Gastrectomy, Eating Behaviour and GLP-1: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gastrectomy, Eating Behaviour and GLP-1 (NCT02971631), a randomized na study. The primary outcome is Nadir Blood Glucose.

2. Clinical Question

3. Trial Design

The registry describes a randomized, quadruple-masked crossover study and lists its phase as NA. It is marked completed. The listed start date is 2017-08-10 and the primary completion date is 2018-09-16. The sponsor is Cambridge University Hospitals NHS Foundation Trust.

The study description says participants who had undergone gastrectomy would attend a screening visit and two whole-day study visits. It describes assessment after a glucose drink and a meal while participants received an infusion of a GLP-1 blocker or placebo. Blood sugar and associated hormones, food consumption and food reward behaviour were among the planned assessments. These details describe the study as recorded; the precise sequence of treatment periods is not reported in the registry record.

The listed primary outcome is nadir blood glucose, measured in mmol/l during a 50g glucose tolerance test while receiving the GLP-1 antagonist. Its stated timeframe is 30-120 minutes of infusion of Exendin 9-39. The primary population description identifies a crossover analysis by paired T test. The posted comparison lists 5 observations for placebo and 5 for Exendin. The registry enrollment is also listed as 5, while the study description refers to a response in 16 patients. The record does not explain this difference, so these figures should not be treated as interchangeable descriptions of the analyzed population.

The outcome record includes 8 outcome measures and 13 analyses; 7 secondary outcomes are posted. They include measures of insulin secretion, eating behaviour, food motivation and attention, meal consumption, hunger and satiety, and infusion-related adverse events. The order of treatment conditions, washout arrangements, details of allocation concealment and the full protocol are not 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.

First Intervention - One Day

Second Intervention - One Day

Analysis population for the primary outcome, as defined in the registry: Crossover study - analysis by paired T test

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 Nadir Blood Glucose, measured over As assessed during a 50g glucose tolerance test while receiving infusion of GLP-1 antagonist. At time 30-120 minutes of infusion of Exendin 9-39..

Nadir blood glucose is the lowest blood sugar reading during the oral glucose tolerance test, according to the endpoint description. The measure is reported as a mean in mmol/l, with standard error as the measure of spread. The stated assessment period is 30-120 minutes of Exendin 9-39 infusion during a 50g glucose tolerance test. The description notes that active compound infusion would last a maximum of four hours and says no effect was expected to persist for more than 30 minutes after infusion. These are descriptions of the intended assessment and expected duration in the registry record, not additional outcome findings.

In general, a nadir captures the lowest observed value during a defined test period. It can be used to describe the lower end of the glucose response rather than its average across the whole test or its change over time. That makes the endpoint relevant to the study’s stated interest in blood sugar responses after gastrectomy. However, the nadir alone does not describe the full glucose profile, associated hormone responses, symptoms, appetite or food intake.

The posted means are 3.0 mmol/l for placebo and 4.4 mmol/l for Exendin, each with a standard error of 0.3. These are summaries for the listed treatment conditions. The registry does not post a confidence interval for the treatment contrast in the reported analysis, nor does the endpoint summary alone describe the spread of individual paired differences.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure summarized by its mean and standard error. In general, a mean describes the central value of a set of measurements. A standard error describes the uncertainty in an estimated mean under the assumptions of the calculation; it is not the same quantity as the variability of individual measurements. The posted standard errors are 0.3 for each condition. The record does not provide standard deviations or the observed paired differences, so those cannot be inferred from the reported summaries.

Why use a paired comparison?

In general, a crossover design allows measurements under different conditions to be compared within the same participants. A paired analysis uses that within-participant structure and can account for stable differences between participants. The registry describes the primary analysis as a paired T test. This is the stated analysis for nadir blood glucose. The treatment sequence, period effects, and any adjustment for them are not reported in the registry record, so the details of how the crossover structure was handled cannot be examined further.

The reported analysis is described as a two-sided t-test, with p = 0.008. In general, a two-sided test considers evidence for a difference in either direction rather than testing only whether one condition is higher. A t-test evaluates the observed comparison relative to its estimated variability under the test’s assumptions. For a paired analysis, that variability concerns the participant-level differences between conditions. The posted arm means and standard errors do not reveal that paired-difference variability on their own.

What does the p-value convey?

In general, a p-value is calculated under a specified null hypothesis and test model. It describes how unusual a result at least as incompatible with that null model would be if the model and null hypothesis were correct. It is not an estimate of the probability that the intervention has no effect, and it is not a measure of clinical importance. Here the registry reports p = 0.008 for the primary comparison. A confidence interval for the contrast is not reported in the registry record, so the p-value should not be mistaken for a direct account of uncertainty in the effect size.

