Drug Interaction Study of Digoxin and BI 10773: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Drug Interaction Study of Digoxin and BI 10773 (NCT01306175), a randomized phase 1 study. The primary outcome is Digoxin: Area Under the Curve 0 to Infinity (AUC0-∞).

2. Clinical Question

The study objective was relative bioavailability: assessing digoxin exposure under the combination condition in relation to digoxin alone. The endpoint concerns measured drug concentration over time, rather than a clinical outcome. The registry identifies the population as healthy volunteers, so the posted results describe pharmacokinetic measurements in that study population and do not by themselves answer questions about outcomes in people with a health condition.

3. Trial Design

This was a phase 1, randomized, open-label, two-way crossover study sponsored by Boehringer Ingelheim. The record reports 20 enrolled participants and a completed study. The stated objective was to investigate relative bioavailability after concomitant multiple oral administration of BI 10773 and a single digoxin dose, compared with digoxin given alone.

In a crossover design, participants receive more than one treatment condition, so treatment comparisons can be made within participants. That general feature can reduce the influence of stable differences between participants on a treatment comparison. The registry identifies the study as a crossover but does not report the sequence assignments, the length or details of any interval between treatment periods, or the conduct of a washout. Those design details are not reported in the registry record and should not be inferred from the design label alone.

The study was open-label, meaning masking was not used. The record describes the reference condition as digoxin alone and the test condition as digoxin with BI 10773. Both are described as oral tablet administration with 240 mL water. The official title specifies a single oral dose of digoxin (0.5 mg) with multiple oral doses of BI 10773 (25 mg once daily). The outcome timeframe lists sampling from 1.5 hours before the first dose through 96 hours after the first dose, with several interim collection times.

The posted primary pharmacokinetic population comprised evaluable participants who had taken at least one dose and had at least one observation for a primary pharmacokinetic endpoint, without important protocol violations relevant to pharmacokinetic evaluation. The record states that one subject was excluded from this set, but the explanation is incomplete: it ends after referring to the pre-dose. The posted primary endpoint counts are 20 for digoxin alone and 19 for the digoxin combination condition. The registry does not explain the difference in those endpoint counts in complete detail.

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.

First Intervention

Washout Period of at Least 14 Days

Second Intervention

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic (PK) set comprised all evaluable subjects who took at least 1 dose of study medication with at least 1 observation for at least 1 primary PK endpoint without any important protocol violations relevant to the PK evaluation. One subject was excluded from the PK set because the pre-dose plasma concentration of digoxin was \>5% of Cmax.

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 Digoxin: Area Under the Curve 0 to Infinity (AUC0-∞), measured over 1.5 hours (h) prior to the first dose and 20 minutes (min), 40 min, 1h, 1.5h, 2h, 2.5h, 3h, 4h, 6h, 8h, 10h, 12h, 24h, 48h, 72h, 96h after the first dose.

The primary endpoint was digoxin area under the plasma concentration-time curve from zero to infinity (AUC0-∞), measured in ng-h/mL. AUC represents accumulated drug exposure over a specified time interval. Extending the interval to infinity includes an estimated contribution after the last measured concentration, rather than restricting the measure to the observed sampling period. The registry describes the endpoint as the area under the digoxin plasma concentration-time curve from zero extrapolated to infinity.

This measure is used in pharmacokinetic comparisons because it condenses concentration measurements over time into an overall exposure measure. In a relative-bioavailability question, comparing AUC under two administration conditions can describe whether measured exposure differs between those conditions. That interpretation depends on the validity of the measurements and analysis; the endpoint alone does not establish a clinical effect or explain the mechanism of any observed difference.

The registry reports the endpoint as a geometric mean, with geometric coefficient of variation as the measure of dispersion. The posted value for digoxin alone is 37.8 ng-h/mL, with a geometric coefficient of variation of 23.2. For the combination condition, the posted value is 39.9 ng-h/mL, with a geometric coefficient of variation of 26.0. These are descriptive summaries. The registry does not post a confidence interval, a treatment ratio, or a hypothesis-test result for the primary comparison, so the precision and inferential uncertainty of the difference between conditions cannot be assessed from these figures alone.

