Ursodiol on Insulin Sensitivity, Gastric Emptying and Body Weight With Type 2 Diabetes on Metformin: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ursodiol on Insulin Sensitivity, Gastric Emptying and Body Weight With Type 2 Diabetes on Metformin (NCT02033876), a randomized phase 2 study. The primary outcome is Change in Area Above Basal (AAB) for Glucose.

2. Clinical Question

This was a short-term physiological study rather than a registry record reporting long-term clinical outcomes. Its stated interests included insulin sensitivity, glycemic control, gastric emptying, and body weight. The primary endpoint focused specifically on the glucose response used to calculate area above basal. The outcome timeframe was baseline and post-treatment, approximately 14 - 17 days.

The central statistical question is how to read the posted arm summaries alongside the sponsor-reported comparison. The median for the Ursodiol arm was 66460.0 mg/dL x min, with an inter-quartile range of 53500.0 to 75480.0. The placebo median was 72570.0 mg/dL x min, with an inter-quartile range of 55650.0 to 93010.0. The ANCOVA p-value is a test result, not an estimate of the size or precision of the arm difference. The registry record does not report the confidence interval or the ANCOVA effect estimate needed to describe those quantities directly.

3. Trial Design

The registry identifies this as a phase 2, randomized, triple-masked, parallel-group trial, with a total enrollment of 24. The posted primary-outcome summaries include 12 participants in the Ursodiol arm and 12 in the placebo arm. The study status is completed. The sponsor is Mayo Clinic. The registry gives a start date of 2013-10 and a primary completion date of 2017-01.

The parallel design compares participants assigned to one of the two arms over the study period. The active intervention was delayed-release ursodeoxycholic acid, described as 600mg taken twice daily; the comparator was matching placebo capsules taken twice daily. The registry identifies masking as triple, but does not report which specific groups or roles were masked. The exact sequence-generation procedure, concealment method, and practical details of the masking are not reported in the registry record.

The primary outcome was change in area above basal for glucose, with results assessed at baseline and post-treatment approximately 14 - 17 days. The record describes mixed-meal glucose results as the basis for calculating the endpoint. The sponsor-reported primary comparison used ANCOVA and was classified as a superiority analysis. The specific ANCOVA model, including any covariates, is not reported in the registry record.

The record also lists secondary outcomes covering fasting glucose, insulin sensitivity, gastric emptying of liquids and solids, weight, body mass index, and FGF-19. Numerical results for these outcomes are not reported in the registry record. No details about the planned sample-size justification, randomization ratio, or analysis populations are reported. The posted equal arm counts describe the primary-outcome summaries; they do not establish every detail of the allocation procedure or participant flow.

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

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 Change in Area Above Basal (AAB) for Glucose, measured over baseline, post-treatment approximately 14 - 17 days.

The primary endpoint is titled change in area above basal (AAB) for glucose and is measured in mg/dL x min. The registry says mixed-meal glucose results are used to calculate the AAB. In general, an area-based glucose measure represents the accumulated response above a basal level across the measurement period: it reflects both the degree of elevation and how long it persists. It therefore describes a meal-related glucose response in a way that a single glucose reading cannot.

The registry record relates the endpoint to the broader concept of the glycemic index, which it describes using an incremental area under the two-hour blood glucose response curve after an overnight fast and ingestion of food with a specified quantity of available carbohydrate. That description provides context for area-based glucose measurement; the registered primary endpoint is specifically AAB calculated from mixed-meal results.

The outcome timeframe is baseline and post-treatment approximately 14 - 17 days. The registry labels the endpoint as a change, but does not report additional details about the calculation of the change, the timing of meal tests, or the individual measurement schedule. The posted arm summaries are medians with inter-quartile ranges. The endpoint can characterize the observed glucose response over the study measurement period, but the registry record does not report longer-term glycemic outcomes or clinical events as the primary endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the median describe?

The median is the midpoint of an ordered set of observed values: half of the observations fall at or below it and half at or above it. Unlike the mean, it is not calculated by adding all values and dividing by the number of observations. A median is often useful when the distribution may be skewed or when unusually high or low observations could strongly affect the mean. The registry identifies the primary outcome summary as a median and its dispersion as an inter-quartile range.

