that the difference is reliable or important. The estimate must be read alongside its uncertainty interval and the underlying study design.

What does the confidence interval describe?

A confidence interval is a range of values compatible with the observed data under the statistical model and its assumptions. The interval for this mean difference runs from -12.72 to 3.055. Since it contains zero, the data do not distinguish clearly between a negative difference, no difference, and a positive difference within the interval. The interval is not a guarantee that the true value lies in the range, and it does not assign probabilities to particular values after the data have been observed. It conveys the precision of the estimate under the method used.

How does the p-value fit in?

The unadjusted p-value is 0.23. In general, a p-value measures how unusual data at least as inconsistent with a specified null hypothesis would be if that hypothesis and the model assumptions were true. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The unadjusted value should be considered together with the estimate, confidence interval, design, and completeness of the reported analysis. A detailed inferential analysis plan is not reported in the registry record.

Why does variation matter?

The experimental group’s standard deviation is 12.78 and the placebo group’s is 20.08. In general, greater variation among individual measurements makes a group mean less precise when the sample size is unchanged. Here, the spread of participant scores is substantial in relation to the observed mean difference. This contributes to uncertainty in the comparison, reflected in the confidence interval. The standard deviation should not be mistaken for a range containing all participants or for the confidence interval around the mean difference.

What is over-dispersion, and is it relevant?

Over-dispersion is a general concern in some statistical models when observed variability exceeds what the model assumes. It can make model-based uncertainty too small if not addressed. The primary endpoint is reported as a continuous measure with means and standard deviations, and the record does not describe a model assessment or an over-dispersion adjustment. It is therefore not possible to say that over-dispersion was present or that a particular adjustment was used. The broader principle is that analysis assumptions should fit the outcome distribution and design.

What can be assessed from the record?

The randomized, parallel-group description supports interpreting the endpoint as a comparison between assigned groups. However, the record does not report how the randomization sequence was generated, whether allocation was concealed, how missing values were handled, or which participants were included in the endpoint analysis. It lists quadruple masking, while the brief summary says triple-blind, without reconciling the descriptions. These gaps prevent a full evaluation of the analysis procedures and potential sources of bias.

The record lists three outcome measures and no posted secondary outcomes in the results information summarized for this comparison. It also lists zero analyses. This does not establish that no analyses were conducted; it means detailed analyses are not reported in the registry record. No adjustment for multiple comparisons or prespecified analysis hierarchy is described. In general, examining multiple outcomes or analyses can increase the chance of a nominally unusual result, so prespecification and transparent reporting help readers interpret findings.

Statistical interpretation should therefore remain tied to what is documented: a mean difference of -4.83 units, its confidence interval, the unadjusted p-value, group summaries, and the stated one-hour assessment. The inferential assumptions and data-handling details beyond those details are not reported in the registry record.

10. Confidence Intervals and Precision

The estimated mean difference is -4.83 units, and its confidence interval extends from -12.72 to 3.055 units. The interval is wide enough to include zero and values on both sides of zero. This means the estimate does not pinpoint a single direction or magnitude of difference with high precision. The point estimate is the center of the reported comparison, but it should not be read without the interval.

In general, a confidence interval expresses the uncertainty around an estimate under a statistical method and its assumptions. A narrower interval indicates greater precision; a wider interval indicates less precision. The interval permits a range that includes an experimental-group mean lower than the placebo-group mean, no difference, and an experimental-group mean higher than the placebo-group mean. It is not evidence that every value in the interval is equally plausible, and it is not a prediction interval for individual participants.

The observed standard deviations were 12.78 in the experimental group and 20.08 in the placebo group. These describe variation in individual scores, not precision of the group difference. The confidence interval incorporates uncertainty in that comparison. The number of participants recorded in each arm is 35, but the analysis population and completeness of endpoint measurements are not reported in the registry record. Those details matter when evaluating precision and potential bias from missing observations.

The unadjusted p-value of 0.23 is consistent with the uncertainty conveyed by the confidence interval. Neither statistic alone determines whether a difference matters clinically. The scale provides a possible interpretation of score direction, but a threshold for a clinically important difference is not reported in the registry record. Consequently, the precision information supports cautious interpretation of the range of possible group differences, not a categorical conclusion about benefit or lack of benefit.

