Glucagon Use in Colonoscopies: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Glucagon Use in Colonoscopies (NCT02078726), a randomized phase 4 study. The primary outcome is Adenoma Detection Rate (ADR) During Colonoscopy Procedure.

2. Clinical Question

The trial record identifies a prospective, double-blind randomized controlled trial of glucagon use in outpatient colonoscopy. The clinical question is whether the recorded adenoma detection rate differed between the glucagon and placebo arms. The comparison is specifically about detecting at least one traditional adenoma as defined in the endpoint description; it is not a comparison of every possible colonoscopy outcome named in the study summary.

The registry describes the study investigators as interested in several procedure-related measures, including procedure times, endoscopist assessment, patient comfort, and willingness to undergo another colonoscopy. However, the primary result posted in the record is adenoma detection rate. No secondary outcomes are posted in the registry record, so the clinical question answered by the available results is narrower than the full set of interests outlined in the study summary.

The record gives an enrollment of 100 and reports 64 participants in the glucagon arm and 36 in the placebo arm for the primary result. It does not report detailed eligibility criteria, participant characteristics, or an explanation of the arm sizes. Those details would help readers judge how well the enrolled population represents other people undergoing outpatient colonoscopy.

3. Trial Design

The registry identifies NCT02078726 as a completed Phase 4 study sponsored by the University of California, San Francisco. It lists the design as randomized, double-masked, and parallel-group. The study title is “Glucagon Use in Colonoscopies,” and the official title describes a prospective, double-blind randomized controlled trial of glucagon in outpatient colonoscopy.

The record lists the study start as 2014-04 and primary completion as 2018-05. It reports enrollment of 100. For the primary endpoint, 64 participants are listed in the glucagon arm and 36 in the placebo arm. These posted arm sizes are not equal; the registry record does not explain the allocation pattern or report whether all enrolled participants contributed to the primary outcome.

The active intervention was 1 mg glucagon given during colonoscopy through an IV. The comparator was 1 mL normal saline given during colonoscopy through an IV. The registry calls the comparator arm Placebo. The masking field is recorded as DOUBLE, but the registry does not specify which people were masked or describe masking procedures in further detail.

The primary outcome is adenoma detection rate during the colonoscopy procedure. Its timeframe description includes average procedure durations of 1139 seconds for the glucagon arm and 1353 seconds for the placebo arm. Those durations are context in the timeframe field; they are not the primary endpoint itself, which is defined as a percentage of participants with at least one qualifying adenoma.

The record lists one outcome measure and no posted secondary outcomes. It lists zero posted analyses. The trial’s original statistical analysis plan, the rationale for the enrollment target, and the planned allocation ratio are not reported in the registry record. Consequently, the posted results support a description of the registered outcome and comparison, but do not establish how closely the reporting reflects every planned analysis.

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 Adenoma Detection Rate (ADR) During Colonoscopy Procedure, measured over During colonoscopy procedure (an average of 1139 seconds for Glucagon Arm and 1353 seconds for Placebo Arm).

The primary endpoint is adenoma detection rate during colonoscopy, expressed as the percentage of participants with at least one traditional adenoma. The registry definition includes tubular or villous adenomas, adenomas with high grade dysplasia, and adenocarcinoma, of any size. In practical statistical terms, each participant is classified according to whether at least one qualifying finding was recorded; the arm-level rate is the proportion meeting that definition.

This endpoint is used to summarize detection at the participant level. It does not count the total number of adenomas found per participant, describe lesion size or location, or distinguish participants with one qualifying finding from those with more than one. The registry’s definition makes clear which types of findings count, but it does not provide further information about how findings were confirmed or classified.

For the glucagon arm, the posted detection rate is 43.75%; for the placebo arm it is 33.33%. The registry reports that the primary endpoint was assessed during the colonoscopy procedure. The timeframe description also records average procedure durations of 1139 seconds for glucagon and 1353 seconds for placebo, but these are not the endpoint measure being compared in the primary result.

