Post Marketing Study in Subjects Who Have Type 2 Diabetes Using the EndoBarrier™ Gastrointestinal Liner: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Post Marketing Study in Subjects Who Have Type 2 Diabetes Using the EndoBarrier™ Gastrointestinal Liner (NCT01114438), a na na study. The primary outcome is Changes in Diabetic Medications at Treatment Completion Compared to Baseline.

2. Clinical Question

The central statistical question is therefore descriptive: among participants with a device who were assessed at treatment completion, how were medication changes distributed across the three listed categories? The endpoint is framed as a comparison with each participant’s baseline medication status rather than as a comparison between treatment groups. The record supplies the endpoint definition and timeframe, but not the category counts, an effect estimate, or an uncertainty interval. Consequently, it does not permit a numerical account of the primary endpoint’s result.

3. Trial Design

This completed post-marketing study evaluated the EndoBarrier Gastrointestinal Liner in subjects with obesity and Type 2 Diabetes. The registry lists Morphic Medical Inc. as the sponsor, a single-group model, and enrollment of 45. Allocation is listed as NA and masking as NONE. The registry lists the phase as NA. The study start is recorded as 2010-10 and primary completion as 2013-01.

A single-group design does not provide a separate concurrent control group for estimating how outcomes might have differed without the device. The primary endpoint instead describes changes from participants’ baseline diabetic medications to treatment completion. Such a before-and-after description can report what was observed among assessed participants, but it cannot by itself distinguish changes associated with the device from changes that could arise over time or alongside other care. The registry record does not describe a randomized comparison, and its allocation information does not identify a control arm.

The primary endpoint timeframe is 12 months, with the assessment described as occurring at EndoBarrier explantation, or treatment completion. The primary analysis population is described as subjects with a device. The number of participants in that population, details of follow-up, and the handling of participants without an assessment are not reported in the registry record. The record lists four outcome measures, but details for the other outcomes are not reported in the information available for this analysis.

The study is marked COMPLETED. That status indicates that the registry records the study as completed; it does not establish that every planned endpoint was reported or that all enrolled participants contributed to every analysis. No protocol-level explanation of the design choices, sample-size justification, or planned statistical analysis is reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was conducted on the Full Analysis Set (FAS) defined as all subjects enrolled and implanted with the EndoBarrier, There were 29 subjects' data available for analysis of HbA1c at the defined 12 month (week 52) endpoint.

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 Changes in Diabetic Medications at Treatment Completion Compared to Baseline, measured over 12 months.

The primary endpoint counts participants according to whether their diabetic medications decreased, increased, or did not change at treatment completion compared with baseline. The assessment is tied to EndoBarrier explantation and has a stated timeframe of 12 months. The analysis population is described as subjects with a device. This is a categorical, participant-count endpoint: it classifies a participant by the direction of medication change rather than measuring a continuous quantity such as dose or a laboratory value.

This kind of endpoint is used to describe medication use over the treatment period. Separating decreases, increases, and no change preserves the direction of change rather than combining all participants into a single average. That can be useful for understanding how medication regimens differed at completion from baseline. However, the registry does not provide the number of participants in each category or the rules used to classify complex changes, such as changes involving more than one medication.

The endpoint description does not establish why a medication changed, whether a change was temporary, or how it related to other clinical measures. The registry record also does not report a comparator group or an inferential estimate for the endpoint. Thus, the measure is suited to a descriptive account of medication changes in the analyzed group, but the posted information does not quantify those changes or support a causal comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count of participants classified into medication-change categories. In general, a count endpoint reports how many people meet a stated outcome definition; it is not the same as a continuous measurement or a rate unless the analysis specifically defines it that way. Here, the categories are decrease, increase, or no change in diabetic medications at treatment completion compared with baseline. The posted information does not include their counts, so no category proportions or contrasts can be interpreted.

What does a confidence interval describe?

In general, a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the analysis. It helps distinguish a relatively precise estimate from one compatible with a wider range of values. No confidence interval for the primary endpoint is reported in the registry record. Without the endpoint counts and an interval or other uncertainty measure, the precision of the primary result cannot be assessed from the posted information.

What can a single-group comparison show?

In general, comparing a participant’s status at follow-up with that same participant’s baseline can describe change over time. It does not provide the same evidence as a randomized comparison with a concurrent control group. Changes may coincide with the intervention without being caused by it, because time, clinical decisions, other care, and participant circumstances can also affect outcomes. This study is recorded as single-group, with allocation listed as NA and masking as NONE. The registry record does not describe a randomized control comparison.

