Evaluating the Safety and Efficacy of Tradipitant vs. Placebo in Idiopathic and Diabetic Gastroparesis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluating the Safety and Efficacy of Tradipitant vs. Placebo in Idiopathic and Diabetic Gastroparesis (NCT04028492), a randomized phase 3 study. The primary outcome is Change From Baseline to Week 12 in the Biweekly Average of Daily Nausea Severity Score Using MMRM Model ITT Population.

2. Clinical Question

The study asks whether symptoms of gastroparesis, particularly nausea, change over 12 weeks among participants assigned to tradipitant compared with participants assigned to placebo. The primary outcome is a change from baseline in the biweekly average daily nausea severity score. It is measured with a patient-reported diary, so the endpoint reflects participants’ ratings of their symptoms rather than an investigator’s rating.

The endpoint description defines a negative change as improvement. The posted estimates are negative in both randomized arms, while the reported between-group analysis gives a p-value of 0.741. These are different pieces of information: changes within each arm describe change from baseline, whereas the comparison addresses whether the groups differed under the reported model. The registry record does not establish the clinical importance of a particular score change or provide a threshold for interpreting it.

3. Trial Design

This is a phase III, randomized, quadruple-masked, parallel-group study. Its official title describes it as multicenter and placebo-controlled. The registry lists Vanda Pharmaceuticals as sponsor and records an enrollment of 979. The study is marked completed, with a start date of 2019-08-20 and a primary completion date of 2025-02-21.

The listed interventions are tradipitant, placebo, and open-label tradipitant. The randomized efficacy comparison shown for the primary outcome is between tradipitant and placebo. The registry specifically says the open-label tradipitant arm was not included in the efficacy analysis. The safety results separately report groups for all three arms. The primary outcome estimates are based on 102 participants in the tradipitant arm and 99 in the placebo arm; these outcome counts should not be treated as the total study enrollment.

The masking designation is quadruple, but the registry record does not identify which roles were masked. It also does not report the allocation ratio, the detailed randomization procedure, or the treatment schedule beyond the primary endpoint’s 12-week timeframe. Those details should not be inferred from the study title or the arm names.

The brief summary states that the study investigated the safety and efficacy of tradipitant versus placebo in relieving nausea and other gastroparesis symptoms. The listed secondary outcomes include change in the Gastroparesis Cardinal Symptom Index total score, change in the biweekly percentage of nausea-free days, and change in the Overall Patient Benefit Scale. The record lists eight outcome measures and eight analyses, but this page focuses on the posted primary endpoint rather than restating the outcome tables.

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

Analysis population for the primary outcome, as defined in the registry: Randomized ITT Population. The Open Label Tradipitant arm was not included in the efficacy analysis.

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 From Baseline to Week 12 in the Biweekly Average of Daily Nausea Severity Score Using MMRM Model ITT Population, measured over 12 weeks.

The primary endpoint is change from baseline to week 12 in the biweekly average of daily nausea severity, measured with the Gastroparesis Core Symptom Daily Diary. The diary asks participants to rate the worst occurrence of nausea severity and other cardinal gastroparesis symptoms in the previous 24 hours. Ratings use a Likert scale from 0, meaning no symptoms, to 5, meaning very severe.

For this endpoint, the change is calculated as the biweekly average post-baseline value minus the baseline value for daily nausea severity. The registry defines a negative change as improvement. The outcome is therefore oriented so that lower, more negative change values correspond to a greater reduction in the recorded nausea severity score from baseline. That describes the score’s direction; it does not establish that a change is clinically important.

A daily diary can capture repeated symptom ratings over time, rather than relying on a single rating at the end of follow-up. Averaging ratings over a biweekly period can also summarize a participant’s reported symptoms across multiple diary entries. These are general features of repeated patient-reported measures, not claims about how complete the diary data were in this study. The registry does not report diary completion rates, a clinically meaningful change threshold, or further details about how individual diary entries were handled.

The endpoint is reported for the randomized ITT population. The open-label tradipitant arm was excluded from the efficacy analysis. The posted arm estimates are least-squares means with 95% confidence intervals, and the primary timeframe is 12 weeks.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous score, and the registry describes the reported arm estimates as least-squares means. In general, a least-squares mean is a model-based estimate of a group mean. Depending on the model, it may account for repeated observations or other terms specified in the analysis. The registry does not list enough model details to say exactly which terms were included, so the estimates should be interpreted as the reported model-based summaries rather than as simple averages calculated from individual participant data.

