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 Gastrointestinal Integrity After Naproxen + Rebamipide Versus Naproxen + Placebo (NCT02632812), a randomized phase 1 study. The primary outcome is Endoscopic Score - Cryer Score.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02632812 |
| Phase | Phase 1 |
| Condition | Gastrointestinal Lesions |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 24 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Biolab Sanus Farmaceutica |
| Results posted | Yes |
2. Clinical Question
Population
Adult male or female volunteers aged 18 years and above with gastrointestinal lesions were eligible. The registry record says all recruited volunteers completed the trial. Enrollment was 24, with 12 participants in each arm.
Intervention
Participants received rebamipide effervescent granules, 100 mg twice daily, for a total daily dose of 200 mg, alongside naproxen tablets at 550 mg twice daily. Treatment lasted 7 days.
Comparator
The comparator was placebo effervescent granules made to mimic rebamipide, alongside naproxen tablets at 550 mg twice daily for 7 days.
Primary question
The trial asked whether adding rebamipide to naproxen was associated with a different endoscopic assessment of gastroduodenal mucosal injury than adding placebo to naproxen. The primary measure was the Cryer score, assessed over a timeframe of 8 days.
This question concerns the short-term endoscopic score under the two treatment assignments, not a broad assessment of long-term gastrointestinal outcomes. The trial also listed safety and tolerability as a secondary objective, with the number of adverse events among its secondary outcomes. Its registry description says gastric integrity was assessed by endoscopy before and after treatment; biopsies, detection of H. pylori, stool occult blood testing, prostaglandin E2 measurement and histopathological analysis were also described. Those assessments do not replace the stated primary endpoint, which is the endoscopic Cryer score.
3. Trial Design
This was a phase 1, randomized, parallel-group trial with quadruple masking and a placebo comparator. The two assigned regimens paired naproxen with either rebamipide or placebo. The registry record lists enrollment of 24 and 12 participants in each arm. Its status is completed, and the primary outcome population is described as all recruited volunteers who completed the trial.
The active regimen was rebamipide effervescent granules, 100 mg twice daily, alongside naproxen tablets, 550 mg twice daily, for 7 days. The comparator regimen used placebo effervescent granules alongside the same naproxen schedule. The brief description also specifies a follow-up visit. The primary outcome timeframe is 8 days.
Random allocation is relevant because, in general, it is intended to make treatment groups comparable on average with respect to factors that could affect outcomes. Masking is intended to reduce the influence of treatment knowledge on participants, clinical care, outcome assessment or analysis, depending on who is masked. The registry identifies masking as quadruple but does not specify in the information available which roles were masked.
The registry describes endoscopy before and after treatment, with stomach biopsies and detection of H. pylori. Stool occult blood was also to be assessed before and after treatment. Biopsy material was to be used for prostaglandin E2 measurement and histopathological analysis. The posted primary outcome, however, is the endoscopic Cryer score; the registry record does not give numerical results for those other assessments in the registry record for this account.
The record identifies the trial sponsor as Biolab Sanus Farmaceutica. Further details about the randomization procedure, allocation concealment, site conduct and protocol deviations are not reported in the registry record. The record also does not state how the planned enrollment was justified.
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 | Rebamipide 200 mg | Placebo |
|---|---|---|
| STARTED | 12 | 12 |
| COMPLETED | 12 | 12 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: All recruited volunteers completed the trial.
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 | Rebamipide 200 mg | Placebo | Total |
|---|---|---|---|---|
| Participants analysed | 12 | 12 | 24 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 12 | 12 | 24 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age, Continuous | years | 23.5 (18 to 49) | 28.5 (19 to 49) | 24.5 (18 to 49) |
| Sex: Female, Male: Female | Participants | 6 | 6 | 12 |
| Sex: Female, Male: Male | Participants | 6 | 6 | 12 |
| Region of Enrollment: Brazil | participants | 12 | 12 | 24 |
6. Primary Endpoint
The registry lists the primary outcome as Endoscopic Score - Cryer Score, measured over 8 days.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Full Range |
| Unit | Cryer score |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the endoscopic Cryer score, reported as a median with a full range, over a timeframe of 8 days. It is a graded assessment of gastroduodenal mucosal injury. The listed categories run from 0, defined as normal or erythema, through increasing findings of submucosal hemorrhage or edema and erosions, to categories that include ulcers. Thus, the score translates endoscopic findings into an ordered scale, with higher categories describing more extensive injury.
An ordinal score can be useful when the clinical assessment consists of ordered categories rather than a measurement on a scale where every numerical step has a known equal meaning. A median describes the central position of the observed scores, while the full range gives the lowest and highest observed values. Neither statistic describes every participant's result or the distribution of scores between those extremes.
