Safety and Efficacy of COV795 in Moderate to Severe Post-Operative Bunionectomy Pain With an Open-label Extension: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy of COV795 in Moderate to Severe Post-Operative Bunionectomy Pain With an Open-label Extension (NCT01484652), a randomized phase 3 study. The primary outcome is SPID48 (Summed Pain Intensity Difference).

2. Clinical Question

3. Trial Design

The registry record describes a phase 3, multicenter, randomized, double-blind, placebo-controlled, parallel-group study followed by an open-label extension. Its title identifies the clinical setting as moderate to severe post-operative pain after bunionectomy. The intervention was COV795, and the control was placebo. The registry lists the study as completed, with enrollment of 329. The record gives a start of 2011-11 and primary completion in 2012-08.

The primary objective was to compare repeated doses of COV795 with placebo using summed pain intensity difference over the first 48 hours. The posted primary endpoint analysis population was the modified intent-to-treat population, described as 303 subjects from Cohort 2. The group outcome counts listed for the primary endpoint are 150 for COV795 and 153 for placebo. Those counts refer to the primary outcome record; they should not be taken to describe every safety analysis or every part of the trial.

The trial's title and design fields report random allocation, double masking, and a parallel-group model. The registry record does not report the randomization method, the details of masking, or the exact procedures used in the open-label extension. It also does not report the treatment dose, the dosing schedule, the eligibility criteria in detail, or the duration of follow-up beyond the primary endpoint timeframe. Those details cannot be inferred from the posted outcome summary.

The record lists one primary outcome measure and no posted secondary outcomes. It contains no posted statistical analyses beyond the primary outcome summaries. Thus, the main quantitative evidence described in the registry record is the comparison of mean SPID48 between the two randomized groups, alongside the safety counts. The trial's original statistical plan and detailed conduct cannot be reconstructed from these summary fields.

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

Analysis population for the primary outcome, as defined in the registry: The analysis population for the primary outcome measure was the modified intent-to-treat population (mITT). The mITT analysis population includes 303 subjects from Cohort 2. Missing pain intensity difference scores were imputed using a multiple imputation method.

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.

Population: The analysis population for baseline characteristics includes all 329 subjects who received at least one dose of study treatment. The subjects were assigned 2 cohort classifications. Cohort 1 included 26 subjects enrolled under the original protocol and amendment 1. Cohort 2 is the remainder of the 303 subjects enrolled after amendment 1.

6. Primary Endpoint

The registry lists the primary outcome as SPID48 (Summed Pain Intensity Difference), measured over 48 hours.

The primary endpoint was SPID48, or summed pain intensity difference over 48 hours. Pain intensity was measured with the 11-point Numeric Pain Rating Scale, ranging from 0 to 10. At each assessment, pain intensity difference was defined as the arithmetic difference between the pain score at that time point and the baseline score. SPID48 sums time-weighted pain intensity differences from 22 measurements over the assessment period.

The endpoint ranges from -480, described as worst, to 480, described as best. A higher SPID value indicates greater pain relief. The measure therefore incorporates both the direction and duration of change in reported pain over the assessment window, rather than relying on a single score at one time point. Its time-weighted construction gives the repeated measurements a combined summary.

Such a continuous summary can be compared between groups using group means, with the mean difference expressing the contrast on the endpoint's scale. The registry record reports mean SPID48 values and a mean difference with a confidence interval. These statistics describe the average contrast in this outcome; they do not, by themselves, describe every participant's experience or establish how individual pain scores changed over time.

7. Results Posted in the Registry

• COV795: mean 114.90 versus 66.90 on control; difference 48.00 (95% CI 26.88 to 69.12), standardized effect d = 0.51; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous: it is a numeric summary that can take a range of values. For a continuous endpoint, a group mean describes the average observed or imputed outcome in that group. The registry reports means of 114.9 for COV795 and 66.9 for placebo. The mean difference of 48.0 expresses the distance between these group averages on the SPID48 scale, with the comparison presented as COV795 relative to placebo.

SPID48 is constructed from repeated pain intensity assessments. Pain intensity difference compares each NPRS score with baseline, and the time-weighted sum combines those differences across 22 measurements during 48 hours. This construction means the outcome reflects a trajectory summarized over time, rather than one assessment alone. Since the registry defines a higher SPID value as greater pain relief, a positive difference in the direction reported means the COV795 mean is higher on this scale.