Confidence intervals, in general, express the precision of an estimate through a range of values compatible with the data and the stated method. A treatment-contrast interval could show whether the data are consistent with a range of effects, but no such interval is reported for the sponsor’s comparison. The means and their standard errors are not a substitute for a confidence interval on a paired treatment contrast because the pairing influences the uncertainty of that contrast.

What does the crossover structure require?

In general, crossover studies rely on meaningful within-participant comparisons and assumptions about treatment order, period effects and carryover. Randomized allocation can help determine treatment order, while masking can reduce the influence of expectations on assessments and interpretation. This registry lists randomization and quadruple masking. The specific randomization procedure, which parties were masked, treatment order, and any washout interval are not reported in the registry record. The study description states that no compound effect was expected to persist for more than 30 minutes after infusion, but that statement is not a report of the actual interval between treatment conditions.

The record’s sample-size assumptions describe a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, a normal approximation, and target effects stated as fractions of the observed effect. These are planning assumptions, not a report of the trial’s original design justification. A planning illustration asks what sample size might be needed under its assumptions; it is not observed power and does not replace an interval estimate or the trial’s own design documentation. The record does not report a sample-size justification for the study.

Finally, the posted analysis is one comparison among a record that lists 13 analyses and 7 secondary outcomes. In general, when multiple outcomes or analyses are considered, the chance of at least one small p-value can depend on how many tests are performed and how their results are interpreted. Whether a multiplicity adjustment was planned or used is not reported in the registry record. The primary p-value should therefore be read as the result of its stated test, without assuming an unreported adjustment or analysis plan.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Placebo", "Exendin")
n   <- c(5, 5)
m   <- c(3, 4.4)
sp  <- c(0.3, 0.3)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Placebo  5  3.000  0.671  2.167  3.833
CARM  Exendin  5  4.400  0.671  3.567  5.233

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a two-sided t-test with p = 0.008. It also lists a paired T test as the primary analysis approach for the crossover study. The posted means are 3.0 mmol/l for placebo and 4.4 mmol/l for Exendin, with standard errors of 0.3 for each condition.

A recalculated comparison based on posted summary statistics is an unadjusted check, not a replacement for the analysis carried out by the sponsor. If the registry identifies a discrepancy, the sponsor’s own analysis takes precedence. In this record, discrepancy is false, so no discrepancy is flagged. No recalculated contrast estimate or recalculated p-value is stated in the registry record details used for this interpretation; accordingly, no additional recalculated number is presented.

Even when a recalculation agrees with a reported result, it cannot recover information absent from the summaries, such as the participant-level paired differences or the precise handling of the crossover periods. The sponsor’s reported test remains the primary statistical result available for interpretation. The posted p-value describes that test, while the missing confidence interval limits direct assessment of the precision of the treatment contrast.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 secondary outcomes and 13 analyses, in addition to the primary outcome description. The secondary outcomes cover insulin secretion, eating rate, hunger and satiety ratings, food attention and motivation, infusion-related adverse events, and meal consumption. This range indicates that the study examined several kinds of physiological and behavioural measures, but detailed results for those outcomes are not available in the record details used for this interpretation.

In general, multiple statistical tests create an interpretive issue: the more outcomes and comparisons examined, the more opportunities there may be to observe a small p-value by chance under null hypotheses. A multiplicity procedure can adjust interpretation or testing to address that issue. Whether the study prespecified a primary testing hierarchy, used any adjustment, or treated secondary results as exploratory is not reported in the registry record. The reported primary p-value should therefore be read as the result of the stated test, not as evidence that multiplicity was addressed.

Missing-data methods are also not reported in the registry record. The record does not describe how incomplete glucose measurements, missed visits, or unavailable outcome measurements were handled. The analysis population is not described beyond the primary outcome’s crossover analysis by paired T test and the posted arm counts. It is not reported whether all enrolled participants contributed usable paired measurements or whether any participants or observations were excluded.

The registry enrollment is 5, while the study description refers to 16 patients. The relationship between these figures is not explained. That leaves uncertainty about the population represented by the posted estimates and should be kept in mind when judging both precision and generalizability. No assumption about missing observations, exclusions, or analysis-set membership can be made from the record.

13. 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 5% reporting threshold.

The posted adverse-event figures list 5 at risk in each treatment condition. No serious adverse events are listed in either condition; the other adverse-event counts are also 0 in both conditions, and deaths are 0 in both. No serious adverse-event terms are listed.

These figures are event counts in a small crossover record. A zero count means no event of the listed kind was recorded in the posted results for that condition; it does not show that an event is impossible or that the condition has no safety risks. With limited observations, uncommon events may not appear, and the record does not provide an interval estimate for event rates.

The registry names infusion-related adverse events as a secondary outcome and states that adverse events were assessed by CTCAE v4. The posted zero counts should be interpreted only as the recorded safety results, not as a comprehensive estimate of risk in broader use. The duration of safety follow-up and the detailed method for collecting or attributing adverse events are not reported in the registry record.

