A Phase 2 Study To Evaluate The Safety Of Apixaban In Atrial Fibrillation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase 2 Study To Evaluate The Safety Of Apixaban In Atrial Fibrillation (NCT00787150), a randomized phase 2 study. The primary outcome is Number of Participants With Major (Per International Society on Thrombosis and Haemostasis [ISTH] Criteria) or Clinically Relevant Non-major Bleeding Adjudicated by Clinical Event Committee During the Treatment Period.

2. Clinical Question

The endpoint combines two categories of bleeding. Major bleeding followed the stated International Society on Thrombosis and Haemostasis criteria; clinically relevant non-major bleeding was clinically overt bleeding that did not meet major criteria but led to hospital admission, physician-guided treatment, or a change in antithrombotic therapy. This composite therefore includes events with different levels of severity and clinical consequences.

The registry record does not report a formal hypothesis, a prespecified superiority or non-inferiority margin, or the trial’s original sample-size justification. The results address the recorded event counts and their uncertainty, rather than establishing a definitive comparative conclusion.

3. Trial Design

This was a phase 2, randomized, parallel-group, active-controlled study of apixaban and warfarin in people with atrial fibrillation. The official title describes a multicenter study lasting 12 weeks and identifies the warfarin group as open label. The registry’s overall masking field is recorded as double, so the record describes both an overall double-masking classification and an open-label warfarin feature. Further details about which participants or study personnel were masked are not reported in the registry record.

The registry records enrollment of 222 and group sizes of 75 for warfarin, 72 for apixaban 2.5mg BID, and 71 for apixaban 5.0 mg BID. The primary outcome timeframe is baseline to Week 12. The study status is completed; the recorded start is 2008-06 and primary completion is 2009-09.

The safety analysis set consisted of all treated participants. The record notes that two participants, one each assigned to an apixaban group, were mistakenly administered warfarin. In accordance with the stated statistical analysis plan, they were included in the warfarin group for the safety analysis. This describes the reported analysis grouping; additional information about treatment adherence, follow-up, or participant flow is not reported in the registry record.

The primary outcome was assessed by a Clinical Event Committee, according to its title. The registry lists 14 outcome measures and records no analyses. Detailed information about randomization procedures, allocation concealment, the original statistical plan, interim analyses, and the sample-size justification is not reported in the registry record. These omissions limit how fully the reported comparisons can be connected to the trial’s prespecified design.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The safety analysis set consisted of all treated participants. Two participants (1 each in the apixaban 2.5 mg BID group and apixaban 5.0 mg BID group) were mistakenly administered warfarin and therefore, included in the warfarin group per the statistical analysis plan.

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 With Major (Per International Society on Thrombosis and Haemostasis [ISTH] Criteria) or Clinically Relevant Non-major Bleeding Adjudicated by Clinical Event Committee During the Treatment Period, measured over Baseline to Week 12.

The primary endpoint counted participants who experienced either major bleeding or clinically relevant non-major bleeding during the treatment period, from baseline to Week 12. It is a participant-count endpoint: a person meeting either component is counted as having the composite outcome. The registry describes the events as adjudicated by a Clinical Event Committee.

Major bleeding was defined using International Society on Thrombosis and Haemostasis criteria. The definition included acute clinically overt bleeding accompanied by a decrease in hemoglobin of 2 g/dL or more over a 24-hour period, transfusion of 2 or more units of packed red blood cells, bleeding at a critical site such as intracranial bleeding, or fatal bleeding. Clinically relevant non-major bleeding was acute or sub-acute, clinically overt bleeding that did not meet major criteria but led to hospital admission, physician-guided medical or surgical treatment, or a change in antithrombotic therapy.

In general, a composite endpoint can capture more events than a single component and bring together outcomes relevant to the study question. Its interpretation depends on the components: events in the composite can differ in severity and frequency. The registry does not report component weighting or a separate primary analysis for each component. The results therefore describe the combined endpoint, while the component outcomes posted as secondary measures offer additional context rather than replacing the primary result.

7. Results Posted in the Registry

• Apixaban 2.5mg BID: 1.4% (1/72) versus 5.3% (4/75) on control; risk difference -3.9 percentage points (95% CI -9.7 to 1.8), risk ratio 0.26 (0.03–2.27); unadjusted two-proportion z-test p = 0.187.
• Apixaban 5.0 mg BID: 1.4% (1/71) versus 5.3% (4/75) on control; risk difference -3.9 percentage points (95% CI -9.7 to 1.9), risk ratio 0.26 (0.03–2.31); unadjusted two-proportion z-test p = 0.192.

