Management of Gastrointestinal and Urogenital (GI/GU) Bleedings in Atrial Fibrillation Patients Using Pradaxa: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Management of Gastrointestinal and Urogenital (GI/GU) Bleedings in Atrial Fibrillation Patients Using Pradaxa (NCT02022020), a registered phase not stated study. The primary outcome is Percentage of Patients With Index Event Safety Outcomes (Ongoing/Resolved/Deceased) at Time of Hospital Discharge.

2. Clinical Question

The study title and brief description focus on the management of gastrointestinal and urogenital bleeding in patients with atrial fibrillation taking dabigatran etexilate. The registry record describes an observational collection of information about these events and their management. It does not provide a detailed account of how patients were selected, how an index event was defined beyond its role in the outcome description, or what clinical decisions were made during care.

The primary outcome uses a practical point in the care pathway: hospital discharge or release. It captures whether bleeding symptoms had resolved by that point, remained ongoing, or the patient had died. The primary-outcome percentage for each status is not reported in the registry record, so the available information does not establish how frequently any of these outcomes occurred.

3. Trial Design

The registry describes an observational study, not a randomized trial. It concerns patients with atrial fibrillation treated with dabigatran etexilate who experienced gastrointestinal or urogenital bleeding. The stated aim was to collect information on management of bleeding events. Enrollment is reported as 220. The sponsor is Boehringer Ingelheim, and the record lists the study status as completed.

The record lists a start date of 2014-01 and a primary completion date of 2015-10. Separately, the primary-outcome timeframe describes data from presentation or admission through in-hospital referrals until discharge, with the data-entry period identified as 28October2010 through 01August2013 and a maximum timeframe of up to 1008 days. The registry record does not explain the relationship between these date descriptions. They should therefore be read as separate statements in the record rather than reconciled by assumption.

The primary-outcome population is described as patients who received treatment in the United States and Canada. The record does not provide country-specific enrollment, site counts, recruitment procedures, or detailed eligibility criteria. It also does not describe a separate control arm, an allocation process, or a blinding procedure. Arm information is limited, and no comparator group suitable for a direct treatment comparison is established by the available record.

The study lists three outcome measures, while analyses are not listed in the registry record. The reported endpoint is the percentage of participants with index-event safety outcomes at hospital discharge. The record explains the status categories, but the percentages for those categories are not reported. Thus, the registry entry gives some information about the study question and outcome definition without supplying the primary result or an inferential 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: 297 subjects were captured in the initial screening. 74 patients were deemed ineligible. A total of 3 patients were withdrawn after being included in the final study cohort following verification of ineligibility leaving 220 subjects eligible for study entry who were enrolled and included in the final study.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Patients who received treatment at two countries (United States and Canada).

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: Patients who received treatment at two countries (United States and Canada).

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Patients With Index Event Safety Outcomes (Ongoing/Resolved/Deceased) at Time of Hospital Discharge, measured over From the time of presentation/admission to an ED/ER or hospitalization through all in-hospital referrals until discharge (between 28October2010 (the date of the first data entry)) and 01August2013 (the date of data entry closure); Up to 1008 days..

The primary endpoint is the percentage of participants whose index-event bleeding status was classified as ongoing, resolved, or deceased at hospital discharge or release. The assessment covers the period from presentation or admission to an emergency department or hospital through in-hospital referrals until discharge. The principal investigator classified status using medical-record information and medical opinion. Ongoing meant bleeding symptoms had not completely resolved at discharge; deceased denoted death; resolved applied otherwise.

This endpoint is useful descriptively because it relates the status of the bleeding event to a recognizable point in care. It distinguishes patients whose symptoms had resolved from those with continuing symptoms and records death as a separate status. It does not, on its own, establish why a status occurred or whether management caused it. Nor does it describe what happened after discharge.

The percentage for each status, the underlying counts, and any uncertainty estimates are not reported in the registry record. Without those results, the endpoint definition can be discussed, but the frequency of ongoing, resolved, or deceased status cannot be assessed from the available information. The record also does not state whether missing discharge assessments occurred or how such cases were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

In general, a percentage outcome expresses how many observations in a defined analysis group meet a stated condition relative to the group used as its denominator. For the primary endpoint, the conditions are ongoing bleeding, resolved bleeding, and deceased status at discharge. To interpret such a percentage, readers need the count in each category and the number of participants assessed for that endpoint. Those primary counts and percentages are not reported in the registry record. Enrollment of 220 is not necessarily the endpoint denominator, so it cannot be substituted without a stated analysis population.

What does the endpoint classification capture?

The classification is based on bleeding status at discharge, rather than on a time-to-event measure or a numerical measurement of blood loss. As defined in the record, ongoing status means symptoms were not completely resolved at discharge, deceased status denotes death, and resolved status applies otherwise. This makes the endpoint a categorical summary of status at a care transition. It does not describe the duration or severity of bleeding in detail, and the record does not specify how disagreements in classification were resolved or whether the classification was assessed independently.

