A Study To Evaluate The Effects Of Celecoxib (Celebrex®) Or Naproxen On Blood Pressure In Pediatric Subjects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study To Evaluate The Effects Of Celecoxib (Celebrex®) Or Naproxen On Blood Pressure In Pediatric Subjects (NCT00807846), a randomized phase 4 study. The primary outcome is Change From Baseline in Systolic Blood Pressure (SBP) at Week 6/Final Visit.

2. Clinical Question

3. Trial Design

The registry describes a Phase 4, randomized, double-blind, multicenter, active-controlled trial with a parallel-group design. The trial title specifies a six-week duration. Masking is recorded as triple, but the registry does not identify which people or roles were masked. The study status is completed; the recorded start is 2009-09 and primary completion is 2012-12.

The two intervention groups were celecoxib and naproxen. The registry reports an enrollment of 201. For the primary safety analysis, it reports 100 participants in the celecoxib group and 98 in the naproxen group. These figures describe the analysis groups and should not be read as an account of every participant’s study course. The arm descriptions do not provide the treatment doses, schedules, or other operational details.

The primary blood-pressure endpoint was assessed at week 6 or the final visit. The safety population included randomized participants who received at least one dose of study medication, and the safety analysis set was used for blood-pressure measurements. For early terminations, missing measurements were imputed using last observation carried forward. The registry does not report details of the randomization procedure, the allocation ratio, the sites, or the trial’s original sample-size justification.

The record lists additional blood-pressure outcomes at earlier visits and outcomes concerning parent and participant assessments of overall well-being. Those listings establish the breadth of the registered outcomes, but do not in themselves provide numerical results for every outcome. The registered analysis plan and details of any prespecified hierarchy among outcomes are 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.

Before assignment: A total of 101 participants were randomized to treatment with Celecoxib and 100 participants to treatment with Naproxen. Of these randomized, 100 participants received treatment with Celecoxib and 98 participants received treatment with Naproxen. Three participants were randomized but did not receive any treatment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Safety population included all randomized participants who received at least one dose of study medication. The Safety Analysis set was used to analyze all BP (blood pressure) measurements. For early terminations the last observation carried forward (LOCF) was used to impute missing data.

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 Change From Baseline in Systolic Blood Pressure (SBP) at Week 6/Final Visit, measured over 6 Weeks/Final Visit.

The primary endpoint is change from baseline in systolic blood pressure (SBP) at week 6 or the final visit, in mmHg. Change is defined as the value at six weeks minus the baseline value. A positive change therefore represents an increase from baseline, while a negative change represents a decrease. The reported group values are least-squares means, with standard errors as their measures of uncertainty.

Blood pressure is a continuous measure, so the endpoint retains information about the size and direction of each group’s average change rather than reducing the result to a category. Comparing changes between the randomized groups focuses the analysis on whether their observed average shifts differed over the study period. The endpoint is relevant to the trial’s stated question about blood pressure, but it represents a measured outcome at a particular follow-up point, not a complete account of blood-pressure patterns beyond the trial.

The primary analysis used the safety population, defined as randomized participants who received at least one dose of study medication. The registry says the safety analysis set was used for all blood-pressure measurements and that last observation carried forward was used for early terminations. The record does not report further details about how the endpoint was measured or how the final visit was defined for participants without a week-6 measurement.

7. Results Posted in the Registry

• Naproxen: mean -0.73 versus 0.37 on control; difference -1.10 (95% CI -3.04 to 0.84), standardized effect d = -0.16; z-test p = 0.266.

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.

Change From Baseline to Week 2 in SBP.

• 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: mean -1.29 versus -0.20 on control; difference -1.09 (95% CI -2.56 to 0.38), standardized effect d = -0.21; z-test p = 0.147.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naproxen and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in SBP at Week 4.

• 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: mean -2.01 versus -0.17 on control; difference -1.84 (95% CI -3.44 to -0.23), standardized effect d = -0.32; z-test p = 0.025.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Naproxen than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Diastolic Blood Pressure (DBP) at Week 2.

