Warm Water for Unsedated Colonoscopy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Warm Water for Unsedated Colonoscopy (NCT00905554), a randomized na study. The primary outcome is Number of Patients Undergoing Complete Unsedated Colonoscopy.

2. Clinical Question

The clinical question concerns the insertion method used during an unsedated colonoscopy and whether the groups differed in completion of the examination without sedation. The investigators’ stated hypothesis was that water irrigation might increase overall tolerability and the proportion of participants completing colonoscopy without sedation. The registered primary outcome measures completion without sedation; pain and tolerability scores are listed among the secondary outcomes.

The outcome counts were 101 of 116 participants in the warm-water group and 89 of 114 in the air-insufflation group. The posted comparison expresses the risk difference as air insufflation relative to warm water: -0.09, with an uncertainty interval that includes no difference. The posted unadjusted p-value is 0.072. These data describe the enrolled groups and the precision of the comparison; they do not by themselves establish why any difference occurred.

The registry record does not specify further eligibility criteria, the setting beyond the named sponsor, or the precise definition used to determine a complete colonoscopy. Those details are not reported in the registry record, so the clinical question should be read at the level of the registered endpoint rather than as a more detailed procedural definition.

3. Trial Design

The registry describes a randomized, parallel-group trial with single masking. The comparison was between warm-water irrigation and air insufflation during the insertion phase of unsedated colonoscopy. Enrollment was 230. The outcome record lists 116 participants in the warm-water arm and 114 in the air-insufflation arm.

The trial status is completed. The registry record gives a start date of 2009-04 and a primary completion date of 2009-07. It lists the primary outcome timeframe as 6-9 months, but does not explain how that timeframe relates to the completion or assessment of the colonoscopy outcome. The phase is listed as not applicable.

Random allocation is recorded, but the method used to generate the allocation sequence, how assignments were concealed, and whether group sizes were set in advance are not reported in the registry record. Masking is recorded as single, but the record does not identify which participants, clinicians, assessors, or other personnel were masked. The record also does not describe deviations from assigned procedures or whether participants could receive sedation after beginning the examination.

The registered interventions concern the procedure used during insertion. The record does not provide detailed instructions for delivering either procedure, such as how the interventions were standardized across clinicians. It names evaluation of pain and tolerability scores as a secondary outcome, but the score instruments and assessment schedule are not reported. These limitations mean that the design can be described as a randomized comparison of two insertion approaches, while several operational details remain unspecified.

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.

Before assignment: no patient excluded from the study flow after being allocated in the study or control arm

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 Number of Patients Undergoing Complete Unsedated Colonoscopy, measured over 6-9 months.

The primary endpoint is the number of participants undergoing complete colonoscopy without sedation. It is a count of participants meeting the registered outcome, rather than a continuous measurement such as a score or a time-to-event outcome. The registry lists the unit as participants and gives a timeframe of 6-9 months. It does not define the procedural criteria for completeness or explain the timing specified by that timeframe.

This endpoint addresses a practical aim stated by the investigators: whether an insertion approach is associated with completion of colonoscopy without sedation. Completion is relevant to the registered question because the trial focuses on an unsedated examination, and the investigators also identified tolerability as an intended area of comparison. The endpoint does not directly measure pain or tolerability; those are named as secondary outcomes. Nor does the endpoint, on its own, establish the reason an examination was or was not completed.

The recorded counts are 101 in the warm-water group and 89 in the air-insufflation group, with group sizes of 116 and 114, respectively. The registry does not report participant-level outcome data, the handling of incomplete or missing outcome status, or a detailed definition of “complete.” Interpretation therefore depends on the registry’s outcome label and summary counts, without additional detail about classification rules.

7. Results Posted in the Registry

• Air Insufflation: 78.1% (89/114) versus 87.1% (101/116) on control; risk difference -9.0 percentage points (95% CI -18.7 to 0.7), risk ratio 0.90 (0.80–1.01); unadjusted two-proportion z-test p = 0.072.

8. Statistical Interpretation

9. Statistical Methods Explained

The registered outcome is binary at the participant level: a participant either underwent a complete unsedated colonoscopy or did not meet that recorded endpoint. The group summaries report counts of participants with the event and the total number in each arm. A comparison of event proportions is a natural way to describe a binary outcome because it relates the event count to the number of participants at risk in each group. The posted risk difference compares those proportions and is expressed with air insufflation relative to warm water.

What does the risk difference describe?

