This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Enterra Therapy Clinical Study (Gastric Stimulation for Gastroparesis) (NCT00157755), a randomized na study. The primary outcome is Percent Reduction in Frequency of Weekly Vomiting Episodes When the Device is Turned ON, Relative to When the Device is Turned OFF.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00157755 |
| Phase | NA |
| Condition | Gastroparesis |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Quadruple |
| Enrollment | 87 |
| Arms in the primary outcome | 2 |
| Lead sponsor | MedtronicNeuro |
| Results posted | Yes |
2. Clinical Question
Population
People with gastroparesis were considered in two etiologic cohorts: diabetic and idiopathic. The primary analysis population comprised subjects who were randomized, completed the blinded crossover period, and had evaluable measurements. The record reports an enrollment of 87, while the primary analysis includes 32 in the diabetic cohort and 20 in the idiopathic cohort. These figures describe different aspects of the record and should not be treated as interchangeable.
Intervention
Enterra Therapy is the gastric stimulation device evaluated in the trial. The primary endpoint compares weekly vomiting frequency when the device was turned ON with frequency when it was turned OFF.
Comparator
The comparator was the device-off period within the crossover design. Participants experienced ON and OFF conditions, so the primary comparison is within a participant rather than between two independently assigned treatment groups.
Primary question
For each etiologic cohort, what was the percent reduction in weekly vomiting frequency during the ON condition relative to the OFF condition? The registry reports medians, full ranges, and Wilcoxon signed rank tests for the cohorts.
The registry describes a randomized, quadruple-masked crossover study. A crossover design allows comparison of conditions experienced by the same participant, which can account for stable differences between people. It also makes the interpretation dependent on the study periods and the availability of evaluable measurements. Details such as the sequence assigned to each participant, the duration of each ON or OFF period, and any washout procedure are not reported in the registry record.
The reported endpoint is specifically about vomiting frequency, not every symptom associated with gastroparesis. The registry also lists secondary outcomes concerning symptom score and vomiting frequency at 12 months, but the primary result is the percent reduction in weekly vomiting frequency at 4.5 months and 7.5 months. The results are presented separately for diabetic and idiopathic gastroparesis; the record does not reduce these cohorts to one overall comparison.
3. Trial Design
The registry describes a randomized, quadruple-masked crossover study of Enterra Therapy for gastroparesis. The enrollment is 87. The study record identifies two etiologic cohorts, diabetic and idiopathic, and lists the device as the intervention. The record's status is terminated. The recorded start is 2002-06 and primary completion is 2008-12; the reason for termination is not reported in the registry record.
In a crossover design, a participant experiences more than one study condition, and outcomes under those conditions can be compared within that participant. Here, the primary endpoint contrasts device ON with device OFF. Because each participant can serve as their own comparator, the analysis differs from a parallel-group comparison in which separate groups remain in separate conditions. The record identifies randomization and quadruple masking, but does not report the allocation sequence, which parties were masked, or how masking was maintained.
The primary endpoint analysis population is narrower than the enrollment description: subjects had to be randomized, complete the study through the end of the blinded crossover period, and provide evaluable measurements. The registry reports 32 analyzed in the diabetic cohort and 20 in the idiopathic cohort for this endpoint. The record does not explain why participants were not included in this analysis or give a complete accounting of missing endpoint data.
The study description says that participants would be followed closely for twelve months and then once a year until the study closed. This is a description of the intended follow-up, not evidence that every participant completed all planned follow-up. The primary endpoint's stated time frame is 4.5 months and 7.5 months. Details about the precise timing and ordering of the crossover periods are not reported in the registry record.
The record lists nine outcome measures and eighteen analyses. It does not report the protocol's sample-size justification, the planned statistical analysis plan, or whether interim analyses were conducted. The absence of those details means the registry results can be described, but the design rationale and all analysis decisions cannot be reconstructed from this record alone.
