Benzocaine Gel Toothache Dose-Response Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Benzocaine Gel Toothache Dose-Response Study (NCT00474175), a randomized phase 4 study. The primary outcome is Percentage of Participants With a Response.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-masked, parallel-group Phase 4 trial and records its status as completed. The study title is Benzocaine Gel Toothache Dose-Response Study. It lists Toothache as the condition and reports an enrollment of 577. The recorded start was 2007-05, and primary completion was 2010-12. The sponsor is identified as Wyeth is now a wholly owned subsidiary of Pfizer.

There were three groups: placebo gel, benzocaine 10% gel, and benzocaine 20% gel. The primary outcome results use 115 participants for placebo, 233 for benzocaine 10%, and 228 for benzocaine 20%. These are the outcome-specific denominators in the posted results; the registry's overall enrollment is a separate figure. Details about the randomization sequence, allocation concealment, and the practical procedures used to maintain masking are not reported in the registry record.

The stated primary analysis population was intent to treat, defined in the record as all randomized participants who received study medication. The primary endpoint was assessed at baseline and at 5, 10, 15, and 20 minutes. The registry lists additional outcomes concerning relief, duration, pain scores, rescue medication or dropout, and satisfaction. Further details of the trial protocol, including the original sample-size justification, prespecified interim analyses, and the full statistical analysis plan, are not reported in the registry record. The record therefore supports describing the design and posted results, but not reconstructing unreported design decisions.

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: Intent to treat (ITT) population included all randomized participants who received study medication.

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 Percentage of Participants With a Response, measured over Baseline, 5, 10, 15 and 20 minutes.

The primary endpoint was the percentage of participants classified as responders. A responder had to show a reduction of at least 1 unit from baseline on the Dental Pain Scale for two consecutive assessments at any time between 5 and 20 minutes. The scale values were 0 (None), 1 (Mild), 2 (Moderate), and 3 (Severe). The assessment schedule included baseline and 5, 10, 15, and 20 minutes.

This endpoint turns repeated pain-intensity assessments into a binary outcome: each participant either met the response definition or did not. The percentage therefore describes the share of participants meeting that specified criterion, rather than the size of each participant's improvement or the duration of relief. The two-consecutive-assessment condition makes the definition depend on a sustained pattern across adjacent assessments, not simply a single favorable score.

General statistical context: binary response outcomes are often used to make group comparisons easy to interpret as differences in the proportion meeting a clinically defined criterion. The result still depends on the chosen response threshold and assessment window. The registry's definition makes those features explicit, but the clinical importance of meeting the criterion, beyond the recorded definition, is not established by the endpoint alone.

7. Results Posted in the Registry

• Benzocaine 10%: 80.7% (188/233) versus 70.4% (81/115) on control; risk difference 10.3 percentage points (95% CI 0.5 to 20.0), risk ratio 1.15 (1.00–1.31); unadjusted two-proportion z-test p = 0.032.
• Benzocaine 20%: 87.3% (199/228) versus 70.4% (81/115) on control; risk difference 16.8 percentage points (95% CI 7.5 to 26.2), risk ratio 1.24 (1.09–1.41); unadjusted two-proportion 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.

Global Satisfaction Assessment

• 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:
  • Benzocaine 10%: mean 2.50 versus 2.10 on control; difference 0.40 (95% CI 0.07 to 0.73), standardized effect d = 0.30; z-test p = 0.018.
  • Benzocaine 20%: mean 2.50 versus 2.10 on control; difference 0.40 (95% CI 0.07 to 0.73), standardized effect d = 0.30; z-test p = 0.018.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Benzocaine 10% than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Benzocaine 20% than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The registry lists secondary outcomes covering time to first confirmed perceptible relief, time to meaningful relief, duration of effect, pain-relief and pain-intensity summary scores, time to dropout due to lack of efficacy or rescue medication, and global satisfaction. These outcomes address related but distinct aspects of the study question. The posted global satisfaction result is measured at 120 minutes, whereas the primary responder endpoint concerns assessments between 5 and 20 minutes. The secondary outcomes therefore should not be treated as interchangeable measures of the primary endpoint.