• 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: mean -0.14 versus -1.35 on control; difference 1.21 (95% CI -0.23 to 2.65), standardized effect d = 0.23; z-test p = 0.101.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naproxen and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes covering systolic and diastolic blood pressure at earlier visits and assessments of overall well-being by parents and participants. Numerical comparative results are available for selected secondary blood-pressure outcomes, but the registry does not provide a complete numerical account of every listed secondary outcome in the results analyzed.

Across the available secondary blood-pressure comparisons, the pattern is not uniform: estimated differences vary in direction and precision across measures and time points. This broader pattern does not overturn the primary result, which concerns systolic blood-pressure change at week 6 or the final visit. Nor should an isolated secondary result be treated as a substitute for the primary endpoint. The listed early and later measurements address related but distinct questions.

The posted secondary comparisons are unadjusted. When several outcomes or time points are examined, the chance of seeing an apparently notable result by chance can increase. Adjusting for multiplicity or prespecifying a hierarchy can help control how such findings are interpreted, but neither an adjustment nor a hierarchy is reported in the registry record. The secondary results are therefore best viewed as supportive or exploratory evidence rather than independent confirmation.

Some registered outcomes concern well-being rather than blood pressure. Those measures address different aspects of the study and cannot be inferred from the blood-pressure findings. The record lists a total of 9 secondary outcomes posted and 10 analyses, but does not report a full multiplicity strategy or explain how all outcomes relate to one another in the statistical plan.

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

• Change From Baseline to Week 2 in SBP.
• Change From Baseline in SBP at Week 4.
• Change From Baseline in Diastolic Blood Pressure (DBP) at Week 2.
• Change From Baseline in DBP at Week 4.
• Change From Baseline in DBP at Week 6/Final Visit
• Change From Baseline in Parent's Assessment of Overall Well-being at Week 6/Final Visit.
• Number of Participants With >= 30% Improvement in the Parent's Global Assessment of Overall Well-being at Week 6/Final Visit.
• Change From Baseline in Participant's Assessment of Overall Well-being at Week 6/Final Visit.

9. Statistical Interpretation

10. Statistical Methods Explained

The primary result is reported as a least-squares mean change for each group. A least-squares mean is a model-based estimate of a group average. Such estimates can differ from simple arithmetic averages when a model accounts for other terms, but the registry record does not give enough detail to reconstruct the sponsor’s full model specification. The arm-level standard errors describe uncertainty in the reported group estimates; they are not standard deviations describing the spread of individual participants’ measurements.

The registry’s own primary comparison uses ANCOVA, a method that relates an outcome to a group indicator while accounting for other model terms. ANCOVA is commonly used for continuous outcomes in trials, and may include baseline outcome values, but the specific covariates and model details are not reported in the registry record. The sponsor reports a mean difference in final values, an interval, and a p-value. Those are the sponsor’s results and are distinct from the unadjusted recalculation based on posted summary statistics.

The recalculated comparison uses the posted group estimates to express naproxen minus celecoxib. Its sign therefore describes that ordering: a negative estimate means the naproxen group’s reported change is lower than the celecoxib group’s reported change. Recalculation from summaries is not a reproduction of the sponsor’s ANCOVA. It does not adjust for the covariates in that analysis, and the standard deviations used in the simple comparison are approximate values derived from the posted standard errors. This makes the recalculation useful as a transparent descriptive check, not a replacement for the sponsor’s analysis.

A confidence interval gives a range of values compatible with the estimate and the statistical model at the stated confidence level. Wider intervals indicate less precision. If an interval includes zero for a difference, the data are compatible with no difference as well as with differences in either direction, within the interval’s limits. An interval is not a guarantee that the true effect lies within its endpoints, and it does not by itself decide whether a difference matters clinically.

A p-value describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. Interpretation should consider the estimate and its interval alongside the p-value. The registry’s primary ANCOVA p-value and the unadjusted recalculated p-value are close, but they arise from different analyses.

The blood-pressure values are continuous outcomes. For continuous measures, comparisons concern averages and the uncertainty around them. A mean can conceal variation among individuals, so a modest average difference does not imply that all participants experienced a similar change. Individual-level distributions and analyses are not reported in the registry record.