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
| Stage | Diabetic: ON First, Then OFF | Diabetic: OFF First, Then ON | Diabetic: Not Randomized | Idiopathic: ON First, Then OFF | Idiopathic: OFF First, Then ON | Idiopathic: Not Randomized |
|---|---|---|---|---|---|---|
| STARTED | 26 | 19 | 10 | 15 | 12 | 5 |
| COMPLETED | 24 | 15 | 3 | 11 | 10 | 0 |
| NOT COMPLETED | 2 | 4 | 7 | 4 | 2 | 5 |
Reasons for not completing this period:
| Reason | Diabetic: ON First, Then OFF | Diabetic: OFF First, Then ON | Diabetic: Not Randomized | Idiopathic: ON First, Then OFF | Idiopathic: OFF First, Then ON | Idiopathic: Not Randomized |
|---|---|---|---|---|---|---|
| Death | 2 | 3 | 3 | 2 | 0 | 0 |
| Lost to Follow-up | 0 | 0 | 1 | 0 | 0 | 0 |
| Withdrawal by Subject | 0 | 0 | 1 | 1 | 1 | 2 |
| Explant due to infection | 0 | 0 | 1 | 0 | 0 | 0 |
| Medical reason | 0 | 0 | 1 | 1 | 0 | 0 |
| Study site closure | 0 | 1 | 0 | 0 | 0 | 0 |
| Subject non-compliance | 0 | 0 | 0 | 0 | 0 | 1 |
| Study completion | 0 | 0 | 0 | 0 | 1 | 2 |
Analysis population for the primary outcome, as defined in the registry: The analysis population included subjects who were randomized, completed the study through the end of the blinded crossover period and provided evaluable measurements.
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.
| Characteristic | Unit | Diabetic | Idiopathic | Total |
|---|---|---|---|---|
| Participants analysed | 55 | 32 | 87 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 55 | 32 | 87 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age Continuous | years | 38.3 (SD 9.8) | 39.4 (SD 10.4) | 38.7 (SD 10.0) |
| Sex: Female, Male: Female | Participants | 36 | 26 | 62 |
| Sex: Female, Male: Male | Participants | 19 | 6 | 25 |
| Region of Enrollment: United States | participants | 55 | 32 | 87 |
6. Primary Endpoint
The registry lists the primary outcome as Percent Reduction in Frequency of Weekly Vomiting Episodes When the Device is Turned ON, Relative to When the Device is Turned OFF, measured over 4.5 months and 7.5 months.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | percent reduction in WVF |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the percent reduction in the frequency of weekly vomiting episodes when the device is ON relative to when it is OFF. Daily vomiting episodes were recorded in diaries for 28 days before each office follow-up visit. Weekly vomiting frequency was based on the average number of weekly episodes recorded in those diaries.
The registry defines the reduction as the OFF-condition frequency minus the ON-condition frequency, divided by the OFF-condition frequency, multiplied by 100%. A positive reduction therefore indicates a lower weekly vomiting frequency in the ON condition; a negative value indicates a higher frequency in the ON condition under that calculation. This scale expresses change relative to the OFF condition, rather than the absolute number of episodes.
The endpoint is measured at 4.5 months and 7.5 months. Vomiting frequency is a direct measure of one symptom named in the study objective, and diary recording is intended to capture episodes over time rather than relying only on a single recollection at a visit. The registry reports the median and full range for each cohort. A median describes the middle observation in the analyzed data, while a full range gives the lowest and highest reported values. A range can be wide even when a median is near zero, so the median alone does not convey how variable individual results were.
The endpoint does not by itself summarize nausea, other gastroparesis symptoms, safety, or longer-term outcomes. The registry lists other outcomes separately. The interpretation of this endpoint is also limited to the participants with evaluable measurements who completed the blinded crossover period.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For the diabetic cohort, the reported median percent reduction in weekly vomiting frequency was 0.0, with a full range from -300.0 to 100.0. The Wilcoxon signed rank test p-value was 0.215. For the idiopathic cohort, the reported median was 17.3, with a full range from -450.0 to 100.0, and the reported p-value was 1.0. These are separate cohort results, not a pooled estimate.