A risk difference is the difference between the event proportions in two groups, with the interpretation depending on which group is subtracted from which. Here, the posted direction is air insufflation relative to warm water, and the estimate is -0.09. A negative value means the observed event proportion is lower in the first-named group under that comparison direction. It is an absolute contrast, not a relative percentage change. The estimate is a summary of the observed groups and does not by itself establish causation or clinical importance.

The record also includes relative measures, including a risk ratio and an odds ratio, each with an interval. These measures compare groups on relative rather than absolute scales. Risk ratios compare event proportions, while odds ratios compare event odds. Their numerical values are not interchangeable with a risk difference. Since the outcome is common in both groups, interpretation should keep the measure’s scale in mind and avoid treating an odds ratio as though it were a risk difference. The risk difference is the measure explicitly used for the posted comparison discussed in the interpretation.

What does a confidence interval add?

A confidence interval describes statistical uncertainty around an estimate under the assumptions of the method used. It gives a range of values compatible with the observed data at the interval’s stated confidence level. A wide interval indicates less precision than a narrow interval. For a risk difference, an interval containing zero includes the possibility of no group difference. It may also include values on either side of zero, which correspond to different directions of effect. The interval does not assign probabilities to individual values after the data are observed, and it does not capture every source of uncertainty, such as errors in outcome classification or limitations in trial conduct.

The posted interval for the risk difference includes zero, as does the interpretation of the relative-effect intervals around their null values. The registry record does not state the confidence level or the precise calculation method for these intervals. The displayed results should therefore be read as the registry’s reported comparison rather than as a reconstruction of an undocumented analysis procedure.

What does the p-value mean?

The unadjusted p-value is 0.072. In general, a p-value describes the compatibility of the observed data with a null hypothesis under a particular statistical model. A smaller p-value signals greater incompatibility with that null model, but it does not tell how large or important an effect is. Nor does it give the probability that the null hypothesis is correct. Its interpretation depends on the analysis model, the outcome definition, and whether the analysis was selected in advance. The registry record does not report the trial’s statistical analysis plan or a prespecified threshold for interpreting the primary comparison.

“Unadjusted” indicates that the posted p-value is not described as adjusted for other variables or for multiple comparisons. It does not establish whether other adjustments were considered in the trial’s original analysis. The registry lists no posted analyses, so the precise analysis workflow and any sponsor-reported analysis details are not reported in the registry record.

Why does over-dispersion matter in general?

Over-dispersion is a general concern for count outcomes when observed variability is greater than a statistical model assumes. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variation. That issue is especially relevant to some count or clustered data analyses. The primary outcome in this trial is recorded as the number of participants with a binary event, and the registry does not report an over-dispersion assessment or a model-based count analysis. No conclusion about over-dispersion in this trial can be drawn from the posted summary counts alone.

More generally, randomized assignment is intended to make groups comparable at baseline, on average, but randomization does not guarantee balance in every realized trial. Details about baseline characteristics and any covariate adjustment are not reported in the registry record. The available summary therefore supports an unadjusted comparison of the posted endpoint counts; it does not permit evaluation of adjusted estimates, subgroup effects, or sensitivity analyses.

10. Confidence Intervals and Precision

Precision concerns how closely the estimated effect is pinned down by the data. The risk-difference estimate is -0.09, and the reported interval extends across zero. Thus the point estimate favors a higher observed completion proportion in the warm-water group, while the uncertainty interval does not rule out no difference or an effect in the opposite direction. The interval is not a guarantee that the true effect lies within it, and it does not account for every possible source of bias or uncertainty.

The sample sizes recorded for the outcome are 116 in the warm-water arm and 114 in the air-insufflation arm. The event counts are 101 and 89, respectively. Those counts provide the basis for the reported comparison, but the registry does not provide individual participant data or detailed outcome-status information. Precision cannot be assessed fully without considering the method used to construct the interval, the endpoint classification process, and possible missing observations.

The unadjusted p-value is 0.072. It should not be read as a measure of precision or as a probability that there is no difference. The interval gives a more direct view of the range of effects compatible with the data, while the p-value summarizes a test against a particular null model. The record does not state the confidence level for the posted interval, so no confidence percentage is assigned in this explanation.

The interval’s inclusion of zero means the findings do not distinguish clearly between no difference and a nonzero difference at the precision represented by the reported analysis. This does not mean the groups are proven equivalent. Equivalence would require an explicitly defined margin and an analysis designed for that question; neither is reported in the registry record.

