A Randomized, Self Controlled Clinical Study of Hemostatic Efficacy of the HemCon Dental Dressing: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Randomized, Self Controlled Clinical Study of Hemostatic Efficacy of the HemCon Dental Dressing (NCT00707486), a randomized phase 3 study. The primary outcome is Time to Hemostasis.

2. Clinical Question

The study’s stated purpose was to assess hemostatic efficacy during dental surgery. Its primary endpoint, time to hemostasis, is a continuous measure: each recorded value represents elapsed time in minutes until bleeding stopped. The central statistical comparison is the mean time associated with HemCon versus the mean time associated with gauze-based care.

The trial is described as randomized and self-controlled, with a crossover model. Participants served as their own controls, so the two treatment-condition results are not simply results from two separately recruited groups. The posted means are 1.9 minutes for HemCon and 9.8 minutes for control: the recalculated mean difference, expressed as HemCon minus control, is -7.9 minutes. The confidence interval for that difference extends from -16.11 to just above zero. These descriptive and recalculated results address the clinical question, but the registry record does not report the sponsor’s own analysis or enough detail to establish how the crossover data were modeled.

3. Trial Design

The study is identified as a Phase 3 randomized, self-controlled clinical study with a crossover model. It enrolled 72 participants and is listed as completed. Tooth extraction is the condition under study. The two interventions were HemCon Dental Dressing and gauze with pressure and/or Gelfoam. The control outcome is labeled Gauze in the posted results. The registry describes participants as serving as their own control, and the primary outcome is reported separately for the two intervention conditions.

A self-controlled crossover design compares treatment conditions within participants rather than treating the conditions as independent groups of people. In general, this can reduce the influence of participant characteristics that remain stable across conditions, because those characteristics are shared by both observations. It also means that statistical analysis should account for the relationship between observations from the same participant. The registry does not report the treatment sequence, the method used to randomize sequences, any washout or timing arrangements, or the model used for the sponsor’s primary analysis. Those details should not be inferred from the design label alone.

Masking is listed as none. The registry record does not report whether outcome assessment was performed by an independent assessor or whether any other steps were used to limit knowledge of treatment condition. The primary completion date is listed as 2009-01, and the status is completed. The registry lists one secondary outcome, incidence of post-surgical sequelae, in addition to the primary outcome. It does not report the trial’s original sample-size justification or detailed protocol assumptions. The posted outcome sample is 71 for each condition, while total enrollment is 72; the reason for that difference is 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants served as their own control thus the total amount of participants is reflected in both of the outcome measures.

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 Time to Hemostasis, measured over Minutes After Application.

The primary endpoint is time to hemostasis, measured in minutes after application. It records how long it takes for an extraction site to stop bleeding. The registry reports the endpoint as a mean for each intervention condition, with standard error as the measure of dispersion. The posted mean is 1.9 minutes for HemCon and 9.8 minutes for control: smaller elapsed time corresponds to bleeding stopping sooner in the recorded observations.

Time to hemostasis is used because it directly quantifies the timing of the event the study set out to assess. A continuous endpoint preserves information about how long participants waited, rather than reducing observations to a yes-or-no classification at an unspecified time. Comparing means offers a concise description of average elapsed time in the two conditions. The measure does not, on its own, describe the full distribution of times, the number of participants with especially long times, or how individual paired results varied. Those features would require additional summaries or analyses, which are not reported in the registry record.

In a crossover study, the endpoint is observed under more than one condition for the same participant. That structure makes within-participant comparisons relevant. The registry confirms the self-controlled design but does not report the exact analysis used to account for paired observations. Thus, the posted mean times describe each condition, while the recalculated mean difference is an unadjusted comparison rather than a reported sponsor analysis.

7. Results Posted in the Registry

• HemCon: mean 1.90 versus 9.80 on control; difference -7.90 (95% CI -16.11 to 0.31), standardized effect d = -0.32; z-test p = 0.059.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is continuous: it measures elapsed time in minutes. The registry reports a mean for each condition and identifies standard error as the dispersion measure. A mean is the arithmetic average of observations. It provides a concise measure of the center of the recorded times, but it does not show whether observations cluster closely around that average or vary widely. A standard error describes the precision of an estimated mean. It is not the same as the standard deviation of individual times and should not be interpreted as the range containing most participants.

What does the mean difference describe?

The posted comparison is a mean difference of -7.9 minutes, with HemCon first and control second in the contrast as represented by the posted result. A difference is expressed in the same units as the outcome, so this estimate is in minutes. A negative difference indicates that the HemCon mean is lower than the control mean. In this study, the condition means are 1.9 and 9.8 minutes, respectively. The contrast gives one summary of the gap between those averages; it does not report how each participant’s two measurements differed.

