Effect of Chlorogenic Acid on Patients With Impaired Glucose Tolerance: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Chlorogenic Acid on Patients With Impaired Glucose Tolerance (NCT02621060), a randomized phase 2 study. The primary outcome is Fasting Plasma Glucose (FPG).

2. Clinical Question

This is a question about a measured laboratory endpoint at the end of the stated treatment period. The registry record identifies fasting plasma glucose as the primary outcome and gives its unit as mmol/L. It describes the measure as reflecting glucose after a 10- to 12-h overnight fast. The listed arm summaries are medians with standard deviations, while the reported between-group analysis uses the Wilcoxon (Mann-Whitney) method and gives a median difference in final values.

The clinical question is therefore narrower than whether chlorogenic acid changes health outcomes generally. It concerns the reported fasting glucose endpoint in this trial population and timeframe. The registry also lists outcomes related to glucose and insulin response, body measures, blood pressure, and triglycerides, but those are secondary measures rather than the stated primary question. The record does not report a broader clinical outcome or establish how a glucose measurement relates to longer-term outcomes.

3. Trial Design

The registry describes a randomized, triple-masked, parallel phase 2 trial, marked completed, with an enrollment of 30. The sponsor is the University of Guadalajara. The condition listed is impaired glucose tolerance. The comparison was chlorogenic acid versus placebo; the registry describes the same daily dose and capsule schedule for both arms, with the intervention differing by its named substance.

The treatment description specifies 1200 mg per day as three capsules of 400 mg, taken before meals for 90 days. The phrase about timing is recorded as “1/ 2 hour” before meals. The primary outcome was assessed at Week 12. The record does not explain the masking procedures, identify which roles were masked, or describe how randomization was generated or concealed. Those details are not reported in the registry record.

Although the enrollment was 30, the primary outcome summaries list 12.0 participants for placebo and 14.0 for chlorogenic acid. The primary outcome description notes 3 and 1 subjects dropped out in the placebo and chlorogenic acid groups, respectively. The registry does not give a full account of participant flow, reasons for withdrawal, or the analysis population used for the primary comparison. It also does not say whether the outcome summaries include all randomized participants or only participants with available measurements.

The registry lists 22 outcome measures, 21 analyses, and 16 secondary outcomes posted. Those counts describe the breadth of the record but do not establish that every outcome was fully analyzed or reported. The record is marked completed; its primary completion is listed as 2016-01. Protocol details such as the planned sample-size justification, prespecified analysis hierarchy, and interim-analysis approach 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: 3 and 1 subjects dropout in the placebo and chlorogenic acid group respectively.

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 Fasting Plasma Glucose (FPG), measured over Week 12..

Fasting plasma glucose (FPG) is the primary endpoint. It measures glucose in plasma after an overnight fast; the registry describes the collection context as a 10- to 12-h overnight fast and gives the unit as mmol/L. A fasting measurement is used to assess glucose under conditions intended to reduce the immediate influence of food intake. In general, using a defined collection condition helps make measurements more comparable across participants and visits.

The recorded timeframe is Week 12. This places the primary assessment at the end of the 90-day dosing period described for both arms. The endpoint is a laboratory measure, not a direct measure of how participants feel or function. It also does not, by itself, describe longer-term glucose control or clinical events.

The registry identifies the arm summaries as medians with standard deviations. It reports a Wilcoxon (Mann-Whitney) comparison of final values, with a median difference of 0.05 and p = 0.004. No confidence interval is listed for that reported comparison. A confidence interval, when available, would convey the range of effect values compatible with the data under the analysis assumptions; without one, the precision of the reported contrast is harder to judge. The registry does not report a fuller explanation of how the primary endpoint analysis handled missing Week 12 measurements.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry’s primary analysis uses the Wilcoxon (Mann-Whitney) method for final fasting plasma glucose values. In general, this is a rank-based method that compares the distributions of observations between groups rather than relying on a direct comparison of arithmetic means. It can be useful when data are not well described by the assumptions of a standard mean-based test. Its interpretation depends on the distributions being compared; it is not automatically a test of medians in every setting. The registry labels its reported parameter as a median difference, but does not explain the calculation in further detail.