Under the endpoint definition, a positive reduction corresponds to lower vomiting frequency during the ON condition relative to OFF, and a negative reduction corresponds to higher frequency during ON. The diabetic cohort's median of 0.0 indicates that the middle reported value was zero on this scale. The idiopathic cohort's median of 17.3 is positive on the defined scale. Neither median describes every participant: the reported full ranges extend on both sides of zero in both cohorts, showing substantial spread in the observed percent reductions. The endpoint's percentage scale can also produce extreme negative values when the OFF frequency used as the reference is small; the registry does not provide individual observations with which to examine that issue.
The p-values are results of the reported Wilcoxon signed rank tests. A p-value measures how compatible the test data are with the test's null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, and it does not quantify the size or clinical importance of an effect. The registry record does not report confidence intervals for these endpoint estimates, so the precision of the estimated cohort effects cannot be assessed from interval estimates. A p-value alone does not show the range of effect sizes compatible with the data.
The record does not state a significance threshold for these analyses or a plan for addressing multiple outcomes and analyses. Consequently, the p-values should be read as the reported cohort-specific test results, not as definitive evidence about a prespecified family of hypotheses. The record lists nine outcome measures and eighteen analyses, but does not explain how those analyses relate to a multiplicity strategy.
These results apply to the analysis population described in the registry: randomized subjects who completed the blinded crossover period and had evaluable measurements. The record does not report how many otherwise eligible subjects lacked measurements, why measurements were unavailable, or how missingness was handled. The endpoint results therefore do not automatically describe all enrolled participants.
The trial record is marked terminated, but the reason and its relationship, if any, to the posted results are not reported in the registry record. The result should be read as a descriptive account of the recorded endpoint and analyses, not as a general conclusion about all people with gastroparesis or every outcome of interest.
9. Statistical Methods Explained
The primary endpoint is expressed as a percent reduction, calculated from the weekly vomiting frequency in the OFF and ON conditions. The OFF value is the reference in the denominator. A positive result means the recorded frequency was lower in the ON condition according to the endpoint formula; a negative result means it was higher. Since this is a relative scale, its values depend on the reference frequency as well as the change between conditions. The registry reports the median and full range rather than a mean and standard deviation.
What does the median describe?
The median is the middle value after observations are ordered. It is often useful when a distribution is skewed or includes extreme values, because it is less affected by those extremes than the arithmetic mean. It does not describe how many participants had values above or below a particular threshold, and it does not show the full shape of the distribution. The full range supplies the minimum and maximum, but it is sensitive to extreme observations and gives no detail about how values are distributed between those endpoints.
The reported ranges are wide and include negative and positive values in both cohorts. This emphasizes that a cohort median is not an individual-level prediction. It also means the median does not capture whether most participants had results near the center or whether the range reflects a small number of extreme results. Individual observations and additional distribution summaries are not reported in the registry record.
What does the Wilcoxon signed rank test assess?
The registry names the Wilcoxon signed rank test for each cohort. In general, this is a nonparametric method for paired observations. It uses the direction and rank of within-pair differences, rather than requiring the differences to follow a normal distribution. In a crossover setting, a paired comparison can be appropriate because each participant contributes measurements under both conditions. The registry does not provide enough detail to establish every aspect of how the test was implemented, including the handling of ties, zero differences, or missing paired observations.
A test p-value summarizes how unusual the test statistic would be under its null model and statistical assumptions. It does not express the probability that the observed result occurred by chance in a simple everyday sense, nor does it measure the probability that the treatment has no effect. The reported p-values are 0.215 for the diabetic cohort and 1.0 for the idiopathic cohort. The registry does not post confidence intervals for these primary endpoint estimates, so the p-values cannot be supplemented with an interval-based assessment of precision from the posted results.
Why does the crossover structure matter?
General statistical explanation: paired analyses use information about differences within participants. This can reduce the influence of stable between-participant variation. However, crossover results can also be affected by period effects, sequence effects, changes over time, or carryover from one condition into another. These are general design considerations, not claims that any such effects occurred in this trial. The registry record does not report tests or adjustments for them, nor does it describe a washout procedure.
The trial is identified as randomized and quadruple-masked. Randomization can help prevent systematic allocation differences, while masking can reduce some forms of expectation or assessment bias. The record does not specify which people were masked or provide enough information to assess the success of either feature. A design label alone does not establish that bias was absent.
