A Study to Learn How Well a Single Dose of the Study Treatment Naproxen Sodium and Caffeine Combined in One Tablet (Fixed-dose Combination) Works to Relieve Pain After Dental Surgeries Compared to the Single Ingredients and to Placebo: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Learn How Well a Single Dose of the Study Treatment Naproxen Sodium and Caffeine Combined in One Tablet (Fixed-dose Combination) Works to Relieve Pain After Dental Surgeries Compared to the Single Ingredients and to Placebo (NCT05485805), a randomized phase 3 study. The primary outcome is Sum of Pain Intensity Difference (SPID) Over 8 Hours Post-dose.

2. Clinical Question

The trial record describes an acute, postsurgical pain setting and a single-dose comparison. Its findings therefore concern the recorded outcomes during the post-dose observation period in this study population. They do not, by themselves, answer questions about repeated use, other kinds of pain, or people who do not match the study population. Those broader questions are not assessed by the primary comparison described in the registry record.

3. Trial Design

The registry record describes NCT05485805 as a completed phase 3 randomized, parallel-group, placebo-controlled trial. Its official title calls it double-blind, while the registry’s masking classification is quadruple. The trial began on 2022-09-21 and reached primary completion on 2024-01-25. Bayer is listed as the sponsor.

Participants were assigned to one of five groups: naproxen sodium/caffeine 220/65 mg, two tablets of naproxen sodium/caffeine 220/65 mg, naproxen sodium 220 mg, caffeine 100 mg, or placebo. Each group received a single dose of two tablets after third-molar extraction. For the one-tablet combination and naproxen-only groups, a placebo tablet was paired with the active tablet. The placebo group received two matching placebo tablets.

The brief description says participants took the dose within 4.5 hours after surgery. It describes pain decrease over 8 hours as the main comparison period. The registry lists additional outcomes concerning pain intensity and relief over several post-dose periods, rescue medication, and the timing of perceived or meaningful relief. The exact procedure for assigning treatment, concealment of assignment, and operational details of masking are not reported in the registry record.

The record gives total enrollment as 541. The primary outcome groups list 147, 148, 147, 50, and 49 participants, respectively. The analysis populations, reasons for any difference between enrollment and the outcome-group counts, and handling of participants with incomplete measurements are not reported in the registry record. The record also does not report the original sample-size justification or the trial’s full prespecified statistical plan.

4. Participant Flow and Accounting

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

Before assignment: A total of 750 participants were screened at a single study center in the United States. 541 participants were randomly assigned to study intervention.

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 Sum of Pain Intensity Difference (SPID) Over 8 Hours Post-dose, measured over Up to 8 hours post-dose.

The primary endpoint is the sum of pain-intensity differences, or SPID, over 8 hours after dosing. Pain intensity is assessed using a Numerical Rating Scale from 0 to 10, with 0 indicating no pain and 10 the worst pain imaginable. At each post-dose time point, pain-intensity difference is calculated relative to baseline. The differences are weighted by the time since the preceding measurement and summed across the relevant period.

This construction uses repeated pain scores rather than relying on a single measurement. General statistical explanation: a weighted sum can reflect both the size of the change and how long it is observed across the measurement period. It is therefore a summary of pain-intensity change over time, not a score taken at one moment. The registry describes SPID from 0 to 8 as having a minimum of 0 when there is no pain relief and a maximum of 80 when pain and relief are very strong.

The endpoint is useful for comparing groups’ recorded pain trajectories during the specified observation window. It does not directly measure every aspect of recovery or establish how a participant felt outside that window. The primary result is a group-level least-squares mean, reported with a standard error. The registry record does not give further detail about the precise measurement schedule or the model specifications used to produce those means.

