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 Feasibility of a Decision Support System to Reduce Glucose Variability in Subject With T1DM (NCT02558491), a randomized na study. The primary outcome is Glucose Variability (Coefficient of Variation).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02558491 |
| Phase | NA |
| Condition | Type 1 Diabetes Mellitus |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Single |
| Enrollment | 33 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Daniel Chernavvsky, MD |
| Results posted | Yes |
2. Clinical Question
Population
People with type 1 diabetes mellitus were included. The registry describes participants using insulin pumps and multiple daily injections in the intervention descriptions, while the primary outcome description specifies participants using an insulin pump. The record reports enrollment of 33 and says 24 participants completed the study; participants who withdrew were excluded from the analysis regardless of how much of the protocol they completed.
Intervention
The intervention was a decision support system intended to guide insulin therapy and reduce glucose variability. Blinded continuous glucose monitor data collected before the experimental admission were analyzed by the study team to determine optimized insulin therapy parameters. The registry description says those parameters would be used during the experimental admission, but the description ends before specifying further implementation details.
Comparator
Usual care was the comparison condition. During the control admission, participants using insulin pumps or multiple daily injections were to use their own insulin parameters, including basal rate, correction factor and carbohydrate-insulin ratio, and determine their own insulin usage.
Primary question
The primary question was whether the glucose variability advisory system could reduce glucose variability, measured by the coefficient of variation, during a 48 hour study admission. The registry record does not report the observed primary outcome values or a treatment comparison, so it does not establish what the study found about the size or precision of any difference.
3. Trial Design
The registry describes a randomized, single-masked crossover study, with enrollment of 33 and completed status. The stated purpose was to demonstrate the safety and feasibility of a decision support system aimed at reducing glucose variability in people with type 1 diabetes using an insulin pump. Crossover designs compare conditions within participants, rather than assigning each participant to only one condition throughout the study. This can help account for stable differences between participants, but interpretation depends on the order and timing of the conditions and on how effects that persist from one condition to another are handled.
The intervention conditions were the decision support system and usual care. The system's parameters were based on blinded continuous glucose monitor data collected before the experimental admission and analyzed by the study team. During the control admission, participants were to use their own insulin parameters and determine their own insulin usage. The record does not specify the sequence of conditions, the duration between admissions, whether a washout period was used, or exactly who was masked. Those details are not reported in the registry record.
The primary outcome timeframe was the duration of the 48 hour study admission. The registry says 24 participants completed the study and that data from participants who withdrew were removed from analysis regardless of protocol completion. It does not explain how many participants contributed to each outcome analysis or describe an analysis that accounts for the crossover structure. The record's intervention descriptions include both insulin pump and multiple daily injection participants, whereas the primary outcome description refers to people using an insulin pump. The scope of the primary analysis is therefore not fully clarified by the registry text.
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: Blinded continuous glucose monitor (CGM) data will be collected for \~28 days prior to the Experimental Admission and analyzed by the study team to determine the optimal insulin therapy parameters for each insulin pump and multiple daily injections (MDI) participant. These optimized parameters will be used during the \~48 hour Experimental Admission.
First Invention
| Stage | Experimental (Decision Support) Admission, Then Control (Usual Care) Admission | Control (Usual Care) Admission, Then Experimental (Decision Support) Admission |
|---|---|---|
| STARTED | 19 | 14 |
| COMPLETED | 14 | 12 |
| NOT COMPLETED | 5 | 2 |
Reasons for not completing this period:
| Reason | Experimental (Decision Support) Admission, Then Control (Usual Care) Admission | Control (Usual Care) Admission, Then Experimental (Decision Support) Admission |
|---|---|---|
| Withdrawal by Subject | 3 | 1 |
| Data collection issues | 1 | 0 |
| Stopping criteria | 0 | 1 |
| Changed parameters during trial | 1 | 0 |
Second Intervention
| Stage | Experimental (Decision Support) Admission, Then Control (Usual Care) Admission | Control (Usual Care) Admission, Then Experimental (Decision Support) Admission |
|---|---|---|
| STARTED | 14 | 12 |
| COMPLETED | 12 | 12 |
| NOT COMPLETED | 2 | 0 |
Reasons for not completing this period:
| Reason | Experimental (Decision Support) Admission, Then Control (Usual Care) Admission | Control (Usual Care) Admission, Then Experimental (Decision Support) Admission |
|---|---|---|
| Withdrawal by Subject | 1 | 0 |
| Stopping Criteria | 1 | 0 |
Third Intervention
| Stage | Experimental (Decision Support) Admission, Then Control (Usual Care) Admission | Control (Usual Care) Admission, Then Experimental (Decision Support) Admission |
|---|---|---|
| STARTED | 12 | 12 |
| COMPLETED | 12 | 12 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: 24 participants completed the study. Data from subjects who withdrew were removed from analysis regardless of the degree of protocol completion.