How should planning calculations be understood?
General statistical explanation: sample-size calculations use assumptions about the effect to be detected, variability, significance level, allocation, and analysis method. A planning illustration based on an observed effect is not the trial's original sample-size justification and is not observed power. The registry's stated planning assumptions use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the planning convention is to report total events, while other outcomes use participants per arm. These conventions describe planning calculations, not additional observed outcomes.
The record does not report the trial's own sample-size justification, and the registry's observed crossover cohorts are not a simple two-arm parallel comparison. Any illustrative sample-size figures should therefore be interpreted only under their stated assumptions, not as evidence that the trial was adequately powered or that its observed results have a particular probability of replication.
What do the cohort-specific results leave unresolved?
The registry reports the cohorts separately, which preserves the distinction between diabetic and idiopathic gastroparesis. It does not post a combined estimate or a formal comparison between the two etiologic cohorts. Differences between cohort medians should not be treated as evidence that the device effect differs by etiology without a suitable direct analysis. The record also does not report confidence intervals, a multiplicity adjustment, or full details of the analysis plan. These omissions limit assessment of precision, confirmatory status, and robustness.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Wilcoxon Signed Rank Test | not reported | 0.215 | |
| Wilcoxon Signed Rank Test | not reported | 1.000 |
The registry reports the sponsor's primary endpoint analyses as Wilcoxon signed rank tests conducted separately for the diabetic and idiopathic cohorts. The posted results are p = 0.215 for the diabetic cohort and p = 1.0 for the idiopathic cohort, alongside the cohort medians and full ranges. No confidence intervals are reported for these primary endpoint results.
No separate recalculated endpoint estimates or test results are included in the record, so there is no recalculation to compare numerically with the sponsor's analyses. The registry does not flag a discrepancy between sponsor and recalculated results. The sponsor's reported analysis is therefore the available inferential result for this endpoint in the record.
Even when sponsor results are the only reported test results, their interpretation remains limited by the information available: the analysis population is restricted to completers with evaluable measurements, and the registry does not report a multiplicity plan, detailed missing-data methods, or confidence intervals. Those limitations do not change the reported numbers, but they constrain what can be concluded about precision and the broader analysis strategy.
11. Multiplicity, Missing Data and Analysis Populations
The record lists nine outcome measures and eighteen analyses. It does not report whether the primary endpoint was designated as the sole confirmatory test, how the other analyses were grouped into families, or whether p-values were adjusted for multiple testing. When many outcomes or analyses are examined, the chance of at least one apparently unusual result can increase. General statistical explanation: multiplicity procedures manage that issue by controlling an error criterion across a defined family of tests. Without the planned family and adjustment details, the reported p-values should be interpreted as the individual test results stated in the registry, rather than as results known to control a specified overall error rate.
The primary analysis population included randomized subjects who completed the blinded crossover period and had evaluable measurements. This definition makes clear that the reported endpoint is not necessarily an analysis of every enrolled participant. The registry record does not report the number or characteristics of participants excluded from this endpoint analysis, the reasons for noncompletion, or the extent and pattern of missing diary data.
Missingness can affect interpretation if availability of measurements is related to symptoms, treatment condition, or other factors. That is a general concern, not evidence that missing data were biased in this trial. The record does not describe imputation, sensitivity analyses, or another method for addressing missing observations. It also does not report whether the test required complete paired ON and OFF data for every analyzed participant, although the named signed rank method is generally used for paired observations.
