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 Threshold Suspend in Pediatrics at Home (NCT02120794), a na na study. The primary outcome is Overall Mean Change in A1C.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02120794 |
| Phase | NA |
| Condition | Type 1 Diabetes |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 194 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Medtronic MiniMed, Inc. |
| Results posted | Yes |
2. Clinical Question
Population
Pediatric patients with type 1 diabetes. The registry reports enrollment of 194. For the primary endpoint, the description says 132 subjects had both baseline and end-of-study A1C measurements.
Intervention
Home use of the 530G insulin pump with the Threshold Suspend (TS) feature for one year.
Comparator
There was no concurrent control group in this single-group study. Allocation is recorded as NA, and the registry identifies the study as non-randomized.
Primary question
Whether the overall mean change in A1C from baseline to one year met the registry’s stated non-inferiority criterion, using an A1C margin of 0.4% and a one-sided significance level of 0.025. The registry record does not report the mean change, confidence interval, or test result.
The question concerns change in a continuous laboratory measure during use of the device at home, rather than a comparison of outcomes between randomized treatment groups. The study objective is described as demonstrating that home use of Threshold Suspend is not associated with glycemic deterioration, as measured by change in A1C. That objective is a statement of what the study set out to assess; it is not itself evidence that the endpoint met its criterion.
Because there is no control group, the central statistical question is whether the observed change satisfies the specified non-inferiority criterion, not whether the device outperforms an alternative. The registry record does not report an endpoint result, so it cannot establish from the posted information whether the criterion was met. Nor does the record provide enough detail to evaluate how changes over time might relate to factors other than device use.
3. Trial Design
The registry describes a multi-center, prospective, observational study of Threshold Suspend with a sensor-augmented pump system in pediatric patients with type 1 diabetes. Its title identifies the setting as home use. The study is marked completed, with enrollment of 194. The recorded start is 2014-06, and primary completion is dated 2017-06-16.
The design is listed as single group, with allocation recorded as NA and masking recorded as NONE. The intervention is the 530G insulin pump, described as being used with the Threshold Suspend feature for one year. There is no concurrent comparison group in the registry record. Consequently, the design does not provide a randomized, like-for-like contrast between people assigned to the device and people assigned to another treatment or usual care.
The primary endpoint is overall mean change in A1C from baseline to one year after screening. The endpoint description says that 136 subjects completed the study and that 132 had both baseline and end-of-study A1C; it states that the primary endpoint was based on those 132 subjects. The record does not explain the reasons that some completers lacked both measurements or describe how missing measurements were handled.
The registry lists four outcome measures and one analysis. Three secondary-outcome entries have the same title concerning mean A1C change stratified by A1C subgroups. The record does not provide the subgroup definitions or results. It also does not report further detail about site-specific procedures, the way participants were recruited, or the protocol governing device use beyond the one-year description.
The study’s observational, single-group structure is important when interpreting change over time. A before-and-after result can describe the measured course of A1C among those with usable measurements, but without a concurrent control it cannot isolate the effect of the device from other influences on A1C. The registry record does not report a strategy for addressing those influences.
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 | 530G Insulin Pump |
|---|---|
| STARTED | 194 |
| COMPLETED | 136 |
| NOT COMPLETED | 58 |
Reasons for not completing this period:
| Reason | 530G Insulin Pump |
|---|---|
| Withdrawal by Subject | 43 |
| Adverse Event | 10 |
| Physician Decision | 4 |
| Due to insurance coverage | 1 |
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 | 530G Insulin Pump |
|---|---|---|
| Participants analysed | 194 | |
| Age, Continuous | years | 11.7 (SD 2.54) |
| Sex: Female, Male: Female | Participants | 70 |
| Sex: Female, Male: Male | Participants | 124 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 29 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 164 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 1 |
| Region of Enrollment: United States | particpants | 194 |
6. Primary Endpoint
The registry lists the primary outcome as Overall Mean Change in A1C, measured over Baseline and 1 year after screening.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | percentage |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the overall mean change in A1C from baseline to one year after screening, with A1C expressed as a percentage. It is a continuous outcome: each participant’s measurement can take a range of values, and the group-level quantity of interest is the average change. The registry describes a 95% confidence interval for this endpoint.
