Assessment of an Education and Guidance Programme for Eliquis Adherence in Non-Valvular Atrial Fibrillation (AEGEAN): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Assessment of an Education and Guidance Programme for Eliquis Adherence in Non-Valvular Atrial Fibrillation (AEGEAN) (NCT01884350), a randomized phase 4 study. The primary outcome is Percentage of Days With a Correct Execution of the Apixaban Dosing Regimen.

2. Clinical Question

The trial addresses an adherence question, rather than a comparison of apixaban with another drug or with no treatment. Both groups received apixaban; the planned difference between them was the accompanying information or education. Correct execution was defined around taking the prescribed tablet twice daily within a 24-hour window. A day could still count as an adherence unit if one dose was missed during that window.

The registry record describes the study purpose as assessing the impact of an educational programme on adherence in participants taking apixaban. It also lists later adherence measures and safety outcomes. The primary question is narrower: it concerns average adherence during the period ending at week 24, not every aspect of adherence over the full treatment period.

3. Trial Design

AEGEAN was a phase 4, randomized, parallel-group trial with no masking. The registry record lists an enrollment of 1217 and a completed status. Its recorded start date is 2013-10-15, and its primary completion date is 2016-01-20. Bristol-Myers Squibb is listed as the sponsor.

The two primary groups received apixaban at either 2.5 mg or 5 mg by mouth twice daily for 48 weeks. One group received primary standard-of-care information; the other received an additional educational programme. The primary adherence endpoint covered day 1 up to week 24, so it did not cover the entire stated treatment period.

Randomization supports a planned comparison between these education conditions, while the parallel design means participants were assigned to a group rather than moving through both primary conditions. No masking was used, so the study was open-label as recorded. The registry record does not report further detail about the randomization procedure, allocation concealment, or how the additional educational programme was delivered.

The primary analysis population is called the primary efficacy analysis set at week 24 and is described as consisting of all randomized participants. The arm-level results list 583 participants in the primary standard-of-care group and 579 in the additional-education group. The registry record does not explain the relationship between those analysis counts and the enrollment figure, or identify the reasons participants may not appear in the reported endpoint analysis.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: 1217 participants were enrolled, 1162 randomized (583 Primary SOC,579 AEP). 55 were enrolled but not randomized. Of the 55, 13 no longer met study criteria, 13 withdrew consent, 3 lost to follow-up, 10 other and 16 unknown.

Primary Efficacy Analysis Set (24 Weeks)

Reasons for not completing this period:

Primary Efficacy Analysis Set (48 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Primary efficacy analysis set (Week 24), which consists of all randomized participants

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Primary Efficacy Analysis Set (24 week)

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Days With a Correct Execution of the Apixaban Dosing Regimen, measured over Day 1 up to week 24.

The primary endpoint is the mean percentage of days on which participants correctly executed the apixaban dosing regimen from day 1 up to week 24. It measures adherence over time rather than a single dose or a clinical event. The reported measure is a group mean, with a standard deviation describing variation among participants.

For this endpoint, an adherence unit is a 24-hour window in which treatment is taken as prescribed: one tablet, at the appropriate 2.5 mg or 5 mg strength, twice a day. If one dose is missed within a 24-hour window, the window still counts as an adherence unit. Adherence is expressed as the percentage of eligible days represented by adherence units.

The endpoint provides a direct way to compare recorded dosing behavior between the two education conditions. It does not by itself measure why a dose was missed, establish a clinical benefit, or describe adherence beyond the endpoint period. The registry record gives special rules for ending the denominator early for some participants who discontinue: the period can be censored at the earlier of the last dose or discontinuation date for specified reasons unrelated to adherence, or otherwise at the minimum of 169 days and the last dose date plus 30 days. The impact of these rules on the reported result is not reported in the registry record.

7. Results Posted in the Registry

• Apixaban (Additional Educational Program): mean 91.88 versus 91.64 on control; difference 0.24 (95% CI -1.67 to 2.15), standardized effect d = 0.01; z-test p = 0.806.

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.