7. Results Posted in the Registry

• Naproxen Sodium/ Caffeine 220/65 mg: mean 31.29 versus 8.62 on control; difference 22.67 (95% CI 16.98 to 28.36), standardized effect d = 1.29; z-test p = <0.001.
• Naproxen Sodium/Caffeine 2x220/65 mg: mean 37.24 versus 8.62 on control; difference 28.62 (95% CI 22.93 to 34.31), standardized effect d = 1.63; z-test p = <0.001.
• Naproxen Sodium 220 mg: mean 31.08 versus 8.62 on control; difference 22.46 (95% CI 16.77 to 28.15), standardized effect d = 1.28; z-test p = <0.001.
• Caffeine 100 mg: mean 5.15 versus 8.62 on control; difference -3.48 (95% CI -10.39 to 3.44), standardized effect d = -0.20; z-test p = 0.325.

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.

Peak Pain Intensity Difference (PID)

• 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 errors.
• Result:
  • Naproxen Sodium/ Caffeine 220/65 mg: mean 6.12 versus 6.33 on control; difference -0.21 (95% CI -0.64 to 0.22), standardized effect d = -0.16; z-test p = 0.339.
  • Naproxen Sodium/Caffeine 2x220/65 mg: mean 6.26 versus 6.33 on control; difference -0.07 (95% CI -0.50 to 0.36), standardized effect d = -0.05; z-test p = 0.750.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naproxen Sodium/ Caffeine 220/65 mg and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naproxen Sodium/Caffeine 2x220/65 mg and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes spanning pain-intensity change, total pain relief, time to first use of rescue medication, the cumulative proportion using rescue medication, time to perceptible relief, time to meaningful relief, and pain-intensity difference. It reports 15 secondary outcomes. These measures address related but distinct aspects of the post-dose experience and observation period; they do not all measure the same construct as the primary SPID endpoint.

The available secondary information includes peak pain-intensity difference through 24 hours and sponsor comparisons for selected groups. The posted peak measure has group means between 6.12 and 6.33. The recalculated comparisons of the combination doses with placebo have confidence intervals crossing zero and p-values of 0.339 and 0.75. The sponsor’s reported comparisons between the two combination doses and between the higher combination dose and naproxen alone also have intervals crossing zero, with p = 0.334 and p = 0.783, respectively. These results do not repeat the primary endpoint: peak pain-intensity difference is a different summary from the summed pain-intensity difference over 8 hours.

The broader secondary pattern cannot be fully characterized from the secondary-outcome names alone. Numerical findings for the other listed outcomes are not reported in the registry record in the information. Consequently, they cannot be used to claim that rescue medication use or timing of relief confirms or conflicts with the primary comparison.

Secondary outcomes can provide context by examining other measurements or time windows, but they generally carry less weight than a prespecified primary outcome when many outcomes and comparisons are examined. General statistical explanation: multiple testing increases the chance of at least one apparently notable result by chance when no adjustment or hierarchy is used. Whether the sponsor specified a hierarchy or adjusted secondary analyses for multiplicity is not reported in the registry record. Secondary p-values should therefore be interpreted cautiously and alongside effect estimates, intervals, and the missing analysis details.

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

• Sum of Pain Intensity Differences From 0 to 2, 4, 6, 12 and 24 Hours Post-dose
• Total Pain Relief (TOTPAR) From 0 to 2, 4, 6, 8, 12 and 24 Hours Post-dose
• Time to First Use of Rescue Medication
• The Cumulative Proportion of Participants Taking Rescue Medication Over the 24 Hour Period
• Time to First Perceptible Relief Measured by a Stopwatch
• Time to Meaningful Relief Measured by a Stopwatch
• Time to First Perceptible Relief Confirmed by Meaningful Relief Defined as the Time to Perceptible Pain Relief
• Pain Intensity Difference (PID)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant’s SPID is a numerical score, and the reported group summaries are least-squares means. A least-squares mean is a model-based group estimate, often adjusted for terms in the analysis model. General statistical explanation: such estimates can differ from simple arithmetic averages because they reflect the model’s adjustment and the way comparisons are defined. The registry reports ANCOVA for the sponsor’s pairwise analyses, but the complete model specification, included covariates, and analysis population are not reported in the registry record.