The inter-quartile range gives the bounds containing the middle half of observed values. For Ursodiol, the median was 66460.0 and the inter-quartile range was 53500.0 to 75480.0 mg/dL x min. For placebo, the median was 72570.0 and the inter-quartile range was 55650.0 to 93010.0 mg/dL x min. These summaries describe each arm separately. They do not by themselves estimate the adjusted contrast between arms, and their overlap or separation is not a formal test of a treatment difference.

What does ANCOVA contribute?

ANCOVA, or analysis of covariance, compares groups while including one or more covariates in a statistical model. In general, covariate adjustment can account for measured variables associated with the outcome and may improve the precision of a comparison. The particular covariates, model specification, and assumptions used for this trial are not reported in the registry record. The posted ANCOVA result is p = 0.43; no adjusted effect estimate or confidence interval accompanies it in the registry record.

A p-value measures how unusual data at least as inconsistent with a specified null hypothesis would be under the model, if that null hypothesis and model assumptions held. It is not the probability that the null hypothesis is true. It also does not quantify the size of an effect or show whether it matters clinically. Interpretation is stronger when the test result is accompanied by an effect estimate and confidence interval, neither of which is reported for the ANCOVA comparison.

How should uncertainty be understood?

In general, a confidence interval gives a range of effect values compatible with the observed data under the analysis procedure. Its width conveys precision: a wider interval indicates less precision, while a narrower interval indicates more. The registry record does not report a confidence interval for the primary comparison. The inter-quartile ranges are not confidence intervals; they describe the spread of individual outcome values within each arm rather than the uncertainty in the between-arm effect.

With 12 participants in each posted arm summary, the arm distributions are based on a limited number of observations. General statistical principle: smaller samples tend to provide less precise estimates and less stable descriptions of distributions than larger samples, all else being equal. That principle does not permit the missing uncertainty interval to be reconstructed from the medians and inter-quartile ranges alone. The ANCOVA result also cannot be recreated from those summaries because the model details and participant-level observations are not reported.

What can the design establish?

Randomization is intended, in general, to make assigned groups comparable on average and to reduce systematic confounding in a treatment comparison. The registry identifies the trial as randomized, but the sequence-generation and allocation-concealment procedures are not reported. Triple masking can reduce expectation or assessment influences in general, but the record does not identify which people were masked or describe how masking was maintained. A parallel design means the arms are compared as distinct groups over the study period.

The sponsor classified the primary analysis as a superiority analysis. A superiority test assesses evidence for a difference in a specified direction or in either direction, depending on its formulation; exact hypotheses are not reported in the registry record. No equivalence or non-inferiority margin is reported. Therefore, the p-value should not be read as demonstrating that the treatments are the same.

What do planning calculations mean?

General explanation: a sample-size calculation estimates how many participants would be needed under stated assumptions to detect a specified effect with a chosen error rate and power. Such a calculation is a planning exercise, not a measure of how much power the completed trial actually had. Any sample-size illustration based on the observed effect is likewise a planning calculation, not observed power. The registry describes planning assumptions of a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. The actual protocol sample-size justification is not reported in the registry record.

Over-dispersion is a general concern when outcome variability is greater than a model assumes; in that case, standard errors can be too small and tests can appear more decisive than warranted. The registry does not report a specific over-dispersion assessment for this endpoint, and the recorded ANCOVA details are insufficient to assess model diagnostics. For this trial, the most defensible interpretation therefore stays close to the posted medians, inter-quartile ranges, and sponsor-reported p-value, while recognizing that important details of the model and uncertainty are not reported.

10. Analyses Reported by the Sponsor

The sponsor-reported primary comparison is an ANCOVA superiority analysis with p = 0.43. The registry record also reports the arm medians and inter-quartile ranges: 66460.0 mg/dL x min (53500.0 to 75480.0) for Ursodiol and 72570.0 mg/dL x min (55650.0 to 93010.0) for placebo. Those descriptive summaries should not be mistaken for the ANCOVA estimate. They summarize the observed outcome values within each arm, whereas ANCOVA is a model-based comparison.