Percentage of Days With a Correct Execution of the Apixaban Dosing Regimen During the 24 to 48 Weeks Period

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Apixaban (Additional Educational Program): mean 88.41 versus 87.59 on control; difference 0.82 (95% CI -2.56 to 4.20), standardized effect d = 0.04; z-test p = 0.634.
  • Apixaban (Secondary SOC): mean 87.51 versus 87.59 on control; difference -0.08 (95% CI -3.35 to 3.19), standardized effect d = -0.00; z-test p = 0.962.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Apixaban (Additional Educational Program) 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 (Secondary SOC) and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results extend the adherence assessment beyond the primary period and include a comparison among primary standard-of-care information, additional education, and secondary standard-of-care conditions. The broad pattern is that the recorded average adherence measures remain close across the groups in the later period, while the reported analyses do not show a clear group difference. This is broadly consistent with the primary comparison's lack of a clearly separated average.

The secondary analyses are not interchangeable with the primary result. They cover a different time window or different group contrasts, and the later-period analysis involves a subset of participants with posted arm counts. The record also lists secondary questions about adherence change, predictors of non-adherence, and adverse events, but detailed results for all those outcomes are not reported in the registry record.

The registry reports an ANOVA across the three groups and a separate two-sided t-test comparing additional education with secondary standard-of-care information. The ANOVA result is classified as non-inferiority or equivalence, but a label alone does not establish that the study demonstrated either. The non-inferiority or equivalence margin and the supporting criteria are not reported in the registry record. A failure to detect a difference is not the same as evidence that groups are equivalent.

Secondary findings generally carry less weight than the primary result because they are additional questions beyond the main endpoint and can involve multiple comparisons. Without a reported multiplicity plan, the chance of at least one apparently unusual result across many tests cannot be evaluated from the registry record. These outcomes can inform the overall pattern, but they do not replace the primary comparison or establish a separate definitive conclusion.

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

• Percentage of Days With a Correct Execution of the Apixaban Dosing Regimen During the 12 to 24 Weeks Period Compared With During the First 12 Weeks
• Percentage of Days With a Correct Execution of the Apixaban Dosing Regimen During the 24 to 48 Weeks Period
• Non-adherence Predictors of 20% or More (vs. at Least 80% Adherence) at 24 Weeks
• Number of Participants With Serious Adverse Events (SAEs), Drug Related Adverse Events (AE), AE Leading to Discontinuation, and Death
• Number of Participants With Serious Adverse Events (SAEs), Drug Related Adverse Events (AE), AE Leading to Discontinuation, and Death

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: it is a percentage measured for each participant and summarized as a mean for each group. A comparison of means is a natural way to describe the average difference between the randomized conditions. The posted arm summaries include each group's participant count, mean, and standard deviation; the recalculation uses those summary statistics to estimate an unadjusted contrast.

What does the mean difference represent?

The recalculated mean difference is 0.24 percentage points, defined as the additional-education mean minus the primary standard-of-care mean. A positive estimate means the observed average was higher in the additional-education group under that direction of comparison; a negative estimate would mean it was lower. This is a difference in percentage points, not a relative percentage change. No relative effect is calculated.

The sponsor's reported estimate is -0.24 because its contrast is presented in the reverse direction. A reversal of contrast direction changes the sign but does not itself change the underlying distance between group means. This directional convention matters when comparing a reported estimate with a recalculation.

Why use a confidence interval?

In general, a confidence interval describes the range of effect estimates compatible with the data and the assumptions of the analysis at the stated confidence level. The primary recalculated interval is -1.674 to 2.154 at 95%. Because zero lies within it, the results are compatible with no mean difference as well as with differences in either direction. The interval is not a probability statement that the true effect lies within those particular limits, and it does not establish equivalence.

What does the p-value add?

In general, a p-value measures how unusual data at least as far from a null value would be under a specified null model and its assumptions. The unadjusted recalculated p-value is 0.806, and the registry's two-sided t-test reports p = 0.812. Neither number is the probability that the null hypothesis is true. A p-value also does not show whether an effect is important; effect size and uncertainty are better read together.

Why can participants vary so much?

In general, the standard deviation describes how dispersed participant measurements are around a group mean. The endpoint summarizes repeated adherence opportunities into a participant-level percentage, and participant-level percentages can vary even when group averages are similar. The standard deviations reported for the primary arms are 17.14 and 16.14. They are descriptive measures of variation, not confidence limits for the group means.