What does the ANCOVA comparison represent?

ANCOVA is a method for comparing group outcomes while accounting for specified variables in a statistical model. The sponsor’s posted results give contrasts between pairs of groups, confidence intervals, and p-values. A contrast’s sign depends on the order in which the groups are subtracted. The registry lists the groups in each comparison but does not give a separate plain-language statement of the subtraction order. Interpret the numerical direction alongside the group means and the named groups, rather than reading a positive or negative sign without context.

The record labels the sponsor’s reported parameters as geometric least-squares means for the primary outcome. It does not explain why that parameterization is used for this outcome or provide enough model detail to reconstruct the adjusted analysis. The recalculated comparisons are explicitly unadjusted and are derived from posted summary statistics. They should not be treated as reproductions of the sponsor’s ANCOVA.

What do the recalculated estimates add?

The recalculated comparisons use the posted group means and standard errors to make simple pairwise comparisons against placebo. The standard deviations used in those calculations are approximated from the reported standard errors and group sizes. This is why the recalculations are described as approximate and unadjusted. They provide a transparent reference based on the group summaries, but they do not account for covariate adjustment, the sponsor’s model structure, or any multiplicity procedure.

A mean difference expresses the estimated separation between two group means in the outcome’s units, here points on the SPID scale. A confidence interval describes a range of estimates compatible with the data and statistical method, subject to the method’s assumptions. A p-value measures how unusual data at least as incompatible with the null comparison would be if that null and the model assumptions held. It is not the probability that the null is true, nor a measure of the size or practical importance of an effect.

Why does the endpoint use repeated measurements?

SPID combines pain-intensity differences measured at post-dose times over a period. Weighting each difference by the time since the preceding measurement makes the summary sensitive to the sequence of observations across that period. General statistical explanation: summarizing repeated observations can make use of more of the recorded trajectory than a single time point, while compressing that trajectory into one number. The resulting summary also depends on the measurement schedule and the endpoint’s definition. The registry describes the calculation in general terms but does not report all implementation details, such as how missing time points were handled.

How should the standard errors be understood?

The primary group estimates are accompanied by standard errors. General statistical explanation: a standard error describes the estimated sampling variability of an estimate, whereas the spread of individual participant outcomes is a different quantity. A smaller standard error generally indicates a more precise estimate under the fitted analysis, but it does not mean individual responses were more uniform. The posted group standard errors are 1.441, 1.436, and 1.443 for the combination and naproxen groups, and 2.47 and 2.521 for the caffeine and placebo groups, respectively.

Because the group sizes differ, the precision of estimates need not be the same across all groups. The caffeine and placebo groups have fewer outcome participants than the other groups, and their reported standard errors are larger. This is a descriptive comparison of the posted estimates; it does not identify why the groups differ in precision or establish anything about the design rationale.

Why are adjusted and unadjusted results not identical?

Adjusted and unadjusted estimates can differ because they use different statistical information and assumptions. ANCOVA estimates depend on the variables included and their relationships with the outcome, while simple recalculations from group summaries do not make those adjustments. Differences can also reflect the exact model parameter or contrast used. The discrepancy flag in the record means the sponsor’s own analysis takes precedence. The unadjusted calculations are useful as a separate check on the posted summary statistics, but should not replace the registered sponsor analyses.

Neither an adjusted model nor a recalculation removes all uncertainty. The validity of each inference depends on its assumptions and the analysis population. Details about model diagnostics, missing-data methods, treatment of protocol deviations, and analysis-set definitions are not reported in the registry record. Those omissions limit independent evaluation of the full statistical analysis.

11. Confidence Intervals and Precision

Confidence intervals show the uncertainty around estimated group contrasts. General statistical explanation: a 95% confidence interval is generated by a procedure that, over repeated studies under its assumptions, would cover the target quantity at the stated rate. It is not a probability statement that a particular fixed effect lies inside the interval. Its width reflects both the observed data and the method used.