Endoscopy directly assesses visible mucosal findings, which is relevant to a trial focused on gastrointestinal integrity. The registry also describes biopsy-based and stool assessments, but those are not the stated primary endpoint. The record does not report the score distribution in enough detail to reconstruct individual observations or to assess how many participants fell into each category.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome reports a median Cryer score of 4 in the rebamipide-plus-naproxen arm, with a full range from 0 to 5. The placebo-plus-naproxen arm also had a median score of 4 and a full range from 0 to 5. These summaries describe the observed endpoint in the two groups; they do not establish that each participant had the same score or that the groups had identical score distributions.
The Cryer categories give context for the median: category 4 includes 5 or more erosions and/or a single ulcer, with or without submucosal hemorrhage or edema. A group median of 4 means the reported middle value fell in that category. The ranges, extending from 0 to 5 in both arms, indicate that observed values covered categories from normal or erythema through the category for multiple ulcers. The ranges do not show how frequently any category occurred.
The same reported medians and ranges do not, by themselves, establish equivalence, non-inferiority or absence of a treatment difference. Establishing such conclusions would require an appropriate analysis and, for equivalence or non-inferiority, a prespecified margin and suitable uncertainty estimates. Those details are not reported in the registry record. A confidence interval, which in general expresses uncertainty around an estimated quantity under a statistical model, is not reported for the primary comparison. A p-value or other inferential test result for this endpoint is also not reported in the registry record.
The registry record lists no posted statistical analyses for the outcome data, and the available results do not include an adjusted or unadjusted treatment-effect estimate. The record's listed medians and ranges are descriptive summaries, not a quantified between-arm contrast. It would therefore be inappropriate to infer a treatment effect from the matching summaries or to claim that a statistical test found no difference.
The scale itself also matters when interpreting the result. A median reduces a set of ordered scores to a central category and can conceal differences in the distribution away from the center. The full range gives only the extremes. With 12 participants per arm, descriptive summaries may be sensitive to individual observations, but the record does not give the individual scores or a fuller distribution. The posted data support a direct description of the reported summaries, not a precise estimate of how the regimens compare across the full outcome distribution.
The record states that all recruited volunteers completed the trial. This is useful information about completion, but it does not establish that every endpoint was available for every participant, nor does it describe how endpoint data were handled. That missing-data detail is not reported in the registry record.
9. Statistical Methods Explained
The registry record identifies a randomized, parallel-group design and reports the primary endpoint as a median with full range. It does not report the method used to compare the groups, an estimated treatment effect, a confidence interval or a p-value. The descriptions below explain general statistical concepts relevant to reading this type of result; they are not claims about analyses that the investigators performed.
What does the median convey?
For ordered observations, the median is a measure of central position. It identifies the middle of the ordered data, with the exact convention depending on the number of observations and the analysis rules. For an ordinal clinical score such as the Cryer score, this is often more interpretable than treating the score as if each step represented an equal numerical change. A median does not reveal whether observations are concentrated at that category or spread across several categories.
The full range gives the smallest and largest observed values. It is easy to understand, but it depends only on the two extreme observations and says little about the pattern between them. Here, each arm's reported median is 4 and the range is 0 to 5. Without a frequency distribution or individual scores, readers cannot determine how participants were distributed across the scale, how many had values near the median or whether the groups differed in other aspects of their distributions.
Why might a treatment-effect estimate be useful?
A comparison estimate expresses the observed contrast between assigned groups in a defined way. For an ordinal endpoint, an analysis could use a method suited to ordered categories, or the protocol could define another appropriate summary. Which approach is suitable depends on the research question and assumptions. The registry information does not specify a comparison method or an effect measure, so no particular model or test can be attributed to this trial.
In general, a confidence interval communicates the uncertainty associated with an estimated effect under the assumptions of the analysis. It can indicate which effect sizes are reasonably compatible with the data and model. A p-value addresses a specified hypothesis under a statistical model; it is not the probability that a treatment works, nor a measure of clinical importance. Neither type of inferential result is reported for this primary comparison in the registry record. Without them, the descriptive summaries cannot show the precision of a between-group estimate.
What do the planning illustrations mean?
The planning illustrations are based on the observed effect, use a two-sided significance level of 0.05, equal allocation between arms and a normal approximation, and consider an effect to detect stated as a fraction of the observed effect. These are planning assumptions, not results from the trial's own design process. Any resulting participant counts are planning illustrations, not observed power and not evidence that the trial had a particular chance of detecting a treatment effect. The registry record does not report a sample-size justification or the design assumptions used by the investigators.