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 events. They are planning illustrations based on the observed effect, rather than the original trial’s design calculation. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row uses the stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true one. The 90% power column gives the corresponding per-arm planning number for a 90% chance. Power describes the behavior of a planned study under assumptions; it is not a measure of the probability that a finding is true.

The rows consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The corresponding figures are 191 and 256 participants per arm; 339 and 454 participants per arm; and 763 and 1,021 participants per arm, respectively, for 80% and 90% power. Smaller effects need more participants to detect under otherwise unchanged assumptions.

These calculations are not observed power. They are hypothetical planning illustrations, not a retrospective measure of how successful the completed trial was. They rely on the observed effect as a reference, which is itself uncertain, and should not be treated as a confirmed effect size or as a recommendation for a future study.

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 = -4.830; pooled standard deviation σ = 16.831.
2. n = 2 × (1.96 + 0.842)² × 16.831² / -4.830² = 190.6, rounded up to 191 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 <- -4.83                      # observed difference in means
sigma <- 16.8305496048109                     # 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(2956070)
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 = 191 (80% power), 256 (90% power)
 75% of observed effect: n per arm = 339 (80% power), 454 (90% power)
 50% of observed effect: n per arm = 763 (80% power), 1,021 (90% power)
power.t.test: 191.6 
Simulated power at n = 191 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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures, while the posted results summarized for the primary comparison include the tooth-sensitivity score. The number of secondary outcomes posted is zero, and the record lists zero analyses. A detailed list of secondary results, analysis populations, and any prespecified hierarchy among outcomes is not reported in the registry record.

Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are considered. In general, the more comparisons made, the greater the chance of finding at least one apparently unusual result by chance. Prespecified primary outcomes and methods for addressing multiple comparisons can help readers distinguish confirmatory analyses from exploratory ones. The registry record does not report whether any multiplicity adjustment was planned or used. The unadjusted p-value should therefore be understood as unadjusted, rather than as a value known to account for other analyses.

The record does not state how missing outcome measurements were handled, whether all randomized participants were included in the analysis, or whether analyses followed an intention-to-treat or another population definition. These details are not reported in the registry record. Without them, the effect of missingness or exclusions on the reported comparison cannot be evaluated.

Randomization is recorded, but the sequence-generation and allocation-concealment procedures are not described. Masking is listed as quadruple, while the brief summary describes the trial as triple-blind; the record does not reconcile those terms or identify who was masked. These reporting gaps do not demonstrate that a procedure was absent. They limit the ability to assess how the trial was conducted and how the analysis handled its data.

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 serious adverse event figures list 35 participants at risk in the experimental group and 35 in the placebo group. The recorded number affected by a serious adverse event is 0 in each group, and deaths are recorded as 0 in both groups. These are the registry’s posted counts for the groups and safety categories listed.

A count of zero means that no serious adverse events were recorded in these entries; it does not prove that no events occurred outside the reporting definitions or observation period. The record does not provide detailed event descriptions, follow-up duration for safety, or information about non-serious events. It also does not report whether every participant contributed complete safety follow-up.

These figures should be read as descriptive safety reporting, not as a definitive comparison of risks. Zero recorded serious events in both groups does not establish equivalence or rule out uncommon harms. The registry lists no serious-event terms among its top serious terms. Interpretation is limited to the categories and counts recorded.

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

The posted results describe lower average tooth-sensitivity scores in the experimental arm than in the placebo arm at the one-hour assessment: 6.31 compared with 11.14. The estimated mean difference is -4.83 units. The confidence interval, from -12.72 to 3.055 units, includes zero and spans values in both directions. The unadjusted p-value is 0.23. Together, these results indicate substantial uncertainty about the size and direction of the group difference.

The results do not establish that the regimens produced identical average scores. A confidence interval containing zero can be consistent with no difference, but it can also be consistent with nonzero differences within its limits. Nor do the results establish a reliable difference. The point estimate alone is not enough to determine how strongly the data support any particular conclusion.