What the confidence intervals show

Each arm estimate is accompanied by a 95% confidence interval: -1.82 to -1.29 for tradipitant and -1.76 to -1.23 for placebo. In general, a confidence interval expresses the uncertainty associated with an estimate under the statistical method used. It is not a range containing every participant’s score, and it is not a probability statement that the fixed underlying value has a 95% chance of being inside this particular interval. The registry does not give confidence intervals for the between-arm contrast, which is the quantity needed to describe the estimated difference between treatments with corresponding uncertainty.

Why a mixed model may be used

In general, mixed models can be useful when outcomes are measured repeatedly for the same participant. Measurements from one person are related, so treating all diary observations as independent would fail to reflect that structure. A mixed model can represent repeated measurements and estimate changes over time while accounting for within-participant correlation. The registry labels the reported method “Mixed Models Analysis,” but does not describe its specific covariance structure, fixed effects, random effects, adjustment variables, or estimation details. It is therefore not possible to verify the exact model from the posted summary alone.

General statistical practice also recognizes that repeated-measures models rely on assumptions about the outcome process and the pattern of missing observations. The registry record does not state the assumptions used for this analysis or how missing diary entries were handled. The method label alone is not enough to establish that missingness was absent, or that a particular missing-data assumption was satisfied.

How to read the p-value

The reported superiority analysis gives p = 0.741 and identifies its method as a mixed models analysis. In general, a p-value measures how compatible the observed data are with a specified null model, under the test’s assumptions. It is not the probability that the null hypothesis is true, and it does not describe the clinical size of an effect. A p-value also does not replace an estimate and confidence interval for the treatment contrast. The registry does not post the contrast estimate or its confidence interval alongside this analysis.

The arm-specific changes and the superiority test answer related but different questions. An arm’s change from baseline describes the estimated change within that group. A superiority analysis evaluates the comparison between groups under the analysis model. Improvement in both arm estimates does not, on its own, demonstrate a difference between the interventions. Conversely, the p-value alone does not show that the two interventions are identical; it reports the result of a test, not proof of equivalence.

Planning illustrations are not observed power

The sample-size section contains planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect. These are hypothetical planning calculations, not the trial’s original design justification and not evidence about the probability that this completed study detected an effect. The registry record does not report the trial’s sample-size justification. The illustrations should therefore be read as conditional planning scenarios, not as observed power or as an explanation of how enrollment was chosen.

In general, power is a design quantity determined by assumptions about an effect, variability, sample size, and analysis. It is not a post hoc measure that can establish whether a result is true. When a calculation is based on an observed effect, it inherits uncertainty from that estimate; the result is best treated as an illustration of planning requirements under its stated assumptions. It does not replace the posted trial analysis.

Finally, the registry’s outcome count and analysis count indicate that there are multiple posted measures and analyses, while the listed secondary outcomes include several symptom and benefit measures. Statistical interpretation across multiple outcomes can depend on whether analyses were prespecified and whether any multiplicity procedure was used. That information is not reported in the registry record. No adjustment should be assumed from the fact that a primary and secondary outcome list is present.

10. Analyses Reported by the Sponsor

The registry lists a mixed models analysis for the superiority comparison and reports p = 0.741. The primary arm results give least-squares mean changes and their 95% confidence intervals, but the posted result does not include a between-arm estimate or confidence interval. The arm estimates cannot be subtracted to create a replacement treatment contrast: the record identifies the estimates as model-based and does not supply the covariance information needed to recover the uncertainty of a contrast.

The result data also indicate that values are missing or not estimable for a two-arm recalculation of this continuous outcome. Accordingly, there is no independent recalculated contrast to set against the registry-reported analysis. The reported p-value remains the available inferential result for the comparison. The record has no discrepancy flag, and there is no separate recalculated result that would conflict with the reported analysis.