In general, sample-size calculations depend on the target effect, outcome variability, chosen statistical test, type I error rate and desired power. Changing these assumptions changes the planning result. A calculation based on an observed effect is especially not a substitute for a prespecified design calculation: the observed effect can be imprecise, and planning around it does not establish whether the study was adequately designed. Numerical planning results and the effect fractions used for the illustrations are not reported in the registry record available for this account.
How should ordinal outcomes be handled?
Ordinal outcomes have a meaningful order, but the gaps between categories need not be equal. Treating the score as a continuous measurement can impose assumptions about those gaps. Methods designed for ordinal outcomes can instead model the ordering, although they also require assumptions that should be stated and assessed. The choice of summary and model should follow the question, the score's structure and the analysis plan. The registry record does not report which approach was used beyond identifying the primary summary as a median and full range.
Small group sizes can make estimates less precise and distributional assumptions harder to assess. In general, over-dispersion means that observed variation exceeds what a chosen model expects; if unaccounted for, this can make uncertainty estimates too narrow. The registry record does not report a model, dispersion assessment or uncertainty estimate, so it is not possible to say whether over-dispersion was relevant or how it was addressed.
Randomization supports a causal comparison in principle, but the strength of inference still depends on how the study was conducted, how outcomes were measured, and how the data were analyzed. The reported design alone does not establish that a particular statistical analysis was valid or that a treatment effect was present. The registry record does not report enough detail to assess those analysis choices.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists the Cryer score as the primary outcome and lists Number of Adverse Events as a secondary outcome. It also reports seven outcome measures, but only one secondary outcome is identified in the information available for this account. The record does not explain how the other outcome measures relate to the listed outcomes or their analyses. A plan for controlling the chance of false-positive findings across multiple comparisons is not reported in the registry record.
In general, testing multiple outcomes or treatment contrasts can increase the chance that at least one result appears statistically significant by chance. A prespecified multiplicity strategy can address that issue, but the registry does not state whether one was planned or used. The absence of a reported strategy should not be interpreted as evidence that no strategy was used; it means that this detail is not reported in the registry record.
The primary outcome population is described as all recruited volunteers who completed the trial. The record does not specify whether the primary analysis followed intention-to-treat, per-protocol or another population definition. It also does not report how missing endpoint values, protocol deviations or participants with incomplete assessments were handled. Completion of the trial by all recruited volunteers does not, on its own, establish that all measurements were complete or that no data were excluded.
No interim-analysis plan, stopping rule or statistical analysis plan is reported in the registry record. Without those details, the timing and prespecification of analyses cannot be assessed. The posted results give descriptive summaries but do not report a treatment-effect estimate, inferential test or missing-data analysis for the primary endpoint.
11. 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 |
|---|---|---|---|
| Rebamipide 200 mg | 0 | 12 | 0.0% |
| Placebo | 0 | 12 | 0.0% |
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 |
|---|---|
| Rebamipide 200 mg | 3 of 12 |
| Placebo | 3 of 12 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Rebamipide 200 mg | Placebo |
|---|---|---|---|
| Abdominal pain | Gastrointestinal disorders | 1 of 12 | 2 of 12 |
| Heartburn | Gastrointestinal disorders | 2 of 12 | 0 of 12 |
| Addominal distress | Gastrointestinal disorders | 0 of 12 | 1 of 12 |
The adverse-event figures report zero participants affected by serious adverse events among 12 at risk in the rebamipide group, and zero among 12 at risk in the placebo group. For other adverse events, the record reports 3 affected participants among 12 at risk in each group. These are the registry's counts of participants affected, not a count of all event episodes.
The serious-event term list is empty. The deaths fields are not populated, so deaths are not reported in the registry record; an empty or unreported field should not be read as a confirmed count of zero. The record does not provide event descriptions, severity breakdowns, timing, attribution to treatment or details of withdrawals due to adverse events.
These figures describe the safety information posted for this trial and should be read in light of its short treatment period and limited enrollment. They do not establish that the regimens have identical safety profiles or exclude uncommon events. The registry lists Number of Adverse Events as a secondary outcome, but no inferential comparison of safety outcomes is reported in the registry record.
12. What the Result Does, and Does Not, Mean
The posted primary endpoint summaries show a median Cryer score of 4, with a range from 0 to 5, in each treatment arm. They provide a compact description of the observed endoscopic score, but not a full account of how scores were distributed among participants. The same median and range are not enough to establish that the regimens produced equivalent outcomes, or to rule out a meaningful difference.