A detection-rate endpoint can be useful for comparing whether participants in each arm had at least one qualifying adenoma detected under the study conditions. It is a binary outcome at the participant level, even though it is reported as a percentage. It does not, on its own, describe diagnostic accuracy, later clinical outcomes, or the effect on any individual participant. The posted endpoint should therefore be interpreted as a group-level comparison of the proportion meeting the specified definition.

7. Results Posted in the Registry

• Glucagon: 43.8% (28/64) versus 33.3% (12/36) on control; risk difference 10.4 percentage points (95% CI -9.2 to 30.0), risk ratio 1.31 (0.77–2.25); unadjusted two-proportion z-test p = 0.307.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is binary at the participant level: a participant either had at least one finding that met the registry’s adenoma definition or did not. The endpoint is then expressed as a percentage within each arm. For a binary percentage outcome, an arm-level estimate describes the observed proportion with the event, while a comparison can express the difference between those proportions. These summaries answer different questions: the arm percentages describe each group, and the risk difference describes the contrast between groups.

What does the risk difference represent?

The registry reports an unadjusted risk difference of 0.104 for glucagon compared with placebo. A risk difference is an absolute contrast on the probability scale: it subtracts one group’s event proportion from the other’s according to the comparison’s stated orientation. A value above zero indicates that the proportion is higher in the arm placed first in the comparison; a value below zero indicates the reverse. The sign is meaningful only with the comparison orientation in mind.

The risk difference is unadjusted. That means the posted estimate is presented as a direct comparison of the observed arm-level proportions rather than a comparison adjusted for participant or procedure characteristics. The registry does not report an adjusted analysis. An unadjusted result is straightforward to describe, but it may not account for differences in measured or unmeasured characteristics between arms. Randomization is intended to support fair group comparisons, though the record does not provide enough information to examine the allocation process or the distribution of participant characteristics.

How should the p-value be read?

The posted unadjusted p-value is 0.307. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions were correct. It is not a measure of the size of the effect and does not directly quantify the chance that a result is due to chance. Nor does it determine whether an observed difference is important in clinical practice. The registry does not report the test specification or statistical analysis plan, so the p-value can be described as posted but its exact calculation cannot be independently reconstructed from the record alone.

What information does the confidence interval add?

The 95% confidence interval for the risk difference extends from -0.092 to 0.300. In general, a confidence interval communicates uncertainty in an estimate by identifying a range of values compatible with the data and the interval procedure. It is not a range that guarantees the true effect lies within its endpoints. Here the interval includes zero, the value corresponding to no risk difference, as well as values in either direction. Its breadth indicates that effects of different sizes remain compatible with the observed data under the method used.

The confidence interval and p-value are related aspects of the comparison, but neither replaces the arm percentages. A reader can use the percentages to understand the observed event frequency in each group, the risk difference to understand the absolute contrast, and the interval to assess uncertainty around that contrast. The registry also lists relative risk and odds ratio summaries, but the primary interpretation focuses on the posted risk difference because it directly expresses an absolute contrast.

What are the limits of the posted analysis?

The registry marks the trial as randomized and double-masked, but does not report details of the allocation sequence, masking procedures, analysis population, or handling of missing data. It lists zero posted analyses and does not provide an analysis plan. The posted comparison is labelled unadjusted. Without further information, it is not possible to determine whether the reported estimate was prespecified, whether any alternative analyses were conducted, or how exclusions and unavailable outcomes were treated.

General statistical practice distinguishes an intention-to-treat analysis, which generally analyzes participants according to assigned group, from analyses based on treatment received or available outcomes. The record does not say which approach was used. This matters because exclusions, crossover, or missing endpoint data can affect the comparability created by randomization. These are general considerations, not claims that any such issue occurred in this trial.

