The Comparison of Outcomes of Antibiotic Drugs and Appendectomy (CODA) Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Comparison of Outcomes of Antibiotic Drugs and Appendectomy (CODA) Trial (NCT02800785), a randomized na study. The primary outcome is Patient-reported Quality of Life as Measured by EuroQol (EQ-5D).

2. Clinical Question

3. Trial Design

The Comparison of Outcomes of Antibiotic Drugs and Appendectomy trial, or CODA, is a completed, randomized, parallel-arm trial registered as NCT02800785. The registry identifies the sponsor as the University of Washington, lists appendicitis as the condition, and records enrollment of 1552. The study start is recorded as 2016-05 and primary completion as 2020-03-05. The phase is listed as NA.

Participants were allocated at random to an antibiotics therapy arm or an appendectomy arm. The registry describes masking as none. Thus, the record describes an unmasked comparison of two treatment strategies, rather than a blinded comparison. The primary endpoint was patient-reported EQ-5D quality of life at four weeks after randomization. The primary endpoint population is described as the participants in each arm who answered the four-week survey; the response counts reported for that outcome are 683 in the antibiotics arm and 664 in the appendectomy arm.

The registry describes the primary evaluation as intention-to-treat, with participants analyzed according to their randomized treatment assignment. That stated approach is important because it preserves the comparison created by randomization, even when the treatment received might differ from the assigned strategy. The record does not report the number of participants who crossed over, the number who received a different treatment from assignment, or how any such cases were handled in the posted endpoint summaries.

The intervention descriptions identify a multi-agent antibiotics strategy and an appendectomy strategy. The antibiotics description includes an initial IV period followed by oral treatment, while the appendectomy description allows an open or laparoscopic approach. The registry record does not report additional details needed to reconstruct every aspect of implementation. It also does not report the original sample-size justification, planned interim analyses, or a complete statistical analysis plan. The existence of posted results should therefore be distinguished from documentation of the full design and analysis rationale.

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 overall number of participants in each arm represents the number of participants who responded to the EQ-5D questions on the 4 week survey

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 Patient-reported Quality of Life as Measured by EuroQol (EQ-5D), measured over Four-weeks after randomization.

The primary endpoint is the mean patient-reported quality-of-life score on the EuroQol EQ-5D at four weeks after randomization. The registry describes EQ-5D as assessing health status across five dimensions. Its score ranges from 0, described as a health state as bad as being dead, to a maximum of 1, described as the best health state. The outcome is a continuous score, and the registry reports a mean and standard deviation for each randomized arm.

Patient-reported quality of life is used because the trial’s stated question concerns outcomes patients care about, rather than only a clinical event or treatment procedure. A score collected at a defined follow-up point provides a way to compare average reported health status between the assigned strategies. The mean describes the group average, while the standard deviation describes how spread out individual scores are around that average. Neither statistic describes every participant’s experience.

The primary endpoint is specifically the four-week EQ-5D response. It should not be treated as a complete measure of later quality of life, symptom resolution, complications, or eventual need for appendectomy. Those are separate outcomes, some of which the registry lists as secondary outcomes. The primary endpoint population is respondents to the four-week survey, and the registry’s intention-to-treat description concerns analysis according to assignment. The record does not report a detailed missing-data procedure for the survey outcome.

7. Results Posted in the Registry

• Appendectomy Arm: mean 0.91 versus 0.92 on control; difference -0.01 (95% CI -0.02 to 0.00), standardized effect d = -0.08; z-test p = 0.158.

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.

Total Number of Patients Who Had Resolution of Appendicitis Symptoms at 30 Days

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Appendectomy Arm: 70.3% (466/663) versus 68.3% (462/676) on control; risk difference 1.9 percentage points (95% CI -3.0 to 6.9), risk ratio 1.03 (0.96–1.10); unadjusted two-proportion z-test p = 0.441.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Appendectomy Arm and the control. The interval gives the range of effects that remain compatible with the data.