Last observation carried forward (LOCF) was used to impute missing data for early terminations. Under this approach, an earlier observed value is carried forward to represent a later missing measurement. It is an assumption about the unobserved follow-up value, not a measurement of that value. If outcomes would have changed after the last observation, LOCF may not represent those unobserved changes. The registry does not report the number of early terminations, the amount of missing data, or sensitivity analyses using alternative assumptions.

Standard errors quantify uncertainty in an estimated mean, while standard deviations describe variability among observations. They answer different questions. The posted group spreads are standard errors, and the recalculation uses approximate standard deviations for its unadjusted comparison. Treating standard errors as if they were standard deviations would mischaracterize the underlying participant-level variability; the stated approximation is therefore important when interpreting the recalculated comparison.

For planning, an assumed effect is compared with the variability and design assumptions to estimate how many participants per arm would be needed to achieve a chosen probability of detecting that effect under the planning model. Smaller target effects require more participants because they are harder to distinguish from variation. Such illustrations are not evidence about the probability that this completed trial detected its true effect. The trial’s own original statistical plan, including its sample-size rationale and any interim-analysis plan, is not reported in the registry record.

11. Confidence Intervals and Precision

The primary recalculated difference is -1.1 mmHg for naproxen relative to celecoxib. Its confidence interval extends across zero, so the data do not pin down the direction of a group difference with high precision. The interval is compatible with a difference favoring either group; its width cautions against treating the point estimate as an exact measure of the underlying effect.

The sponsor’s ANCOVA result gives a 90% confidence interval from -0.56 to 2.76 for the reported contrast. That interval also includes zero. Its reported estimate is 1.100, with p=0.274. Because the registry’s ANCOVA contrast is presented in the reverse direction from the unadjusted naproxen-minus-celecoxib recalculation, the signs should not be compared without attending to the contrast order.

Confidence intervals reflect sampling uncertainty under the analysis assumptions. They do not account automatically for every source of uncertainty, including missing-data assumptions, model choices, or the selection of outcomes for reporting. The sponsor’s interval and the recalculated interval also come from different analyses, so their endpoints should not be treated as interchangeable.

The primary result is therefore more informative about what values remain plausible than about a single definitive difference. A point estimate near zero with an interval that crosses zero does not demonstrate equivalence. The registry does not report an equivalence margin, a non-inferiority margin, or a precision target against which to judge whether the interval is sufficiently narrow for such a conclusion.

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 participants per arm, not total events. They are illustrations based on the observed effect rather than the trial’s original design assumptions. The calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers detecting the stated fraction of the observed effect.

The 80% power column gives the number of participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the assumed effect under those assumptions. The 90% power column gives the corresponding estimate for a 90% chance. Power in this context is a prospective planning probability conditional on the assumed effect and model; it is not a measure of whether a result is true or clinically important.

Rows for 100%, 75%, and 50% of the observed effect represent progressively smaller target effects. Smaller effects are harder to distinguish from random variation, so the planning figures increase as the assumed effect becomes smaller. These figures are not observed power for the completed trial.

The registry record does not report the trial’s original sample-size justification or the assumptions used when its enrollment target was set. The planning illustration should therefore be read separately from the recorded enrollment and analysis-group sizes. It does not establish what the study was designed to detect under its original protocol.

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 = -1.100; pooled standard deviation σ = 6.965.
2. n = 2 × (1.96 + 0.842)² × 6.965² / -1.100² = 629.4, rounded up to 630 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 <- -1.1                      # observed difference in means
sigma <- 6.96527099257452                     # 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(807846)
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 = 630 (80% power), 843 (90% power)
 75% of observed effect: n per arm = 1,119 (80% power), 1,498 (90% power)
 50% of observed effect: n per arm = 2,518 (80% power), 3,371 (90% power)
power.t.test: 630.4 
Simulated power at n = 630 per arm: 0.802 (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 primary analysis is ANCOVA on final values. It reports a mean difference of 1.100, a 90% confidence interval from -0.56 to 2.76, and p=0.274. The unadjusted recalculation from the posted group estimates gives -1.1 for naproxen minus celecoxib, with p=0.266. The signs are reversed because the comparisons are expressed in opposite directions. Their magnitudes and p-values are similar, and both intervals include zero.