The listed primary summaries are medians with standard deviations. A median describes the middle of an ordered set of values, while a standard deviation describes spread around a mean and is most directly interpretable for data summarized by means. Pairing these two summaries is unusual enough that readers should avoid treating the standard deviation as a confidence interval or assuming it describes uncertainty in the median. The registry does not report confidence limits for the arm summaries.

A p-value measures how compatible the observed data are with a specified null model, under the assumptions of the analysis. It does not state the probability that the null hypothesis is true, and it does not measure the size, importance, or precision of an effect. The reported primary p-value is 0.004. Because the registry does not provide a confidence interval for the primary contrast, the range of effect sizes compatible with that analysis is not directly available from the reported result.

What do the recalculated secondary comparisons represent?

For selected continuous secondary outcomes, recalculated comparisons use posted arm means and standard deviations to form an unadjusted mean difference and an approximate confidence interval. These are simple, unadjusted re-calculations from summary statistics, not the trial’s full individual-participant analysis. They do not account for covariates, repeated measurements, missing-data procedures, or multiplicity unless those features are explicitly included; they are not reported as adjusted analyses.

A mean difference is the contrast between group averages, expressed in the outcome’s original unit. A confidence interval describes the uncertainty around an estimated contrast under the calculation’s assumptions. If it includes zero, the data are compatible with no mean difference as well as with differences in either direction at the interval’s confidence level. It should not be used alone to decide whether an effect matters clinically. For the recalculated secondary comparisons, the intervals are broad, which signals substantial uncertainty in the estimates.

Those re-calculations do not reproduce the registry’s Wilcoxon analyses. The sponsor-reported Wilcoxon tests and the mean-based re-calculations use different statistical approaches and may target different features of the data. A difference in p-values between them is not, by itself, evidence that either calculation is erroneous. The re-calculations are best treated as a transparent secondary view of the posted summary statistics, not as replacements for analyses based on participant-level data.

Why do over-dispersion and variability matter?

In general, over-dispersion means that observed variability is greater than a model assumes. When variability is larger than assumed, standard errors may be too small and confidence intervals too narrow if the model is not adapted. This trial’s posted summaries include standard deviations for the continuous outcomes, but the registry does not report a model-based analysis that evaluates over-dispersion. The general concept is relevant when selecting models for count or rate outcomes, but the posted outcomes considered in the re-calculated comparisons are continuous measures.

The standard deviations for some outcomes are large relative to the differences between the listed arm means. This means individual values may vary considerably within each group, making a group contrast imprecise. The recalculated intervals reflect uncertainty using the posted summary statistics and stated calculation assumptions. They do not show the distribution of individual values and cannot identify influential observations or departures from model assumptions.

How should planning illustrations be read?

Planning illustrations translate an observed effect into hypothetical participants per arm for a future study under the stated assumptions: two-sided significance level 0.05, equal allocation, and a normal approximation. The illustrations consider 100%, 75%, and 50% of the observed effect. They are not the trial’s design justification, and they are not observed power. Sample-size estimates based on an unstable effect from a small study can be especially uncertain. The registry does not report the original sample-size rationale, so these illustrations should be treated only as planning scenarios.

For instance, planning for 75% of the observed effect in the glucose area-under-the-curve comparison gives 462 participants per arm for 80% power and 619 participants per arm for 90% power. For the insulin area-under-the-curve comparison, the corresponding planning figures are 842 participants per arm for 80% power and 1,128 participants per arm for 90% power. These are hypothetical calculations, not estimates of what the completed trial had power to detect.

What information is missing for a fuller analysis?

The registry record does not report individual-level data, baseline outcome values, adjustment variables, the missing-data method, or a detailed statistical analysis plan. It also does not report whether analyses were prespecified beyond the designation of a primary outcome. Without those details, the statistical summaries cannot establish robustness to analytic choices or clarify whether observed differences reflect baseline imbalance, incomplete follow-up, or other factors. General statistical principles can explain what different methods do, but they cannot fill those gaps in the registry record.

10. Analyses Reported by the Sponsor

The registry’s primary sponsor-reported analysis is a Wilcoxon (Mann-Whitney) comparison of final fasting plasma glucose values. It reports a median difference of 0.05 and p = 0.004, without a confidence interval. The arm summaries list medians of 5.8 mmol/L for placebo and 5.5 mmol/L for chlorogenic acid. The registry does not explain the calculation behind the reported median difference, so the relationship between that contrast and the displayed medians is not clear from the record alone.