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.

Number of Participants With Total Bleeding Events During the Treatment Period

• 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:
  • Apixaban 2.5mg BID: 12.5% (9/72) versus 17.3% (13/75) on control; risk difference -4.8 percentage points (95% CI -16.3 to 6.6), risk ratio 0.72 (0.33–1.58); unadjusted two-proportion z-test p = 0.412.
  • Apixaban 5.0 mg BID: 23.9% (17/71) versus 17.3% (13/75) on control; risk difference 6.6 percentage points (95% CI -6.5 to 19.7), risk ratio 1.38 (0.72–2.63); unadjusted two-proportion z-test p = 0.323.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 2.5mg BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 5.0 mg BID and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Major (Per International Society on Thrombosis and Haemostasis [ISTH] Criteria) Bleeding Events During the Treatment Period

• 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:
  • Apixaban 2.5mg BID: 0.0% (0/72) versus 1.3% (1/75) on control; risk difference -1.3 percentage points (95% CI -3.9 to 1.3); unadjusted two-proportion z-test p = 0.326.
  • Apixaban 5.0 mg BID: 0.0% (0/71) versus 1.3% (1/75) on control; risk difference -1.3 percentage points (95% CI -3.9 to 1.3); unadjusted two-proportion z-test p = 0.329.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 2.5mg BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 5.0 mg BID and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Clinically Relevant Non-major Bleeding Events During the Treatment Period

• 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:
  • Apixaban 2.5mg BID: 1.4% (1/72) versus 4.0% (3/75) on control; risk difference -2.6 percentage points (95% CI -7.8 to 2.6), risk ratio 0.35 (0.04–3.26); unadjusted two-proportion z-test p = 0.331.
  • Apixaban 5.0 mg BID: 1.4% (1/71) versus 4.0% (3/75) on control; risk difference -2.6 percentage points (95% CI -7.8 to 2.6), risk ratio 0.35 (0.04–3.31); unadjusted two-proportion z-test p = 0.338.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 2.5mg BID and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban 5.0 mg BID and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary bleeding results add detail to the composite primary endpoint by separating total bleeding, major bleeding, and clinically relevant non-major bleeding. In broad terms, these outcomes show that the observed pattern depends on which bleeding measure is considered and which apixaban dose is compared. The counts for major bleeding are particularly sparse. This variation is a reason to interpret the primary composite together with its definition, rather than assuming that every component follows the same pattern.

The secondary measures relate to the primary result but do not independently confirm it. The primary endpoint combines major and clinically relevant non-major bleeding; secondary outcomes can help describe those components or broaden the picture to other events. They do not remove the uncertainty in the primary estimates, and the registry does not report a joint model linking the outcomes or a prespecified hierarchy among them.

The registry lists outcomes beyond bleeding, including stroke or systemic embolism, a combined stroke, systemic embolism, or all-cause death outcome, and myocardial infarction or all-cause death. Numerical results for these outcomes are not reported in the registry record. A missing numerical result is not evidence that an event did not occur.

In general, repeated testing across several endpoints raises the chance that at least one comparison will appear notable by chance. The registry lists 14 outcome measures and reports no analyses in its analysis record; a multiplicity adjustment or testing hierarchy is not reported in the registry record. Accordingly, unadjusted secondary comparisons carry less weight than a prespecified, multiplicity-controlled confirmatory analysis. The secondary results are best treated as descriptive and exploratory, especially given the small event counts and the lack of reported analysis details.

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

• Number of Participants With Total Bleeding Events During the Treatment Period
• Number of Participants With Major (Per International Society on Thrombosis and Haemostasis [ISTH] Criteria) Bleeding Events During the Treatment Period
• Number of Participants With Clinically Relevant Non-major Bleeding Events During the Treatment Period
• Number of Participants With Stroke or Systemic Embolism During the Intended Treatment Period
• Number of Participants With Stroke, Systemic Embolism, or All-Cause Death During the Intended Treatment Period
• Number of Participants With Myocardial Infarction or All-Cause Death During the Intended Treatment Period

9. Statistical Interpretation

10. Statistical Methods Explained

In general, randomization is intended to make comparison groups similar on average with respect to factors that could affect outcomes. This study is recorded as randomized and parallel group, with warfarin as the active comparator. The registry record does not describe the sequence-generation method or allocation concealment, so those procedural details cannot be assessed from the posted information.

What does the risk difference describe?