Rate of Participants With Perforated Appendicitis

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Appendectomy Arm: 15.9% (99/621) versus 31.9% (59/185) on control; risk difference -15.9 percentage points (95% CI -23.3 to -8.6), risk ratio 0.50 (0.38–0.66); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Appendectomy Arm than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Participants With at Least One Complications From Treatment

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Appendectomy Arm: 3.2% (21/656) versus 5.5% (37/676) on control; risk difference -2.3 percentage points (95% CI -4.5 to -0.1), risk ratio 0.58 (0.35–0.99); unadjusted two-proportion z-test p = 0.042.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Appendectomy Arm than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes cover distinct clinical and care-use domains: symptom resolution, perforated appendicitis, treatment complications, appendiceal cancer, hospital days, clinic or emergency visits, and eventual appendectomy incidence. The registry lists seven secondary outcomes, while numerical comparisons in the posted results describe symptom resolution, perforated appendicitis, and treatment complications. Numerical results for the other listed secondary outcomes are not reported in the registry record.

Across the numerical secondary results, the estimates do not form a single uniform pattern. The symptom-resolution comparison is close to zero and its confidence interval includes zero. The perforation comparison points in the opposite direction from the treatment-complication comparison, and the complication interval narrowly excludes zero. These outcomes have different definitions, follow-up times, and analysis denominators, so they should not be collapsed into one overall judgment. In particular, the perforation analysis uses different numbers of participants in the two arms, and the registry does not explain the difference.

These secondary findings complement, but do not replace, the primary quality-of-life result. The primary endpoint estimates average EQ-5D at four weeks, whereas the secondary measures address events and treatment experience. A result on one domain does not automatically predict the result on another. Nor does a secondary event comparison explain the primary score difference.

The reported secondary p-values are unadjusted. When several outcomes are examined, the chance of obtaining at least one apparently notable result by chance can increase. Without a stated multiplicity plan or adjustment, isolated secondary p-values carry less confirmatory weight than a prespecified primary analysis. The registry does not report a hierarchy for testing the secondary outcomes, so the results are best read as outcome-specific estimates that may motivate questions, rather than as a single definitive statistical verdict across all outcomes.

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

• Total Number of Patients Who Had Resolution of Appendicitis Symptoms at 30 Days
• Rate of Participants With Perforated Appendicitis
• Number of Participants With at Least One Complications From Treatment
• Rates of Participants With Appendiceal Cancer
• Days in Hospital After Index Treatment Within 90 Days
• Number of Clinic Visits or Emergency Room Visits
• Eventual Appendectomy Incidence Proportion

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes the primary evaluation as an intention-to-treat analysis and reports continuous EQ-5D scores using means and standard deviations. The methods below explain how to read those statistics and the posted comparisons. General statistical explanations are labeled as general; they should not be taken as evidence that a method beyond the one expressly described was used.

What does the mean describe?

For a continuous outcome, the mean is the arithmetic average of the observed scores in a group. In this trial, the reported arm means are 0.92 for antibiotics and 0.91 for appendectomy. The mean offers a compact comparison of average reported health status, but it does not show the distribution of scores, identify individual experiences, or describe how many participants had a particular score. The registry also reports a standard deviation of 0.13 in each arm. In general, a standard deviation describes how dispersed observations are around their group mean; it is not the uncertainty interval for the difference between groups.

How should the mean difference be read?

The comparison is expressed as the appendectomy mean relative to the antibiotics mean. The reported difference is -0.01. In general, a difference estimate gives the direction and size of the contrast according to the stated subtraction order. A value near zero indicates that the estimated group averages are close on the outcome’s scale, but closeness alone does not demonstrate equivalence. To support an equivalence conclusion, a study would need an appropriate margin and an analysis designed for that question; such a margin is not reported in the registry record.

What does intention-to-treat mean?

In general, intention-to-treat analysis compares participants according to the treatment group to which they were randomized, rather than regrouping them according to treatment received. This approach aims to retain the comparability created by random assignment. The registry explicitly says the primary EQ-5D evaluation uses this approach. The record does not provide enough detail to establish how every secondary analysis handled deviations from assigned treatment, or how missing outcome responses were handled.

What does the p-value say?

The unadjusted p-value for the primary comparison is 0.158. In general, a p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the null hypothesis is true, and it does not quantify the size or importance of the difference. The word unadjusted indicates that this result is not presented as corrected for a family of multiple comparisons. The registry does not report a multiplicity adjustment for the primary result.

Why is the confidence interval useful?

The primary 95% confidence interval runs from -0.024 to 0.004. In general, a confidence interval conveys the uncertainty around an estimate under the method used to calculate it. Its span is informative alongside the point estimate: the interval includes zero and allows values on either side of zero. This supports a cautious reading of the estimated difference rather than treating the point estimate as exact. The interval is not a range containing a known proportion of individual participants’ scores; it concerns the estimated contrast.