The stated sampling timeframe begins 1.5 hours before the first dose and continues through 96 hours after the first dose, with measurements at specified intervals. The record does not explain the method used to estimate the portion of AUC beyond the last quantifiable concentration.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the design as randomized, open-label, and crossover, and reports the primary endpoint using a geometric mean and geometric coefficient of variation. It does not report the statistical analysis plan, the model used for the primary comparison, or a formal inferential result. The statistical concepts below explain how these methods are generally interpreted; they are not claims about analyses that were conducted in this trial.

What does a geometric mean describe?

A geometric mean is calculated on a multiplicative scale: conceptually, measurements are combined through their logarithms and then returned to the original scale. This approach is often suitable for pharmacokinetic measures that are positive and whose variation is better described proportionally than additively. A geometric mean can be read as a central summary of the distribution on that scale. It is not the same as an arithmetic mean, and the two types of averages need not be equal.

Here the posted geometric means are 37.8 ng-h/mL for digoxin alone and 39.9 ng-h/mL for the combination condition. Those values identify the central summaries but do not quantify uncertainty in a treatment comparison. The registry does not report a geometric-mean ratio or another comparative estimate.

What does the geometric coefficient of variation convey?

A coefficient of variation expresses spread relative to a central value, rather than in the measurement's original units. The geometric coefficient of variation is associated with a log-scale description of variation. The posted values are 23.2 for digoxin alone and 26.0 for the combination condition. They convey variability in the observations summarized for each condition; they are not confidence limits and do not give the uncertainty of a difference or ratio between conditions.

In general, a confidence interval gives a range of values compatible with the data and the assumptions of a specified statistical procedure. Its width conveys precision: a narrower interval indicates greater precision under that procedure, while a wider interval indicates less. No confidence interval for the primary comparison is reported in the registry record. Thus, the extent of uncertainty around the relative exposure cannot be read from the geometric coefficients of variation alone.

Why is the crossover structure relevant?

In general, crossover studies compare treatment conditions within the same participant. A paired analysis can account for stable participant characteristics shared across periods, which may make a within-person comparison more precise than a comparison that treats all observations as independent. That advantage depends on suitable design and analysis. Period effects, treatment sequence, carryover, missing period data, and the handling of incomplete pairs can matter. The registry labels this study a two-way crossover but does not report the sequence details, interval between conditions, or primary analysis model. No conclusion about how these issues were addressed can be drawn from the posted record.

The endpoint counts are 20 for digoxin alone and 19 for the combination condition. These counts do not establish how many participants had complete paired observations, and the registry does not post a participant-level flow by sequence or period. The primary population description says that one subject was excluded from the pharmacokinetic set, but the accompanying reason is truncated. Consequently, the relationship between enrollment, crossover completion, and endpoint availability is not fully described.

Why can over-dispersion matter?

In general, over-dispersion occurs when observed variation is greater than a statistical model expects. If a model understates variability, standard errors may be too small and confidence intervals too narrow, making evidence appear more precise than it is. The appropriate response depends on the outcome and analysis model and may involve a model that accommodates extra variation. The registry does not report an analysis model or an assessment of over-dispersion for this endpoint. The geometric coefficients of variation describe spread but do not establish whether a model was adequate.

What can be inferred from the reported analysis information?

The record posts endpoint summaries but no analyses. It therefore does not identify a test statistic, degrees of freedom, confidence level for a treatment estimate, or adjustment for the crossover structure. The absence of an analysis entry is not evidence that the study team performed no analysis; it means that inferential details are not reported in the registry record. Without those details, a reader cannot reproduce or evaluate a formal comparison.

The registry also describes sample-size assumptions for planning illustrations: a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects expressed as fractions of the observed effect. Such illustrations are not the trial's original design justification and are not observed power. In general, planning calculations describe sample sizes under stated assumptions; they do not show the probability that a completed study detected its effect. The registry does not report the original sample-size justification, and no numerical planning results are stated in the available record.