For global satisfaction, the posted means are 2.1 for placebo, 2.5 for benzocaine 10%, and 2.5 for benzocaine 20%, with standard deviations of 1.6, 1.2, and 1.2, respectively. The sponsor's reported analyses give treatment differences of 0.17 for placebo versus benzocaine 20% and 0.16 for placebo versus benzocaine 10%. The registry lists additional secondary outcomes, but results for those outcomes are not reported in the registry record.

The primary result concerns a binary early-response definition; satisfaction at 120 minutes is a different construct and time point. Secondary findings can provide context about other aspects of participants' experience, but they do not replace or directly confirm the primary endpoint. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for secondary testing. General statistical context: when many outcomes or comparisons are examined, unadjusted p-values can overstate the strength of evidence because the chance of at least one apparently notable result increases with the number of tests. For that reason, unadjusted secondary results generally carry less weight than a clearly prespecified primary analysis, and should be interpreted as supportive or exploratory unless the analysis plan establishes otherwise.

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

• Time to First Confirmed Perceptible Relief
• Time to Meaningful Relief
• Duration of Effect
• Sum of Pain Relief Combined With Pain Intensity Differences (SPRID) Scores
• Time to Dropping Out Due to Lack of Efficacy or Rescue Medication
• Pain Relief Combined With Pain Intensity Difference (PRID) Scores
• Global Satisfaction Assessment

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary because every participant is classified as meeting or not meeting the responder definition. The posted group percentages describe the observed fraction of participants who met that definition. In general, a binary outcome can be compared using an absolute difference in proportions, a relative risk, or an odds ratio. These measures express different aspects of a group contrast and should not be treated as interchangeable.

What does a risk difference describe?

A risk difference subtracts one group's event proportion from the other group's event proportion. For this endpoint, the event is being classified as a responder. The simple unadjusted recalculation gives risk differences of 0.102 for benzocaine 10% versus placebo and 0.169 for benzocaine 20% versus placebo. The associated 95% confidence intervals are 0.005 to 0.200 and 0.074 to 0.262. These recalculations use the posted responder counts and group sizes without adjustment. The registry's reported treatment differences are separate estimates from its stated Cochran-Mantel-Haenszel analysis.

General statistical context: an absolute difference is often easier to connect to the observed proportions than a relative measure, but it does not show how much an individual participant's pain changed. It also depends on the event definition and the comparator group's event proportion. Here, the event is not any improvement; it is the specific response pattern recorded for the Dental Pain Scale across the specified assessments.

How do relative risk and odds ratio differ?

The simple recalculation also reports relative risks of 1.146 for benzocaine 10% versus placebo and 1.239 for benzocaine 20% versus placebo. The corresponding odds ratios are 1.754 and 2.88. General statistical context: relative risk compares the proportions directly, while an odds ratio compares odds, where odds are the event proportion divided by the non-event proportion. An odds ratio is not the same quantity as a relative risk; the size of the numerical distinction depends in part on how common the event is. Neither measure changes what the endpoint means, and neither should be read as a measure of pain reduction for an individual.

What do the confidence intervals and p-values contribute?

A confidence interval gives a range of values compatible with the observed data and the method's assumptions. It conveys uncertainty as well as the estimate's direction and scale. A p-value assesses how incompatible the data are with a specified null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. The recalculated p-values are unadjusted: 0.032 for benzocaine 10% versus placebo and <0.001 for benzocaine 20% versus placebo. The registry reports p = 0.038 and p < 0.001 for those placebo comparisons using its own analysis.

The sponsor's posted primary method is Cochran-Mantel-Haenszel. The registry names that method and reports treatment differences, confidence intervals, and p-values. Further detail about the exact analysis specification, including any stratification variables or other adjustments, is not reported in the registry record. It would therefore be inappropriate to infer those details from the method name alone. The sponsor's estimates take precedence because the simple recalculations do not reproduce the posted analysis.

Why can analysis methods yield different estimates?

General statistical context: an unadjusted calculation from group summaries can differ from a trial analysis that uses a method accounting for additional structure in the data. Differences can also arise from analysis populations, missing-data handling, or the precise definition of the estimated contrast. The record does not identify which, if any, of these considerations explains the discrepancy in this case. The appropriate conclusion is limited to noting that the posted analyses are not identical and that the registry's own result is the primary reported analysis.

What does the response definition add?