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 planning figures are participants per arm, not total events. They are planning illustrations based on the observed effect, rather than the trial’s original design assumptions. The calculation assumptions listed with them are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect.

For planning at 100% of the observed effect, the figures are 278 participants per arm for 80% power and 372 participants per arm for 90% power. At 75% of the observed effect, the figures are 470 participants per arm for 80% power and 628 participants per arm for 90% power. At 50% of the observed effect, the figures are 998 participants per arm for 80% power and 1,335 participants per arm for 90% power.

In plain language, the 80% power column gives the per-arm sample size estimated to detect the specified effect under the stated assumptions with an 80% chance of statistical detection. The 90% power column gives the corresponding estimate with a higher chance of detection. Smaller effects need more participants to detect under the same assumptions, which is why the planning figures increase as the target fraction decreases. The curve represents this same general relationship between target effect size and required sample size; it does not establish what the trial was designed to detect. These figures are not observed power.

The registry record does not report the original sample-size justification, the effect size used to design the trial, or whether these assumptions match the investigators’ planned analysis. The displayed values should therefore be read only as an illustration based on the observed comparison and the listed calculation assumptions.

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 = 0.871; treatment proportion p₁ = 0.781; average p̄ = 0.826.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5365 = 1.0515.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5327 = 0.4484.
4. n = (1.0515 + 0.4484)² / (-0.0900)² = 277.8, rounded up to 278 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 <- 0.870689655172414                          # control proportion
p1 <- 0.780701754385965                          # 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(905554)
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 = 278 (80% power), 372 (90% power)
 75% of observed effect: n per arm = 470 (80% power), 628 (90% power)
 50% of observed effect: n per arm = 998 (80% power), 1,335 (90% power)
power.prop.test: 277.8 
Simulated power at n = 278 per arm: 0.807 (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

Multiplicity arises when several outcomes, comparisons, subgroups, or analysis choices are tested. Conducting many tests can increase the chance of at least one apparently noteworthy result by chance if no adjustment or prespecified testing strategy is used. The registry lists the primary endpoint as complete unsedated colonoscopy and names pain and tolerability scores as a secondary outcome. It records one posted secondary outcome, but the exact component measures and their results are not included in the outcome information summarized for the primary comparison.

The registry record does not report a multiplicity strategy, such as a hierarchy among outcomes or an adjustment to account for multiple tests. It also does not report a prespecified statistical analysis plan, so it is not possible to determine which outcomes, subgroups, or comparisons were planned in advance. The displayed unadjusted p-value should consequently be interpreted as an unadjusted result, without assuming that it was the only test conducted.

Missing data and analysis populations are also not described. The record does not state whether every randomized participant contributed to the primary outcome analysis, how unknown outcome status was handled, whether any participants were excluded after randomization, or whether analyses followed assigned group regardless of procedure received. The arm-level outcome denominators are 116 and 114, but the registry does not explain the relationship between those denominators and the full analysis population beyond the recorded enrollment total of 230.

These absences matter because missing outcomes or post-randomization exclusions can affect a group comparison, and different analysis populations can answer different questions. The posted counts remain interpretable as the counts recorded for each arm, but the record does not provide enough detail to evaluate missingness, exclusions, adherence, or sensitivity to alternative analysis choices.

14. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The serious adverse event figures record 0 events among 116 participants at risk in the warm-water group and 0 among 114 at risk in the air-insufflation group. The figures for other adverse events also record 0 in each group, with the same group sizes at risk. These are the event counts recorded in the registry safety information.

Zero recorded events in these summaries does not establish that no adverse event occurred in any participant or that the groups have identical safety profiles. Interpretation depends on how events were defined, collected, and reported. The registry record does not provide those details, event narratives, or information about the duration and completeness of safety surveillance.

Deaths are not reported as a numerical result. They should not be interpreted as zero based on the absence of a value. No serious adverse event terms are listed. The figures describe the registry record’s safety entries and do not support broader claims beyond those recorded counts.

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

The posted data show that 101 of 116 participants in the warm-water group and 89 of 114 in the air-insufflation group underwent complete unsedated colonoscopy. The comparative estimate is reported as a risk difference of -0.09 for air insufflation relative to warm water. The point estimate is in the direction of fewer completed examinations in the air-insufflation group, but the uncertainty interval includes zero and the unadjusted p-value is 0.072. The results therefore leave considerable uncertainty about the size and direction of the underlying difference.