Finally, multiplicity can arise when several outcomes or comparisons are tested. In general, testing many hypotheses without accounting for the number of tests can increase the chance of at least one false-positive finding. The registry lists one secondary outcome and three outcome measures, but does not report a multiplicity strategy or a hierarchy of analyses. The primary comparison's inferential result is also absent, so the consequences of multiplicity cannot be evaluated from the posted information.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists AUC from zero to infinity as the primary endpoint and AUC from zero to the last quantifiable data point as a secondary outcome. It reports three outcome measures and zero analysis entries. The presence of multiple measures makes multiplicity a relevant general consideration: when several hypotheses are tested, a plan may specify which result is primary and whether adjustments or other controls will be used. The multiplicity strategy, outcome hierarchy, and any adjustment are not reported in the registry record.

Missing-data handling is also not described in detail. The primary population is defined as evaluable participants who took at least one dose, had at least one observation for a primary pharmacokinetic endpoint, and had no important protocol violations relevant to the pharmacokinetic evaluation. The record states that one subject was excluded from this set, but its explanation ends after mentioning the pre-dose. The complete exclusion reason, the number and reasons for other missing observations, and any imputation or sensitivity analyses are not reported in the registry record.

The primary endpoint counts are 20 for digoxin alone and 19 for the combination condition, while the adverse-event groups each include 20 participants at risk. These counts refer to different posted summaries and should not be treated as interchangeable. The registry does not explain all differences in the analysis populations or show how participants with incomplete crossover data were handled.

It is also not reported whether interim analyses were planned or conducted, or what statistical analysis plan governed the primary endpoint. The registry record does not specify whether analyses were prespecified, how protocol deviations were treated beyond the population definition, or whether missingness was considered related to observed outcomes. These omissions limit the ability to assess selection, missing-data, and multiplicity risks.

11. 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 non-serious adverse event terms

The serious adverse event figures list 20 participants at risk in each condition and zero participants affected by a serious adverse event in each. Read narrowly, the posted record reports no serious adverse events among the 20 participants at risk in the digoxin-alone group and none among the 20 at risk in the combination group. The record does not provide serious adverse event terms; the list of top serious terms is empty.

These figures are counts in a study population, not proof that serious adverse events cannot occur. A zero count in a study of this size does not establish the absence of risk, and the figures alone do not support a precise comparison between conditions. The registry does not post a comparative safety analysis or confidence interval for serious adverse events.

The deaths field is null for both groups. A null entry is not the same as a reported count of zero; the number of deaths is not reported in the registry record. The non-serious adverse event counts are two among 20 at risk in the digoxin-alone group and zero among 20 at risk in the combination group. Those counts are separate from serious adverse events and do not provide a complete safety profile or a formal treatment comparison.

Because this was a phase 1 study in healthy volunteers, the safety figures describe the recorded study groups. They do not establish safety in other populations or under other treatment conditions.

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

The result supports a descriptive statement about the primary pharmacokinetic summaries: the registry reports a geometric mean AUC from zero to infinity of 37.8 ng-h/mL for digoxin alone and 39.9 ng-h/mL for digoxin with BI 10773. The corresponding geometric coefficients of variation are 23.2 and 26.0. The endpoint counts are 20 and 19, respectively. These figures describe the posted observations and their reported variability.

They do not, on their own, establish whether concomitant treatment changed digoxin exposure by a meaningful amount. A formal comparison would ordinarily report an estimate of the relative effect and its uncertainty, with the analysis reflecting the crossover design. Neither a treatment ratio nor a confidence interval nor a p-value is reported for the primary endpoint. Similar-looking group summaries are not a substitute for a prespecified equivalence or interaction analysis.

The study's stated aim was relative bioavailability in healthy male and female volunteers. Its scope is therefore specific to the participants, dosing conditions, and endpoint described in the record. The endpoint measures plasma exposure over time; it is not itself a clinical outcome. The posted results do not establish effects in people with health conditions, nor do they establish outcomes under dosing schedules other than those described.

The record's population description and endpoint counts also leave some uncertainty about which observations contributed to the comparison. One participant was excluded from the pharmacokinetic set, but the reason is incompletely recorded. The available summaries do not show complete paired data or explain all missing observations. These limitations matter because the interpretation of a crossover comparison depends on the participant-level data and analysis, not just the separate geometric means.