The stated primary analysis population is subjects with a device. Defining an analysis population is important because it indicates who contributes to an endpoint estimate. The number analyzed, the specific inclusion rules, and the treatment of participants with missing endpoint assessments are not reported in the registry record. Enrollment of 45 should therefore not be treated as the primary endpoint denominator without a reported analysis count.

How should planning calculations be understood?

General sample-size planning asks how many participants might be needed to detect an assumed effect under specified statistical assumptions. The planning illustration associated with the posted material is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect defined as a fraction of an observed effect. These are planning assumptions, not a description of the actual single-group study design. They do not establish that the study enrolled two arms or used equal allocation.

Any sample-size figures from such an illustration are hypothetical planning quantities, not observed power and not evidence that the study had a particular probability of finding an effect. Observed power should not be inferred from the completed study or from the number enrolled. The registry record does not report the study’s own sample-size justification, the effect assumed in its original design, or a prospective power calculation. The primary outcome counts needed to describe an observed effect are also not reported in the registry record.

What does over-dispersion mean in general?

For count outcomes, over-dispersion means that observed variability is greater than a statistical model’s assumed variability. If it is present and ignored, uncertainty estimates can be too narrow and statistical tests can be misleading. Whether over-dispersion is relevant depends on the outcome and analysis model. The primary endpoint is described as participant counts in categories, but no model, dispersion estimate, or inferential analysis is reported in the registry record. There is therefore no basis to claim that over-dispersion was assessed or to identify a particular count model.

More broadly, interpretation depends on the analysis plan, definitions, denominators, and uncertainty estimates. The registry record does not report an inferential method for the medication-change endpoint, a statistical adjustment strategy, or criteria for selecting analyses. These omissions mean that the endpoint can be described from its definition, but its numerical result and statistical precision cannot be reconstructed from the posted information.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Decrease in Glucose-Lowering Meds", "Increase in Glucose-Lowering Meds at Week 52", "No Change in Glucose-Lowering Meds at Week 52")
n      <- c(45, 45, 45)
events <- c(23, 4, 18)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Decrease in Glucose-Lowering Meds  45  23  51.1  35.8  66.3  37.0  65.0  36.9  65.1
ARM  Increase in Glucose-Lowering Meds at Week 52  45  4  8.9  2.5  21.2  3.5  20.7  3.6  20.8
ARM  No Change in Glucose-Lowering Meds at Week 52  45  18  40.0  25.7  55.7  27.0  54.5  27.0  54.6

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists four outcome measures, but the information available for this analysis gives details only for the primary medication-change endpoint. The other outcome descriptions and their results are not reported in the registry record. A multiplicity plan explaining whether and how statistical testing across multiple outcomes was controlled is also not reported. Without that plan and the associated analyses, it is not possible to assess how multiple comparisons may have been handled.

Missing-data methods are not reported in the registry record. The primary analysis is described as being conducted on subjects with a device, but the record does not state how many participants had an endpoint assessment, how missing medication information was handled, or whether participants were excluded for incomplete follow-up. It also does not report an imputation method, sensitivity analysis, or assumptions about missingness. These details matter because the analyzed group may differ from all enrolled participants if outcome information is unavailable for some people.

The record’s description of the analysis population is not enough to establish that all enrolled participants were included in the primary analysis. Enrollment was 45; the endpoint denominator is not reported. Similarly, the completed study status does not demonstrate that every participant completed the stated timeframe or that every outcome was observed. No statistical analysis population beyond subjects with a device is described in the registry record.

These reporting gaps limit assessment of the robustness of the primary result. Since the category counts themselves are not reported, the effects of missing information or multiple testing cannot be quantified from the posted data.

12. 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: abdominal pain upper (6.7%), atrial flutter (2.2%), myocardial infarction (2.2%), abdominal pain lower (2.2%), abdominal pain (2.2%).

The registry’s safety figures list 45 participants at risk for serious adverse events and 13 affected, and 45 at risk for other adverse events and 40 affected. These are reported as participant counts. They describe adverse-event reporting in the device group; they are not estimates of a difference from a control group, because no separate control group is described.

The serious-event terms listed most often are Abdominal pain upper at 6.7%, followed by Atrial Flutter, Myocardial infarction, Abdominal pain lower, Abdominal pain, and Chest pain, each at 2.2%. These terms and percentages provide a registry description, but do not by themselves explain event timing, duration, recurrence, clinical course, or attribution to the device. The registry record does not report those details in the information available for this analysis.

Deaths are not reported in the registry record. That should not be interpreted as a confirmed count of zero. Safety figures are distinct from the primary endpoint and do not answer the question about changes in diabetic medications. Because this is a single-group study, the listed counts cannot establish how safety outcomes compare with what would have occurred in a control group.