For the sponsor-reported comparison of the one-tablet combination with placebo, the estimate is 22.671 and the 95% confidence interval runs from 16.966 to 28.375. For the two-tablet combination with placebo, the estimate is 28.62 and the interval runs from 22.921 to 34.32. For naproxen sodium alone compared with placebo, the estimate is 22.46 and the interval runs from 16.753 to 28.166. Each of these intervals is entirely above zero, and each sponsor-reported p-value is below 0.001.

For caffeine compared with placebo, the sponsor-reported estimate is -3.476, with a 95% confidence interval from -10.41 to 3.457 and p = 0.325. Because that interval crosses zero, the estimate is compatible with differences in either direction under the reported method. It should not be interpreted as proof of no difference. The width also indicates that the comparison is less precise than a statement of exact equality would imply.

The sponsor-reported difference between the two combination doses is -5.95, with an interval from -9.946 to -1.954; the comparison of the two-tablet combination with naproxen alone is 6.161, with an interval from 2.161 to 10.161. These intervals describe model-based estimates, and their meaning depends on the contrast definitions and model used. The registry record does not report a clinically important-difference threshold, so statistical precision cannot by itself determine whether an interval represents a practically important range.

Confidence intervals also do not account automatically for the number of comparisons. If many hypotheses are assessed, the chance of at least one apparently unusual result can increase unless a suitable multiplicity approach is used. Whether and how the sponsor adjusted for multiple comparisons is not reported in the registry record.

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 are illustrations based on the observed comparison effects, not the trial’s original design calculations. For the continuous outcome comparisons, the figures refer to participants per arm. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The planning rows consider 100%, 75%, and 50% of the observed effect. The sample-size assumptions for the original trial are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm under those assumptions for an analysis with an 80% chance of detecting the specified effect if that effect and the assumptions hold. The 90% power column gives the corresponding number per arm for a 90% chance. General statistical explanation: power is a property of a planned study under assumptions; it is not a measure of the probability that a particular trial result is true. Smaller effects require more participants to detect with the same target power, which is why the planning figures increase when the assumed effect is reduced from 100% to 75% or 50% of the observed effect.

For time-to-event outcomes, the listed planning convention is total events rather than participants per arm; the other listed calculations use participants per arm. These illustrations use observed effects and should not be read as evidence that the completed trial had a particular probability of detecting its result. This is not observed power.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 22.671; pooled standard deviation σ = 17.515.
2. n = 2 × (1.96 + 0.842)² × 17.515² / 22.671² = 9.4, rounded up to 10 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 <- 22.671                      # observed difference in means
sigma <- 17.5148585601324                     # 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(5485805)
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 = 10 (80% power), 13 (90% power)
 75% of observed effect: n per arm = 17 (80% power), 23 (90% power)
 50% of observed effect: n per arm = 38 (80% power), 51 (90% power)
power.t.test: 10.4 
Simulated power at n = 10 per arm: 0.779 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s analyses are reported as ANCOVA comparisons of geometric least-squares means, with 95% confidence intervals and p-values. The registry includes sponsor-reported comparisons of the combination doses, the individual ingredients, and placebo. These adjusted analyses are distinct from the simple recalculations, which compare posted group summaries without the sponsor’s model adjustment.

For the combination-dose and naproxen-only comparisons with placebo, the sponsor estimates and confidence intervals are close in scale to the unadjusted recalculations and each sponsor-reported interval is above zero. The caffeine-placebo sponsor comparison also has an interval that includes zero, with p = 0.325. These are broad points of agreement in the direction and uncertainty of the posted comparisons, not evidence that the methods are interchangeable.

The record flags a discrepancy, so the sponsor’s own analysis takes precedence. The unadjusted recalculations are supplementary checks, not a correction to the sponsor’s results. The registry does not report the complete ANCOVA specification, the adjustment variables, or the full analysis-population rules. Without those details, the sponsor analysis cannot be independently reconstructed from the record alone.