The endpoint requires a reduction of at least 1 unit for two consecutive assessments between 5 and 20 minutes. The scale has four recorded levels, from 0 (None) to 3 (Severe). General statistical context: defining a binary response threshold can make a result straightforward to communicate, but it discards information about differences among participants who do and do not cross that threshold. A participant just meeting the criterion and one with a larger change are both counted as responders. Likewise, participants with some improvement that does not satisfy the full rule are not classified as responders under this endpoint.

The assessment schedule and rule mean that the endpoint addresses early response, not the full course of toothache relief. The registry lists other outcomes related to timing, duration, pain scores, rescue medication, and satisfaction, but the primary endpoint alone cannot answer those separate questions. The registered measurement is useful for its defined comparison, provided its scope is kept clear.

What can planning calculations say?

The sample-size illustrations use the observed effect as a reference and consider planning for 100%, 75%, and 50% of that observed effect. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For non-time-to-event outcomes, the listed figures are participants per arm; time-to-event calculations, if present, would instead report total events. The figures are planning illustrations, not results from the trial's original design. The original design assumptions are not reported in the registry record. These calculations are not observed power.

General statistical context: smaller effects generally require more information to detect with a given power and significance level. Power is a design-stage probability under specified assumptions, not a measure of how likely the observed result is to be true. A planned sample size based on an effect estimate can be sensitive to that estimate and to assumptions about variability or event frequency. The registry record does not report the trial's own sample-size justification, so the illustrations should not be mistaken for evidence of what the investigators planned.

11. Confidence Intervals and Precision

The confidence intervals for the registry's primary placebo comparisons are 0.3 to 19.3 for benzocaine 10% versus placebo and 7.2 to 25.7 for benzocaine 20% versus placebo, each at the 95% level. The registry reports treatment differences of 9.80 and 16.46, respectively. These intervals indicate a range of values compatible with the data and the analysis assumptions; they are not guarantees that the true value lies in a particular range. Their widths convey uncertainty in the estimated contrasts.

The simple unadjusted recalculations have 95% confidence intervals of 0.005 to 0.200 for benzocaine 10% versus placebo and 0.074 to 0.262 for benzocaine 20% versus placebo. These are risk-difference intervals based on posted summary counts, and they are not identical to the registry's own treatment-difference intervals. Because the sponsor's analysis is the primary posted analysis, its estimates and intervals should lead interpretation. The discrepancy is a reason to keep the two analyses distinct, not to select whichever interval appears more favorable.

General statistical context: intervals are more informative when considered alongside the estimate, outcome definition, and analysis method. A confidence interval does not account for every source of uncertainty, such as unreported analysis choices or possible missing data. Nor does a narrower interval by itself demonstrate clinical importance. The registry does not report additional details about missing outcome handling or the full analysis specification, so precision should be interpreted within those limits.

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 planning figures describe participants per arm for the non-time-to-event comparisons in this study. They are not total participants across both arms. A time-to-event calculation would instead report the total number of events, but the listed primary endpoint is a percentage-based binary outcome. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation.

The rows consider planning for 100%, 75%, or 50% of the observed effect. The 80% power column gives the estimated participants per arm under those assumptions for a design with 80% power to detect the stated effect; the 90% power column gives the corresponding estimate for 90% power. In plain language, higher planned power generally requires a larger sample under the same other assumptions. Planning for a smaller effect also requires more participants per arm than planning for a larger effect. The numerical entries are illustrations based on the observed effect, not the trial's own original sample-size calculation. The original design assumptions and sample-size justification are not reported in the registry record.

These figures are not observed power. They do not measure the probability that the trial's findings are true or whether the study succeeded. They show how sample-size requirements would vary under the stated planning assumptions and effect fractions. The displayed planning values should be read as conditional estimates, not as enrollment targets that can be attributed to the investigators.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.704; treatment proportion p₁ = 0.807; average p̄ = 0.756.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6077 = 1.1911.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6034 = 0.5078.
4. n = (1.1911 + 0.5078)² / (0.1025)² = 274.6, rounded up to 275 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
p0 <- 0.704347826086957                          # control proportion
p1 <- 0.8068669527897                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(474175)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 275 (80% power), 368 (90% power)
 75% of observed effect: n per arm = 507 (80% power), 678 (90% power)
 50% of observed effect: n per arm = 1,177 (80% power), 1,575 (90% power)
power.prop.test: 274.6 
Simulated power at n = 275 per arm: 0.802 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 registry's primary comparisons report Cochran-Mantel-Haenszel treatment differences. For placebo versus benzocaine 10%, the reported estimate is 9.80, with a 95% confidence interval from 0.3 to 19.3 and p = 0.038. For placebo versus benzocaine 20%, the estimate is 16.46, with a 95% confidence interval from 7.2 to 25.7 and p < 0.001. The registry also reports a treatment difference of 6.72 for benzocaine 10% versus benzocaine 20%, with a 95% confidence interval from 0.2 to 13.3 and p = 0.047.