In general, an unadjusted recalculation from summary statistics can be useful as a check when sufficient statistics are available, but it may not reproduce a model-based analysis. Here, the needed between-arm summary is not posted in the outcome results. The most transparent reading is therefore to report the registry’s mixed models p-value alongside the two arm-specific changes, while keeping clear that they are not the same estimand.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists eight outcome measures and eight analyses. The secondary outcomes named in the record include change in the Gastroparesis Cardinal Symptom Index total score, change in the biweekly percentage of nausea-free days, and change in the Overall Patient Benefit Scale. The primary outcome is the biweekly average daily nausea severity score at week 12. The presence of several outcomes raises a general interpretive issue: repeated statistical testing can increase the chance of at least one apparently notable result when many comparisons are considered.

The registry record does not report whether a multiplicity adjustment was used, whether a hierarchy among endpoints was prespecified, or how secondary analyses were intended to support inference. It does not report a detailed statistical analysis plan. Therefore, the p-value for the primary comparison should be read as the posted result for that analysis, without assuming that it was or was not adjusted for other tests.

The primary efficacy population is described as the randomized ITT population, and the open-label tradipitant arm was excluded from efficacy analysis. ITT generally refers to analyzing participants according to their randomized assignment, but the registry record does not give further details about the exact analysis set or participant flow for this endpoint. The arm counts reported for the outcome are 102 for tradipitant and 99 for placebo; they should not be interpreted as proof that every randomized participant contributed all scheduled observations.

MMRM is the named method for the primary endpoint, but the record does not describe missing-data assumptions, diary completion, imputation, sensitivity analyses, or the handling of participants with incomplete follow-up. These omissions matter because repeated-measures analyses can depend on the amount and pattern of missing observations and on assumptions about why data are missing. The registry record also does not report interim analyses or the sample-size justification. Their absence from the posted information means they cannot be characterized here, not that they did or did not occur.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: procedural pain (1.0%), pulmonary embolism (1.0%), hypoglycaemia (1.0%), myocardial infarction (1.0%), small intestinal obstruction (1.0%).

The safety results list serious adverse events for the tradipitant, placebo, and open-label tradipitant groups. The posted figures are 3 of 102 for tradipitant, 2 of 99 for placebo, and 35 of 778 for open-label tradipitant. The record lists two deaths in the open-label group and none in either randomized group. These figures describe the groups as posted; they do not establish that an event was caused by study treatment.

The record also lists other adverse events affecting 48 of 102 participants in the tradipitant group, 25 of 99 in the placebo group, and 267 of 778 in the open-label group. These counts should not be read as a direct comparison of risk without further context about exposure, follow-up, event definitions, and analysis. In particular, the open-label group differs from the randomized comparison, and the registry does not provide the information needed to make it a comparable control group.

Serious adverse event terms listed include procedural pain, pulmonary embolism, hypoglycaemia, myocardial infarction, and small intestinal obstruction, each shown as 1.0%, as well as diverticulitis at 0.4%. The record does not assign these listed term percentages to a particular treatment arm in the displayed information. Small counts and the lack of detail on event timing, exposure, and attribution limit interpretation. These safety summaries are descriptive and should be considered separately from the efficacy endpoint.

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

The posted primary outcome results show negative estimated changes from baseline in both randomized arms at the week 12 endpoint. The registry defines negative change as improvement on the nausea-severity score. The registry-reported mixed models superiority analysis gives p = 0.741. Together, these results describe the observed arm summaries and the reported statistical comparison; they do not, on their own, settle the clinical value of either intervention.

The p-value should not be interpreted as the probability that there is no difference, nor as proof that the groups are equivalent. Equivalence or noninferiority requires an appropriate design and prespecified margins, which are not reported. Nor does the p-value communicate the size of a possible treatment difference. The registry record does not post a between-arm estimate with a confidence interval, making it difficult to assess the range of comparative effects compatible with the data from the posted summary alone.

The negative within-arm estimates should also not be treated as evidence that the intervention caused the change. Participants in both arms have estimated changes, and a treatment effect is a comparison between groups under the analysis plan. The placebo arm is essential to that comparison. The diary scale describes symptom ratings, but the registry does not report a threshold for a clinically meaningful change or establish how changes in this score relate to other outcomes important to participants.

More broadly, a registry result is a compact account rather than a complete analysis report. The record gives the endpoint, arm estimates, confidence intervals, analysis label, and p-value, but not the full model specification, missing-data details, multiplicity approach, or participant flow for the endpoint. These gaps do not invalidate the posted results; they limit how much can be independently assessed from them.