The study’s crossover, self-controlled design matters for analysis. In general, measurements from one participant are correlated because they share participant characteristics and study circumstances. A paired analysis uses the within-participant relationship, whereas an analysis that treats the conditions as independent may not. The registry does not identify the sponsor’s analysis model or provide the participant-level paired differences or their standard deviation. The displayed comparison is a recalculated unadjusted result from posted summaries, and should not be taken as proof that the within-participant structure was modeled in the same way as in the trial’s intended analysis.

How should the p-value be read?

In general, a p-value describes how incompatible the observed data, or data more extreme, are with a specified null model, assuming that model and the analysis assumptions are appropriate. It is not the probability that the null hypothesis is true, nor is it the probability that the result happened “by chance.” The recalculated unadjusted p-value is 0.059. It is close to, but above, 0.05, the significance level stated for the planning illustration. This does not prove there is no difference; it indicates that the recalculated test does not meet that conventional cutoff. A p-value also does not convey effect size, clinical importance, or precision by itself.

For a paired or crossover analysis, the test statistic depends on the variability of within-participant differences. The registry does not report that variability or the sponsor’s inferential method. Therefore, the recalculated p-value should be interpreted together with the fact that it is unadjusted and based on posted summary statistics. The trial’s own analysis results are not reported in the registry record.

What does the confidence interval add?

In general, a confidence interval describes a range of values compatible with the data and the method’s assumptions, while reflecting sampling uncertainty. It is not a guarantee that the true value lies inside the particular interval. The recalculated interval for the mean difference extends from -16.11 minutes to just above zero. Because zero lies within that interval, the posted result does not rule out no mean difference under the recalculation’s assumptions. The interval also includes differences favoring either direction, which signals substantial uncertainty about the size and direction of the underlying mean contrast.

The interval is wide in relation to the point estimate, so the point estimate alone should not be treated as a precise estimate of the difference. The registry record does not report a sponsor-generated interval or clarify whether its primary analysis used the same approach as the recalculation. General interpretation of an interval always depends on the chosen model, the data structure, and the assumptions used to estimate uncertainty.

Why can over-dispersion matter?

In general, over-dispersion means that observed variation is greater than a statistical model expects. When it is present but ignored, standard errors can be too small and confidence intervals too narrow, making results appear more precise than warranted. The endpoint is continuous, and the registry does not report a count model or an over-dispersion assessment. This concept is relevant as a general statistical caution, not as a finding about this trial. The registry’s standard errors for the condition means are 1.36 and 3.96 minutes, but they do not by themselves establish whether a chosen model adequately represents the data.

Other general considerations for continuous time outcomes include skewness and influential long times. Means can be affected by unusually large observations. The registry record does not provide distributions, medians, ranges, or participant-level observations, so those features cannot be assessed from the posted summaries. No assumptions about the shape of the time distribution should be inferred.

The design and reporting context therefore governs the methods reading: the means describe the recorded conditions, while the recalculated contrast and inferential statistics are not the sponsor’s posted analysis. The registry does not report details about sequence effects, period effects, missing-data handling, covariate adjustment, analysis populations, or multiplicity procedures. Those omissions prevent a fuller assessment of whether the statistical model matched the crossover design and the trial’s analysis plan.

10. Confidence Intervals and Precision

The recalculated mean difference is -7.9 minutes, and its confidence interval runs from -16.11 minutes to an upper limit just above zero. The interval is the most direct indication of the uncertainty attached to that estimate. Its lower end represents a sizeable negative contrast, while the upper end permits a value slightly greater than zero. Thus, the posted interval spans a range that includes no difference and does not locate the underlying mean contrast precisely on one side of zero.

In general, a confidence interval is a range generated by a statistical procedure that, under its assumptions, has a specified long-run coverage property. It does not mean there is a particular probability that the true value lies in this one realized interval. Nor does it account for every possible source of uncertainty, such as bias, missing outcomes, or a mismatch between the model and the study design.

Here, the interval belongs to an unadjusted recalculation from summary statistics. The registry does not report a sponsor-generated interval or the full paired data needed to evaluate the crossover structure. The interval’s precision should therefore be interpreted in the context of that limitation. Its inclusion of zero is consistent with the recalculated p-value of 0.059, though the p-value and interval express uncertainty in different forms. The interval is not evidence that the conditions are equivalent, and its width does not establish that any particular difference is clinically important. The registry record does not report a threshold for clinical importance.