Randomization, masking, and a crossover structure are reported, but the registry does not provide a full statistical analysis plan. Interim analyses, prespecified handling of protocol deviations, and the rationale for the analyzed population are not reported in the registry record. These gaps limit reconstruction of the complete inferential strategy and make it difficult to distinguish planned analyses from analyses chosen after data became available.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Diabetic | 8 | 55 | 14.5% |
| Idiopathic | 4 | 32 | 12.5% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Diabetic | 13 of 55 |
| Idiopathic | 9 of 32 |
Most frequent serious adverse event terms
| Event term | Organ system | Diabetic | Idiopathic |
|---|---|---|---|
| Lead migration/dislodgment | General disorders | 1 of 55 | 1 of 32 |
| Apnoeic attack | Respiratory, thoracic and mediastinal disorders | 1 of 55 | 0 of 32 |
| Central venous catheterisation | Surgical and medical procedures | 1 of 55 | 0 of 32 |
| Chest pain | General disorders | 1 of 55 | 0 of 32 |
| Death | General disorders | 0 of 55 | 1 of 32 |
| Diabetic ketoacidosis | Metabolism and nutrition disorders | 1 of 55 | 0 of 32 |
| Gastrointestinal haemorrhage | Gastrointestinal disorders | 1 of 55 | 0 of 32 |
| High lead impedance | General disorders | 1 of 55 | 0 of 32 |
| Impaired gastric emptying | Gastrointestinal disorders | 1 of 55 | 0 of 32 |
| Implant site haematoma | General disorders | 1 of 55 | 0 of 32 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Diabetic | Idiopathic |
|---|---|---|---|
| Lead impedance NOS or high lead impedance | General disorders | 11 of 55 | 6 of 32 |
| Abdominal pain | Gastrointestinal disorders | 4 of 55 | 0 of 32 |
| Paraesthesia | Nervous system disorders | 0 of 55 | 3 of 32 |
Serious events with the highest arm-level rate: death (3.1%), lead migration/dislodgment (3.1%), neurostimulator migration (3.1%), paraesthesia (3.1%), apnoeic attack (1.8%).
The registry reports serious-event figures separately for the diabetic and idiopathic groups. For the diabetic group, the serious-event figures list 8 affected among 55 at risk, and the other-event figures list 13 affected among 55 at risk. For the idiopathic group, the serious-event figures list 4 affected among 32 at risk, and the other-event figures list 9 affected among 32 at risk. These are the registry's reported counts and denominators; they should not be combined into a new total.
The serious adverse event term list includes Death, Lead migration/dislodgment, Neurostimulator migration, and Paraesthesia, each at 3.1%; it also lists Apnoeic attack and Central venous catheterisation, each at 1.8%. The registry does not give the denominator or further detail needed to interpret these percentages as risks for a particular group or follow-up interval. The separate deaths field is not reported, so the death term percentage cannot be reconciled with a death count from this record.
These figures do not establish that an event was caused by the device, nor do they describe timing, duration, severity beyond the serious-event classification, or outcomes after the event. The record also does not provide a comparison of safety outcomes between ON and OFF periods. The counts and listed terms describe what was posted; they are not a complete assessment of causal safety.
13. What the Result Does, and Does Not, Mean
The results describe a specific registry endpoint: percent reduction in weekly vomiting frequency during the device-ON condition relative to device OFF, among randomized subjects who completed the blinded crossover period and had evaluable measurements. The diabetic cohort's reported median was 0.0, and the idiopathic cohort's reported median was 17.3. The full ranges extend from -300.0 to 100.0 in the diabetic cohort and from -450.0 to 100.0 in the idiopathic cohort. These results convey both cohort-level summaries and broad variation in the observed endpoint values.
The reported Wilcoxon signed rank p-values are 0.215 and 1.0, respectively. They are test results, not measures of effect size, clinical importance, or probability that a treatment works. Without confidence intervals, the registry record does not give an interval estimate of the precision of the medians or paired effects. The results also do not establish whether the differences between the two etiologic cohorts are meaningful; no direct comparison between those cohorts is reported.
The endpoint does not represent all symptoms of gastroparesis, all outcomes of the device, or all people with the condition. The registry lists other outcome measures, but their results are not described by this primary endpoint. The analyzed population is limited to participants who completed the blinded crossover period and had evaluable measurements, and the approach to unavailable data is not reported.
General statistical explanation: a study result is interpreted in the context of its design, measurement, analysis population, and uncertainty. A crossover comparison can provide within-participant information, but its interpretation also depends on period and sequence considerations. The record does not report enough detail to evaluate all such considerations. The trial's terminated status is recorded, but the reason is not reported; the status alone does not explain the findings.