How do binary comparisons differ?

Several secondary outcomes count participants with or without an event. Their comparisons are reported as risk differences, which express the difference between the event proportions in the two groups according to the listed comparison order. The corresponding confidence intervals describe uncertainty in those differences. In general, a risk difference is distinct from a relative risk or odds ratio: it describes an absolute contrast on the event-proportion scale. The registry also posts relative measures for some comparisons, but the presence of more than one measure does not remove uncertainty or resolve concerns about outcome ascertainment and analysis populations.

Why can over-dispersion matter?

General: over-dispersion occurs when observed variability is greater than a statistical model assumes. If a model understates variability, its standard errors can be too small and its confidence intervals too narrow, making the evidence look more precise than it is. The registry record does not report an over-dispersion assessment or model-based adjustment for these results. This concept is relevant when evaluating count or clustered data, but it should not be assumed to have affected the posted analyses in this trial.

What do the registry methods leave unresolved?

The primary endpoint description states the intention-to-treat approach, and the posted results give group summaries and comparisons. The registry record does not report a complete analysis plan, the original sample-size justification, planned interim analyses, or detailed methods for missing data. It also lists no reported analysis entries for the secondary results. Consequently, the posted comparisons can be interpreted as the estimates and unadjusted tests shown, but the record does not support claims about unreported model choices, prespecified secondary testing sequences, or additional sensitivity analyses.

11. Confidence Intervals and Precision

The primary comparison is estimated as a mean difference of -0.01, with a 95% confidence interval from -0.024 to 0.004. This interval is the main guide to precision: it shows the uncertainty around the estimated contrast under the calculation used. Because the interval includes zero, the posted result does not rule out no difference in average EQ-5D score. It also permits values on either side of zero, so the estimate is not precise enough to establish a unique direction for the underlying difference.

In general, a confidence interval is produced by a method that would cover the target parameter at its stated rate over repeated studies under its assumptions. It is not a probability statement about the fixed effect in this one trial, and it does not describe the spread of individual participant scores. The standard deviation of 0.13 in each arm describes that individual-score variability; the confidence interval instead concerns the between-arm mean difference.

Precision depends on the amount of information available and on variability. For this endpoint, the registry says the arm counts represent participants responding to the four-week EQ-5D survey: 683 in the antibiotics arm and 664 in the appendectomy arm. The trial enrollment was 1552, but the registry does not provide a detailed accounting of why the primary analysis response counts are smaller than enrollment or how missing responses were handled. It is not appropriate to infer a missing-data mechanism from the counts alone.

The reported p-value of 0.158 is consistent with the interval including zero, but the p-value and interval answer different questions. The interval gives a range of effects compatible with the analysis, while the p-value summarizes incompatibility with a zero-difference model. Neither establishes equivalence or the absence of an effect. The registry does not state a clinically important difference threshold for EQ-5D, so the clinical importance of values within the interval cannot be determined from the record alone.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design assumptions. For the continuous primary endpoint and the binary secondary comparisons, the displayed quantities are participants per arm, not total participants and not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; each scenario targets the stated fraction of the observed effect. Time-to-event calculations, if present, would instead be reported as total events using the Schoenfeld method, but the outcomes with numerical planning figures in this record are not identified as time-to-event outcomes.

The 80% power column means the number of participants per arm estimated to give an 80% chance of detecting the specified effect under those planning assumptions, if that effect were the true one and the analysis assumptions held. The 90% column uses a higher target chance of detection and therefore generally calls for a larger planned sample. These are design calculations, not a description of what happened to participants in the completed trial. They are not observed power.

Each row uses a different fraction of the observed effect: the full observed effect, 75% of that effect, or 50% of that effect. The illustrations show that smaller effects require more participants to detect, because a smaller contrast is harder to distinguish from random variation. The figures are therefore conditional on the selected effect size and stated assumptions; they are not a recommendation or a retrospective judgment about whether the trial was adequately designed.

The registry record does not report the trial’s own original sample-size justification or its intended design effect. The planning values should not be confused with the enrollment of 1552, nor should a row be read as the number of people actually enrolled or analyzed. The curve traces the same planning relationship between the assumed effect size and the required per-arm sample size; it is a visual aid to that relationship, not an additional observed result.