What is an unadjusted re-calculation?

In general, a simple unadjusted comparison uses the posted group summaries without incorporating additional covariates or a more complex model. It can serve as a transparent check on the scale and direction of the contrast. It does not reproduce unreported modeling choices, data handling, or adjustments that might have been used in the trial analysis. The registry lists the sponsor's primary method as a two-sided t-test. Further details of the analysis model are not reported in the registry record.

The same broad distinction applies to secondary outcomes. An analysis across three groups using ANOVA tests for evidence of group differences under that method, while a two-sided t-test can compare a specified pair. A non-significant omnibus test or pairwise test does not demonstrate that groups are equivalent. Equivalence or non-inferiority interpretation requires a prespecified margin and an appropriate analysis; the margin and supporting details are not reported in the registry record.

General statistical caution also applies to adherence percentages: observations within a participant may be related because they come from the same person over time. The registry's endpoint description reduces the dosing record to a participant-level percentage, but does not report further modeling of repeated observations. It is therefore not possible to assess from the record whether any additional longitudinal modeling or sensitivity analysis was used.

11. Confidence Intervals and Precision

The primary recalculated 95% confidence interval for the mean difference is -1.674 to 2.154 percentage points, using the direction additional education minus primary standard-of-care information. It spans zero. In practical statistical terms, the observed data do not pin down a single direction for the average difference: values on either side of no difference remain compatible with the interval.

The interval is relatively wide compared with the very small observed mean difference of 0.24. That combination means the point estimate alone should not be treated as a precise measurement of the effect. The confidence interval gives a more informative account of the range of uncertainty than the point estimate or p-value alone. It should not be read as a guarantee that all effects inside the limits are equally plausible, or as a statement that effects outside the limits are impossible.

The registry's own primary t-test reports a 95% interval from -2.18 to 1.7 for a difference of -0.24, in the reverse contrast direction. Its limits also include zero. The recalculated and reported intervals are not identical, and the registry record does not describe the reason for that difference. The trial's reported analysis and the unadjusted recalculation should therefore be kept distinct.

Precision for a mean difference depends on the amount of information in the comparison and the variability of the measure. The primary standard deviations are 17.14 and 16.14, while the arm-level analysis counts are 583 and 579. These summary figures help contextualize the interval, but do not remove uncertainty about missing endpoint data or unreported analysis choices. The interval concerns the mean contrast, not individual adherence or the possible effect for any particular participant.

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 own design calculation. For these continuous outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect.

The 80% power column describes the number of participants per arm that a hypothetical study would require, under those assumptions, to have an 80% chance of detecting the specified effect if that effect were true. The 90% power column describes the corresponding planning target for a 90% chance. These are design probabilities under assumptions, not probabilities that the observed trial result is correct.

Each row changes the target effect relative to the observed effect. Planning for a smaller effect requires more participants because a smaller difference is harder to distinguish from the natural variation in the outcome. When the observed effect is very small, the resulting planning counts can become very large. This is a property of the illustrative calculation and should not be mistaken for a recommendation about an appropriate future trial.

These figures are not observed power. The original trial's sample-size justification, target effect, and design assumptions are not reported in the registry record. The planning illustration also does not account for details not included in its stated assumptions, such as missing data or other design features.

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 = 0.240; pooled standard deviation σ = 16.651.
2. n = 2 × (1.96 + 0.842)² × 16.651² / 0.240² = 75,558.7, rounded up to 75,559 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 <- 0.239999999999995                      # observed difference in means
sigma <- 16.6507831988831                     # 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(1884350)
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 = 75,559 (80% power), 101,152 (90% power)
 75% of observed effect: n per arm = 134,327 (80% power), 179,826 (90% power)
 50% of observed effect: n per arm = 302,235 (80% power), 404,607 (90% power)
power.t.test: 75559.7 
Simulated power at n = 75,559 per arm: 0.801 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The registry reports a two-sided t-test for the primary comparison. It gives a difference in percentage of -0.24, a 95% confidence interval from -2.18 to 1.7, and p = 0.812. This is the sponsor's own posted analysis of the trial.