15. Multiplicity, Missing Data and Analysis Populations

The record lists 16 outcome measures, 125 analyses, and 15 posted secondary outcomes. This is a substantial set of outcomes and comparisons to consider alongside the primary endpoint. General statistical explanation: when many statistical tests are conducted, some may produce small p-values by chance even if there are no corresponding underlying differences. Procedures such as family-wise error control, a prespecified testing hierarchy, or false-discovery-rate control can address different multiplicity goals, but they are not interchangeable.

The registry record does not report whether the sponsor adjusted the primary or secondary p-values for multiplicity, whether testing followed a prespecified hierarchy, or which comparisons were designated confirmatory. The sponsor-reported p-values should therefore be read as posted, without assuming that they are adjusted. In particular, a p-value below a conventional threshold for one comparison does not establish that all comparisons were protected against the consequences of multiple testing.

The registry reports ANCOVA for the sponsor’s pairwise analyses, but the full statistical analysis plan, covariate list, model diagnostics, and contrast definitions are not reported in the registry record. The record also does not report the primary analysis population, treatment of protocol deviations, handling of rescue medication in the endpoint analysis, missing-data rules, or sensitivity analyses for missing observations. These details matter because analysis choices can affect estimates and uncertainty.

Interim analyses and any associated stopping rules are not reported in the registry record. The original sample-size justification and design assumptions are also not reported. The posted planning illustrations use observed effects and are separate from evidence about how the trial was originally planned. These gaps do not invalidate the posted results, but they limit assessment of the analysis’s prespecification, robustness, and control of false-positive findings.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry’s adverse-event figures list zero serious adverse events in each of the five treatment groups and zero deaths in each group. The numbers at risk for serious adverse events were 147, 148, 49, 50, and 147 across the listed groups. The registry also gives counts for other adverse events: 19, 11, 4, 3, and 23, respectively.

These are counts from the registry record, not a complete description of event type, severity, timing, duration, or relationship to treatment. The record lists no top serious adverse-event terms. It does not provide enough detail to explain the definition or collection window for the other adverse events. Counts alone do not establish whether groups differ in risk, particularly when group sizes differ, and no rates or comparative safety analysis should be inferred from the counts.

The absence of reported serious events in these group summaries is limited to the recorded trial data and its observation period. It does not establish that serious events cannot occur, nor does it answer broader safety questions beyond this study. The registry record does not report a detailed safety-analysis plan in the information described.

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

The posted primary-outcome results describe group differences in a summed pain-intensity measure during the period up to 8 hours after a single dose. The sponsor-reported comparisons with placebo for the combination doses and naproxen sodium have confidence intervals above zero, while the caffeine-placebo interval crosses zero. The registry also reports comparisons between active groups. These findings are statistical comparisons of trial outcomes; they are not a complete account of all possible experiences or outcomes.

The results apply to the study setting described in the registry: participants had 3 or 4 third molars removed and moderate to severe postoperative dental pain, and received a single dose. The findings do not automatically generalize to other procedures, other pain conditions, repeated dosing, different populations, or longer-term outcomes. Such settings are not evaluated by the primary result.

A statistically distinguishable group estimate is not equivalent to a judgment of clinical importance. The registry does not report a threshold for a clinically important SPID difference. It also does not provide the full statistical analysis plan or complete missing-data methods, which limits interpretation of how robust the estimates are to analysis choices.

Conversely, an interval that crosses zero does not prove that two treatments have identical effects. It means the reported analysis does not clearly distinguish the group contrast from zero at the stated confidence level; the range of compatible estimates still matters. The caffeine-placebo comparison is one example of why a non-significant result should not be translated into proof of no effect.

The record’s adverse-event counts are limited summaries and should not be treated as a definitive account of safety. Overall, the results describe measured outcomes in a particular randomized trial, under its recorded conditions. They do not determine an individual’s expected outcome or answer clinical questions that were not studied.

