Oral vs. Intravenous TXA Study Proposal: TJA: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Oral vs. Intravenous TXA Study Proposal: TJA (NCT02233101), a randomized na study. The primary outcome is Number of Participants Who Required Blood Transfusion.

2. Clinical Question

3. Trial Design

The registry lists a randomized, parallel-group study with no masking. It names oral TXA and intravenous TXA as the two interventions. The record gives an enrollment of 167 and lists the study status as completed. The study start is 2014-06, and primary completion is listed as 2015-08. The phase is recorded as NA.

The primary outcome is the number of participants who required a blood transfusion during or within 24 hours after surgery. The outcome description says that hemoglobin was measured during and after surgery for the first 24 hours to determine whether a transfusion was indicated. The posted outcome groups list 74 participants in the oral arm and 80 in the intravenous arm, with 4 and 2 participants requiring transfusion, respectively. The record does not explain how those outcome group counts relate to the listed enrollment.

The brief study summary also describes interest in postoperative hemoglobin reduction and postoperative blood loss. The registry record does not report results for those measures in the information available for this analysis. It lists Other Complications as a secondary outcome, but the details and results for that outcome are not reported in the registry record. The sponsor is Rush University Medical Center. Recruitment procedures, site details, treatment adherence, and the original design assumptions 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.

Overall Study

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 Number of Participants Who Required Blood Transfusion, measured over during or within 24 hours after surgery.

The primary endpoint counts participants who required a blood transfusion during or within 24 hours after surgery. It is a binary outcome at the participant level: each person either required a transfusion in the stated timeframe or did not. The registry describes hemoglobin measurement during and after surgery for the first 24 hours as the basis for determining whether transfusion was indicated.

Transfusion is a clinically interpretable event following surgery and connects with the study's stated interest in blood loss and postoperative hemoglobin reduction. Counting participants with the event allows the two treatment groups to be compared using event risks and a risk difference. The endpoint does not, by itself, measure the amount of blood lost, the degree of hemoglobin change, or the number of transfusions per participant. Those are distinct measurements. Although blood loss and hemoglobin reduction appear in the study summary, corresponding posted results are not reported in the registry record.

The timeframe matters to interpretation: the posted endpoint concerns transfusion during or within 24 hours after surgery, rather than transfusion at any later point. The record does not provide a transfusion threshold, adjudication procedure, or additional operational definition beyond hemoglobin measurement and the stated timeframe. These details are therefore not reported in the registry record.

7. Results Posted in the Registry

• Intravenous Tranexamic Acid: 2.5% (2/80) versus 5.4% (4/74) on control; risk difference -2.9 percentage points (95% CI -9.1 to 3.3), risk ratio 0.46 (0.09–2.45); unadjusted two-proportion z-test p = 0.352.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the risk difference represent?

A risk difference subtracts the event risk in one group from the event risk in the reference group. For this comparison, the estimate is expressed as intravenous TXA relative to oral TXA. The reported value is -0.029. Its negative sign indicates that the observed transfusion risk was lower in the intravenous group. The size of the estimate is in risk units, not a count of participants and not a relative measure. It should be read alongside its confidence interval, because the point estimate alone does not show how uncertain the comparison is.

How do risk and odds ratios differ?

A risk ratio compares the event risks between groups by division, while an odds ratio compares the event odds. Both describe relative rather than absolute contrasts. The registry reports a risk ratio of 0.463 with interval 0.087 to 2.451 and an odds ratio of 0.449 with interval 0.080 to 2.525. These are additional descriptions of the same binary endpoint, not separate outcomes. Relative measures can look more dramatic or less intuitive than an absolute risk difference, especially when events are uncommon. Their uncertainty intervals are important: each reported interval includes values on either side of no relative difference.

What does the unadjusted p-value say?

The unadjusted p-value is 0.352. In general, a p-value describes how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It is not the probability that the null hypothesis is true, and it does not quantify the chance that a conclusion is wrong. The registry calls this result unadjusted. Whether any correction for multiple testing was planned or performed is not reported in the registry record. The p-value should not be used alone to infer equivalence, clinical importance, or the probability of benefit.

