Effect of Metabolic Control at Onset of Diabetes on Progression of Type 1 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Metabolic Control at Onset of Diabetes on Progression of Type 1 Diabetes (NCT00891995), a randomized phase 2, phase 3 study. The primary outcome is C-peptide Average Area Under the Curve (AUC) in Response to a Mixed Meal at 1 Year Following Enrollment..

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group study with no masking and lists the study phases as PHASE2 and PHASE3. The condition was type 1 diabetes. Enrollment was 71, and the listed sponsor was Jaeb Center for Health Research. The record gives a start of 2010-09 and primary completion of 2013-10. Its current status is TERMINATED; a reason for termination is not reported in the registry record.

Participants were assigned to intensive treatment or standard treatment. The intensive regimen included 4-6 days of closed-loop therapy, then an insulin pump, continuous glucose monitoring, and home glucose monitoring for 2 years. The standard regimen consisted of home glucose monitoring for 2 years. The registry lists the devices used, but does not provide further detail about the treatment procedures in the posted information.

The main endpoint was C-peptide average area under the curve from a mixed-meal tolerance test at 1 year following enrollment. The reported primary analysis population was described as 68 participants who completed the one-year visit and had positive autoantibody. One participant in the intensive group did not complete the MMTT and was excluded from all C-peptide analyses. The posted arm counts for the primary result are 47 and 20. The registry does not explain the relationship between these counts and the stated population description in greater detail.

The record also lists secondary outcomes including other C-peptide measures, HbA1c, adverse events, continuous-glucose-monitor outcomes, daily insulin dose, and BMI percentile. Detailed results are not posted for every listed outcome. The registry does not report the reason for termination, the randomization procedure, or further operational details of masking beyond stating NONE.

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: Among the 68 participants who completed 1 year visit and with positive autoantibody, 1 participant in the intensive treatment group did not complete MMTT test, thus was excluded from all C-peptide analyses.

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.

Population: Only 48 participants in the intensive treatment group and 20 participants in the standard treatment group who show any positive autoantibody in the first year were included in the analyses.

6. Primary Endpoint

The registry lists the primary outcome as C-peptide Average Area Under the Curve (AUC) in Response to a Mixed Meal at 1 Year Following Enrollment., measured over At baseline, MMTT data were collected at 0 and 90 min; at 12 months, MMTT data were collected at 0 to 240 min post meal.

The primary endpoint was C-peptide average area under the curve (AUC) in response to a mixed meal at 1 year following enrollment. AUC summarizes measurements across the test period; the record defines the endpoint as AUC divided by time, with units of pmol/ml. At baseline, MMTT data were collected at 0 and 90 min. At 12 months, data were collected from 0 to 240 min post meal.

C-peptide is a marker of the body's own insulin production. In general, a mixed-meal test provides a standardized stimulus for assessing the C-peptide response, while an average AUC represents the response across the specified measurement period rather than at a single time point. This endpoint was used to examine beta cell function, which the study summary identifies as the biological question motivating the trial.

The posted group results are geometric means with 95% confidence intervals. In general, a geometric mean is a measure of central tendency suited to positive measurements that may be analyzed on a multiplicative scale. A confidence interval describes uncertainty around an estimated quantity; it is not a range containing every participant's measurement. The registry reports an ANCOVA p-value for the primary comparison, but does not report an adjusted between-group estimate or its confidence interval. The group intervals therefore describe each treatment group separately and should not be read as an interval for the treatment contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the geometric mean describe?

The primary endpoint is reported as a geometric mean rather than an arithmetic mean. In general, the geometric mean summarizes positive measurements on a multiplicative scale. It is often useful when values are right-skewed or when proportional differences are more meaningful than additive ones. That is a general explanation of the statistic, not a claim about the observed distribution in this trial. The registry does not describe the distribution of the endpoint or provide diagnostic information supporting a particular scale.

The reported geometric means are 0.43 pmol/ml for intensive treatment and 0.52 pmol/ml for standard treatment. Each has a separate 95% confidence interval. In general, a confidence interval expresses uncertainty in an estimate under the analysis method; it does not mean that most individual measurements fall inside the interval. Since these are intervals for the group summaries, they do not directly give the uncertainty in the adjusted difference between groups.