For selected secondary outcomes, the registry also reports Wilcoxon p-values, while the recalculated contrasts use posted means and standard deviations. These are not the same analysis: a rank-based test and a mean-based comparison can yield different p-values because they use the data differently and may address different aspects of the distributions. For insulin area under the curve, the sponsor-reported p-value is 0.074; the recalculated unadjusted mean comparison has p = 0.645. For body weight, the corresponding values are 0.022 and 0.893. These comparisons should not be treated as direct replications of one another.

The registry marks no discrepancy. The re-calculations are therefore best understood as additional, simple checks based on posted summary statistics rather than a replacement for the sponsor’s analysis. The record does not provide the participant-level data or enough analysis detail to determine why the methods yield different results.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary outcome and numerous secondary outcomes, but the registry record does not report a multiplicity strategy. Multiplicity arises when several outcomes or comparisons are tested: even if no true differences exist, some results may appear statistically notable by chance. Common approaches include controlling the overall error rate or designating analyses as exploratory, but the record does not say whether either approach was used. Accordingly, secondary p-values should not be interpreted as if they were the sole planned test.

Missing data are also relevant. The primary outcome description records 3 dropouts in the placebo group and 1 in the chlorogenic acid group. The endpoint summaries list 12.0 and 14.0 participants, respectively, while the trial enrollment is 30. The record does not provide the reasons for dropout, the timing of missing measurements, or the method used to handle them. It is therefore not possible to determine from the registry whether the primary analysis was based on all randomized participants, participants with observed Week 12 values, or another analysis population.

The registry does not report imputation, sensitivity analyses, or assumptions about why measurements were missing. In general, analyses can be affected if the likelihood of missingness is related to outcomes or group assignment. This is a general concern, not a claim that such bias occurred in this trial. Without a description of the missing-data approach, readers cannot assess how robust the results are to plausible alternatives.

The registry lists 21 analyses and 22 outcome measures, but does not report a detailed statistical analysis plan that would clarify prespecification, analysis populations, or how outcomes were selected for reporting. One other endpoint is noted as having values missing or not estimable for a two-arm comparison. These reporting gaps make it difficult to distinguish unavailable results from analyses that were not conducted. No interim analyses or sample-size justification are reported in the registry record.

12. 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.

Most frequent non-serious adverse event terms

The registry lists no serious adverse events in either group among 15 participants at risk per group. It reports 3 participants with other, non-serious adverse events in the placebo group and 2 in the chlorogenic acid group, also among 15 at risk per group. No deaths are listed for either group. These counts describe the adverse-event information posted in the registry record; they do not establish that no unreported events occurred.

Serious adverse events and other adverse events are distinct categories in the posted information. The registry does not provide serious-event terms because none are listed, and it does not provide the event descriptions or severity details for the other events. It also does not explain how events were collected, whether events were assessed as related to treatment, or whether participants with missing safety follow-up were included in the at-risk counts.

These figures should be read as descriptive safety reporting for this trial, not as a definitive comparison of risk. In general, a small trial may be unable to characterize uncommon harms precisely. The registry does not report a formal statistical comparison of adverse events, and the posted counts alone do not support a conclusion about safety beyond the events recorded.

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

The posted results describe a randomized comparison of chlorogenic acid with placebo in people with impaired glucose tolerance, with fasting plasma glucose designated as the primary endpoint at Week 12. The registry reports different arm medians and a Wilcoxon comparison with p = 0.004. This is evidence of a difference under the stated test, but the report does not include a confidence interval for the primary contrast, making its precision difficult to assess.

The result does not establish the clinical importance of the difference. The registry does not state a threshold for a meaningful change in fasting glucose, and it does not provide baseline fasting glucose values or a change-from-baseline analysis. The final-value summaries therefore cannot show how much either group changed over time. A p-value alone cannot answer whether a difference is large enough to matter to patients or clinicians.

The finding also does not establish an effect on outcomes beyond those measured and reported. The registry lists multiple secondary outcomes, but many are incomplete or have limited analysis detail, and no multiplicity approach is reported. The secondary analyses should be treated cautiously, particularly when different analytic methods yield different inferential results.