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. The outcome is continuous rather than time-to-event, so the figures represent participants per arm. The registry’s planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the stated fraction of the observed effect as the effect to detect. The illustration is based on the observed effect, not the trial’s own design; the original sample-size assumptions are not reported in the registry record.

For planning at 100% of the observed effect, the estimates are 157 participants per arm for 80% power and 210 participants per arm for 90% power. For planning at 75% of the observed effect, they are 279 and 373 participants per arm, respectively. For planning at 50% of the observed effect, they are 627 and 839 participants per arm, respectively. These are illustrations under the stated assumptions, not observed trial outcomes.

In general, power is the probability that a specified analysis will detect an effect of a chosen size if that effect is present and the assumptions hold. The 80% column describes a planning scenario with an 80% chance of detecting the assumed effect; the 90% column uses a higher target chance. Smaller effects are harder to distinguish from random variation, so planning for a smaller effect requires more participants. The plotted sample-size relationship should be read in that same way: the required sample grows as the target effect becomes smaller. This is not observed power.

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 = -7.900; pooled standard deviation σ = 24.947.
2. n = 2 × (1.96 + 0.842)² × 24.947² / -7.900² = 156.5, rounded up to 157 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 <- -7.9                      # observed difference in means
sigma <- 24.9470960233852                     # 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(707486)
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 = 157 (80% power), 210 (90% power)
 75% of observed effect: n per arm = 279 (80% power), 373 (90% power)
 50% of observed effect: n per arm = 627 (80% power), 839 (90% power)
power.t.test: 157.5 
Simulated power at n = 157 per arm: 0.801 (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 the primary outcome, time to hemostasis, and one secondary outcome, incidence of post-surgical sequelae. It reports two outcome measures and zero analyses. No analysis results for the secondary outcome are posted. The registry record does not report a multiplicity adjustment, a testing hierarchy, or which outcomes were included in any family of statistical tests.

In general, multiplicity arises when several outcomes, time points, subgroups, or comparisons are tested. The more tests conducted, the greater the chance that at least one result appears statistically notable even when no corresponding effect exists. Procedures such as adjusting a significance threshold or specifying a hierarchy can address this issue, but no such plan is reported in the registry record. The posted recalculated p-value should therefore be understood as an unadjusted result, not as a multiplicity-controlled inference.

The outcome count shows 71 participants in each condition, compared with enrollment of 72. The reason for the difference is not reported in the registry record. It is also not reported whether the same 71 participants contributed measurements under both conditions, how missing measurements were handled, or whether any participant was excluded from the analysis. The trial’s analysis populations, such as an intention-to-treat or per-protocol population, are not reported.

For a self-controlled crossover study, missing one condition’s measurement can affect whether a participant contributes to a paired comparison. The registry does not state whether incomplete pairs were included in any analysis or how they were treated. It also does not report an analysis plan for period or sequence effects. These are not findings that such issues occurred; they are details needed to evaluate the analysis that are absent from the registry record. With zero analyses posted, the registry does not allow a full assessment of multiplicity, missing-data methods, or the population underlying the sponsor’s intended inference.

14. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The posted adverse-event group is labeled Hemcon Dental Dressing and Gauze With Pressure. It lists 71 participants at risk for serious adverse events and zero affected by a serious event. For other adverse events, 71 participants were at risk and 20 were affected. No serious adverse-event terms are listed. Deaths are not reported in the registry record.

These counts describe the posted safety entries for that combined group. They do not provide event-level detail for the 20 participants with other adverse events, and the registry does not report separate safety counts for HemCon and control conditions. Because the group label includes both HemCon Dental Dressing and gauze with pressure, the counts should not be interpreted as a condition-by-condition comparison. The record also does not report the nature, timing, severity, or relatedness of the other events in this summary.

In general, zero reported serious events in a group is a count, not proof that no serious event could occur. The registry’s entry indicates that none were recorded as affecting participants in the listed serious-event category. It does not establish a broader safety conclusion beyond the information recorded. The posted figures do not support a comparative safety estimate between the two intervention conditions.

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

The results describe a lower posted mean time to hemostasis for HemCon than for gauze-based control: 1.9 minutes compared with 9.8 minutes. The recalculated mean difference is -7.9 minutes, with HemCon minus control as the direction of comparison. This is the observed summary contrast in the registry. It does not, by itself, establish that the underlying difference is precisely estimated or that it would reproduce in other participants or settings.