Accordingly, the defensible conclusion is limited: the registry posts descriptive digoxin AUC summaries for the two conditions, but does not report enough comparative statistical information to determine the precision or inferential status of their relationship. The data as posted do not justify claiming equivalence, absence of an interaction, or a definitive difference.

13. Why This Trial Matters Statistically

This trial illustrates a common distinction in clinical pharmacology between an endpoint summary and an interpretable treatment comparison. AUC from zero to infinity condenses plasma concentration measurements into an exposure measure, and geometric summaries are commonly used for this type of data. Yet two condition-specific means do not by themselves answer whether exposure differs, how large a relative difference might be, or how uncertain that estimate is.

The crossover design makes the analytic structure especially important. When participants receive both conditions, the comparison can use within-participant information. Separate group summaries do not reveal the paired differences or their variability. Without a treatment estimate, a confidence interval, and a description of the model, readers cannot assess the precision gained or lost through the design, or evaluate how period and sequence features were handled.

The registry also distinguishes an endpoint's population from enrollment. Although 20 participants were enrolled, the primary pharmacokinetic summaries include counts of 20 for digoxin alone and 19 for the combination condition. The record notes an exclusion from the pharmacokinetic set but leaves its explanation incomplete. This makes transparent reporting of analysis populations and missing data important to interpreting even a relatively focused phase 1 study.

Finally, the study demonstrates why descriptive safety counts should be read separately from inferential evidence. No serious adverse events are posted among the participants at risk in either condition, but those counts are not proof of no risk and are not a comparative safety analysis. The statistical value of the record lies in its clearly named endpoint and descriptive summaries, while its limited reporting of comparative uncertainty constrains what can be concluded.

14. 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 most important limitation of the posted results is the absence of a formal comparative estimate for the primary endpoint. The registry gives geometric means and geometric coefficients of variation but no treatment ratio, confidence interval, p-value, or primary analysis model. It is therefore not possible to determine the precision or inferential status of the comparison from the record alone.

Although the design is identified as a randomized two-way crossover, the registry does not report treatment sequences, period-specific results, the interval between conditions, or the handling of potential carryover or period effects. It also does not show paired participant-level summaries. These details are important for understanding how the crossover structure informed the analysis.

The pharmacokinetic population description is incomplete in one specific respect: the record says one subject was excluded because of a pre-dose issue, but the explanation is truncated. The endpoint counts differ between conditions, and the record does not fully explain missing observations, exclusions, or completeness of paired data. No imputation approach or sensitivity analysis is reported.

The record lists one secondary outcome and three outcome measures, but does not describe a multiplicity plan, analysis hierarchy, or interim analyses. The original sample-size justification and statistical analysis plan are not reported. Planning illustrations based on stated assumptions do not replace the trial's original design rationale or provide observed power.

Safety reporting is also limited for interpretation. The serious adverse event counts are zero in both groups, but death counts are null rather than stated as zero, and there is no comparative safety analysis. The study population consists of healthy volunteers, which limits the scope of the posted results. These registry entries do not establish what results may appear in other records or reports.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Drug Interaction Study of Digoxin and BI 10773 — NCT01306175.
• Registry record last updated 2014-06-17; status last verified 2014-05.
• Results first posted 2014-06-17.
• 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.

18. Record Summary

NCT01306175 is a completed, randomized, open-label, phase 1, two-way crossover study sponsored by Boehringer Ingelheim. It enrolled 20 healthy male and female volunteers to assess relative bioavailability of a single oral dose of digoxin given alone or with multiple oral doses of BI 10773.

The primary endpoint was digoxin AUC from zero to infinity, reported as a geometric mean and geometric coefficient of variation. The posted summaries are 37.8 ng-h/mL and 23.2 for digoxin alone, and 39.9 ng-h/mL and 26.0 for the combination condition. The corresponding endpoint counts are 20 and 19. No formal comparative estimate, confidence interval, or p-value is reported in the registry record.

The registry reports zero serious adverse events among 20 participants at risk in each condition; death counts are not reported. Important details of the statistical analysis, missing-data handling, and crossover implementation are not reported or are incomplete.