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

The primary endpoint asks how diabetic medications changed between baseline and treatment completion for participants with a device. The registry defines three possible categories and gives a 12-month timeframe. It does not report how many participants fell into any category. As a result, the posted primary information does not show whether decreases, increases, or no changes were more common, and it does not quantify the precision of any such pattern.

The completed status, the enrollment of 45, and the endpoint definition do not substitute for an endpoint result. The analysis denominator is not reported, nor are a confidence interval or a statistical test. It is therefore not possible to infer the result from enrollment alone. Likewise, the adverse-event counts are separate safety descriptions and cannot be used to fill in the missing primary outcome information.

Even if a within-group change were reported, the design would constrain its interpretation. A single-group before-and-after comparison can describe what changed among observed participants, but it cannot by itself show that the device caused the change. The registry does not describe a randomized control group or an adjusted causal analysis. Other influences on medication decisions and health status could coincide with the study period.

The record supports a limited account of design and reported safety counts. It does not support a conclusion that the device did or did not produce a particular medication change, nor a comparative statement about benefit or harm. Such conclusions would require appropriately reported outcomes and a design or analysis capable of addressing the relevant comparison. The distinction is between what the registry defines, what it reports numerically, and what its design allows readers to infer.

14. Why This Trial Matters Statistically

This study illustrates why registry design details and outcome reporting need to be read together. The medication-change endpoint is understandable as a description of how diabetic medication use differed between baseline and treatment completion. But without the counts in the decrease, increase, and no-change categories, readers cannot characterize the primary result. An endpoint definition alone does not establish what was observed.

The single-group structure is also central to statistical interpretation. A baseline comparison can be informative about change within the observed group, but it does not supply a concurrent counterfactual. Without a separate control group, the study cannot distinguish changes associated with the device from changes occurring for other reasons during follow-up. This limitation is not resolved by the completed status or by knowing the overall enrollment.

The registry’s adverse-event counts add a descriptive safety record, while leaving unanswered questions about timing, attribution, and comparison. Keeping these safety data separate from the medication-change endpoint prevents one type of result from being treated as evidence about another. The lack of reported analysis details, missing-data procedures, and other outcome results further limits independent statistical assessment.

These features make the record useful for understanding what was intended to be measured and what information is available, while also showing why transparent outcome counts and analysis details matter. Clear denominators, category results, uncertainty estimates, and methods for handling missing observations would allow readers to judge the precision and limitations of the findings more fully. In their absence, interpretation should remain descriptive and bounded by the registry record.

15. 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 specific study is that the registry defines the medication-change endpoint but does not report its category counts. The number of participants included in the primary analysis is also not reported. Although overall enrollment is 45, that value does not establish the denominator for the endpoint. The record therefore does not allow readers to describe the primary result or its precision.

The single-group design has no separate concurrent control comparison described. Baseline-to-completion changes may be described, but they cannot by themselves establish that the device caused those changes. Allocation is listed as NA and masking as NONE. The registry record does not describe methods to address confounding or provide an adjusted causal estimate.

Other important details are not reported in the registry record: the number completing the 12-month timeframe, missing-data handling, endpoint classification rules, the statistical analysis plan, and the sample-size justification. The record lists four outcome measures, but information about the other measures is not reported in the registry record for this analysis. Multiplicity handling is also not reported.

The safety information is limited to the listed counts and terms. The registry record does not report further details about timing, duration, recurrence, event attribution, or deaths. These limitations prevent a more complete assessment of both the primary endpoint and safety findings. They do not establish that unreported methods or results were absent from the full study materials; they indicate that they are not reported in the registry record considered for this analysis.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Post Marketing Study in Subjects Who Have Type 2 Diabetes Using the EndoBarrier™ Gastrointestinal Liner — NCT01114438.
• Registry record last updated 2017-04-24; status last verified 2017-04.
• Results first posted 2017-02-20.
• 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.

19. Record Summary

The registry record describes a completed, single-group post-marketing study of the EndoBarrier Gastrointestinal Liner in subjects with obesity and Type 2 Diabetes. Morphic Medical Inc. is listed as sponsor. Enrollment was 45; the study start is recorded as 2010-10 and primary completion as 2013-01. Allocation is listed as NA, masking as NONE, and phase as NA.

The primary endpoint counts participants with decreased, increased, or unchanged diabetic medications at treatment completion compared with baseline, at 12 months. The primary analysis is described as involving subjects with a device. The category counts and primary analysis denominator are not reported in the registry record. Adverse-event entries report 13 participants with serious adverse events and 40 with other adverse events, each among 45 at risk. These safety entries do not provide a control comparison or the missing primary endpoint result.