The analyses are not identical. The sponsor’s result is model-based, while the recalculation is a simple unadjusted comparison of posted summary statistics and uses an approximate standard deviation. Differences in modeling and interval construction mean the recalculation should not be substituted for the sponsor’s analysis. The registry record does not provide enough detail to reproduce every aspect of the ANCOVA.

The record does not flag a discrepancy between the reported and recalculated results. The sponsor’s analysis remains the trial’s reported analysis; the recalculation is a supplementary check that helps make the direction and approximate scale of the comparison transparent.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists 14 outcome measures and 10 analyses, with 9 secondary outcomes posted. It includes a primary blood-pressure endpoint and several secondary outcomes spanning repeated blood-pressure measurements and well-being assessments. When multiple outcomes and time points are tested, some results may appear notable by chance even if there is no underlying difference for those outcomes. Statistical procedures for handling this issue, such as a prespecified hierarchy or multiplicity adjustment, are not reported in the registry record.

Accordingly, the unadjusted secondary p-values should not be read as if each were the sole confirmatory test. The registry does not say whether a particular secondary outcome was intended to support a formal claim, how testing order was determined, or whether any additional analyses were planned. The original statistical analysis plan and any interim-analysis plan are not reported in the registry record.

For missing data, the record states that last observation carried forward was used for early terminations. This means a prior observed measurement was used in place of a later missing measurement. The method relies on an assumption about how measurements would have behaved after the last observation. The number of early terminations, the amount of missingness at each time point, and sensitivity analyses under different missing-data assumptions are not reported.

The primary blood-pressure analysis used the safety analysis set, described as randomized participants who received at least one dose of study medication. The registry does not report whether an intention-to-treat analysis, a per-protocol analysis, or other analysis populations were used for the primary endpoint. It also does not give the analysis population for every secondary result in the information summarized in the record.

The record identifies ANCOVA for the sponsor’s reported primary analysis but does not provide a full model specification, covariate list, or detailed handling of all analysis decisions. The trial’s own sample-size rationale is also not reported. These omissions limit assessment of prespecification and robustness, but do not change what the posted estimates and intervals state.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hand fracture (1.0%).

The serious adverse-event results list 0 affected participants among 100 at risk in the celecoxib group and 1 among 98 at risk in the naproxen group. The registry also lists hand fracture at 1.0% among the top serious-event terms. The record does not explicitly connect that term to a particular group in the information, so it should not be assigned to an arm on that basis alone.

These are counts of participants recorded with serious adverse events during the trial, not a complete measure of all possible harms or a comparison of long-term safety. The death counts are not reported in the registry record. The small number of recorded serious events means these figures provide limited information for comparing uncommon events. They should be read descriptively and alongside the trial’s duration and population, rather than as proof that serious events cannot occur.

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

The primary result describes average systolic blood-pressure change over six weeks in the trial’s safety population. The group estimates were close in scale, and the recalculated comparison was uncertain, with an interval that includes zero. The sponsor’s ANCOVA result also has an interval including zero. Taken together, the posted results do not establish a clear between-group difference at the primary endpoint.

That conclusion is not the same as showing the treatments are equivalent or that there is no possible difference. Equivalence requires a prespecified margin and an analysis designed to determine whether the full range of plausible differences lies within that margin. The registry record does not report such a margin or an equivalence objective. A non-significant result, by itself, cannot establish equivalence.

The endpoint is a group average, so it does not describe every participant’s individual blood-pressure change. It is measured at week 6 or the final visit and does not establish what happens after the trial period. The registry’s LOCF procedure also means measurements missing after early termination were imputed rather than directly observed at the later time point.