The registry reports standard errors of 7.64 for COV795 and 7.6 for placebo as the measures of dispersion around the means. A standard error describes the precision of an estimated mean, rather than the spread of individual participants' values. It is therefore different from a standard deviation. The pooled standard deviation used in the posted recalculation is not necessary to interpret the sponsor's reported mean contrast and is not a direct description of the spread of individual outcomes in each group.

The confidence interval gives a range of values compatible with the estimate under the statistical procedure used. In general, a confidence interval communicates both the estimated direction and its precision: a narrower interval indicates less uncertainty about the estimate, while a wider interval indicates more. For this outcome, the reported interval extends from 26.88 to 69.12 for the mean difference. The registry record does not state the exact interval construction or the full model used to derive it.

The p-value is reported as <0.001. In general, a p-value measures how unusual data at least as extreme as those observed would be under a specified null hypothesis and model assumptions. It is not an effect-size measure, does not measure the chance that a treatment effect is real, and is not a substitute for the estimate and interval. The registry record does not provide the test statistic, degrees of freedom, or the details of the testing model.

The primary analysis population was modified intent-to-treat and included 303 subjects from Cohort 2. The record states that missing pain intensity difference scores were imputed using multiple imputation. In general, multiple imputation represents uncertainty about missing values by creating multiple plausible completed datasets, analyzing each dataset, and combining the estimates. Its validity depends on the imputation model and assumptions about the missingness process. The record does not report the variables in the imputation model, the number of imputations, the missingness pattern, or sensitivity analyses, so the impact of those choices cannot be evaluated from the registry information.

The posted outcome is an average contrast. A mean can be influenced by the distribution of values and does not disclose whether most participants had similar outcomes or whether responses varied widely. The registry does not report medians, outcome distributions, subgroup estimates, or participant-level data. General statistical interpretation should therefore distinguish the average group result from individual response patterns.

The registry describes randomized allocation and double masking, design features intended in general to reduce systematic differences between groups and the influence of expectations on outcome assessment. However, the record does not report how allocation was generated or concealed, who was masked, whether masking was assessed, or how deviations from assigned treatment were handled. The design labels alone do not establish that these procedures were implemented without difficulty.

The record lists one primary outcome and no secondary outcome results or posted analysis entries. This limits assessment of how the primary endpoint fits into the full set of planned analyses. The original statistical plan, including any multiplicity control, covariate adjustment, interim analysis, or sample-size justification, is not reported in the registry record. No such procedures should be assumed from the design description or the p-value alone.

10. Confidence Intervals and Precision

The reported mean difference in SPID48 was 48.0, and its 95% confidence interval ranged from 26.88 to 69.12. The interval lies above zero, so the estimate and its reported uncertainty are consistent with a positive average difference on the SPID48 scale. Since higher values indicate greater pain relief, the direction corresponds to a higher mean in the COV795 group.

Precision concerns how narrowly the effect is estimated, not whether the effect is clinically important. In general, a confidence interval combines information about the estimate and its uncertainty. The interval does not imply that every value inside it is equally likely, nor does it guarantee that the true value lies within it. It describes the range produced by the stated confidence procedure under its assumptions.

The registry does not report the full model or interval-construction method, and it does not provide a minimum clinically important difference for SPID48. Consequently, the interval can be read as statistical uncertainty around the group mean contrast, but it cannot establish whether the values within it correspond to small, moderate, or large clinical importance. That judgment would require context not reported in the registry record.

The posted standard errors, 7.64 and 7.6, describe precision of the group mean estimates rather than variation among individual participants. The record does not provide group standard deviations or a distributional display in its outcome summary. Precision of the mean difference should therefore not be confused with the degree to which individual outcomes may overlap between groups.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm for 80% power and participants per arm for 90% power. They are not total event counts. The curve expresses the same planning relationship across the listed effect assumptions: as the effect to detect becomes smaller, more participants are needed in each arm.

The planning illustration assumes a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as its reference and considers planning for 100%, 75%, and 50% of that observed effect. The registry record does not report the assumptions used for the trial's original design or its original sample-size justification.

The 80% power column gives the planned number of participants per arm associated with an 80% chance of rejecting the null hypothesis under the illustration's assumptions, if the assumed effect is the true effect. The 90% power column conveys the corresponding planning target for a 90% chance under those assumptions. These are planning quantities, not guarantees about a future study's result. This is not observed power.