Why are confidence intervals useful?

In general, a confidence interval represents uncertainty around an estimate under a specified statistical procedure. A narrow interval indicates greater precision than a wide interval, while an interval spanning substantively different effects leaves more uncertainty about the size and direction of the contrast. Here, the risk-difference interval runs from -0.091 to 0.033 and includes zero. It therefore does not isolate a single direction of effect. The interval is not a guarantee that the true value lies within its limits, and its interpretation depends on the method and assumptions used to calculate it. The registry record does not report the interval method.

How does equivalence differ from a conventional difference test?

In general, equivalence is assessed against a prespecified margin representing the largest difference considered acceptable for the question. The evidence must be sufficiently precise to rule out differences beyond that margin under the chosen procedure. A conventional test that fails to reject a no-difference null does not demonstrate equivalence. The study summary states an equivalence hypothesis, but the margin, formal equivalence analysis, and original statistical plan are not reported in the registry record. Therefore, the posted unadjusted comparison should be read as an estimate of the difference with uncertainty, not as a completed equivalence assessment.

What assumptions can affect a binary-outcome comparison?

General methods for comparing binary outcomes rely on how participants enter the analysis, how outcomes are defined, and how missing observations are handled. They may also depend on the statistical model and the treatment of repeated or clustered observations, if relevant. The registry record does not report the analysis population, missing-data procedure, or full model specification. With a relatively small number of observed transfusions in the posted group counts, the estimate can be sensitive to each event. That general feature makes the confidence interval especially important, without changing the observed counts.

The registry reports no discrepancy flag for the posted comparison. The sponsor's own analysis therefore does not need to take precedence over a recalculated comparison under the stated discrepancy rule. However, no trial analysis records are posted, so detailed sponsor methods and adjusted estimates are not available in the registry record. The posted comparison remains an unadjusted summary rather than a substitute for a complete statistical analysis plan or report.

10. Confidence Intervals and Precision

The risk-difference estimate is -0.029, with a confidence interval from -0.091 to 0.033. The interval includes zero and extends in both directions. This means the posted data do not precisely identify the direction of the risk difference. The point estimate favors a lower observed risk in the intravenous group, but the interval also permits a positive risk difference. The interval should be read as a range of values compatible with the data under the method used, not as a range of guaranteed true effects.

The relative estimates are also uncertain. The risk ratio is 0.463, with an interval from 0.087 to 2.451. The odds ratio is 0.449, with an interval from 0.080 to 2.525. Each relative interval spans values below and above the no-difference value for that measure. The estimates do not support a precise claim about the relative effect's direction or magnitude.

Precision is distinct from statistical significance and from clinical equivalence. An interval that includes zero does not establish no effect, while a nonsignificant p-value does not demonstrate that any difference is acceptably small. Assessing equivalence would require the prespecified margin and corresponding analysis, neither of which is reported in the registry record. The interval method and any adjustment for multiple comparisons are also not reported in the registry record.

The posted event counts are 4 and 2, so the comparison rests on few observed transfusions. That context helps explain why uncertainty matters, but it does not justify replacing the reported estimates with a new calculation. The outcome group sizes and enrollment entry are not reconciled in the registry record, and no additional information about missing outcome data is reported.

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 sample-size figures are planning illustrations expressed as participants per arm, not total event counts. They use the observed effect as a reference and show planning for 100%, 75%, and 50% of that observed effect. For 100% of the observed effect, the figures are 705 participants per arm for 80% power and 944 participants per arm for 90% power. For 75% of the observed effect, they are 1,365 participants per arm for 80% power and 1,826 participants per arm for 90% power. For 50% of the observed effect, they are 3,317 participants per arm for 80% power and 4,440 participants per arm for 90% power.