The results concern pediatric participants with juvenile idiopathic arthritis in this randomized comparison of celecoxib and naproxen. They do not automatically apply to other populations, durations, outcomes, or treatment regimens. The record does not report dosing details in the arm descriptions, and the posted endpoint should not be treated as a comprehensive assessment of health outcomes or safety.

Statistically, the most appropriate reading is bounded: the posted data estimate a small difference in average change, but leave uncertainty about its direction and size. Secondary findings add context but are unadjusted and cover multiple outcomes and time points. The registry record does not provide the details needed to interpret them as definitive evidence beyond the primary endpoint.

18. Why This Trial Matters Statistically

This trial illustrates why a point estimate, its uncertainty, and the analysis method need to be read together. The primary comparison has an estimate near zero, but the interval also permits differences in either direction. Reporting only the point estimate would obscure that uncertainty; reporting only a p-value would conceal the estimated scale and direction.

It also illustrates the distinction between a sponsor’s adjusted analysis and an unadjusted recalculation from posted summaries. Both help describe the primary comparison, but they answer it through different statistical procedures. Contrast order matters as well: the same comparison can have opposite signs depending on which group is subtracted from which.

The repeated secondary outcomes show why multiplicity matters. Several related measures and time points can produce a varied set of estimates. Without a reported hierarchy or adjustment, an isolated secondary result deserves less weight than a prespecified primary result. Missing-data handling matters too: LOCF carries an assumption about measurements after early termination, and the record does not give information on the amount of missing data or alternative analyses.

Finally, the planning illustrations distinguish the size of a study from the certainty of its findings. Required participants per arm depend on the assumed effect and planning assumptions; they are not observed power and do not reveal the trial’s original design rationale. These considerations help readers judge what the registry’s statistics can support—and where they remain limited.

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 doses or dosing schedules in the arm descriptions, so the treatment regimens cannot be fully characterized from those descriptions. It also does not give the randomization procedure, allocation ratio, site details, or the identities of the roles covered by triple masking. These omissions limit assessment of trial conduct from the registry entry alone.

The primary analysis uses the safety population, and early-termination measurements were imputed using LOCF. The record does not report the number of early terminations or the amount of missing data, and no sensitivity analyses are described. The primary ANCOVA is identified, but its complete model specification and covariates are not reported. The trial’s original sample-size rationale and any interim-analysis plan are also not reported.

The primary comparison is limited to a six-week endpoint in pediatric participants with juvenile idiopathic arthritis. The record does not establish whether findings extend to longer follow-up or other populations. The arm-level standard errors are not individual-level variability measures, and the recalculated comparison uses an approximate standard deviation; it is therefore not a full reconstruction of the sponsor’s model-based analysis.

Finally, multiple secondary outcomes are listed, while the reported multiplicity strategy is not available. The safety record gives a small number of serious adverse events and does not report death counts. These data are descriptive and limited in their ability to characterize uncommon events. The limitations concern what can be concluded from the registry record; they do not, on their own, establish that the trial was conducted improperly.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study To Evaluate The Effects Of Celecoxib (Celebrex®) Or Naproxen On Blood Pressure In Pediatric Subjects — NCT00807846.
• Registry record last updated 2021-02-02; status last verified 2021-01.
• Results first posted 2015-11-11.
• 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 describes a completed Phase 4, randomized, double-blind, multicenter, active-controlled, parallel-group trial comparing celecoxib with naproxen in pediatric subjects with juvenile idiopathic arthritis. Its primary endpoint was change from baseline in systolic blood pressure at week 6 or the final visit.

The primary group estimates were 0.366 mmHg for celecoxib and -0.734 mmHg for naproxen. The unadjusted recalculated difference was -1.1 mmHg for naproxen relative to celecoxib, with p=0.266; the sponsor’s ANCOVA reported a 90% confidence interval that includes zero and p=0.274. The posted results do not establish a clear difference at the primary endpoint.

The registry also lists secondary blood-pressure and well-being outcomes, reports LOCF for early terminations, and records 0 serious adverse events in the celecoxib group and 1 in the naproxen group. Multiplicity procedures, the original sample-size rationale, and several analysis details are not reported in the registry record.