The recalculated confidence interval includes zero, and the unadjusted p-value is 0.059. Those statistics indicate uncertainty under the recalculation’s assumptions. They do not prove that the interventions have equal mean times, and they do not establish a clinically important difference. The registry does not report a threshold for clinical importance or the sponsor’s own primary analysis. The evidence should therefore be described as a posted pattern in the means accompanied by an imprecise, unadjusted comparison, rather than as a definitive efficacy conclusion.

General statistical interpretation also distinguishes absence of evidence from evidence of absence. A result that does not meet a conventional significance threshold may reflect a small effect, substantial variability, limited information, or a combination of factors. The posted summaries do not identify which explanation applies. The trial enrolled 72 participants, with 71 listed for each condition, and the reason for the unmatched enrollment count is not reported.

The crossover design makes the participant-level pairing relevant. Since the registry does not report the sponsor’s model, sequence details, or paired data, the recalculated comparison cannot answer every design-specific question. The outcome results do not establish how the pattern varies between participants, whether all enrolled participants contributed complete measurements, or whether the same results hold across different clinical circumstances. The record supports a careful description of its posted means and recalculated statistics, but not a medical recommendation or a general claim about treatment performance.

16. Why This Trial Matters Statistically

This trial is statistically notable because its randomized, self-controlled crossover design makes the within-participant structure central to interpretation. Participants served as their own controls, so each participant’s results under the two conditions are related rather than independent observations. A statistical analysis that accounts for that relationship may use information that is not visible in separate condition means. The registry confirms the design but does not report the sponsor’s model or paired-data summaries.

The posted condition means differ, but the recalculated interval includes zero and the unadjusted p-value is 0.059. This illustrates why an effect estimate, its uncertainty, and its analysis context should be read together. The mean difference gives the direction and scale in minutes; the interval conveys uncertainty; the p-value summarizes compatibility with a null model under specific assumptions. None of these alone establishes clinical importance.

The registry record also illustrates the difference between a trial’s posted outcome summaries and a full statistical report. It lists 72 enrolled participants and outcome counts of 71 per condition, but does not explain the discrepancy. It identifies one secondary outcome but posts no analyses. It does not report the original sample-size justification, missing-data procedures, or multiplicity strategy. These gaps matter because they shape how confidently readers can connect a trial design to an inferential conclusion.

Finally, the sample-size illustration gives a way to understand how planning changes with the target effect and desired power. It is distinct from the trial’s original design assumptions and from an assessment of what the completed study achieved. Keeping those ideas separate helps prevent the observed result from being treated as a direct statement about power or as a definitive answer to the clinical question.

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 main limitation for statistical interpretation is that the registry does not report the sponsor’s primary analysis. The mean difference, confidence interval, and p-value are recalculated unadjusted results from posted summaries. Because this is a self-controlled crossover study, the relationship between each participant’s measurements can affect the appropriate estimate and its uncertainty. Participant-level paired results and the model used for the intended analysis are not reported, so the recalculation cannot be treated as a substitute for that analysis.

The registry reports 72 enrolled participants but lists 71 for each condition. The reason for the difference, the number with complete paired measurements, and the handling of missing data are not reported. Analysis populations and exclusions are also not reported. These omissions make it difficult to determine exactly which participants contributed to the displayed comparisons.

The registry does not report the treatment sequence, randomization details, timing between conditions, or procedures for addressing period or sequence effects. Masking is listed as none, and additional details about outcome assessment are not reported. The record also does not provide distributions or participant-level values that would allow assessment of skewness, outliers, or variation in paired differences.

Only one secondary outcome is identified, but its results are not posted. Multiplicity procedures and the trial’s original sample-size justification are not reported. The adverse-event summary combines HemCon Dental Dressing and gauze with pressure, so it does not provide a separate safety comparison by condition. These limitations constrain interpretation of the posted registry results; they do not establish that any particular methodological problem occurred.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Randomized, Self Controlled Clinical Study of Hemostatic Efficacy of the HemCon Dental Dressing — NCT00707486.
• Registry record last updated 2013-01-09; status last verified 2012-12.
• Results first posted 2013-01-09.
• 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

NCT00707486 is a completed Phase 3 randomized, self-controlled crossover study of HemCon Dental Dressing versus gauze-based care for tooth extractions. The primary endpoint was time to hemostasis in minutes after application. The registry reports means of 1.9 minutes for HemCon and 9.8 minutes for control, with 71 participants listed for each condition.

The recalculated unadjusted mean difference is -7.9 minutes, with a confidence interval extending from -16.11 to just above zero and a p-value of 0.059. The registry does not report the sponsor’s own analysis. It lists one secondary outcome and zero analyses. The original sample-size justification, analysis populations, missing-data methods, and multiplicity procedures are not reported in the registry record.