The outcome is a binary percentage, so its interpretation depends on the endpoint definition and on complete, consistent classification of participants. The registry specifies the qualifying adenoma types, but does not provide further adjudication or measurement details. No secondary outcome results are posted. As a result, the statistical interpretation should remain centered on the single registered primary endpoint and should not infer findings about other measures.

10. Confidence Intervals and Precision

The 95% confidence interval for the unadjusted risk difference runs from -0.092 to 0.300, around a point estimate of 0.104. This interval indicates substantial uncertainty about the size and direction of the contrast. Its endpoints include a negative difference as well as a positive difference, meaning the posted data are compatible with either a lower or higher detection rate in the glucagon arm under the interval method.

A confidence interval is not a guarantee that the true effect lies between its endpoints, nor does it assign a probability to each possible effect value. In general, it is a procedure for expressing estimation uncertainty: across repeated use under its assumptions, intervals constructed in the stated way have the named coverage property. The exact interpretation depends on the method and assumptions, which are not fully described in the registry record.

The arm percentages, 43.75% for glucagon and 33.33% for placebo, are point summaries and do not convey precision by themselves. The risk-difference interval is more informative about uncertainty in the comparison. Because the interval spans no difference and values in both directions, the point estimate should not be treated as a precise estimate of an underlying effect. The unadjusted p-value of 0.307 is consistent with the interval’s inclusion of zero, but the interval adds information about the range of effect sizes compatible with the data.

The registry reports arm sizes of 64 and 36 for the primary result. It does not report confidence intervals for the separate arm percentages, the interval calculation method, missing outcome counts, or an adjusted interval. These omissions constrain assessment of precision and robustness. In particular, the reported interval alone does not reveal whether all randomized participants contributed endpoint data or whether any assumptions were made about unavailable outcomes.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size estimates are planning illustrations based on the observed effect, not the original trial’s documented design calculation. They are expressed as participants per arm, not as total events. The assumptions listed for the illustration are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect.

The estimates are given for planning to detect 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For each effect assumption, the display gives the number of participants per arm for 80% power and for 90% power. In plain language, 80% power means that, if the assumed effect and calculation assumptions were correct, repeated studies of the planned size would have an 80% chance of meeting the specified statistical criterion. The 90% column describes the corresponding planning target with a higher chance of meeting that criterion under the same assumptions.

The estimates increase as the target effect becomes smaller: detecting 100% of the observed effect is associated with 342 participants per arm for 80% power and 457 participants per arm for 90% power; planning for 75% of the observed effect gives 600 per arm and 803 per arm, respectively; and planning for 50% of the observed effect gives 1,332 per arm and 1,782 per arm, respectively. This pattern reflects a general statistical principle: smaller differences are harder to distinguish from random variation, so more participants are needed to achieve a given planning target, all else equal.

These are planning illustrations, not observed power for the completed trial. The original trial’s sample-size assumptions and justification are not reported in the registry record. The illustration uses equal allocation even though the posted primary-result arm sizes are 64 and 36, so it should not be read as a reconstruction of the trial’s actual design or as a prediction of its results.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.333; treatment proportion p₁ = 0.438; average p̄ = 0.385.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6883 = 1.3490.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6843 = 0.5760.
4. n = (1.3490 + 0.5760)² / (0.1042)² = 341.5, rounded up to 342 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.333333333333333                          # control proportion
p1 <- 0.4375                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2078726)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 342 (80% power), 457 (90% power)
 75% of observed effect: n per arm = 600 (80% power), 803 (90% power)
 50% of observed effect: n per arm = 1,332 (80% power), 1,782 (90% power)
power.prop.test: 341.5 
Simulated power at n = 342 per arm: 0.803 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no posted secondary outcomes. The study summary describes investigator interest in several other measures, including procedure times, adenoma detection rate, endoscopist assessment of difficulty, patient comfort, and willingness to undergo repeat colonoscopy. However, the registry record posts no secondary outcome results. It is therefore not possible to assess how those measures were analyzed or what they showed.