Accordingly, the posted data support a description of the observed cohort summaries and reported tests. They do not, by themselves, support a broad claim about effectiveness, safety, or outcomes outside the specified population and endpoint. No treatment recommendation follows from these statistical results.
14. Why This Trial Matters Statistically
This trial is statistically informative because it uses a randomized, masked crossover design and evaluates a symptom frequency outcome within two etiologic cohorts. A crossover design makes the participant-level comparison central: the ON and OFF conditions are contrasted within participants rather than represented solely by separate parallel groups. That structure can be useful when outcomes vary substantially between people, although it also makes the handling of period, sequence, and possible carryover effects important.
The endpoint illustrates why scale and distribution matter. The primary measure is a percent reduction relative to the OFF frequency, and the registry reports medians and full ranges. The medians are 0.0 for diabetic gastroparesis and 17.3 for idiopathic gastroparesis, while the ranges include both negative and positive values. Reading the median without the ranges could obscure the variability; reading the ranges without recognizing that they are extremes could also misrepresent the typical participant.
The separate cohort results matter because the registry defines the study around diabetic and idiopathic etiologies. Yet reporting cohorts separately is not the same as testing whether their results differ. The record does not provide a formal between-cohort comparison or a combined estimate. Likewise, the reported p-values do not replace confidence intervals or an explicit multiplicity plan.
The analysis population also matters. The endpoint results concern randomized participants who completed the blinded crossover period and had evaluable measurements. Without a detailed account of missing outcomes and exclusions, readers cannot fully assess how closely that population represents everyone enrolled. These distinctions are useful in clinical-trial interpretation: design labels and p-values are only part of the evidence, alongside endpoint definition, distribution, analysis population, and uncertainty.
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 details needed to assess the primary result are not reported in the registry record. The record does not give the crossover sequence, the duration of each condition, the timing of any transition, or whether a washout procedure was used. It also does not identify who was masked or describe how masking was maintained. These omissions prevent a full evaluation of potential period, sequence, or carryover effects.
The primary analysis includes randomized subjects who completed the blinded crossover period and provided evaluable measurements. The registry does not explain exclusions, provide a flow of participants through the analysis, or describe the amount and pattern of missing diary data. No missing-data method or sensitivity analysis is reported. The posted cohort results may therefore not represent all enrolled participants.
The record reports medians, full ranges, and p-values, but no confidence intervals for the primary endpoint. Individual-level values and other distribution summaries are also not reported. This limits assessment of precision and of how the wide ranges arose. The registry gives separate cohort analyses but does not report a direct comparison between diabetic and idiopathic gastroparesis.
The record lists nine outcome measures and eighteen analyses but does not report a multiplicity strategy, a complete statistical analysis plan, interim analyses, or the study's sample-size justification. The trial is marked terminated, but the reason for termination is not reported. Safety figures are not accompanied by sufficient detail about event timing, attribution, or the denominator for listed serious-event percentages. These limitations restrict interpretation to the results and definitions that the registry actually records.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Enterra Therapy Clinical Study (Gastric Stimulation for Gastroparesis) — NCT00157755.
- Registry record last updated 2010-03-10; status last verified 2009-12.
- Results first posted 2010-03-10.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
The registry identifies NCT00157755 as a randomized, quadruple-masked crossover study of Enterra Therapy for gastroparesis. It lists diabetic and idiopathic cohorts, enrollment of 87, and a terminated status. The recorded start is 2002-06 and primary completion is 2008-12; the reason for termination is not reported in the registry record.
The primary endpoint is percent reduction in weekly vomiting frequency when the device is ON relative to OFF, assessed at 4.5 months and 7.5 months. The reported cohort medians are 0.0 for diabetic gastroparesis and 17.3 for idiopathic gastroparesis, with full ranges of -300.0 to 100.0 and -450.0 to 100.0, respectively. Reported Wilcoxon signed rank p-values are 0.215 and 1.0.
The analysis population includes randomized subjects who completed the blinded crossover period and had evaluable measurements. Confidence intervals, detailed missing-data methods, multiplicity handling, and a direct comparison between the cohorts are not reported. Interpretation should remain tied to this endpoint, these analyzed participants, and the limitations recorded in the registry.