A1C is used in this study as the measure of glycemic change named in the objective. Assessing the mean change allows the study to ask whether the observed course over the follow-up period is compatible with the prespecified non-inferiority criterion. The registry states an A1C margin of 0.4% and a significance level of 0.025 (one-sided). The margin defines the boundary used in the stated non-inferiority assessment; it is not an observed change or a result.
The endpoint description says the primary endpoint was based on the 132 subjects who had both baseline and end-of-study A1C. It does not report the mean change, the confidence interval limits, or the non-inferiority test outcome. Thus the endpoint’s definition and analysis population are described, but its numerical result cannot be interpreted from the registry record.
In general, a mean summarizes the average change but does not show how individual changes are distributed. A confidence interval communicates uncertainty around the estimated mean. Neither quantity is reported numerically in the registry record, so the magnitude and precision of the primary endpoint remain unknown from the posted information.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record defines a non-inferiority question but does not report the numerical result needed to answer it. It gives the endpoint, the 0.4% A1C margin, the one-sided significance level of 0.025, and the number of subjects with both measurements. It does not give the mean change in A1C, the 95% confidence interval, or a test result. The stated objective should therefore not be read as a finding.
In general, a non-inferiority analysis asks whether the data rule out an amount of worsening beyond a prespecified margin. Its conclusion depends on the estimated effect and its uncertainty in relation to that margin, with the direction of the contrast and the test definition made explicit. The registry supplies the margin but not the estimate, interval limits, or detailed hypothesis specification. It is not possible to determine whether the criterion was met, how close the result was to the margin, or how precise the estimate was.
The study is prospective and observational, and the design is listed as single group. A measured change from baseline to follow-up would describe change among the analyzed participants. Without a concurrent control group, however, that change would not by itself show what would have happened over the same period without the device. Changes could coincide with other influences; the registry does not describe an approach for separating those influences from device use.
The endpoint analysis was based on 132 subjects with both measurements, according to the registry description. The same description reports that 136 subjects completed the study. The record does not explain why measurements were unavailable for some completers, whether participants without paired measurements differed from those included, or how missing data may have affected the estimate. This limits assessment of the representativeness of the analyzed group.
The registry also reports safety counts, but those figures do not fill the gap in the primary efficacy result. The serious adverse-event field reports 4 affected among 194 at risk; the other-adverse-event field reports 121 among 194 at risk, and deaths are recorded as 0. The record does not give enough detail in these summary figures to characterize event timing, duration, severity beyond the serious classification, or individual exposure. These entries should be read as registry safety summaries, not as a comparison of risk against an alternative group.
Overall, the record supports a description of the study’s design, endpoint, and analysis population. It does not support a conclusion about whether the primary non-inferiority criterion was met or how large and precise the A1C change was. The distinction between a study objective and a reported result is central to interpretation.
9. Statistical Methods Explained
What does the mean change represent?
The primary measure is the mean change in A1C from baseline to one year after screening. For a continuous outcome, a mean expresses the average value of the measured changes in the analysis group. It is a useful summary of the group’s overall direction and magnitude, but it does not describe every participant’s change or show the spread of individual values. The registry identifies the parameter as a mean and specifies a 95% confidence interval, but it does not report the mean estimate or interval limits.
In general, a confidence interval gives a range of values compatible with the data and the statistical procedure under its assumptions. Its width conveys precision: a narrower interval indicates greater precision than a wider one. The interval is not a guarantee that the true value lies within its limits, and it does not convey the distribution of individual outcomes. Since the numerical interval is not reported in the registry record, neither precision nor compatibility with the non-inferiority margin can be assessed from this record.
How does non-inferiority differ from a superiority question?
In general, a non-inferiority test evaluates whether an outcome is sufficiently close to a reference standard that it does not cross a prespecified boundary for unacceptable difference. The boundary is called the margin. Here, the registry states an A1C margin of 0.4% and a one-sided significance level of 0.025. The margin is a criterion for the planned assessment, not a result observed in participants.
A test result cannot be inferred from the margin alone. Interpretation requires the estimated change or contrast, its uncertainty, and the direction in which the margin is applied. The registry describes an overall mean change and a non-inferiority test, but does not report the test statistic, p-value, confidence interval limits, or a full specification of the contrast and its direction. The record therefore does not show whether the analysis met the stated criterion.