In plain terms, the 80% power column gives the planned number of participants per arm for a design intended to have 80% power to detect the stated effect under the listed assumptions. The 90% power column gives the corresponding planning figure for a design intended to have 90% power. These are design targets, not probabilities that the observed result is correct. Smaller effects require more participants to detect, as reflected by the increasing figures for the smaller fractions of the observed effect.

The listed assumptions are a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The trial's own original sample-size assumptions and justification are not reported in the registry record. This is not observed power.

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

Worked calculation for this trial

1. Control proportion p₀ = 0.054; treatment proportion p₁ = 0.025; average p̄ = 0.040.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2756 = 0.5401.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2748 = 0.2313.
4. n = (0.5401 + 0.2313)² / (-0.0291)² = 704.8, rounded up to 705 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.0540540540540541                          # control proportion
p1 <- 0.025                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2233101)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 705 (80% power), 944 (90% power)
 75% of observed effect: n per arm = 1,365 (80% power), 1,826 (90% power)
 50% of observed effect: n per arm = 3,317 (80% power), 4,440 (90% power)
power.prop.test: 704.8 
Simulated power at n = 705 per arm: 0.808 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary outcome and identifies Other Complications as a secondary outcome. The study summary also names postoperative hemoglobin reduction and blood loss as areas of interest. Details and results for those additional measures are not reported in the registry record. The record lists one secondary outcome, but does not provide a complete account of all outcomes that may have been planned or measured.

Multiplicity arises when several outcomes, time points, groups, or analyses are tested. In general, conducting many tests can increase the chance of at least one apparently unusual result, so a statistical plan may specify which outcome is primary and whether adjustments or a testing hierarchy will be used. The registry record does not report a multiplicity strategy, a full statistical analysis plan, or a rationale for the number of tests. The posted p-value is explicitly unadjusted, and no adjusted result is reported.

The registry lists zero analysis records, while a comparison estimate and outcome counts are posted. This means the registry's analysis-record listing does not supply detailed analysis documentation for the reported comparison. It does not establish that no analysis was performed. The analysis population, whether all randomized participants were analyzed as assigned, and whether any per-protocol or other population was used are not reported in the registry record.

Missing-data handling is also not reported in the registry record. The outcome group counts cannot be assumed to represent a particular missing-data method, and the record does not explain whether every enrolled or randomized participant had outcome ascertainment. Reasons for missing outcomes, protocol deviations, treatment discontinuation, and adherence are not reported in the registry record. These omissions limit assessment of potential bias and prevent a fuller reconstruction of the analysis.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

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

The serious adverse event listing reports 0 serious events in the oral TXA group, with 86 at risk, and 0 in the intravenous TXA group, with 81 at risk. Other adverse events are also listed as 0 in each group, using the same respective at-risk counts. Deaths are listed as 0 in each group. These are the registry's posted safety figures; they should not be interpreted as proof that no risk exists.

The at-risk counts differ from the outcome group counts reported for the primary transfusion endpoint. The registry record does not explain the difference or specify the safety analysis population. It also does not give detailed event definitions, ascertainment procedures, or the observation period for the adverse-event entries. The record lists no serious adverse-event terms. Consequently, the figures describe what was posted for the reported categories, but do not support detailed comparisons of event rates or conclusions about safety beyond those entries.

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

The posted data describe a randomized comparison of oral and intravenous TXA for transfusion during or within 24 hours after surgery. The observed transfusion counts were 4 among 74 participants in the oral group and 2 among 80 in the intravenous group. The reported risk difference is -0.029, with a confidence interval from -0.091 to 0.033, and the unadjusted p-value is 0.352. The estimate points toward a lower observed event risk in the intravenous group, but its uncertainty is broad enough to include a difference in either direction.

These results do not establish that the routes are equivalent. The registry summary states an equivalence hypothesis, but the equivalence margin and the formal analysis are not reported in the registry record. A failure to show a statistically significant difference is not the same as evidence that any difference is acceptably small. The posted confidence interval and p-value address a conventional comparison; they do not supply the missing equivalence criteria.