In general, a confidence interval communicates the uncertainty around an estimate, under the assumptions of the statistical method used. For a percentage, that uncertainty depends in part on the number assessed and the number meeting the outcome definition. A confidence interval would not remove limitations arising from selection, measurement, or lack of a comparator. No confidence interval or other uncertainty estimate for the primary outcome is reported in the registry record, and the information available does not support recalculating one.

Why does study design affect interpretation?

General statistical reasoning distinguishes descriptive summaries from comparative estimates. A descriptive percentage characterizes the participants included in an analysis. A comparative estimate asks how outcomes differ between groups and requires a defined comparison. This record describes an observational study and does not establish a like-for-like control group. It does not report a randomization procedure. The posted information therefore does not support interpreting the primary outcome as a treatment effect. Confounding is a general concern in observational comparisons: patient features related to both treatment or care choices and outcomes can make a difference between groups difficult to attribute to treatment. Since a suitable comparison and relevant adjusted analyses are not reported, that concern cannot be addressed from this record.

The record names a primary outcome and lists three outcome measures, but it does not list statistical analyses. It does not report whether estimates were adjusted, which model was used, how uncertainty was calculated, or whether any planned subgroup analyses were conducted. It is not appropriate to infer these choices from the endpoint label or from standard practice. The registry record also does not report an interim-analysis plan or a sample-size justification.

What does over-dispersion mean in general?

Over-dispersion is a general feature of some count-data settings in which observed variability exceeds what a simple statistical model assumes. If it is ignored, standard errors may be too small and uncertainty may appear more limited than it is. That concept is relevant when analysing variable event counts, but the primary endpoint is described as a percentage of participants in categories, and no model for count data is reported. The registry record does not indicate that over-dispersion was assessed or that an over-dispersion adjustment was used.

The registry also gives planning assumptions for sample-size illustrations based on an observed effect: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect stated as a fraction of the observed effect. Such calculations, when presented, are planning illustrations rather than observed power. They do not establish the study's original design rationale. In this record, the primary outcome estimate needed to characterize an observed effect is not reported, and the design does not establish a randomized, like-for-like two-arm comparison. The record does not report the trial's own sample-size justification. Any planning illustration should therefore be kept separate from the study's observed findings and design.

Finally, the timeframe and data-entry dates are part of the outcome description, but the registry record does not explain their relationship to the listed study start and primary completion dates. Statistical interpretation depends on knowing which participants and observations fall within an analysis period. Without further detail on that relationship, readers should not infer a recruitment timeline or follow-up schedule beyond the statements actually recorded.

10. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns arise when several outcomes, comparisons, or subgroup analyses are tested. In general, conducting many tests can increase the chance that at least one result appears unusual by chance. A statistical analysis plan may specify which outcome is primary and how secondary analyses are interpreted. The registry record identifies a primary outcome and lists three outcome measures, but it does not report a multiplicity strategy, a testing hierarchy, or which outcomes were formally tested. Analyses are not listed in the registry record, so it is not possible to determine whether any adjustment for multiple testing was used.

Missing-data methods are also not reported. The registry record does not state whether discharge status was unavailable for any participant, how missing status was handled, or whether participants with incomplete records were excluded. It does not give the number assessed for the primary endpoint. Enrollment of 220 should not be assumed to be the denominator for the primary outcome, because the endpoint-specific analysis population is not stated.

The analysis populations are not described beyond the primary-outcome population of patients treated in the United States and Canada. The record does not define an intention-to-treat population, a per-protocol population, or another analysis set. Because this was observational and no treatment allocation is stated, those labels should not be assumed to apply in their usual randomized-trial sense. The record also does not report whether all enrolled participants were included in each outcome analysis or how repeat events were handled. These omissions limit reproducibility and make it difficult to assess whether the reported adverse-event figures and primary outcome refer to the same participants or observation period.

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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gastrointestinal haemorrhage (23.6%), haematuria (14.5%), rectal haemorrhage (14.5%), anaemia (10.5%), lower gastrointestinal haemorrhage (10.0%).

The registry record lists 184 affected participants among 220 at risk for serious adverse events and 79 affected participants among 220 at risk for other adverse events. These are descriptive figures. They do not establish that dabigatran caused the events, and they are not a comparison with an untreated group or another treatment. The record does not report the definitions used for serious and other adverse events, the event-collection procedures, or the observation period for these figures.

The serious-event terms listed with percentages are gastrointestinal haemorrhage at 23.6%, haematuria at 14.5%, rectal haemorrhage at 14.5%, anaemia at 10.5%, lower gastrointestinal haemorrhage at 10.0%, and upper gastrointestinal haemorrhage at 10.0%. The denominator for these term percentages is not reported in the registry record. The terms should therefore be read as listed event percentages, not as rates with a confirmed denominator or as mutually exclusive categories. The record does not say whether a participant could contribute to more than one term.

A death count is not reported in the adverse-event information. The primary endpoint includes deceased as a discharge-status category, but its result is not reported either. These pieces of information should not be combined or treated as interchangeable. The serious-event figures describe reported adverse events; the primary endpoint describes a status classification at discharge.