What does ANCOVA contribute?

The registry identifies ANCOVA as the method for the primary comparison and reports p=0.49. In general, analysis of covariance compares groups while adjusting for one or more covariates. Adjustment can account for measured baseline characteristics or other specified variables, potentially improving precision or accounting for imbalance. The registry does not identify the covariates, model specification, transformation, or handling of model assumptions. Those details cannot be inferred from the method label alone.

The reported p-value is tied to the sponsor's stated ANCOVA, not to a simple comparison reconstructed from the two group summaries. A p-value describes how compatible the data are with a specified null model under the analysis assumptions; it is not the probability that the null hypothesis is true, nor a measure of the magnitude or clinical importance of an effect. Without an adjusted estimate and interval, the direction and precision of the adjusted contrast are less fully described than they would be with those quantities.

What do unadjusted recalculations mean?

Where posted secondary summary statistics permit it, simple recalculations describe unadjusted contrasts between the reported groups. For continuous outcomes, a mean difference compares the group means. For a binary outcome, a risk difference compares the observed proportions with the event. These are descriptive calculations from the posted summaries, not replacements for an adjusted analysis. They may not match a trial analysis that uses different populations, covariate adjustment, repeated observations, or another model.

For example, the secondary HbA1c result has a reported ANCOVA p-value of 0.40, while the unadjusted recalculation has p=0.74. These values answer related but not necessarily identical inferential questions. The registry does not report an ANCOVA effect estimate or confidence interval for that endpoint. For other secondary outcomes, a sponsor-reported comparison may be absent. A missing analysis is not evidence that a comparison was nonsignificant; it means the corresponding analysis is not reported in the registry record.

How should confidence intervals and p-values be read together?

In general, an estimate gives a magnitude and direction, while its confidence interval describes statistical uncertainty around that estimate. A p-value summarizes compatibility with a null hypothesis under a specified model. Neither alone establishes whether a difference matters clinically. For unadjusted binary comparisons, an interval spanning zero for a risk difference is compatible with both directions of the contrast as well as no difference, under that calculation. The posted secondary calculations should therefore be read as uncertain estimates, not as definitive evidence for or against a treatment effect.

Over-dispersion is a general issue for count or binary data when the observed variability exceeds what a simple model assumes. In that situation, standard errors from the simpler model may be too small, making intervals too narrow and p-values too optimistic. The registry does not report an over-dispersion assessment or a model for the secondary binary endpoint. The simple recalculation is consequently best regarded as an unadjusted illustration, not a substitute for a prespecified analysis.

What do the planning illustrations represent?

The planning illustrations use the observed effect as a basis for considering hypothetical future sample sizes. The stated assumptions are a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as stated fractions of the observed effect. They are planning illustrations, not observed power and not the trial's original sample-size justification. The sample-size assumptions explicitly distinguish participants per arm for most outcomes from total events for time-to-event outcomes. A sample-size justification for this trial is not reported in the registry record.

Planning based on an observed effect can be unstable, particularly when an estimate is imprecise or near zero. The illustrations should not be interpreted as showing that this trial had a particular probability of detecting an effect. The registry does not report the study's original target enrollment, assumptions used in designing the trial, or whether those assumptions changed.

What analysis details remain unknown?

The registry does not report a multiplicity adjustment, a prespecified hierarchy for secondary outcomes, interim analyses, or a detailed missing-data strategy. It also does not provide all model specifications or a complete account of analysis populations for every endpoint. Such omissions limit reconstruction and appraisal of the analyses; they do not establish that a particular procedure was or was not used outside the registry record.

10. Analyses Reported by the Sponsor

The primary analysis reported by the registry is an ANCOVA comparison with p=0.49. The record gives geometric means and separate 95% confidence intervals for each group, but no adjusted between-group estimate or confidence interval. This reported analysis is the trial's stated primary inferential result; the group summaries alone cannot reconstruct its adjusted contrast.