18. Why This Trial Matters Statistically

This trial offers a useful example of how a continuous, repeated-measure pain endpoint can be summarized and compared across randomized groups. SPID combines pain-intensity differences over a defined period, while other listed outcomes examine related dimensions such as relief and rescue medication. Distinguishing the endpoint from the broader set of secondary measures is important: each addresses a different statistical question.

The record also illustrates why sponsor analyses and recalculations from group summaries should be kept separate. An unadjusted comparison can provide a transparent reference, but it does not reproduce an ANCOVA model. Where the record flags a discrepancy, the sponsor’s own analysis takes precedence. Clear separation of estimands and methods helps prevent readers from treating two different analyses as if they were identical.

Precision is as important as the direction of an estimate. The confidence intervals show which comparisons are estimated more clearly and which remain compatible with a range of possibilities. The caffeine-placebo interval, for example, crosses zero, so the posted analysis does not establish a clear difference. Such interpretation is more informative than relying on a p-value alone.

Finally, the registry record highlights limits of public statistical reporting. The full model details, multiplicity procedures, missing-data rules, analysis populations, interim analyses, and original sample-size assumptions are not reported. Their absence makes it harder to assess prespecification and robustness. For biostatisticians and clinical researchers, recognizing both the information available and the unresolved methodological questions is part of responsible 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.

The registry record does not report the full ANCOVA specification, including the model covariates, diagnostics, or detailed contrast definitions. The sponsor’s adjusted results therefore cannot be reproduced from the record alone. The recalculations use approximate standard deviations derived from posted standard errors and group sizes, and they are unadjusted; they are not substitutes for the sponsor’s analysis.

The record does not report the analysis populations, handling of missing primary-outcome measurements, treatment of protocol deviations, or sensitivity analyses. It also does not describe how rescue medication may have been handled in the primary analysis. These omissions matter in a study where rescue medication is itself a listed outcome and the primary endpoint summarizes measurements over time.

The trial examines a single dose in participants with postoperative dental pain after removal of 3 or 4 third molars. It does not establish how results would extend to other clinical settings, repeated dosing, or longer observation periods. The primary endpoint covers up to 8 hours, while some listed secondary outcomes extend to 24 hours; results from one period should not be treated as results for the other.

The record lists many analyses but does not report a multiplicity strategy or identify a complete testing hierarchy. The lack of a stated clinically important-difference threshold also limits interpretation of the practical importance of score differences. Finally, the adverse-event summary gives counts but not detailed event descriptions, timing, severity, or comparative analysis. These features constrain conclusions that can be drawn from the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Learn How Well a Single Dose of the Study Treatment Naproxen Sodium and Caffeine Combined in One Tablet (Fixed-dose Combination) Works to Relieve Pain After Dental Surgeries Compared to the Single Ingredients and to Placebo — NCT05485805.
• Registry record last updated 2025-03-25; status last verified 2025-03.
• Results first posted 2025-03-25.
• 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.

23. Record Summary

NCT05485805 is a completed randomized, parallel, placebo-controlled phase 3 trial sponsored by Bayer. It enrolled 541 participants with postoperative dental pain after third-molar removal and compared single doses of two naproxen sodium/caffeine combination regimens, naproxen sodium alone, caffeine alone, and placebo.

The primary outcome was SPID over up to 8 hours after dosing. Sponsor-reported ANCOVA comparisons with placebo favored the combination regimens and naproxen sodium on the measured endpoint; the caffeine-placebo confidence interval crossed zero. Because the record flags a discrepancy, sponsor analyses take precedence over unadjusted recalculations.

Important details—including the full analysis plan, multiplicity procedure, missing-data methods, analysis populations, and original sample-size rationale—are not reported in the registry record. Safety figures list zero serious adverse events and zero deaths in each group, alongside counts of other adverse events.