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

The record establishes that the study focused on management of gastrointestinal and urogenital bleeding events in patients with atrial fibrillation treated with dabigatran etexilate. It defines a primary outcome that classifies bleeding status at discharge as ongoing, resolved, or deceased, and identifies the clinical records and medical opinion used for classification. It also reports serious and other adverse-event counts and lists several serious-event terms with percentages.

Those details do not establish the frequency of each primary-outcome status, because the associated counts and percentages are not reported in the registry record. They also do not establish a comparative treatment effect. An observational description of patients receiving a treatment cannot, on its own, show what would have occurred under a different treatment or management strategy. No like-for-like comparator, random allocation, or comparative analysis is described.

The adverse-event figures likewise do not prove that the treatment caused the recorded events. The record does not give the event definitions, the term-percentage denominator, or enough detail about observation and analysis populations to place those figures in a comparative risk framework. The absence of these details does not mean that an outcome did not occur; it means that the registry record does not report enough information to characterize it further.

In general, statistical evidence is best interpreted by considering the estimate, its uncertainty, the study design, and the way outcomes were measured. Here, the primary estimate and its uncertainty are unavailable, and the observational design limits causal interpretation. The most supportable reading is therefore about the scope and structure of the registry entry: it documents a descriptive study question and outcome definitions, but the reported information does not answer how often each discharge status occurred or whether one management approach was better than another.

13. Why This Trial Matters Statistically

This record is statistically relevant because it illustrates the distinction between a clearly named outcome and a reported outcome estimate. The endpoint specifies a clinically interpretable status at discharge and defines its categories. Yet without the category counts, percentages, or uncertainty estimates, readers cannot evaluate the observed distribution. A precise endpoint definition is valuable, but it cannot substitute for reporting the result and denominator.

The study also demonstrates why design labels matter. The registry describes observational data collection and does not establish a randomized comparison. In general, observational data can describe events and care in the population under study, but comparison and causal interpretation require attention to how patients came to receive treatment, how outcomes were measured, and what other factors may influence them. The record does not provide enough detail to resolve those issues or support a comparative estimate.

Its adverse-event reporting provides counts for affected participants and percentages for selected serious-event terms, but the context needed to interpret those percentages is incomplete. Denominators, event definitions, and observation details are not reported in the registry record. This makes it important to distinguish a listed percentage from a fully specified risk estimate.

The record also contains different date descriptions: study start and primary completion dates, and a separate data-entry period in the endpoint timeframe. When records contain such differences, statistical readers should preserve the distinction rather than construct a timeline that is not documented. Overall, the study matters as an example of how outcome definition, analysis populations, denominators, and design determine what can be learned from a registry entry.

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 primary limitation for interpreting this record is that the primary outcome results are not reported. Although the endpoint categories and assessment point are described, the category counts, percentages, analysis denominator, and uncertainty estimates are absent. The registry record therefore does not show the observed discharge-status distribution.

The observational design and lack of a described comparator prevent a direct treatment comparison. The record does not state an allocation method or describe a control group that can be considered like-for-like. It also does not provide detailed eligibility criteria, recruitment procedures, baseline characteristics, or information about the severity of bleeding at presentation. These omissions limit assessment of which patients the results would represent and how patient differences might affect outcomes.

The outcome classification relied on medical-record information and medical opinion, but the record does not describe classification procedures beyond those terms. It does not report whether assessors followed a standardized adjudication process, whether assessments were checked for consistency, or how incomplete information was handled. The timeframe includes data-entry dates that are not reconciled with the listed study start and primary completion dates.

For adverse events, the record does not report definitions, event-collection methods, observation period, or the denominator for the listed term percentages. It does not specify whether event categories could overlap. The analysis populations, missing-data approach, and statistical analysis methods are not reported, and no analyses are listed in the registry record. These limitations constrain interpretation of both the primary endpoint and the adverse-event summaries.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Management of Gastrointestinal and Urogenital (GI/GU) Bleedings in Atrial Fibrillation Patients Using Pradaxa — NCT02022020.
• Registry record last updated 2017-01-31; status last verified 2016-12.
• Results first posted 2017-01-31.
• 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

NCT02022020 is a completed observational study sponsored by Boehringer Ingelheim, concerning management of gastrointestinal and urogenital bleeding events in patients with atrial fibrillation treated with dabigatran etexilate. Enrollment is reported as 220. The primary endpoint classifies bleeding status at hospital discharge or release as ongoing, resolved, or deceased, based on medical-record information and medical opinion.

The registry record reports serious adverse events in 184 affected participants among 220 at risk, and other adverse events in 79 among 220 at risk. It lists selected serious-event terms with percentages. The primary outcome counts and percentages, analysis methods, endpoint denominator, and analysis populations are not reported in the registry record. No like-for-like control comparison is described, so the entry supports descriptive interpretation rather than a comparative treatment-effect conclusion.