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

The posted result indicates that the mean nadir blood glucose was 3.0 mmol/l in the placebo condition and 4.4 mmol/l in the Exendin condition. The standard error was 0.3 in each. The sponsor-reported two-sided t-test gave p = 0.008. These figures describe the primary glucose endpoint under the study’s assessment conditions and the stated crossover comparison.

They do not establish the size of the paired treatment contrast with its uncertainty, because a confidence interval for that contrast and the standard deviation of paired differences are not reported. A p-value is not a measure of clinical importance and does not indicate the probability that the result will recur. The registry result also does not establish what happened for each participant or how consistent individual responses were.

The study description presents the infusion as a way to investigate the physiological role of GLP-1 after gastrectomy, not as a test of Exendin 9-39 as a medication. The primary endpoint is a glucose measurement during a defined test period. It does not, by itself, answer questions about longer-term glucose control, eating behaviour, quality of life, or other clinical outcomes. Although the registry lists several secondary outcomes, their findings are not included in the registry record for this interpretation.

The small posted enrollment and the unexplained difference between enrollment 5 and the study description’s reference to 16 patients limit conclusions about the population represented. The record also does not give enough detail to assess missing data, period effects or multiplicity. The result is therefore best understood as a statistical finding for the reported endpoint and comparison, with important uncertainty about its precision and broader applicability.

15. Why This Trial Matters Statistically

This study is statistically notable because it uses a randomized, masked crossover design to examine a physiological question. In general, a crossover comparison can control for stable differences between participants by comparing treatment conditions within the same person. That structure can be useful when the scientific question concerns an acute response under contrasting conditions. The registry identifies paired analysis as the primary approach, though it does not report enough detail to assess treatment order or period effects.

The primary outcome is a low point in a glucose response rather than a broad summary of the full response. That distinction matters statistically: a nadir and an average over time represent different features of the measurements, and one cannot stand in for the other. The registry’s stated assessment window and reported means make the target of the comparison clear, while the absent contrast interval limits assessment of precision.

The record also illustrates why statistical significance and study scope must be considered together. The reported p-value is small under the stated test, but the posted enrollment is 5 and the record does not resolve its difference from the study description’s reference to 16 patients. The record lists several secondary outcomes and analyses, but does not report multiplicity methods. These details constrain how broadly the result can be interpreted.

Finally, the study’s stated purpose is physiological investigation rather than evaluation of Exendin 9-39 as a medication. Statistical interpretation should preserve that distinction. The observed endpoint comparison may inform understanding of the test response in this study, but it is not a substitute for evidence about a clinical treatment strategy or outcomes not measured by the primary endpoint.

16. 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 main limitation for interpreting the primary result is the small posted enrollment of 5, alongside the unexplained reference to 16 patients in the study description. The registry does not clarify whether the latter is a planned number, a different population description, or related to the posted analysis in another way. The listed counts of 5 for each treatment condition are compatible with the crossover description, but do not resolve that inconsistency.

The registry does not provide a confidence interval for the treatment contrast, the standard deviation of paired differences, or participant-level results. The means and standard errors therefore give only a partial account of uncertainty. The stated paired test is identified, but the treatment sequence, period effects, washout arrangements and handling of carryover are not described in the record. The statement that an effect was not expected to persist more than 30 minutes after infusion does not establish how the study periods were separated.

The registry does not report missing-data handling, exclusions, the analysis population beyond the paired-test description, or a sample-size justification. It also does not state whether multiplicity was addressed despite listing 13 analyses and 7 secondary outcomes. The detailed findings for those secondary outcomes are not included in the registry record used here, limiting assessment of whether the primary finding aligns with the other measured domains.

Lastly, the study is described as a physiological investigation in people after gastrectomy. Its primary endpoint concerns nadir blood glucose during a defined test, so conclusions should not be extended to other outcomes or treatment uses that the record does not establish.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Gastrectomy, Eating Behaviour and GLP-1 — NCT02971631.
• Registry record last updated 2019-08-29; status last verified 2019-08.
• Results first posted 2019-08-29.
• 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.

20. Record Summary

The registry record describes a completed randomized, quadruple-masked crossover study sponsored by Cambridge University Hospitals NHS Foundation Trust. It lists enrollment as 5 and phase as NA. The study description refers to 16 patients, but the record does not explain how that figure relates to the listed enrollment.

The primary outcome was nadir blood glucose during a 50g glucose tolerance test while receiving Exendin 9-39. The posted means were 3.0 mmol/l for placebo and 4.4 mmol/l for Exendin, with a standard error of 0.3 in each condition. The sponsor-reported two-sided t-test gave p = 0.008.

The registry lists 7 secondary outcomes and 13 analyses. No serious adverse events, other adverse events or deaths are recorded in either listed condition. Details on multiplicity, missing-data handling and the full analysis population are not reported in the registry record.