The registry record does not report the ANCOVA effect estimate, confidence interval, covariates, or full model specification. It is therefore not possible to describe the adjusted effect’s magnitude or precision from the posted analysis. Nor can the model result be independently reconstructed from the arm medians and inter-quartile ranges. The p-value alone does not provide either missing quantity.

No recalculated estimate for this primary comparison is reported in the registry record. The information available supports describing the sponsor’s test result and the group summaries, but not replacing the model-based comparison with a different numerical estimate. The statistical reading should keep the distinction clear: the p-value is the reported test result, the medians describe the arm outcomes, and neither supplies a confidence interval for the treatment contrast.

11. Multiplicity, Missing Data and Analysis Populations

The primary outcome is change in area above basal for glucose. The registry also lists 7 secondary outcomes: change in fasting glucose, change in insulin sensitivity, gastric emptying of liquids, gastric emptying of solids, change in weight, change in body mass index, and change in FGF-19. Numerical results for these secondary outcomes are not reported in the registry record. The record does not state which outcomes, if any, were treated as confirmatory beyond the primary outcome.

When researchers test several outcomes, the chance of obtaining at least one apparently notable result can increase. General statistical practice may address this through a pre-specified hierarchy, adjustment of significance thresholds, or careful labeling of secondary and exploratory findings. Whether any such approach was planned or used is not reported in the registry record. The primary p-value should therefore be read as the reported result for the registered primary analysis, without assuming that it was adjusted for multiple comparisons.

Information about missing data is also not reported in the registry record. The record does not describe whether any outcome observations were missing, how missing values were handled, or whether analyses used imputation or another method. It does not report participant flow by analysis, exclusions from the primary analysis, or the analysis population definition. The 12 participants listed in each arm’s primary summary identify the reported group counts but do not establish how missing data or exclusions were handled.

The registry record does not report whether the primary analysis followed an intention-to-treat, per-protocol, or other population definition. It also does not describe sensitivity analyses for missingness or deviations from the assigned intervention. These details matter because the analysis population and missing-data assumptions can influence the interpretation of a treatment comparison. In their absence, the posted ANCOVA result can be described, but its handling of these issues cannot be evaluated from the registry record.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports no serious adverse events and no deaths in either arm. In the Ursodiol arm, 0 of 12 participants were reported as affected by a serious adverse event, and 0 of 12 by another adverse event. In the placebo arm, the corresponding figures were also 0 of 12 for serious adverse events and 0 of 12 for other adverse events. The record lists no top serious adverse event terms.

These are counts from a trial with 12 participants at risk in each arm. General statistical principle: observing no events in a limited group does not establish that an event cannot occur, and it does not precisely estimate the frequency of uncommon events. The reported counts describe what was recorded during this trial; they do not support a broad comparison of safety risks.

The registry does not report further detail about adverse-event definitions, collection procedures, severity grading, or the observation period for safety outcomes. The absence of recorded events should therefore be read narrowly as a description of the posted counts, not as evidence that either intervention is free of adverse effects.

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

The primary outcome summaries describe the observed mixed-meal glucose area above basal in each arm, while the sponsor’s ANCOVA provides a test result for the primary comparison. The median was 66460.0 mg/dL x min in the Ursodiol arm and 72570.0 mg/dL x min in the placebo arm. The reported ANCOVA p-value was 0.43. The registry record does not include an adjusted effect estimate or confidence interval, so the magnitude and precision of the model-based comparison cannot be stated from the posted analysis.

A p-value is not a measure of clinical importance. It does not say how much the outcome differed, and it does not give the probability that the null hypothesis is true. Likewise, the arm medians are not estimates of the adjusted contrast and their inter-quartile ranges are not uncertainty intervals for that contrast. The available summaries and test result answer different statistical questions.

The result concerns the registered endpoint and its post-treatment measurement, approximately 14 - 17 days after baseline. It does not by itself establish effects on longer-term glycemic control, insulin sensitivity, gastric emptying, body weight, or any other listed secondary outcome. Numerical secondary-outcome results are not reported in the registry record. The short outcome timeframe also limits what can be inferred about persistence beyond the reported measurement period.