The unadjusted recalculations are separate descriptive analyses based on posted summary statistics. They do not adjust for covariates and need not reproduce an ANCOVA result. For HbA1c, the registry reports an ANCOVA p-value of 0.40, while the unadjusted recalculation reports p=0.74. The distinction reflects different analysis approaches, not necessarily a contradiction. The registry does not report a sponsor p-value for every secondary endpoint, so no comparison with a sponsor result is possible for those outcomes.

No discrepancy is flagged between the posted primary result and a recalculated primary result. There is no recalculated primary comparison to set against the sponsor's ANCOVA p-value in the posted information. The recalculations are therefore supplementary descriptions, not a replacement for the reported primary analysis. For a fuller interpretation of the sponsor analysis, an adjusted effect estimate, confidence interval, and model details would be needed; these are not reported in the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes, including C-peptide measures, HbA1c, adverse events, continuous-glucose-monitor outcomes, daily insulin dose, and BMI percentile. It does not report whether the investigators prespecified a hierarchy among those outcomes or adjusted for multiple comparisons. In general, testing several outcomes creates more opportunities for chance findings. Without a reported multiplicity plan, an isolated secondary p-value should be treated cautiously and should not be given the same confirmatory weight as a clearly prespecified primary analysis.

The available primary population description is specific: 68 participants completed the one-year visit and had positive autoantibody, and one participant in the intensive group did not complete the MMTT and was excluded from all C-peptide analyses. The posted primary arm counts are 47 and 20. The registry does not provide a fuller accounting that reconciles those counts with the stated population, or a detailed participant flow for every endpoint. The enrolled total of 71 should not be assumed to be the analysis population for each result.

The record does not report a detailed missing-data method, assumptions about missing outcomes, or sensitivity analyses. That absence matters because incomplete follow-up or a missed test can affect which participants contribute to an endpoint and may limit how confidently results represent all randomized participants. The record explicitly identifies one MMTT exclusion, but does not give enough detail to assess all possible missingness across outcomes.

Interim analyses and stopping rules are not reported in the registry record. The status is TERMINATED, but the reason for termination and any relation to the conduct or interpretation of the analyses are not reported. A sample-size justification for the original design is also not reported. The planning illustrations based on observed effects do not fill that gap; they are not evidence about the original design assumptions.

These omissions limit independent assessment of confirmatory status, completeness, and robustness. They should be distinguished from evidence that a procedure was not used: the registry simply does not report those details.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gastroenteritis (5.0%), anaphylactic reaction (2.1%), suicidal ideation (2.1%).

The registry reports serious adverse events affecting 2 of 48 participants in the intensive treatment group and 1 of 20 in the standard treatment group. Other adverse events affected 6 of 48 and 2 of 20, respectively. These are counts of participants recorded in each category, not a statistical comparison of risk. No inferential safety comparison is reported in the registry record.

The listed serious adverse event terms include gastroenteritis (5.0%), Anaphylactic reaction (2.1%), and Suicidal ideation (2.1%). The record does not provide enough detail alongside these terms to establish a comparative treatment-related risk. The registry lists no death counts for the groups; deaths are not reported in the registry record.

These figures describe reported events in the recorded groups. They do not establish that an event was caused by a treatment, and the small group counts make simple numerical comparisons especially limited. They also do not give a complete picture of timing, duration, severity beyond the serious-event classification, or relatedness. Such details are not reported in the registry record.

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

The posted results describe a randomized comparison of intensive and standard management and report C-peptide geometric means at 1 year, with separate confidence intervals for each group. The registry's primary ANCOVA p-value is 0.49. Taken together, these registry entries show what was measured and the reported inferential result, but they do not provide an adjusted effect estimate or its confidence interval. The precision of the adjusted treatment contrast therefore remains difficult to assess from the registry record.

A p-value of 0.49 is not proof that the treatments are equivalent, and it is not proof that no difference exists. A non-small p-value does not identify the size of a possible effect or establish that the study could reliably exclude effects of meaningful size. Equivalence would require a defined margin and an analysis designed to test it; neither is reported in the registry record.

The result also should not be taken as a direct measure of every aspect of diabetes management. C-peptide response is the primary endpoint, while secondary outcomes cover other measures. The posted information does not establish that a result on one measure predicts results on all the others, and the incomplete reporting of secondary analyses limits broad conclusions.