Multiplicity is the statistical issue that arises when several outcomes, comparisons, or subgroup analyses are tested. In general, testing more hypotheses can increase the chance of at least one result appearing notable by chance, unless the analysis plan accounts for the number and relationships of tests. The registry record does not report a multiplicity strategy, nor does it report a complete analysis plan. Because no secondary analyses are posted, the available record does not allow a reader to determine whether or how multiplicity was addressed.

The record reports a primary result with an unadjusted comparison, and lists zero posted analyses. It does not state whether this analysis was prespecified, whether interim analyses occurred, or whether additional analyses were conducted but not registered. Those details are not reported in the registry record and should not be inferred from the posted p-value.

Missing-data handling is also not described. The record does not report the number of participants with missing primary endpoint values, the reasons for any missing values, or any imputation or sensitivity analysis. It also does not identify the analysis population, such as whether the comparison included all randomized participants or only those with available outcome data. The reported arm sizes alone cannot establish how many participants contributed to the endpoint analysis or whether the enrolled and analyzed populations were the same.

These omissions matter because outcome availability and analysis-population choices can affect a comparison. In general, preserving participants in their assigned groups is one way to retain the advantages of randomization, while excluding participants or analyzing only observed outcomes can change the composition of the groups. The record does not indicate that exclusions occurred; it simply does not report the details needed to evaluate this issue.

14. 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 adverse-event information lists zero serious adverse events in the glucagon arm and zero in the placebo arm. The registry reports 64 at risk in the glucagon arm and 36 at risk in the placebo arm for serious events. It also reports zero other adverse events and zero deaths in each arm, with the same at-risk counts recorded for the other-event category.

These are counts as recorded in the registry, not estimates of the probability of harm or evidence that no adverse events of any kind occurred. The record does not provide event definitions, the period of safety follow-up, the method of collecting adverse events, or additional detail about non-serious events. It also does not list serious event terms. The reported zero counts should therefore be read as no events recorded in these registry categories, rather than as proof of an absence of risk.

The at-risk figures indicate the group sizes used in the adverse-event record, but the registry does not explain how participants were included in safety monitoring or whether any data were unavailable. These safety entries should be interpreted separately from the primary adenoma detection result: they describe the posted adverse-event counts and do not alter the uncertainty around the primary comparison.

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

The registry result describes adenoma detection rates of 43.75% in the glucagon arm and 33.33% in the placebo arm. The unadjusted risk-difference estimate is 0.104. Its 95% confidence interval ranges from -0.092 to 0.300, and the unadjusted p-value is 0.307. Together, these figures describe the observed comparison and the uncertainty around it; they do not establish a definitive direction or magnitude for the underlying difference.

The result does mean that the posted arm percentages favored glucagon numerically for this particular endpoint in this trial. It does not mean that the underlying detection rate is known to be higher with glucagon: the confidence interval includes both a negative and a positive risk difference. Nor does the p-value state the probability that glucagon has no effect or that the finding is due to chance. Statistical estimates are subject to uncertainty, and interpretation depends on the study design, endpoint measurement, and analysis choices.

The finding applies to the registered endpoint: whether a participant had at least one traditional adenoma as defined in the record during colonoscopy. It does not describe the number of adenomas found, other colonoscopy outcomes, longer-term health outcomes, or the experience of individual participants. Although the study summary mentions procedure times and patient-centered measures, results for those outcomes are not reported in the registry record.

The registry identifies the trial as randomized and double-masked, which are design features intended to support fair comparison and reduce bias. However, details of the randomization and masking procedures, endpoint analysis population, missing data, and statistical plan are not reported. Those limits prevent a complete appraisal of how the posted analysis was conducted. The record also does not report participant characteristics or eligibility criteria in detail, limiting assessment of how broadly the result applies.

In short, the posted data provide a numerical comparison with considerable uncertainty, not a conclusive estimate of the effect on adenoma detection and not evidence about every potential effect of glucagon during colonoscopy. They should be read as a result from this recorded trial under its stated endpoint, with the missing methodological details kept in view.