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 | All Participants |
|---|---|---|
| Participants analysed | 24 | |
| Age, Continuous | years | 37 (SD 11) |
| Sex: Female, Male: Female | Participants | 15 |
| Sex: Female, Male: Male | Participants | 9 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 |
| Race (NIH/OMB): Asian | Participants | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 |
| Race (NIH/OMB): White | Participants | 24 |
| Race (NIH/OMB): More than one race | Participants | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 |
| Type 1 Diabetes Mellitus (T1DM) duration | years | 21 (SD 11.1) |
| Hemoglobin A1c (HbA1c) | percentage | 7.2 (SD 1) |
| Height | cm | 172 (SD 10.7) |
| Weight | kg | 81.8 (SD 19.9) |
| Body Mass Index (BMI), kg/m2 | kg/m^2 | 27.4 (SD 4.6) |
| Total Daily Insulin (TDI) | U | 46.7 (SD 22.3) |
6. Primary Endpoint
The registry lists the primary outcome as Glucose Variability (Coefficient of Variation), measured over Duration of the 48 hour study admission.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | coefficient of variation |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was glucose variability, measured as the coefficient of variation during the 48 hour study admission. A coefficient of variation expresses the spread of measurements relative to their mean, making it a way to describe variability on a scale that accounts for the measurement's overall level. This is a general explanation of the measure; the registry identifies the endpoint as a mean of a continuous measure and lists standard deviation as its dispersion statistic.
Glucose variability is relevant to the study's stated objective because the decision support system was intended to reduce variability, rather than to assess only a single glucose reading. The endpoint's meaning depends on the measurement procedure, the period represented, and the analysis used. The registry specifies the admission timeframe but does not report the observed mean coefficient of variation, its standard deviation, or an estimated comparison between conditions. It also does not report a confidence interval or other measure of uncertainty for the primary endpoint. Accordingly, the endpoint and its intended role can be described, but the magnitude and precision of the observed outcome cannot be assessed from the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinical question and a quantitative primary endpoint, but it does not report the primary outcome results. There are no posted values for mean glucose variability under either condition, no estimate of the within-participant comparison, and no confidence interval or test result. The direction, magnitude, and precision of the primary outcome comparison therefore cannot be determined from this record. An absence of reported results is not evidence that the conditions were equivalent or that the system had no effect.
The study is described as a feasibility study with a safety objective and a glucose variability objective. Feasibility and an endpoint comparison are related but distinct questions. A feasibility assessment would ordinarily describe whether study procedures and intervention delivery could be carried out; specific feasibility criteria and results are not reported in the registry record. The primary outcome is described as an assessment of effectiveness, but that description alone is not an outcome finding.
The registry reports enrollment of 33 and states that 24 participants completed the study. It also says that participants who withdrew were excluded from analysis regardless of the degree of protocol completion. This matters when interpreting any outcome estimate because the analyzed participants may not represent everyone enrolled. The record does not give the reasons for withdrawal, the number included in each analysis, or an assessment of how missing outcomes may affect the results.