Nor does this record settle questions about longer-term outcomes, other populations, or different dosing schedules. It describes one phase 2 parallel trial with a 90-day treatment description and a Week 12 primary assessment. The registry does not report the original sample-size rationale, detailed analysis populations, missing-data handling, or a fuller protocol. Those gaps restrict how confidently the results can be generalized or how robustly they can be evaluated.

In general, randomized comparisons can help distinguish group outcomes when allocation and follow-up are adequately conducted and analyzed. The registry identifies the trial as randomized and triple-masked, but does not provide the procedures needed to evaluate those features in detail. The findings should therefore be understood as the results recorded for this trial, with uncertainty about their magnitude, precision, and sensitivity to analytic choices.

14. Why This Trial Matters Statistically

This trial is statistically interesting because the primary result is reported using a rank-based test, while several secondary recalculations use mean differences from summary statistics. These methods are not interchangeable. Comparing them highlights why the choice of estimand and statistical method matters: an analysis of ranks and an analysis of means can emphasize different features of a distribution, especially when variability is substantial.

The record also illustrates the distinction between statistical evidence and estimation. The primary p-value is small, but there is no reported confidence interval for the primary contrast. Without an interval, readers have less information about the range of plausible effect sizes. The displayed medians give context, but they do not replace an interval or explain the reported median-difference calculation.

Another important issue is the relationship between enrollment, outcome availability, and missingness. The record gives an enrollment of 30, notes dropouts for the primary outcome, and reports smaller arm counts for the primary summaries. It does not explain the analysis population or the missing-data method. In general, these details matter because the interpretation of an estimate depends not only on the observed values but also on which participants contribute data and why others are missing.

The multiple secondary outcomes make transparent reporting especially important. Without a reported multiplicity plan, isolated secondary p-values are less persuasive than a coherent, prespecified pattern across outcomes. The registry includes several outcome domains, yet not all results are available in the posted summaries. This makes the trial useful as an example of how analysis choices and reporting completeness shape what can be concluded from a clinical study record.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several limitations are specific to the information recorded for this trial. The enrollment is 30, and the primary endpoint summaries include 12.0 participants for placebo and 14.0 for chlorogenic acid. The record notes dropouts but does not explain their reasons or report a detailed participant-flow account. It does not state whether the primary analysis included all randomized participants or only those with an observed Week 12 value.

The primary endpoint is reported as medians with standard deviations, and the reported Wilcoxon analysis gives a median difference of 0.05 without a confidence interval. The record does not explain the construction of that difference, provide baseline values, or report a change-from-baseline comparison. This limits interpretation of both the size and precision of the primary result.

The registry does not report the randomization procedure, allocation concealment, masking procedures, or a detailed statistical analysis plan. It also does not report a sample-size justification, interim-analysis plan, or a multiplicity strategy. These omissions do not prove that procedures were absent, but they prevent evaluation from the registry record.

Secondary outcomes are numerous, and results are not fully available for all listed measures. The record identifies one other endpoint with values missing or not estimable for a two-arm comparison. Recalculated comparisons from summary statistics cannot recover participant-level distributions, adjust for baseline factors, or reproduce the sponsor’s rank-based analyses. The safety information is also limited to event counts, without event descriptions, relatedness assessments, or detailed collection methods.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effect of Chlorogenic Acid on Patients With Impaired Glucose Tolerance — NCT02621060.
• Registry record last updated 2019-08-02; status last verified 2019-06.
• Results first posted 2019-08-02.
• 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.

19. Record Summary

The registry describes a completed, randomized, triple-masked, parallel phase 2 trial sponsored by the University of Guadalajara. It enrolled 30 participants with impaired glucose tolerance and compared chlorogenic acid with placebo. Both arm descriptions specify 1200 mg per day in three capsules of 400 mg for 90 days.

The primary outcome was fasting plasma glucose in mmol/L at Week 12. The posted arm medians are 5.8 for placebo and 5.5 for chlorogenic acid. The reported Wilcoxon (Mann-Whitney) analysis gives a median difference of 0.05 and p = 0.004, without a confidence interval. The endpoint description notes dropouts in both groups.

The registry lists multiple secondary measures and posts results for 16 secondary outcomes. It does not report a multiplicity plan, detailed missing-data methods, or a full statistical analysis plan. Safety reporting lists no serious adverse events and gives counts for other adverse events. The results are informative as a record of this trial, but important details needed to judge precision, robustness, and broader interpretation are not reported in the registry record.