Simple unadjusted recalculations from the posted responder counts produce risk differences of 0.102 and 0.169 for the two placebo comparisons. Those recalculations and the sponsor's reported treatment differences are not identical. The sponsor's own analysis takes precedence; the unadjusted calculations are useful as transparent checks on the posted group summaries, not replacements for the registry analysis. The reason for the difference between estimates is not reported in the registry record, so it should not be attributed to a particular adjustment, missing-data rule, or modeling choice.

The registry's method name and estimates can be stated, but a fuller account of the analysis specification is not reported in the registry record. Interpretation should therefore rely on the reported method and results without inferring additional procedures.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary endpoint, several secondary outcomes, and multiple comparisons. It does not report whether the statistical plan adjusted for multiplicity, whether a testing hierarchy was prespecified, or how the family of hypotheses was defined. These details are not reported in the registry record. The reported p-values therefore should not be described as multiplicity-adjusted, and no correction should be assumed.

General statistical context: testing several endpoints or group contrasts creates multiple opportunities to observe a small p-value even when the underlying effects are absent. Multiplicity procedures can control that risk, but they may change the threshold for interpreting each result. Without a reported plan, the p-values are best read as the values reported for their individual analyses, with particular caution for secondary findings and comparisons beyond the primary question.

The registry defines the primary analysis population as all randomized participants who received study medication. It does not provide further detail about how participants with missing endpoint assessments were handled, whether every randomized participant contributed to each assessment, or whether any imputation method was used. Those missing-data procedures are not reported in the registry record. The outcome-specific group sizes are posted, but they do not by themselves explain the handling of missing measurements or deviations from the stated population definition.

The trial's original sample-size justification, interim-analysis plan, and full statistical analysis plan are also not reported in the registry record. These omissions limit assessment of whether the reported analyses followed prespecified choices. They do not demonstrate that the analyses were inappropriate; they mean only that the registry record does not document the information needed to evaluate those questions fully. The findings can still be described using the reported methods, estimates, and intervals, while keeping these uncertainties in view.

16. 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 serious adverse event figures show 0 affected participants in the placebo group, with 115 at risk; 0 in the benzocaine 10% group, with 233 at risk; and 0 in the benzocaine 20% group, with 228 at risk. These are counts of participants with serious adverse events, not a complete description of every safety outcome. The registry does not list serious adverse event terms in the posted results.

For other adverse events, the record reports 4 affected participants among 115 in the placebo group, 6 among 233 in the benzocaine 10% group, and 6 among 228 in the benzocaine 20% group. The posted figures give counts, not a description of event types or severity. Death counts are not reported in the registry record.

General statistical context: a count of zero observed serious adverse events does not establish that an event is impossible, especially when the number of participants observed is limited. Likewise, these figures alone do not provide a complete comparison of safety, duration of exposure, or the nature of adverse events. They should be read as the specific adverse-event counts recorded, not as a comprehensive safety assessment.

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

The primary result describes how often participants met a particular early-response rule: a reduction of at least 1 unit on the Dental Pain Scale for two consecutive assessments between 5 and 20 minutes. The posted percentages were 70.4% for placebo, 80.7% for benzocaine 10%, and 87.3% for benzocaine 20%. The registry's own reported treatment differences and confidence intervals provide the primary statistical comparisons. The simple unadjusted recalculations differ from those reported analyses, and the sponsor's analysis takes precedence.

These results do not describe the amount of pain relief for every participant, how long relief lasted, or whether pain later returned. The responder definition combines repeated assessments into a yes-or-no outcome and does not distinguish among participants who meet the criterion to different degrees. A participant who narrowly meets the rule and one with a larger reduction are both classified as responders under this endpoint.