The result also does not answer whether the routes differ in postoperative blood loss or hemoglobin reduction. Those measures are mentioned in the study summary, but their results are not reported in the registry record. The posted comparison is limited to the transfusion endpoint and its stated timeframe. It does not establish what happens after that timeframe or describe treatment effects on other complications.

The adverse-event entries report zero serious events, zero other adverse events, and zero deaths in each listed group. Those entries are limited by the lack of further safety methods and by the unexplained difference between the at-risk counts and primary-outcome group counts. They should be read as registry reporting, not as proof of absence of harms.

Overall, the result is an imprecise estimate of the transfusion risk difference in this trial record. It is not a clinical recommendation and does not support a claim that either route works better or that the routes are interchangeable.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate and a significance test are not enough to answer an equivalence question. The study summary states that oral and intravenous TXA were hypothesized to be equivalent, but the registry record does not report an equivalence margin or formal equivalence procedure. Without those elements, a conventional nonsignificant result cannot demonstrate that any difference is acceptably small.

The endpoint also shows how event counts and precision interact. The posted transfusion counts are 4 and 2, and the confidence interval for the risk difference extends across zero. The observed direction is not settled by the point estimate alone; the interval communicates the uncertainty relevant to the comparison. This is a useful distinction for researchers interpreting binary outcomes in small event-count settings.

The record further highlights the value of complete reporting. It names blood loss and hemoglobin reduction in the study summary, but does not report their results. It does not provide the original statistical plan, analysis population, or missing-data methods. These omissions make it difficult to assess how the stated hypothesis was operationalized and how broadly the posted result can be interpreted.

Finally, the planning illustrations show that the sample size needed for a design depends on the effect targeted and the desired power. The figures are not evidence that the completed trial had a particular power, and the original design assumptions are not reported. Separating planning calculations from observed results helps prevent overinterpretation of a nonsignificant finding.

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 does not report an equivalence margin, although the study summary describes an equivalence hypothesis. The posted estimate therefore cannot establish equivalence. The original statistical plan, sample-size justification, and any interim analyses are not reported in the registry record.

The outcome group counts are 74 for oral TXA and 80 for intravenous TXA, while the enrollment entry is 167. The registry record does not explain how these figures relate to each other. Safety entries use at-risk counts of 86 and 81, and their relationship to the primary endpoint groups is also not explained. No reconciliation or participant flow information is reported.

Results for postoperative blood loss and hemoglobin reduction are not reported in the registry record, despite those measures appearing in the study summary. The secondary outcome is identified as Other Complications, but its detailed definition and results are not reported. The record does not report outcome ascertainment beyond the primary endpoint timeframe or provide detailed adverse-event methods.

Information about eligibility criteria, dose and timing of TXA, adherence, missing data, protocol deviations, and the analysis population is not reported in the registry record. The comparison has few observed transfusions and a wide confidence interval, limiting precision. These limitations restrict both evaluation of potential bias and interpretation beyond the posted primary endpoint.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Oral vs. Intravenous TXA Study Proposal: TJA — NCT02233101.
• Registry record last updated 2017-11-28; status last verified 2017-10.
• Results first posted 2017-04-28.
• 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

The registry identifies NCT02233101 as a completed, randomized, unmasked, parallel-group study sponsored by Rush University Medical Center. It compares oral with intravenous TXA in people undergoing total joint arthroplasty. The listed enrollment is 167; the study start is 2014-06, and primary completion is 2015-08.

The primary endpoint is the number of participants requiring transfusion during or within 24 hours after surgery. The posted outcome groups report 4 events among 74 participants for oral TXA and 2 among 80 for intravenous TXA. The risk difference is -0.029, with a confidence interval from -0.091 to 0.033 and an unadjusted p-value of 0.352. The record states an equivalence hypothesis but does not report an equivalence margin or formal equivalence analysis. Results for other measures named in the study summary are not reported in the registry record.