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 = -0.010; pooled standard deviation σ = 0.130.
2. n = 2 × (1.96 + 0.842)² × 0.130² / -0.010² = 2,652.9, rounded up to 2,653 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 <- -0.01                      # observed difference in means
sigma <- 0.13                     # 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(2800785)
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 = 2,653 (80% power), 3,552 (90% power)
 75% of observed effect: n per arm = 4,717 (80% power), 6,314 (90% power)
 50% of observed effect: n per arm = 10,612 (80% power), 14,207 (90% power)
power.t.test: 2653.9 
Simulated power at n = 2,653 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. Multiplicity, Missing Data and Analysis Populations

The registry describes the primary EQ-5D evaluation as intention-to-treat, but does not report a complete statistical analysis plan. No planned interim analyses, sample-size justification, or multiplicity strategy are reported in the registry record. The record lists seven secondary outcomes, and numerical comparisons are posted for some of them. It does not describe a testing hierarchy or a correction for examining multiple outcomes. The unadjusted p-values should therefore be interpreted as individual comparisons, not as results protected against the possibility of chance findings across a family of tests.

Missing data are especially relevant to the primary endpoint because its analysis population is defined by response to the four-week survey. The registry reports 683 respondents in the antibiotics arm and 664 in the appendectomy arm, but does not report the number of missing surveys by arm, reasons for nonresponse, or a procedure for handling missing scores. The full disposition of all randomized participants is not reported in the record available for this comparison. No imputation method, missing-at-random assumption, or sensitivity analysis is reported, and none should be inferred.

The secondary outcomes also use outcome-specific denominators. For example, the perforation comparison reports 185 participants in the antibiotics arm and 621 in the appendectomy arm. The registry does not explain the different denominators, so it is not possible to determine from the record whether they reflect different eligibility, ascertainment, or missingness. The distinction matters because the denominator defines the population contributing to an event proportion.

The primary description’s intention-to-treat statement clarifies the intended assignment-based comparison, but it does not by itself resolve missing outcome data or show how every secondary endpoint was analyzed. The registry reports no secondary analysis entries, and detailed analysis populations and missing-data methods are not reported in the registry record. These omissions limit assessment of robustness and should temper comparisons that rely on incomplete or differing endpoint populations.

15. 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 0.5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: unplanned hospitalization not for appendectomy (2.9%).

The adverse-event record reports serious adverse events affecting 19 participants in each arm. The at-risk counts are 676 in the antibiotics arm and 656 in the appendectomy arm. For other adverse events, the record reports 26 affected participants in the antibiotics arm and 5 in the appendectomy arm, with the same arm-specific at-risk counts. Deaths are reported as 0 in both arms. A listed serious-event term is Unplanned hospitalization not for appendectomy, reported as 2.9%; the registry does not provide an arm-specific attribution for that term in the information presented.

These figures are counts of participants recorded with events, not a complete account of event severity, duration, timing, or causal relationship to treatment. The available summary does not establish that every event was caused by the assigned strategy. Nor should the counts alone be converted into a broad safety conclusion: the at-risk populations differ, and the registry does not provide further detail about event collection or adjudication in this record.

Serious and other adverse-event categories are reported separately. They should not be added together, because the record does not state whether participants could appear in both categories. The posted data describe recorded events in the stated groups and should be interpreted as a limited safety summary rather than an exhaustive characterization of all possible harms.

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

The primary result estimates the difference in average four-week EQ-5D score between participants assigned to antibiotics therapy and those assigned to appendectomy. The means were 0.92 and 0.91, and the estimated difference was -0.01. The 95% confidence interval, from -0.024 to 0.004, includes zero. The unadjusted p-value was 0.158. Together, these results do not provide a clear statistical separation between the arm means in the posted primary comparison, while still allowing a range of possible differences.

This is not proof that the strategies are identical. A nonsignificant result does not demonstrate equivalence, and the registry does not report an equivalence margin or a noninferiority design. It is also not a direct statement about whether either strategy is preferable for an individual participant. The endpoint is a group average at a specific follow-up time and does not capture every outcome that could matter to patients.

The secondary findings address other outcomes and have their own limitations. Their estimates do not all point in the same direction, and the secondary p-values are unadjusted. The record does not report full missing-data procedures, a secondary testing hierarchy, or an explanation for the unequal analysis denominators in the perforation comparison. These factors make it important to avoid using one secondary result as a stand-alone summary of the entire trial.