This result does not establish that the two approaches are equivalent. A nonzero difference may remain compatible with the data, and the registry record does not report an equivalence margin or a trial designed to demonstrate equivalence. It also does not establish that either procedure improves tolerability, because the posted primary endpoint is completion without sedation, while numerical results for the named pain and tolerability outcomes are not included in the primary comparison information.

The outcome counts cannot explain why a participant did or did not complete colonoscopy without sedation. The record does not describe causes of incomplete examinations, sedation after procedure initiation, or detailed criteria for completion. It also does not report the statistical analysis plan, analysis population rules, or handling of missing data. Those omissions limit how far the comparison can be generalized or interpreted.

The findings are specific to this randomized trial’s enrolled groups, procedures, and recorded outcome. The registry lists 230 enrolled participants and the trial as completed, but additional details about eligibility, clinical setting, and procedural delivery are not reported. The results should be understood as an estimate with uncertainty from the recorded comparison, not as a universal prediction or a recommendation for individual care.

16. Why This Trial Matters Statistically

This trial illustrates why an effect estimate and its uncertainty should be read together. The observed counts differ between groups, and the reported risk difference gives the contrast a direction and scale. However, the confidence interval includes zero, while the unadjusted p-value is 0.072. Looking only at the point estimate would understate the uncertainty; looking only at the p-value would discard information about the possible range and direction of effects.

The distinction between statistical evidence and clinical meaning is also important. The primary endpoint is completion without sedation, while the investigators’ stated rationale also concerns tolerability. A completion count does not measure pain directly, and the record does not include numerical pain or tolerability results in the primary comparison information. Different outcomes can address related but distinct questions, and their results should not be treated as interchangeable.

The record also shows why transparent methods matter. Randomization and single masking are recorded, but sequence generation, concealment, identity of masked parties, missing-data handling, and analysis populations are not described. Without those details, readers cannot fully assess the conduct and analysis of the trial or determine how robust the reported comparison is to alternative choices.

Finally, planning calculations based on an observed effect are not the same as the assumptions used to design a trial. The planning figures illustrate how target effect size and desired power relate to participant requirements under stated assumptions. They do not reveal the original design rationale. Keeping those distinctions clear prevents retrospective planning illustrations from being mistaken for evidence that the original trial was adequately or inadequately sized.

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.

The registry record leaves several trial-specific details unresolved. It does not state the eligibility criteria, the precise clinical setting, or detailed procedures for delivering warm-water irrigation and air insufflation. It records single masking without identifying who was masked, and it does not describe the randomization sequence or allocation concealment. These details would help readers assess implementation and the potential for bias.

The primary endpoint is named as complete unsedated colonoscopy, but the record does not define completion criteria or explain the 6-9 months timeframe attached to the outcome. It does not describe how pain and tolerability scores were measured, despite listing their evaluation as a secondary outcome. Numerical secondary-outcome results are not included in the posted information summarized for the primary comparison.

The record does not report a statistical analysis plan, original sample-size assumptions, multiplicity handling, analysis populations, missing-data methods, or sensitivity analyses. It does not explain whether the recorded arm denominators represent all randomized participants or how deviations from assigned procedures were handled. These omissions make it difficult to assess the extent to which the comparison reflects the intended randomized analysis.

Safety information is limited to the recorded zero counts for serious and other adverse events in both groups. Definitions, collection procedures, surveillance duration, and event narratives are not reported, and deaths are not given as a numerical result. The findings should therefore be interpreted as the information recorded in the registry, with these limitations in mind.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Warm Water for Unsedated Colonoscopy — NCT00905554.
• Registry record last updated 2011-10-25; status last verified 2011-09.
• Results first posted 2011-10-25.
• 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

The registry describes a completed, randomized, parallel-group trial with single masking comparing warm-water irrigation with air insufflation during the insertion phase of unsedated colonoscopy. Enrollment was 230. The primary endpoint was the number of participants undergoing complete unsedated colonoscopy, with a recorded timeframe of 6-9 months.

The outcome counts were 101 of 116 in the warm-water group and 89 of 114 in the air-insufflation group. The risk difference for air insufflation relative to warm water was -0.09, with an interval that includes zero; the unadjusted p-value was 0.072. The registry lists pain and tolerability scores as a secondary outcome. Serious and other adverse event counts were recorded as 0 in both groups; deaths are not numerically reported.

Details of the original sample-size rationale, statistical analysis plan, missing-data handling, and analysis populations are not reported in the registry record.