In general, a risk difference is the event proportion in one group minus the event proportion in the comparison group. A negative estimate in these comparisons means the observed proportion was lower in the apixaban group than in the warfarin group. The primary estimates are -0.039 for apixaban 2.5mg BID and -0.039 for apixaban 5.0 mg BID. These are unadjusted recalculations based on the posted participant counts and group sizes, rather than adjusted estimates from a reported model.

For the primary endpoint, the recorded event counts are four of 75 in the warfarin group, one of 72 in the apixaban 2.5mg BID group, and one of 71 in the apixaban 5.0 mg BID group. The recorded group proportions are 0.053, 0.014, and 0.014, respectively. When an endpoint occurs infrequently, a small change in the count can have a noticeable effect on the estimated group proportion and the comparison. Sparse outcomes also tend to leave substantial uncertainty about the underlying difference.

How should the p-values be read?

In general, a p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The unadjusted p-values for the primary comparisons are 0.187 and 0.192. The registry does not report a full primary analysis plan or a formal decision threshold, so the p-values should be read alongside the estimated differences, confidence intervals, endpoint definitions, and limitations of the record.

The term “unadjusted” is important: these are simple comparisons, with no adjustment for other variables or for testing multiple outcomes. The registry records no analyses in its analysis record. It therefore does not show whether the trial’s original analysis used a different model, covariate adjustment, or other inferential procedure. Such details are not reported in the registry record.

Why does the endpoint definition matter?

The primary outcome combines major bleeding and clinically relevant non-major bleeding. These components have different criteria and may represent different levels of clinical seriousness. The composite count is useful for describing whether a participant met either category, but it does not make the components interchangeable. In general, a combined endpoint should be interpreted with awareness of its definition and of how much each component contributes. Component-specific results are posted as secondary outcomes, but the registry record does not report a separate adjusted primary analysis for them.

The outcome title identifies adjudication by a Clinical Event Committee. Adjudication can support consistent classification of events against defined criteria. The registry does not provide further detail about adjudication procedures, blinding of adjudicators, or disagreement resolution, so those methods cannot be evaluated from the record.

What can be learned from secondary comparisons?

In general, an unadjusted secondary comparison is less conclusive when many outcomes are examined, because the chance of finding a seemingly notable result by chance can increase across repeated tests. This is the multiplicity problem. The registry lists 14 outcome measures, while the analysis record reports no analyses and no multiplicity procedure. The secondary estimates therefore describe the posted counts and simple comparisons, but the record does not establish that they were planned or adjusted as a family of confirmatory tests.

For the secondary outcomes with group counts and comparison estimates posted, the estimates are likewise unadjusted. The registry also lists outcomes involving stroke, systemic embolism, myocardial infarction, and all-cause death, but numerical results for these outcomes are not reported in the registry record. Their absence from the reported numerical results should not be interpreted as evidence that no such events occurred.

What is the role of the analysis population?

The primary safety analysis set included all treated participants, and the record describes the handling of two participants mistakenly given warfarin. The posted group sizes and event counts are presented according to this safety grouping. The registry record does not report a separate intention-to-treat analysis, a per-protocol analysis, or details of missing primary outcome data. In general, analysis populations answer different questions; without those additional results, robustness to alternative grouping or missing-data assumptions cannot be assessed.

11. Confidence Intervals and Precision

In general, a confidence interval gives a range of values compatible with the observed data under the method used to calculate it. It conveys uncertainty around an estimate; it is not a guarantee that the true effect lies within the interval, nor does it describe the range of individual participant outcomes.

For the primary comparison of apixaban 2.5mg BID with warfarin, the recalculated risk difference is -0.039, with a confidence interval from -0.097 to 0.018. For apixaban 5.0 mg BID compared with warfarin, the estimate is -0.039, with a confidence interval from -0.097 to 0.018. Each interval crosses zero, the value corresponding to no risk difference. Thus the posted data do not pin down the direction of the difference with precision.

The intervals are broad relative to the estimates, reflecting limited precision in comparisons based on the small number of primary events. Values within each interval include possible differences favoring either group. The estimates are not proof that the groups are the same, and the intervals do not establish equivalence. The registry does not report an equivalence margin or a non-inferiority margin.

In general, confidence intervals depend on the statistical method and assumptions used. The posted comparisons are unadjusted recalculations from the group summaries; the registry record does not provide a reported inferential analysis method for the primary outcome. The intervals should therefore be understood as precision measures for these recalculated comparisons, not as a substitute for a fully documented trial analysis.