The endpoint is a graded measure of gastroduodenal mucosal injury, and category 4 includes 5 or more erosions and/or a single ulcer, with or without submucosal hemorrhage or edema. The primary outcome therefore concerns the endoscopic findings captured by that scale over the specified timeframe. It does not, by itself, describe longer-term clinical consequences, symptoms or outcomes outside the assessment period.
The results also do not establish a precise treatment effect. A between-arm estimate, uncertainty interval and inferential test are not reported in the registry record. Without such information, the data cannot be used to determine the range of effects compatible with the observations under a specified analysis. Nor do the posted summaries show that no difference was found: absence of a reported test is not evidence of a null result.
The design offers a randomized comparison of two short-term regimens, with naproxen in both arms and rebamipide or placebo as the additional assigned treatment. That design helps frame the intended comparison, but conclusions still depend on the conduct of the trial, completeness of outcome measurement and the analysis used. Those details are not fully reported in the registry record.
Accordingly, the most direct interpretation is descriptive: both groups have the same reported median and range on the primary endpoint. The record does not support a claim that rebamipide reduced, increased or had no effect on endoscopic injury. It also does not support medical advice or conclusions about use beyond the specific regimens and short observation described in this trial.
13. Why This Trial Matters Statistically
This trial is statistically informative as an example of the difference between a descriptive result and an inferential conclusion. The registry reports the median and full range of an ordinal endoscopic score in each randomized arm. Those summaries make the central category and observed extremes visible, but do not quantify a treatment contrast or its precision. Readers should distinguish a matching set of summaries from evidence of equivalence.
The choice of endpoint summary matters. The Cryer score orders findings from normal or erythema through increasing evidence of erosions and ulcers. An ordinal scale has meaningful rank, but the distance between categories is not necessarily uniform. A median and range are interpretable summaries, yet they cannot reveal the detailed distribution. A fuller account of category counts or an analysis appropriate to ordered data would help characterize the comparison; such details are not reported in the registry record.
The design also illustrates why randomized allocation and masking are not sufficient statistical information by themselves. They describe important features of how the comparison was intended to be conducted, but interpretation also depends on outcome availability, analysis populations, prespecified methods, uncertainty estimates and multiplicity handling. Several of these details are not reported in the registry record.
Finally, planning calculations and observed results answer different questions. The registry describes planning illustrations based on the observed effect under specified assumptions. Such illustrations can help explore the participant count needed for a future study under those assumptions, but they are not observed power and do not retroactively establish the adequacy of this trial's design. A planning justification for this trial is not reported in the registry record.
14. 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 primary results are limited to a median and full range for each arm. No individual-level data, category frequencies, treatment-effect estimate, confidence interval or p-value are reported in the registry record. As a result, the amount of uncertainty around a between-group comparison and the distribution of scores within each arm cannot be assessed from these summaries.
Enrollment was 24, with 12 participants per arm, and the treatment period was 7 days. This is a small, short trial, so the posted information provides limited evidence about less common outcomes or longer-term effects. The record does not report a sample-size justification, making it impossible to assess the intended precision or target effect from the design information.
The registry describes quadruple masking but does not state which roles were masked. It does not provide details of allocation concealment, endpoint assessment procedures, protocol deviations or adherence. It also does not report the statistical comparison method, the analysis population beyond saying all recruited volunteers completed the trial, or methods for handling missing endpoint data.
The record describes additional assessments, including biopsies, H. pylori detection, stool occult blood, prostaglandin E2 and histopathological analysis. Numerical results for those measures are not included in the registry record for this account. Multiplicity handling and an analysis plan are not reported in the registry record. These gaps limit assessment of the completeness, prespecification and precision of the findings, without establishing that any particular procedure was or was not carried out.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Gastrointestinal Integrity After Naproxen + Rebamipide Versus Naproxen + Placebo — NCT02632812.
- Registry record last updated 2017-03-20; status last verified 2017-01.
- Results first posted 2017-03-20.
- 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.
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.
18. Record Summary
NCT02632812 was a completed phase 1 randomized, parallel-group, placebo-controlled trial with quadruple masking. It compared rebamipide effervescent granules plus naproxen with placebo effervescent granules plus naproxen. Enrollment was 24, with 12 participants in each arm; all recruited volunteers were reported to have completed the trial.
The primary outcome was the endoscopic Cryer score over a timeframe of 8 days, reported as a median and full range. Both arms had a median of 4 and a range from 0 to 5. The registry also lists Number of Adverse Events as a secondary outcome. The posted record does not report an inferential comparison or an estimate of uncertainty for the primary endpoint, and does not report methods for multiplicity or missing data.