The unadjusted recalculation describes the contrast in the opposite direction: additional education minus primary standard-of-care information. Its estimate is 0.24, with a 95% confidence interval from -1.674 to 2.154 and an unadjusted p-value of 0.806. The reversed sign is consistent with the contrast being written in the other direction. The reported and recalculated results are close in overall interpretation: both intervals include zero and both p-values are large. They are not numerically identical.

The registry record does not explain the difference between the reported and recalculated interval limits or p-values. The sponsor's result comes from the trial's reported t-test, whereas the recalculation uses the posted arm means, standard deviations, and counts without adjustment. They should not be presented as though they came from an identical analysis. The discrepancy indicator is false; no separate discrepancy between the reported result and the recalculation is flagged in the record.

15. Multiplicity, Missing Data and Analysis Populations

The registry record lists multiple outcome measures and analyses, including the primary adherence endpoint, later adherence periods, non-adherence predictors, and safety outcomes. It does not report a multiplicity adjustment or a hierarchy specifying which secondary comparisons should be treated as confirmatory. The number of analyses recorded is 26. Without a stated testing strategy, p-values for additional outcomes should be interpreted cautiously, particularly when many comparisons may be considered.

In general, testing multiple hypotheses increases the opportunity to see a small p-value by chance if all tested null hypotheses are true. Adjustment procedures or a prespecified hierarchy can address that issue, but the record does not say whether such methods were used. The secondary findings are therefore best viewed as supportive or exploratory unless further analysis documentation establishes otherwise.

The registry describes the primary efficacy analysis set as all randomized participants, but the posted primary arm counts are 583 and 579. Enrollment is listed as 1217. The record does not provide a participant flow accounting, reasons for missing endpoint values, or the number excluded from the primary endpoint calculation. It also does not explain whether the denominator censoring rules led to differing amounts of observation across participants.

For some discontinuations, the endpoint description specifies how the eligible period is censored. However, the record does not report how many participants were affected by each rule, how missing adherence information was handled, or whether sensitivity analyses were performed. It does not report a general missing-data assumption or imputation method. As a result, the potential effect of incomplete adherence records on the group means cannot be assessed from the registry record alone.

The analysis details beyond the listed two-sided t-test are not reported in the registry record. In particular, a full statistical analysis plan, covariate adjustment details, interim-analysis plans, and the original sample-size justification are not reported in the registry record. These omissions limit assessment of whether all analyses followed prespecified methods.

16. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cardiac failure (2.3%), atrial fibrillation (1.8%), cardiac ablation (1.3%), myocardial infarction (0.9%), dyspnoea (0.9%).

The safety record presents serious adverse event counts and counts for other adverse events across named groups and periods. For example, the primary standard-of-care period lists 69 participants with serious adverse events among 604 at risk, and 40 with other adverse events among 604 at risk. The additional educational programme period lists 79 participants with serious adverse events among 558 at risk, and 46 with other adverse events among 558 at risk. These are counts within the registry's stated groups and periods, not the primary adherence comparison.

Other listed groups include continued additional education, secondary standard-of-care information, and primary standard-of-care information during period 2. Their numbers at risk and event counts are recorded separately; participants and observation periods should not be assumed to be directly comparable across these entries without further design details. The registry also lists cardiac failure, atrial fibrillation, cardiac ablation, myocardial infarction, dyspnoea, and pneumonia among the serious-event terms, with percentages recorded for each.

Deaths are shown as not reported in the registry record. The safety counts alone do not establish whether an event was caused by apixaban or by the education condition, and they do not support a treatment recommendation. The registry's group and period definitions should be retained when interpreting these figures.

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

The posted primary results show similar average adherence percentages in the two randomized education groups through week 24, with a small observed mean difference and an uncertainty interval that includes zero. This is a description of the recorded comparison, not proof that the programmes produce identical adherence or that any possible difference is clinically unimportant.

The result does not establish the absence of an effect. The confidence interval includes differences in both directions, and the registry record does not state a threshold for what size of adherence change would matter. It also does not establish that education has no effect on other outcomes, on adherence outside the primary time window, or on particular participants.