The finding applies to the primary score measured one hour after bleaching in the trial population and under the recorded procedures. It does not answer questions about later sensitivity, other whitening settings, other populations, or outcomes not reported. The registry record does not report a threshold for a clinically important score difference, so the clinical importance of the estimated difference cannot be determined from these statistics alone.

The absence of recorded serious adverse events in the listed safety entries is limited to those counts and categories. It does not prove the absence of all adverse events or establish comparative safety. The registry’s missing details about missing data, analysis populations, and analysis procedures also constrain interpretation. The most defensible reading is descriptive: the two recorded group means differed by the posted estimate, with an interval that leaves considerable uncertainty.

16. Why This Trial Matters Statistically

This trial illustrates why a comparison should be read through its estimate and uncertainty, rather than through a single p-value or the direction of the point estimate alone. The experimental group’s mean score was lower, and the mean difference was -4.83 units. Yet the confidence interval extends across zero, and the unadjusted p-value is 0.23. A reader who focuses only on the lower experimental mean could overstate what the data establish; a reader who treats the interval as proof of no difference could also overstate the evidence.

The comparison also highlights the distinction between participant-level variation and uncertainty in a treatment contrast. The group standard deviations describe the spread of scores among participants, while the interval for the mean difference expresses uncertainty about the estimated group contrast. In general, noisy measurements and limited sample sizes can make estimates imprecise, even when the point estimate appears to favor one group.

Planning illustrations based on the observed effect demonstrate how the required sample size depends on the effect a future study aims to detect. Smaller effects require more participants under the stated assumptions. These illustrations are not observed power and do not substitute for a prospectively justified design calculation.

Finally, registry reporting matters for statistical appraisal. The record gives a randomized, parallel-group design and outcome summaries, but it does not describe the analysis population, missing-data handling, or multiplicity plan. It also uses different masking descriptions in different parts of the entry. Clear reporting of design and analysis decisions is essential for judging how much confidence to place in a result.

17. Limitations

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

The registry record leaves several trial-specific details unresolved. It lists an enrollment of 70 and 35 participants per arm, while the brief summary says 66 volunteers would be selected and recruited. The record does not explain this difference or specify how many participants contributed observed primary-endpoint measurements. The analysis population and missing-data handling are not reported in the registry record.

The masking description is also inconsistent within the record: the listed status is quadruple masking, while the brief summary calls the trial triple-blind. The record does not explain the terminology or identify which participants, clinicians, assessors, or other roles were masked. Randomization is recorded, but sequence generation and allocation concealment are not described.

The intervention descriptions use different levels of detail. The experimental intervention is named dexamethasone acetate, while the summary describes dexamethasone capsules and a placebo gel. The placebo arm description names potassium nitrate, and the summary specifies a desensitizing gel with 6% potassium nitrate and 0.10% fluoride. The registry does not provide enough information to resolve any differences between these descriptions.

The primary endpoint is measured at one hour, so the posted comparison is limited to that assessment time. A clinically important difference threshold, detailed outcome-assessment procedures, and results for the other listed outcome measures are not reported in the registry record. The record also does not report a detailed statistical analysis plan, multiplicity approach, or the trial’s original sample-size assumptions. These reporting limitations restrict evaluation of the result’s precision, robustness, and broader applicability.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Anti-inflammatory Steroids in the Prevention of Tooth Sensitivity — NCT02956070.
• Registry record last updated 2020-11-25; status last verified 2020-11.
• Results first posted 2020-11-25.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT02956070 is a completed randomized, parallel-group trial sponsored by Universidade Federal Fluminense. The record lists an enrollment of 70, with 35 participants in each of the experimental and placebo arms. The listed masking status is quadruple, although the brief summary describes the study as triple-blind.

The primary outcome is the mean visual analog scale score for tooth sensibility one hour after bleaching, on a scale from 0 to 100. The experimental-group mean was 6.31, compared with 11.14 in the placebo group. The posted mean difference was -4.83 units, with a confidence interval from -12.72 to 3.055 and an unadjusted p-value of 0.23.

The registry records zero serious adverse events and zero deaths in each group. Details on the original sample-size assumptions, analysis populations, missing-data handling, and multiplicity procedures are not reported in the registry record.