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 for these binary outcomes, not total events. The planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers effects equal to 100%, 75%, or 50% of the observed effect. The registry record does not report the trial’s original design assumptions or sample-size justification.

The 80% power column indicates the planned number of participants per arm under the stated assumptions for a design intended to have an 80% probability of detecting the specified effect if that effect is truly present and the assumptions hold. The 90% power column uses a higher probability target under the same framework. For example, for the primary comparison of apixaban 2.5mg BID with warfarin, planning for 100% of the observed effect gives 327 participants per arm for 80% power and 437 participants per arm for 90% power. Planning for 75% of the observed effect gives 664 participants per arm for 80% power and 888 participants per arm for 90% power. Planning for 50% of the observed effect gives 1,677 participants per arm for 80% power and 2,245 participants per arm for 90% power.

Smaller effects require more participants to detect under these assumptions, because a smaller separation between groups is harder to distinguish from random variation. The figures are illustrations based on the observed effect, not the trial’s own design requirements. They are not observed power: observed power is not a useful substitute for interpreting the estimated effect and its confidence interval after the results are known. These planning values should not be treated as evidence that the trial was adequately or inadequately designed, since its original assumptions are not reported in the registry record.

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

Worked calculation for this trial

1. Control proportion p₀ = 0.053; treatment proportion p₁ = 0.014; average p̄ = 0.034.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2549 = 0.4996.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2533 = 0.2132.
4. n = (0.4996 + 0.2132)² / (-0.0394)² = 326.5, rounded up to 327 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 <- 0.0533333333333333                          # control proportion
p1 <- 0.0138888888888889                          # 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(787150)
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 = 327 (80% power), 437 (90% power)
 75% of observed effect: n per arm = 664 (80% power), 888 (90% power)
 50% of observed effect: n per arm = 1,677 (80% power), 2,245 (90% power)
power.prop.test: 326.5 
Simulated power at n = 327 per arm: 0.828 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 14 outcome measures, including the primary composite and multiple secondary outcomes. The analysis record contains no reported analyses, and the registry record does not report a multiplicity adjustment, a testing hierarchy, or which comparisons were intended to be confirmatory. When several outcomes or contrasts are examined, the chance of a nominally notable result arising by chance can increase. This general concern means that unadjusted secondary results should be interpreted cautiously and should not be treated as a set of independent confirmations.

The registry provides no detailed statistical analysis plan, no reported interim-analysis information, and no sample-size justification. These items are not reported in the registry record. It is therefore not possible to determine from the posted record whether interim monitoring or other design decisions affected interpretation, or how the reported comparisons relate to prespecified analysis choices.

The safety analysis population is described as all treated participants. Two participants, one each in an apixaban group, were mistakenly administered warfarin and were included in the warfarin group according to the stated statistical analysis plan. A separate intention-to-treat result or alternative analysis retaining those participants in their randomized groups is not reported in the registry record.

Missing outcome data, withdrawals, and follow-up completeness are not described in the posted results. The record does not report how missing values were handled, whether any participants lacked endpoint assessment, or whether sensitivity analyses were conducted. Without that information, the potential effect of missing data on the comparisons cannot be evaluated. These are reporting limitations, not evidence that missingness or bias occurred.

15. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cerebral infarction (2.7%), accidental overdose (1.4%), overdose (1.4%), hypoglycaemia (1.4%), back pain (1.4%).

The serious adverse event figures record four participants in the warfarin group, one in the apixaban 2.5mg BID group, and five in the apixaban 5.0 mg BID group as affected, with 75, 72, and 71 participants at risk, respectively. These figures are separate from the primary bleeding composite: a serious adverse event is not necessarily a bleeding event, and the registry does not state that all serious events were caused by treatment.

The registry also reports other adverse-event counts of 17 for warfarin, 24 for apixaban 2.5mg BID, and 22 for apixaban 5.0 mg BID. Serious-event terms listed in the record include Cerebral infarction at 2.7%, and Accidental overdose, Overdose, Hypoglycaemia, Back pain, and Rectal cancer at 1.4% each. The listed terms should be read as reported events, not as evidence of causation. The registry record does not report deaths.

These safety counts are descriptive. The record does not provide a detailed comparative safety analysis, adjustment for multiple event types, or causal attribution for the listed terms. Differences in counts alone do not establish a difference in risk.

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

The posted results describe a phase 2 comparison in which the recorded primary composite event counts were lower in each apixaban group than in the warfarin group. The recalculated risk differences are similar for the two doses, and both confidence intervals include zero. The intervals also allow differences in either direction, so the registry results do not establish a definitive comparative effect.