The appropriate conclusion is therefore bounded: the record reports the stated symptom-score changes and a mixed models p-value for the randomized comparison. It does not provide enough information to independently reconstruct the model-based contrast, determine clinical importance, establish equivalence, or attribute safety events to treatment. No medical conclusion should be drawn from these statistical summaries alone.

14. Why This Trial Matters Statistically

This trial is statistically informative because it separates within-arm change from the randomized comparison. Both reported arm estimates are negative, but the question of comparative effect depends on the modeled contrast between tradipitant and placebo. Treating a within-group change as if it were the treatment effect is a common interpretive error, especially when participants in the comparator group also change from baseline.

The endpoint also illustrates the value and complexity of repeated patient-reported measurements. A daily diary can capture symptoms repeatedly, and an average over a biweekly period can summarize reported experience across time. At the same time, repeated observations from the same participant are correlated, which is one reason a mixed model may be used. A model label, however, does not fully describe the analysis. The missing specification and diary-completion information make it important to distinguish the posted result from an independently reproducible analysis.

The result is a useful reminder that a p-value and arm confidence intervals answer different questions. Here, the registry provides uncertainty intervals around the arm estimates, while the reported superiority test is summarized by a p-value without a contrast estimate or interval. That limits assessment of the comparative effect’s magnitude and precision. A fuller report of the contrast and its uncertainty would make the statistical result easier to interpret.

Finally, the study record lists multiple outcome measures and analyses, as well as a separate open-label arm. These features make clear population definitions and multiplicity methods important to interpretation. The record does not report those methods in detail. For biostatisticians and clinical researchers, the case underscores the need to read endpoints, analysis populations, model summaries, and safety groups in relation to one another, rather than treating a single number as a complete account of the trial.

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 arise from the amount of detail available in the registry record. The primary result includes arm-specific least-squares mean changes and 95% confidence intervals, but no between-arm estimate or confidence interval. As a result, the comparative effect and its precision cannot be directly evaluated from the posted primary outcome summary. The mixed models p-value does not fill that gap.

The model is identified only as a mixed models analysis. The registry does not report its terms, covariance structure, estimation procedure, or assumptions. It also does not report the handling of missing diary entries, the completeness of diary data, or sensitivity analyses. These are particularly relevant details for an endpoint based on repeated patient-reported ratings.

The record describes the efficacy population as randomized ITT and excludes the open-label arm, but it does not provide a detailed participant flow or explain how the analysis set was constructed. The overall enrollment of 979 and primary outcome arm counts of 102 and 99 refer to different summaries; the record does not explain their relationship in detail. The open-label safety group should not be treated as interchangeable with either randomized arm.

Multiplicity procedures, the statistical analysis plan, interim analyses, and the sample-size justification are not reported in the registry record. The sample-size illustrations on the page use the observed effect under stated planning assumptions and are not a substitute for the original design rationale. The record also does not give a clinically meaningful threshold for the nausea score, which limits interpretation of the scale changes beyond their direction and statistical uncertainty.

These limitations concern what can be learned from the posted information; they do not show that the underlying study lacked a particular procedure or that the analysis was necessarily inappropriate. They define the boundary of what can be independently assessed.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Evaluating the Safety and Efficacy of Tradipitant vs. Placebo in Idiopathic and Diabetic Gastroparesis — NCT04028492.
• Registry record last updated 2026-09-24; status last verified 2026-08.
• Results first posted 2026-09-24.
• 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

NCT04028492 is a completed phase III randomized, quadruple-masked, parallel-group study sponsored by Vanda Pharmaceuticals. The official title describes a multicenter, placebo-controlled trial of tradipitant for gastroparesis symptoms. The registry records 979 enrolled participants, a study start date of 2019-08-20, and a primary completion date of 2025-02-21.

The primary endpoint is change from baseline to week 12 in biweekly average daily nausea severity, measured with the Gastroparesis Core Symptom Daily Diary. The posted least-squares mean changes are -1.55 for tradipitant and -1.49 for placebo, with their respective 95% confidence intervals. The registry-reported mixed models superiority analysis gives p = 0.741. The open-label tradipitant arm was excluded from efficacy analysis.

The record also lists secondary outcomes and safety results. It does not report a between-arm estimate and confidence interval for the primary comparison or several details needed to reconstruct the analysis, including the model specification and missing-data handling. The results are therefore best read as the posted arm summaries and reported test, with those limits kept in view.