Nor does the comparison test apixaban against a different medicine or no medicine. Both groups received apixaban, and the contrast concerns the accompanying educational and standard-of-care information. The adherence endpoint measures a dosing behavior; it is not itself a clinical outcome such as an event related to atrial fibrillation.

The later adherence results provide additional context but cover different periods and comparisons. The registry record does not report complete results for every listed secondary outcome, a multiplicity plan, detailed missing-data handling, or all analysis specifications. These limits mean the evidence should be interpreted as a registry-reported comparison of average adherence, with uncertainty and reporting gaps, rather than as a complete account of every possible benefit, harm, or mechanism.

Finally, group averages do not characterize each participant. The reported standard deviations indicate variation around the mean, and the analysis does not identify individual reasons for missed doses. The results therefore describe the groups' average recorded adherence under the trial conditions, not what any particular person should do.

18. Why This Trial Matters Statistically

This trial is statistically useful because it isolates an educational contrast while keeping the medication regimen in both groups. That design makes the primary comparison a question about average adherence under two information conditions, rather than a comparison of different drugs. The endpoint also turns dosing records across time into a participant-level percentage that can be compared between randomized groups.

The result illustrates why estimates and confidence intervals should be read together. The means are close, but the interval remains broad enough to include values on either side of no difference. Reporting only a p-value would conceal the range of effects compatible with the data; reporting only the point estimate could make the result seem more precise than it is.

The record also illustrates distinctions that matter in clinical research reporting: a registry-listed trial analysis versus an unadjusted recalculation, a primary endpoint versus multiple secondary questions, and a failure to demonstrate a difference versus evidence of equivalence. The later analysis includes an equivalence or non-inferiority classification, but the record does not include the margin needed to interpret that label.

Finally, the adherence endpoint and safety reporting raise practical statistical questions about missing observations, censoring, analysis populations, and repeated measurements. The registry explains some endpoint rules but not the extent to which those rules affected participants. Recognizing what the record does not report is part of interpreting the numerical result responsibly.

19. Limitations

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

The primary endpoint is a behavioral measure constructed from dosing windows and eligible days. The registry describes its definition and censoring rules, but does not report the source or completeness of the adherence records, the number of participants affected by each censoring rule, or sensitivity analyses under alternative missing-data assumptions. The influence of incomplete recording on the means is therefore not assessable from the registry record.

The analysis population description and the arm counts require care. The primary efficacy set is described as all randomized participants, while the posted arm-level counts are 583 and 579 and enrollment is 1217. The record does not reconcile these numbers or provide a participant flow diagram. The reasons for missing endpoint information and any exclusions are not reported in the registry record.

The study was unmasked. The record does not describe how that might have affected adherence recording, participant behavior, or other aspects of measurement. It also gives limited detail about the educational programme itself, the randomization procedure, and implementation of standard-of-care information.

The statistical reporting is incomplete for some interpretive purposes. A full analysis plan, the original sample-size rationale, multiplicity handling, details of the equivalence or non-inferiority margin, and any interim analyses are not reported in the registry record. The listed secondary outcomes do not all have detailed posted results. Safety counts are reported by period and group, but deaths are not reported in the registry record, and those counts alone do not establish causation.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Assessment of an Education and Guidance Programme for Eliquis Adherence in Non-Valvular Atrial Fibrillation (AEGEAN) — NCT01884350.
• Registry record last updated 2019-08-19; status last verified 2019-07.
• Results first posted 2019-08-19.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

AEGEAN was a completed, phase 4, randomized, parallel-group trial without masking, with an enrollment of 1217. It compared apixaban plus primary standard-of-care information with apixaban plus an additional educational programme in participants with non-valvular atrial fibrillation.

The primary outcome was mean percentage of days with correct apixaban dosing from day 1 up to week 24. The posted group means were 91.64 and 91.88. The recalculated unadjusted difference was 0.24 percentage points, with a 95% confidence interval from -1.674 to 2.154; the registry's two-sided t-test reported p = 0.812.

The primary comparison does not show a clearly separated average adherence result, but its uncertainty does not establish equivalence. Later adherence and safety outcomes are also listed, with limits from incomplete reporting of analysis and missing-data details.