The findings also do not establish what happens beyond the reported time points or provide a complete account of treatment crossover, later procedures, or all aspects of safety. Although the registry lists eventual appendectomy incidence and several other outcomes, numerical results for all listed secondary outcomes are not reported in the registry record. The posted results are most appropriately read as estimates tied to the specified endpoint definitions, analysis populations, and follow-up periods.

In general, clinical interpretation requires both statistical uncertainty and context about what difference would matter to patients. The registry does not state a threshold for a clinically important EQ-5D difference. The confidence interval and point estimate can be described, but the record alone does not establish whether values within that interval would be meaningful in practice. This interpretation is statistical, not medical advice.

17. Why This Trial Matters Statistically

CODA addresses a comparison of treatment strategies rather than a narrow comparison between individual drugs. Its randomized, parallel-arm design creates a framework for comparing assignment to an antibiotics approach with assignment to appendectomy. The primary outcome is patient-reported quality of life, which reflects the trial’s stated interest in outcomes that matter to patients as well as clinical events.

Statistically, the trial illustrates why an estimate, a confidence interval, and a p-value need to be read together. The primary point estimate is close to zero, but the interval shows the range of uncertainty around it. The p-value alone does not establish the size or importance of a difference, and a result that does not show a clear separation is not the same as evidence of equivalence. That distinction is central when interpreting comparative trials.

The registry also makes visible the challenge of interpreting multiple outcomes. Quality of life, symptom resolution, perforation, complications, hospital days, and later procedures are not interchangeable measures. Their analysis populations and time points may differ, and unadjusted secondary tests require caution. The mismatch in denominators for the posted perforation comparison is a reminder that a result’s denominator is part of its meaning, not a minor detail.

Finally, the record illustrates the limits of registry-level reporting. It identifies the primary endpoint and its intention-to-treat approach, but the original sample-size rationale, interim analysis plans, multiplicity strategy, and missing-data methods are not reported. Transparent interpretation separates what the numerical results estimate from what the registry record does not establish. That separation supports more disciplined statistical reading without turning one endpoint or p-value into a claim about every outcome.

18. 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.

Several limitations specific to this record constrain interpretation. The primary endpoint is based on respondents to a four-week EQ-5D survey, and the registry does not provide a full account of nonresponse, reasons for missing scores, or the method used to address missing data. The reported response counts should therefore be understood as the primary endpoint analysis populations, not as a complete disposition of everyone randomized.

The registry reports only a subset of numerical secondary comparisons, although it lists seven secondary outcomes. Numerical results for appendiceal cancer, days in hospital, clinic or emergency visits, and eventual appendectomy incidence are not reported in the registry record. The absence of those results prevents a complete synthesis across the listed secondary domains. The available secondary p-values are unadjusted, and the record does not describe a multiplicity plan.

The perforation comparison has markedly different arm denominators, 185 and 621. No explanation for that difference is reported in the registry record. This makes the analysis population and the reason for the denominator difference important unresolved questions. The registry also does not report detailed outcome ascertainment, crossover, adherence, or methods for handling deviations from assigned treatment across all endpoints.

The posted information does not include the trial’s original sample-size justification, interim-analysis plan, complete analysis plan, or a stated clinically important difference for EQ-5D. The planning illustrations based on observed effects cannot replace those design details. The record also reports no masking. These limitations do not invalidate the posted estimates, but they restrict what can be concluded about robustness, clinical importance, and the interpretation of secondary findings.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: The Comparison of Outcomes of Antibiotic Drugs and Appendectomy (CODA) Trial — NCT02800785.
• Registry record last updated 2023-05-06; status last verified 2023-05.
• Results first posted 2023-04-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT02800785 is the completed CODA randomized, unmasked, parallel-arm trial sponsored by the University of Washington. It enrolled 1552 participants with appendicitis and compared an antibiotics therapy strategy with appendectomy. The primary endpoint was mean EQ-5D quality-of-life score four weeks after randomization, analyzed according to randomized assignment as described in the registry.

The posted primary means were 0.92 for antibiotics and 0.91 for appendectomy. The estimated mean difference was -0.01, with a 95% confidence interval from -0.024 to 0.004 and an unadjusted p-value of 0.158. The interval includes zero and does not establish equivalence. Numerical secondary results are reported for a subset of the listed outcomes; their unadjusted analyses and differing denominators warrant caution.

The registry record does not report the original sample-size justification, planned interim analyses, a complete multiplicity strategy, or detailed missing-data methods. Those omissions limit assessment of the full analysis plan and the robustness of some endpoint comparisons.