These data do not show that the treatments are equivalent. A failure to establish a clear difference is not the same as evidence that two groups have the same outcome. The registry does not report a non-inferiority or equivalence margin, and the primary comparisons are imprecise. Nor do these results establish a definitive advantage for either dose: the low number of observed primary events and the width of the confidence intervals limit what can be concluded.

The findings apply to the recorded study population, treatment period, endpoint definition, and safety grouping. The study evaluated two apixaban doses against warfarin over 12 weeks in participants with NVAF. It does not, from the registry record alone, answer questions about other populations, durations, outcomes, or treatment settings. It also does not provide numerical results for several listed secondary outcomes.

The registry result is a statistical description, not medical advice. It should not be used alone to make an individual treatment decision. The record’s limited reporting of analysis methods, missing-data handling, and multiplicity further constrains interpretation. A fuller assessment would require the complete trial report and analysis details, which are not reported in the registry record.

17. Why This Trial Matters Statistically

This trial is statistically instructive because the primary outcome is a clinically defined composite with relatively few observed events. When event counts are sparse, group estimates can look separated while still having wide uncertainty. The primary confidence intervals include zero, illustrating why the numerical ordering of event counts is not enough to establish a comparative conclusion.

The case also shows why endpoint construction matters. Major bleeding and clinically relevant non-major bleeding are combined for the primary outcome but are not identical events. Secondary component counts can help readers understand the composite, while remaining limited by their own sparse counts and by the absence of reported multiplicity control.

Another relevant feature is the safety analysis grouping. The registry states that two participants assigned to apixaban but mistakenly administered warfarin were included in the warfarin group for the safety analysis. A reader therefore needs to distinguish the reported safety grouping from a presumed grouping based solely on randomized assignment. The record does not report an alternative analysis.

Finally, this record distinguishes posted numerical results from the trial’s original statistical plan. Recalculated unadjusted comparisons and planning illustrations can help explain the observed data, but they do not recreate unreported sponsor analyses or design assumptions. For this trial, the lack of reported analysis details, missing-data information, and multiplicity procedures is part of the statistical context, not a basis for filling in what the record leaves unknown.

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.

The primary result is based on few observed events, which limits precision. Both recalculated primary confidence intervals include zero and extend across values on either side of no difference. The comparisons therefore do not establish the direction or size of a definitive effect.

The registry’s numerical comparisons are unadjusted recalculations from posted group summaries. Although the record states that the primary endpoint was adjudicated by a Clinical Event Committee, it does not report the full statistical analysis, adjudication procedures, or whether the reported comparisons were the sponsor’s prespecified analyses. A full analysis plan is not reported in the registry record.

The primary endpoint combines major and clinically relevant non-major bleeding. Those categories have different definitions and clinical implications. Their component counts are sparse, so the composite result should not be assumed to represent an identical pattern for each component. The registry also lists other secondary outcomes, but numerical results for several are not reported in the registry record.

Multiplicity procedures, interim analyses, and the trial’s original sample-size justification are not reported in the registry record. Nor are missing-data handling, follow-up completeness, withdrawals, or a separate intention-to-treat analysis described. The safety analysis grouping includes two participants who were mistakenly administered warfarin, as stated in the record; results using another grouping are not reported.

The study’s 12-week treatment period and phase 2 design also define the scope of these observations. The results should not be generalized to unreported populations, longer periods, or outcomes without additional evidence. These limitations identify what the registry can and cannot establish; they do not prove that bias or an adverse consequence occurred.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Phase 2 Study To Evaluate The Safety Of Apixaban In Atrial Fibrillation — NCT00787150.
• Registry record last updated 2013-05-01; status last verified 2013-04.
• Results first posted 2013-03-06.
• 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

The registry record describes NCT00787150, a completed phase 2 randomized, parallel-group, active-controlled study of apixaban 2.5 mg BID and 5 mg BID versus warfarin in participants with NVAF. The official title describes a 12-week multicenter study and identifies warfarin as open label. The record lists enrollment of 222 and a primary outcome of major or clinically relevant non-major bleeding from baseline to Week 12.

The primary event counts were four in the warfarin group and one in each apixaban group. Recalculated unadjusted risk differences are -0.039 and -0.039, with confidence intervals that include zero. These results are imprecise and do not establish a definitive comparative conclusion. The registry does not report the original sample-size justification, detailed analysis plan, multiplicity procedures, or missing-data methods.