16. Why This Trial Matters Statistically

This trial illustrates why a point estimate should not be read in isolation. The observed percentages differ numerically, and the risk-difference estimate gives a direct summary of that contrast. But the confidence interval is broad and crosses zero, so the data remain compatible with effects in either direction. Reporting both the estimate and its uncertainty is essential for avoiding an overconfident interpretation of a binary outcome.

It also highlights the distinction between an outcome’s name and its operational definition. “Adenoma detection rate” could be interpreted in different ways unless the record specifies the unit and threshold. Here, the measure is the percentage of participants with at least one qualifying traditional adenoma, including the listed histologic categories and any size. That definition makes clear that the analysis is participant-based rather than a count of all lesions.

The record shows how trial design labels and result transparency provide different kinds of information. Randomized and double-masked design fields indicate features of the stated design, while the result record does not describe the randomization method, masking procedures, missing-data handling, or analysis population. Statistical interpretation depends on those implementation and analysis details as well as on the design label itself.

Finally, the trial is a useful example of cautious planning interpretation. The sample-size illustrations use fractions of the observed effect and specified assumptions, and show that smaller target effects require more participants per arm. They are not evidence about the trial’s observed power and do not document the original design rationale. Keeping planning calculations distinct from observed results helps prevent a retrospective planning illustration from being mistaken for a property of the completed study.

17. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record contains limited detail about the enrolled population. It lists Colon Cancer as the condition and identifies outpatient colonoscopy in the title and study description, but does not report eligibility criteria, baseline characteristics, or the clinical reasons for colonoscopy. Without those details, it is difficult to judge the representativeness of the participants or the applicability of the result to other colonoscopy populations.

The posted arm sizes for the primary endpoint are 64 for glucagon and 36 for placebo. The record does not explain why the sizes differ, whether these are randomized or analyzed counts, or whether any participants were excluded or had unavailable endpoint data. It also does not report the randomization sequence, allocation concealment, or which parties were masked. The design fields alone do not resolve these operational questions.

Endpoint reporting is focused on whether at least one qualifying adenoma was detected. The record does not provide additional details on confirmation or adjudication of findings, nor does it report secondary outcome results. The brief summary names several other outcomes of interest, but the registry lists no posted secondary outcomes. The primary result therefore cannot support conclusions about procedure duration, patient comfort, endoscopist-rated difficulty, or willingness to return.

The statistical plan, test specification, missing-data approach, and analysis population are not reported. There is no documented multiplicity strategy or information about interim analyses. The record also does not give a sample-size justification for the original trial. As a result, the comparison can be described using the posted estimate, interval, and p-value, but the full analysis process cannot be evaluated.

Safety entries record zero serious events, other events, and deaths in each arm, but the registry does not report event definitions or follow-up duration. These data are limited to the categories and counts posted and should not be treated as a comprehensive safety assessment.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Glucagon Use in Colonoscopies — NCT02078726.
• Registry record last updated 2025-07-03; status last verified 2025-07.
• Results first posted 2021-04-20.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT02078726 is a completed Phase 4 randomized, double-masked, parallel-group trial sponsored by the University of California, San Francisco. The registry lists a study start of 2014-04, primary completion of 2018-05, and enrollment of 100. The study examined 1 mg intravenous glucagon versus 1 mL intravenous normal saline during outpatient colonoscopy.

The primary endpoint was adenoma detection rate, defined as the percentage of participants with at least one traditional adenoma of any size. The posted rates were 43.75% for glucagon and 33.33% for placebo. The unadjusted risk difference was 0.104, with a 95% confidence interval from -0.092 to 0.300 and an unadjusted p-value of 0.307.

The registry lists no posted secondary outcomes and zero posted analyses. Safety records report zero serious adverse events, zero other adverse events, and zero deaths in both arms. Details of the original statistical plan, analysis population, missing-data handling, and sample-size justification are not reported in the registry record.