The numbers should be interpreted as hypothetical design guidance based on the observed effect, not as proof that the completed trial was adequately or inadequately sized. The observed result is itself uncertain, so planning directly from it can produce optimistic or unstable targets. Smaller assumed effects require more participants because a smaller signal is harder to distinguish from random variation. The figures do not account for details not stated in the planning assumptions, such as attrition, a different analysis model, or any adjustments to the study design.

12. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = 48.000; pooled standard deviation σ = 93.791.
2. n = 2 × (1.96 + 0.842)² × 93.791² / 48.000² = 59.9, rounded up to 60 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 <- 48                      # observed difference in means
sigma <- 93.791086146454                     # 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(1484652)
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 = 60 (80% power), 81 (90% power)
 75% of observed effect: n per arm = 107 (80% power), 143 (90% power)
 50% of observed effect: n per arm = 240 (80% power), 321 (90% power)
power.t.test: 60.9 
Simulated power at n = 60 per arm: 0.795 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome measure, SPID48, and no posted secondary outcome results. It lists no analysis entries. This means the posted record does not provide a complete inventory of planned or conducted hypothesis tests. The original statistical analysis plan, any hierarchy of outcomes, and any multiplicity adjustment are not reported in the registry record.

Multiplicity arises when several hypotheses or outcomes are tested. In general, conducting many tests can increase the chance of at least one apparently positive result by chance. Prespecified priorities and adjustment procedures can address that issue, but their use cannot be inferred from a single reported p-value. The record's unadjusted p-value of <0.001 should be understood as the reported result for this endpoint, not as evidence that the broader multiplicity strategy is known.

The primary analysis population is described as the modified intent-to-treat population, comprising 303 subjects from Cohort 2. The registry states that missing pain intensity difference scores were imputed using multiple imputation. It does not report the imputation model, the number of imputed datasets, the amount or pattern of missing data, or the assumptions used for the missingness process. Nor does it report sensitivity analyses that would show how the estimate changes under alternative missing-data assumptions.

The record does not give a detailed definition of the modified intent-to-treat population, beyond the Cohort 2 membership and stated population size. It does not report how participants were assigned to that population, whether any randomized participants were excluded from the primary analysis, or how protocol deviations and treatment discontinuations were handled. The listed primary outcome group counts are 150 for COV795 and 153 for placebo, but those counts alone do not explain the analysis-set rules.

General interpretation of a randomized trial depends on both the comparison and the analysis population. Missing-data handling can affect the result if missingness is related to outcomes, while analysis-set decisions can affect how closely the estimate reflects the randomized comparison. The posted information establishes that imputation was used, but it is not sufficient to assess the sensitivity of the result to imputation assumptions or analysis-population choices.

14. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: allergic reaction (1.1%), gastrooesophageal reflux disease (0.4%), deep vein thrombosis (0.4%), pregnancy test urine positive (0.4%).

The registry reports serious adverse events in separate safety groups. The COV795 group lists 3 affected among 235 at risk, and the placebo group lists 1 affected among 94 at risk. For other adverse events, the record lists 89 affected among 166 at risk for COV795 and 35 among 163 at risk for placebo. These are counts with their corresponding at-risk counts as reported; the group denominators differ across the serious and other adverse-event summaries.

The registry lists allergic reaction at 1.1%, gastrooesophageal reflux disease at 0.4%, deep vein thrombosis at 0.4%, and pregnancy test urine positive at 0.4% among the leading serious-event terms. It does not assign these percentages to a treatment group in the posted terms. Death counts are not reported in the registry record.

Safety counts should be interpreted with attention to the denominators and the period of observation. The record does not report the full adverse-event coding details, duration of exposure, event-level participant histories, or the reason for the different at-risk counts. The counts are descriptive summaries and do not, on their own, establish causation or support a treatment recommendation.

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

The primary endpoint result describes an average difference in time-weighted pain intensity difference during the 48-hour assessment period. COV795 had a mean SPID48 of 114.9, compared with 66.9 for placebo; the reported mean difference was 48.0. The confidence interval was 26.88 to 69.12, and the unadjusted p-value was <0.001. The direction is consistent with a higher average SPID48 in the COV795 group, and the endpoint definition interprets higher values as greater pain relief.

This result does not establish how much any one participant's pain changed, how responses were distributed, or whether the mean contrast meets a threshold for clinical importance. The registry does not report such a threshold. Nor does the result summarize pain after the 48-hour primary assessment window or resolve outcomes in the open-label extension. It is a comparison on the specified primary endpoint, within the stated modified intent-to-treat population and its imputed data.