There are 13 outcome measures and 32 analyses reported in the record. These counts indicate that multiple outcomes and analyses were planned or recorded, but do not establish how results were selected, whether statistical adjustment for multiple comparisons was used, or which analyses were considered primary beyond the named endpoint. Such details are not reported in the registry record. Without the outcome estimates and analysis details, statistical significance, uncertainty, and robustness cannot be evaluated.
The safety entries list no serious adverse events and no deaths in either condition, and list one other adverse event in each. These entries describe the recorded counts, not a complete safety interpretation. The record does not provide event descriptions or explain how adverse events were attributed in a crossover study. The posted safety counts should therefore be read alongside the limits of the reporting, not as proof that either condition is free of risk.
Overall, the registry establishes that a randomized crossover feasibility study was conducted and identifies its intended comparison and endpoint. It does not provide enough outcome data to conclude how glucose variability differed between conditions or how precisely any difference was estimated.
9. Statistical Methods Explained
The registry describes a randomized, single-masked crossover design. In a randomized study, the assignment process is intended to reduce systematic differences in how conditions are allocated. In a crossover study, participants receive or experience the comparison conditions in sequence, so each participant can contribute information under more than one condition. This general design explanation does not establish the sequence used in this study: the sequence and allocation details are not reported in the registry record.
What does a crossover comparison require?
In general, crossover analyses need to reflect that measurements from the same participant are related. Treating those measurements as if they came from unrelated participants can misrepresent uncertainty. The analysis may also need to account for period effects, condition order, and carryover, meaning an effect of an earlier condition that remains during a later period. The registry record does not report whether these features were assessed or how they were handled. It does not identify the statistical model used for the primary endpoint.
What does the endpoint summary mean?
The primary endpoint is described as a mean of a continuous measure, with standard deviation as the dispersion statistic. A mean describes the average value in the analyzed observations. Standard deviation describes how spread out individual observations are around their mean. These are descriptive summaries; by themselves they do not show whether a condition comparison is statistically distinguishable from no difference, nor do they describe the precision of that comparison. The record does not state the endpoint values or provide an estimated contrast and its uncertainty.
A confidence interval is a general statistical tool for expressing the precision of an estimate. It gives a range of values compatible with the data and the method's assumptions. A narrow interval indicates greater precision than a wide interval, while an interval spanning values on both sides of a comparison's null value may leave the direction or size of the effect uncertain. No confidence interval for the primary comparison is reported in the registry record. No p-value or other inferential result for that comparison is reported either.
Why can variability affect analysis?
General statistical explanation: an outcome's variability affects how precisely a comparison can be estimated. Greater variability can make it harder to distinguish a real pattern from random fluctuation, while lower variability can permit greater precision for the same amount of information. In crossover studies, the variability of within-participant differences is particularly relevant. The registry names standard deviation as the dispersion measure but does not report the observed standard deviation or the variability of within-participant differences.
Glucose measurements can also show patterns that differ across participants or across time. General methods may account for repeated measurements, clustering, or over-dispersion when observations vary more than a simple model expects. Those approaches require appropriate data and assumptions. The registry record does not identify the measurement-level analysis, describe a model for repeated glucose observations, or report whether over-dispersion was considered. It would not be appropriate to infer a specific method from the endpoint name alone.
How should planning illustrations be read?
General statistical explanation: a sample-size calculation asks how much information a future study might need under stated assumptions. The record describes planning illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It also states that the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not evidence that the study was designed using those assumptions, and not a measure of the power observed after the study.
The record characterizes these calculations as planning illustrations rather than the trial's own design. Sample-size illustrations are sensitive to the effect size and variability assumed, as well as to the analysis and allocation assumptions. A smaller target effect generally requires more information to detect with comparable precision, all else equal; this is a general relationship, not a finding from this trial. The registry record does not report a sample-size justification for the original study. Without the primary outcome estimate and the specific illustration values, no numerical planning conclusion can be drawn from the record text.
What can be concluded from missing inferential details?
The absence of a reported statistical method does not show that no analysis was performed. It means the registry record does not let readers assess whether the analysis matched the randomized crossover design, how uncertainty was calculated, or how assumptions were checked. The outcome estimates and analysis details are needed to judge the strength and precision of the evidence. General explanations of methods cannot substitute for those trial-specific results.