The trial is a randomized phase 2 study with 12 participants in each posted arm summary. Randomization is a valuable design feature, but the limited number of participants and missing model details warrant caution about precision and reproducibility. The record does not report an equivalence analysis, so the p-value cannot be interpreted as showing that the arms are equivalent. Nor should the event counts be treated as proof of absence of safety risk. The appropriate reading is bounded: these are the reported summaries and analysis for this study, with substantial information about effect size, precision, and secondary outcomes not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically instructive because its registry record pairs descriptive summaries with a model-based test but does not report the adjusted effect estimate or confidence interval. That distinction is important in clinical research. A p-value can indicate how the data relate to a null hypothesis under an analysis model, but it cannot replace an estimate of the effect or its precision. The arm medians provide an understandable description of observed outcomes, yet they do not supply the adjusted contrast tested by ANCOVA.

The outcome is an area-based measure derived from mixed-meal glucose results. It captures a response over a measurement period rather than relying on a single reading. That makes it useful as a physiological endpoint for a short-term study, while also meaning that interpretation depends on how the measurement and change were defined. The registry identifies the measure and timeframe, but the calculation details beyond the mixed-meal basis are not reported.

The record also illustrates why a non-significant-looking test result should not be converted into a claim of no effect or equivalence. The ANCOVA p-value is 0.43, but an interval estimate and a stated equivalence margin are not reported. Without those, readers cannot assess which effect sizes remain compatible with the data or whether a clinically meaningful difference is excluded.

Finally, the trial has a small posted count in each arm and reports no numerical secondary results. This constrains what can be learned about the endpoint distributions, model stability, and broader outcome pattern. Statistical interpretation benefits from keeping separate the primary outcome, the descriptive summaries, the model-based test, and the information that the registry does not report.

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.

Several trial-specific limitations affect interpretation of the posted result. The primary summaries are based on 12 participants per arm, and the registry record does not report a confidence interval for the ANCOVA comparison. Without an interval estimate or adjusted effect estimate, readers cannot assess the precision or plausible range of the primary contrast. The medians and inter-quartile ranges describe the arms but do not resolve that limitation.

The registry identifies ANCOVA as the analysis method but does not report its covariates, model specification, assumptions, or diagnostic checks. Details of random sequence generation and allocation concealment are also not reported. Although the study is described as triple-masked, the specific people masked and the procedures used to preserve masking are not reported. These omissions limit assessment of the analysis and implementation from the registry record.

The endpoint timeframe is baseline and post-treatment approximately 14 - 17 days. This supports interpretation of a short-term measurement, not an inference about persistence over a longer period. The registry lists secondary outcomes, but their numerical results are not reported. Details of missing data, analysis populations, exclusions, and sensitivity analyses are not reported either. The record also does not report the protocol’s sample-size justification or whether multiplicity was addressed.

These limitations do not erase the posted summaries or sponsor-reported p-value, but they narrow the claims that can be made from them. They are especially relevant when interpreting a test result without an effect interval, assessing other outcomes, or generalizing beyond the recorded endpoint and timeframe.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Ursodiol on Insulin Sensitivity, Gastric Emptying and Body Weight With Type 2 Diabetes on Metformin — NCT02033876.
• Registry record last updated 2019-05-08; status last verified 2019-04.
• Results first posted 2019-05-08.
• 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 describes a completed, randomized, triple-masked, parallel-group phase 2 trial sponsored by Mayo Clinic, with enrollment of 24. The study compared delayed-release ursodiol with matching placebo in overweight or obese patients with type 2 diabetes on metformin. The primary outcome was change in area above basal for glucose, calculated from mixed-meal glucose results and measured at baseline and post-treatment approximately 14 - 17 days.

The posted primary summaries include 12 participants per arm. The median was 66460.0 mg/dL x min for Ursodiol and 72570.0 mg/dL x min for placebo. The sponsor-reported ANCOVA comparison had p = 0.43. No adjusted effect estimate or confidence interval is reported. Numerical results for the listed secondary outcomes, the primary model details, and information about missing data and analysis populations are not reported in the registry record.