The statistical evidence is also not the same as certainty about every aspect of the effect. A small p-value does not quantify clinical importance, establish the probability that the hypothesis is true, or remove uncertainty associated with missing observations, analysis choices, and trial conduct. The confidence interval communicates uncertainty around the mean difference but does not show individual-level variation.

The safety summaries record serious and other adverse events in their respective at-risk groups, along with selected serious-event terms. They do not provide enough detail to determine the cause of an event, compare risks under a common follow-up period, or account for all features of exposure. Death counts are not reported in the registry record. These data should be read as descriptive registry summaries rather than as a complete safety evaluation.

Overall, the posted record supports a statistical description of the SPID48 group contrast and the listed adverse-event counts. It does not, by itself, answer questions about longer-term outcomes, all planned endpoints, individual benefit, clinical importance, or the effect of unreported design and analysis decisions. Such questions require information beyond the results posted in the registry record.

16. Why This Trial Matters Statistically

This trial illustrates how a clinical endpoint can compress repeated measurements into one estimand. SPID48 combines time-weighted pain intensity differences across 22 measurements during 48 hours. The resulting mean difference summarizes the average contrast over the assessment period, but it depends on the endpoint's construction and does not preserve the full time course of pain scores.

It also highlights the distinction between a standard error and participant-level variability. The posted standard errors describe uncertainty around the group means; they are not measures of how widely participants' outcomes varied. Interpreting group averages alongside confidence intervals is useful, but those statistics do not substitute for distributions or individual-level data.

Missing-data handling is another important statistical feature. The primary population included 303 subjects from Cohort 2, and missing pain intensity difference scores were addressed using multiple imputation. Because the imputation model and sensitivity analyses are not reported, readers can identify the approach but cannot evaluate its assumptions or robustness from the registry record.

Finally, the posted result is numerically clear but sits within a limited registry report: one primary outcome is listed, no secondary results or analysis entries are posted, and the original statistical plan is not reported. That makes it especially important to separate the reported endpoint estimate from unanswered questions about multiplicity, design assumptions, and broader outcomes. The trial provides an example of why an estimate, its interval, the analysis population, and the limits of the record all matter together.

17. Limitations

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

The registry record gives summary results rather than participant-level data or a detailed statistical report. It does not report the full analysis model, the method used to construct the confidence interval, covariates, handling of protocol deviations, or any sensitivity analyses. The reported p-value is identified as unadjusted, but the record does not explain whether the primary test was part of a wider testing strategy.

The modified intent-to-treat population is identified as 303 subjects from Cohort 2, but the operational definition and exclusion rules are not reported. Missing pain intensity difference scores were imputed using multiple imputation, yet the imputation model, missingness pattern, number of imputations, and diagnostics are not reported. The sensitivity of the estimate to the assumptions behind that procedure cannot be assessed from the posted information.

The record reports standard errors for the means but does not report group standard deviations or participant-level distributions. It also does not provide a clinical-importance threshold for SPID48, so the numerical mean difference cannot be classified against such a threshold from this record alone. No secondary outcome results are posted, and the full outcomes planned for the trial are not described in the available summary.

Additional design details are absent, including the randomization procedure, masking procedures, treatment dose and schedule, detailed eligibility criteria, and the statistical assumptions used for the original sample-size design. Safety summaries have differing at-risk counts across event categories, and the record does not explain those denominators or report deaths. These gaps limit detailed evaluation of the robustness and scope of the reported findings.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Safety and Efficacy of COV795 in Moderate to Severe Post-Operative Bunionectomy Pain With an Open-label Extension — NCT01484652.
• Registry record last updated 2016-10-21; status last verified 2016-09.
• Results first posted 2014-07-02.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT01484652 is a completed phase 3, multicenter, randomized, double-blind, placebo-controlled, parallel-group trial of COV795 for moderate to severe post-operative bunionectomy pain, followed by an open-label extension. The registry lists enrollment of 329 and a study start of 2011-11, with primary completion in 2012-08.

The primary endpoint was SPID48, a time-weighted summary of pain intensity differences over 48 hours. In the modified intent-to-treat population described as 303 subjects from Cohort 2, the posted means were 114.9 for COV795 and 66.9 for placebo. The mean difference was 48.0, with a 95% confidence interval from 26.88 to 69.12 and an unadjusted p-value of <0.001. The record also reports serious and other adverse-event counts, while key analysis and design details are not reported in the registry record.