10. Multiplicity, Missing Data and Analysis Populations
The record lists 13 outcome measures and 32 analyses. Multiple outcomes or analyses can create a multiplicity issue: when many statistical comparisons are made, the chance of finding an apparently noteworthy result by chance can increase. General statistical approaches may control that risk, for example by adjusting significance thresholds or by specifying a hierarchy of outcomes in advance. The registry record does not report whether any adjustment was made, which outcomes were included in the 32 analyses, or how the analyses were prioritized. The primary endpoint is identified, but the record does not provide a broader statistical analysis plan.
Missing data are addressed only in part. The registry states that 24 participants completed the study and that data from participants who withdrew were removed from analysis regardless of how much of the protocol they completed. It does not report the reasons for withdrawal, the amount of missing outcome data, or whether any imputation or sensitivity analysis was used. It also does not give the number included in each outcome analysis. Excluding participants with incomplete participation can affect the interpretation if those participants differ from those retained; whether that occurred cannot be determined from the record.
Analysis populations are not described in enough detail to identify an intention-to-treat population or a per-protocol population. The record also does not explain how the crossover structure was addressed when participants withdrew or had incomplete periods. These omissions limit assessment of potential bias and the robustness of any comparison. They do not establish that the analyses were inappropriate; the relevant information is simply not reported in the registry record.
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 |
|---|---|---|---|
| Decision Support System | 0 | 33 | 0.0% |
| Usual Care | 0 | 33 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Decision Support System | 0 of 33 |
| Usual Care | 0 of 33 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Decision Support System | 1 of 33 |
| Usual Care | 1 of 33 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Decision Support System | Usual Care |
|---|---|---|---|
| Hypoglycemia requiring glucagon administration | Endocrine disorders | 0 of 33 | 1 of 33 |
| Ketones | Endocrine disorders | 1 of 33 | 0 of 33 |
| Respiratory illness | Respiratory, thoracic and mediastinal disorders | 0 of 33 | 1 of 33 |
| Acute Herpetic Gingivitis | Infections and infestations | 0 of 33 | 1 of 33 |
| Ankle fracture | Musculoskeletal and connective tissue disorders | 0 of 33 | 1 of 33 |
| Bump on head | Injury, poisoning and procedural complications | 0 of 33 | 1 of 33 |
The adverse-event entries list zero serious adverse events in the decision support system condition and zero in usual care. They also list one other adverse event in each condition and zero deaths in each. The record gives 33 participants at risk for serious events and 33 at risk for other events in each condition. These are the registry's reported group figures; because the study is described as a crossover, the group entries should not be interpreted as separate sets of participants.
Counts alone do not describe the nature, timing, duration, severity, or relatedness of the events. The record does not provide event terms for serious adverse events, and it does not give a description of the other adverse events. It also does not explain how events were attributed to a condition or whether the same participant could contribute events under both conditions. The listed figures therefore provide a limited account of recorded safety events, not a comprehensive comparison of safety. They do not establish that either condition is free of risk or that the conditions differ in safety.
12. What the Result Does, and Does Not, Mean
The registry record establishes that a randomized, single-masked crossover feasibility study evaluated a decision support system against usual care, with glucose variability measured by the coefficient of variation as the primary endpoint. It also describes the intended intervention: study-team analysis of blinded continuous glucose monitor data to determine optimized insulin parameters for use during the experimental admission. The control condition involved participants using their own insulin parameters and determining their own insulin usage.
What the record does not establish is the primary outcome result. No condition-specific mean, comparison estimate, uncertainty interval, or test result is reported. It is therefore not possible to determine the direction or size of any difference in glucose variability, or how precisely a difference was estimated. The lack of a posted result should not be interpreted as evidence for no difference, and the study's objective should not be mistaken for its finding.
The reported completion and withdrawal information gives some context for the analyzed data, but does not reveal how missing observations affected the results. Likewise, the reported adverse-event counts do not describe event details or establish a comparative safety profile. Additional outcome and analysis information would be needed to assess the clinical and statistical meaning of the findings.