The registry status of TERMINATED does not, by itself, explain why the study ended or determine whether the posted results are biased. The reason for termination, interim analyses, and detailed missing-data methods are not reported. The study's enrollment and analysis populations also differ in description, so the primary results should be understood as applying to the participants included in that analysis rather than automatically to every enrolled participant.

Accordingly, the record supports a restrained statistical interpretation: it reports group-level one-year C-peptide summaries and an ANCOVA p-value, but leaves important information about the adjusted effect and analysis conduct unavailable. It does not establish equivalence, clinical benefit, or absence of effect.

14. Why This Trial Matters Statistically

This trial illustrates why endpoint summaries and inferential comparisons answer different questions. The group geometric means and their confidence intervals describe each arm, while ANCOVA provides the reported adjusted comparison. Without the adjusted estimate and its interval, the p-value alone conveys limited information about the magnitude and precision of the contrast. Readers need both effect estimates and uncertainty to judge what the data can distinguish.

It also shows why analysis populations matter. The study enrolled 71 participants, but the primary result concerns a described subset of participants who completed the one-year visit and had positive autoantibody, with an additional MMTT exclusion. The posted primary arm counts are 47 and 20. Distinguishing enrollment, follow-up, and endpoint-specific analysis populations is essential when assessing how broadly a result applies.

The secondary outcomes demonstrate another statistical challenge: multiple outcomes may provide a broad description of a trial, yet they also complicate interpretation when multiplicity handling and an outcome hierarchy are not reported. Unadjusted comparisons can be useful descriptive checks, but they are not equivalent to prespecified adjusted analyses.

Finally, planning illustrations based on observed effects are not evidence of the original design's power or rationale. Keeping planning calculations separate from observed results prevents a common misreading. Together, these features make the record a useful example of how to separate posted summaries, sponsor-reported tests, recalculations, and unknown analysis details.

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.

The registry record leaves several trial-specific uncertainties. It does not report a reason for termination, despite listing the study status as TERMINATED. It also does not provide a detailed participant flow or fully reconcile the described primary population with the posted arm counts. The impact of these gaps on representativeness cannot be determined from the record.

For the primary endpoint, the group geometric means and intervals are reported, but the adjusted between-group estimate and confidence interval are not. The ANCOVA covariates and model details are also absent. This limits interpretation of the magnitude, direction, and precision of the primary adjusted comparison beyond the reported p-value.

The registry does not report a multiplicity strategy, a missing-data approach, interim analyses, or a detailed sample-size justification. Secondary results are unevenly reported, and some sponsor analyses are absent. The simple recalculations cannot resolve these omissions because they use unadjusted posted summaries and may not reproduce the trial's intended analyses.

The study is listed as unmasked, which is relevant when considering outcomes that may be influenced by behavior or assessment. The registry does not describe additional safeguards or how outcome assessment was conducted. No assumption about such procedures can be made from the record.

Finally, the primary population description and the available endpoint results do not establish how well findings generalize beyond participants represented in the analysis. This is a limitation of what can be concluded from the posted record, not evidence that the observed results are invalid.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effect of Metabolic Control at Onset of Diabetes on Progression of Type 1 Diabetes — NCT00891995.
• Registry record last updated 2016-12-30; status last verified 2016-11.
• Results first posted 2016-10-27.
• 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 randomized, parallel-group, unmasked study of intensive versus standard treatment for type 1 diabetes, with enrollment of 71 and status TERMINATED. The primary endpoint was C-peptide average AUC in response to a mixed meal at 1 year. The posted geometric means were 0.43 pmol/ml for intensive treatment and 0.52 pmol/ml for standard treatment, with an ANCOVA p-value of 0.49.

The primary adjusted effect estimate and confidence interval, the reason for termination, and several analysis details are not reported in the registry record. Secondary outcomes cover a range of metabolic and safety measures, but their analyses are not uniformly available. Unadjusted recalculations and planning illustrations should be read as supplementary descriptions, not as substitutes for the reported primary analysis or the trial's original design rationale.