The findings also do not, on their own, establish the clinical importance of the observed differences or answer every question about safety, satisfaction, or compliance with proposed label directions. The registry lists other outcomes, but results for several are not reported in the registry record. The posted safety counts are limited to the adverse-event figures recorded and do not replace a fuller account of event types or follow-up.

General statistical context: an estimate and its confidence interval describe uncertainty in a group comparison under an analysis method. They do not predict an individual participant's experience, and a p-value is not the probability that a treatment effect is real. The statistical results should be understood in the context of the defined population, endpoint, short primary assessment window, and the unreported details of missing-data and multiplicity procedures. These are descriptions of the posted trial results, not medical advice.

18. Why This Trial Matters Statistically

This study illustrates how a clinical trial can turn repeated measurements into a prespecified binary endpoint. The response rule requires both a minimum change on the Dental Pain Scale and confirmation across two consecutive assessments within a defined early period. That makes the outcome easy to state as a percentage, while also showing why the exact threshold and timing matter to interpretation.

The three-group design allows comparisons of placebo with two benzocaine formulations, as well as a direct comparison between the formulations. The posted group percentages and the registry's reported treatment differences convey the observed pattern in different ways. The recalculated absolute differences offer a straightforward check from the posted counts, while the sponsor's Cochran-Mantel-Haenszel analysis remains the primary reported analysis. The discrepancy is a useful reminder that a simple summary-data calculation need not reproduce a trial's reported analysis.

The record also illustrates why precision and analysis transparency matter. Confidence intervals communicate uncertainty around estimates, while p-values alone do not convey the size or importance of a difference. Multiple endpoints and comparisons make it important to know whether multiplicity was addressed, yet that plan is not reported in the registry record. Similarly, missing-data handling and the original sample-size justification are not reported. These gaps constrain the strength of conclusions that can be drawn from the registry record, without erasing the value of the posted outcome data.

Finally, planning calculations based on observed effects can help readers understand how assumptions about effect size influence sample requirements. Such illustrations are distinct from the study's original design rationale and from observed power. Keeping those distinctions clear is central to responsible statistical interpretation.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The primary endpoint captures a narrowly defined early response during assessments at 5, 10, 15, and 20 minutes. It does not, by itself, describe response beyond that period or establish duration of relief. The binary classification also compresses differences in the size of participants' changes into responder and non-responder categories. Interpretation should remain tied to the registry's threshold and assessment rule.

The posted group percentages have outcome-specific denominators, while the registry reports overall enrollment separately. Further detail on participant flow and the handling of missing endpoint data is not reported in the registry record. The definition of the analysis population includes randomized participants who received study medication, but the record does not explain how missing assessments or other analysis-population issues were addressed.

The registry's reported primary estimates differ from simple unadjusted recalculations based on posted summary counts. The sponsor's own analysis takes precedence, but the reason for the discrepancy is not reported in the registry record. The registry names the Cochran-Mantel-Haenszel method without reporting the full analysis specification, so additional adjustments or implementation details should not be inferred.

The multiplicity plan, original sample-size justification, interim-analysis plan, and full statistical analysis plan are not reported in the registry record. Results for several listed secondary outcomes are also not reported in the registry record. These gaps limit assessment of prespecification and the full pattern across outcomes. The safety information includes counts of participants affected by serious and other adverse events, but does not provide a full account of event types; death counts are not reported in the registry record. Each limitation narrows interpretation rather than proving a particular source of bias.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Benzocaine Gel Toothache Dose-Response Study — NCT00474175.
• Registry record last updated 2013-02-28; status last verified 2013-02.
• Results first posted 2012-01-18.
• 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, double-masked, parallel-group Phase 4 study of placebo gel and benzocaine gel in 10% and 20% formulations for toothache. Overall enrollment was 577. The primary endpoint was the percentage meeting a response definition based on a reduction of at least 1 Dental Pain Scale unit for two consecutive assessments between 5 and 20 minutes.

Posted response percentages were 70.4% for placebo, 80.7% for benzocaine 10%, and 87.3% for benzocaine 20%. The registry reports Cochran-Mantel-Haenszel comparisons and confidence intervals; simple unadjusted recalculations differ, so the sponsor's own analysis takes precedence. The record lists additional outcomes and provides global satisfaction results, but results for several secondary outcomes are not reported in the registry record. Details about multiplicity, missing-data handling, the original sample-size justification, interim analyses, and the full statistical plan are not reported in the registry record.