More generally, a feasibility study can inform whether study procedures and an intervention can be evaluated, while a treatment-effect conclusion depends on the outcome estimates, uncertainty, analysis quality, and study context. The registry record states a safety and feasibility purpose, but does not report feasibility criteria or results. Thus the design and intended question are identifiable, while the empirical result and its implications remain unreported in the registry record.
13. Why This Trial Matters Statistically
This trial is statistically interesting because it combines a randomized comparison with a crossover design and a continuous glucose variability endpoint. A crossover structure can make within-participant comparison possible, which is useful when participants differ in their usual glucose patterns. At the same time, it requires an analysis that recognizes repeated observations within participants and considers condition order and possible carryover. Those design and analysis details are not reported in the registry record, making the example a reminder that the design label alone does not show how the data were analyzed.
The endpoint also illustrates the distinction between a meaningful measurement and a reported result. The coefficient of variation is intended to characterize glucose variability relative to its mean, but that endpoint name does not reveal the observed effect or its precision. A usable interpretation requires condition-specific summaries, a comparison estimate, and uncertainty information. Those results are not reported in the registry record.
The study's feasibility purpose and its stated primary outcome raise related but separate questions. Feasibility concerns whether the study and intervention can be carried out; an endpoint comparison concerns the measured outcome. The record does not specify feasibility criteria or their results, and it does not provide the primary outcome comparison. This makes clear why a trial's objective, its endpoint definition, and its findings must be distinguished.
Finally, the registry lists numerous outcome measures and analyses but does not report how multiplicity was handled. This is important for interpreting evidence across a set of comparisons, especially when the analysis plan is not available. In this record, the main statistical lesson is not a particular treatment effect: it is that design information and safety counts cannot replace outcome estimates, uncertainty, and transparent analysis details.
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 principal limitation for interpreting this study is that the registry record does not report the primary outcome results. It provides no mean coefficient of variation by condition, no comparison estimate, and no measure of uncertainty. The primary question therefore cannot be answered from the reported record. Feasibility criteria and results are also not described, despite the stated feasibility purpose.
Important design details are absent. The sequence of conditions, timing between admissions, any washout procedures, and the identity of the masked party are not reported in the registry record. The analysis model and its handling of the crossover structure are also not stated. These omissions prevent assessment of period effects, carryover, and whether the statistical analysis reflects the repeated nature of the measurements.
The population description is not fully consistent in scope: intervention descriptions refer to both insulin pump and multiple daily injection participants, while the primary outcome description refers to people using an insulin pump. The record does not clarify the primary analysis population. It reports that 24 participants completed the study and that withdrawal data were excluded, but does not describe reasons for withdrawal, outcome-specific analysis counts, or sensitivity analyses for missing data.
The record lists 13 outcome measures and 32 analyses without explaining their prioritization or whether multiplicity was addressed. The adverse-event entries also lack event descriptions and attribution details. Finally, the posted planning assumptions do not substitute for a documented original sample-size justification, which is not reported in the registry record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Feasibility of a Decision Support System to Reduce Glucose Variability in Subject With T1DM — NCT02558491.
- Registry record last updated 2024-08-07; status last verified 2024-08.
- Results first posted 2023-03-23.
- 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
The registry record describes a completed, randomized, single-masked crossover feasibility study of a decision support system versus usual care for people with type 1 diabetes. Enrollment was 33, and 24 participants completed the study. The primary endpoint was mean glucose variability, measured by the coefficient of variation during the 48 hour study admission.
The intervention used blinded continuous glucose monitor data analyzed by the study team to determine optimized insulin parameters; the control condition used participants' own parameters and insulin usage decisions. Participants who withdrew were excluded from analysis regardless of protocol completion. The record lists zero serious adverse events and zero deaths in each condition, and one other adverse event in each.
Primary outcome results, analysis methods, confidence intervals, feasibility results, and detailed missing-data procedures are not reported in the registry record. As a result, the magnitude and precision of the primary comparison cannot be assessed from the record.