For a one-sided test, the inferential question is framed in one direction rather than assessing departures in both directions. That choice must match the scientific question and the definition of the outcome. The registry reports the one-sided significance level, but does not provide the detailed statistical hypothesis or explain how the direction was encoded. It is not appropriate to reconstruct those details or the conclusion from the registry description alone.
What can the single-group structure establish?
The record lists the model as single group and allocation as NA. In general, a baseline-to-follow-up analysis in one group estimates change over time in the participants included in the analysis. It does not create the same counterfactual comparison as random assignment to a concurrent control group. Even if an estimate were available, the change could reflect influences other than the intervention, and the design would not separate those influences through a randomized between-group contrast.
The study is described as prospective and observational. These features describe how the study was organized, but they do not remove the interpretive limitation of having no concurrent comparator. Masking is recorded as NONE. The registry does not report additional methods for limiting measurement or analysis bias, so further safeguards should not be assumed.
Why do paired measurements matter?
The endpoint description states that 132 subjects had both baseline and end-of-study A1C and that the primary endpoint was based on those subjects. Paired measurements allow change to be evaluated using measurements from the same person at both time points. The description also says that 136 subjects completed the study. The registry does not explain why some completers lacked a pair of measurements or report whether the analysis used any other information from participants without paired values.
In general, missing outcome data can affect an estimate if the participants with observed measurements differ systematically from those without them. The size and direction of any such effect depend on the missingness pattern and the data. The registry record does not report a missing-data method, sensitivity analysis, or assumptions about why values were unavailable. It is therefore not possible to assess the effect of missing measurements on the endpoint.
What do the planning illustrations mean?
In general, sample-size calculations are planning tools that describe the number of participants a hypothetical design might require under specified assumptions. They are not observed power and do not replace analysis of the trial’s outcome. The planning illustration associated with the posted material is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with an effect defined as a fraction of the observed effect. These assumptions do not describe the actual single-group design listed in the registry, and the illustration is not the trial’s own design justification.
The registry record does not report a sample-size justification, the assumptions used to plan enrollment, interim analyses, or a full statistical analysis plan. It also does not give the observed effect needed to evaluate any planning illustration. Accordingly, such calculations should be treated only as hypothetical planning examples, not as evidence about the achieved precision or a basis for claiming that the study had adequate power.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists four outcome measures and one analysis. The secondary-outcome list contains three entries with the same title: mean change in A1C from baseline to one year, stratified by different A1C subgroups. The subgroup definitions and results are not reported in the registry record. The record does not explain whether these repeated entries refer to distinct analyses or how the subgroup assessments were specified.
When multiple outcomes or subgroup analyses are evaluated, the chance of at least one apparently notable result can increase. In general, a multiplicity plan can define how that issue is handled, for example by prioritizing outcomes or adjusting inference. The registry does not report a multiplicity adjustment or other strategy for interpreting the secondary analyses. The stated one-sided significance level applies to the described primary non-inferiority assessment; the record does not explain how it relates to secondary outcomes.
For the primary endpoint, the record states that 136 subjects completed the study and that 132 had both baseline and end-of-study A1C. The endpoint was based on those 132 subjects. Reasons for the missing paired measurements, methods for handling missing data, and sensitivity analyses are not reported in the registry record. The record also does not define additional analysis populations or describe whether all enrolled participants were included in any other efficacy analysis.
These omissions matter because outcome availability can affect both the number of observations contributing to an estimate and the interpretation of that estimate. Without information on why measurements were missing and how the analysis treated them, the potential influence of missingness cannot be assessed. Likewise, without a stated multiplicity approach or subgroup details, secondary findings cannot be evaluated for their inferential status from 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 |
|---|---|---|---|
| 530G Insulin Pump | 4 | 194 | 2.1% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| 530G Insulin Pump | 0 of 194 |
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 |
|---|---|
| 530G Insulin Pump | 121 of 194 |
Most frequent serious adverse event terms
| Event term | Organ system | 530G Insulin Pump |
|---|---|---|
| Dabetic Ketoacidosis | Metabolism and nutrition disorders | 2 of 194 |
| Severe Hyperglycemia | Metabolism and nutrition disorders | 1 of 194 |
| Gastroenteritis | Infections and infestations | 1 of 194 |
| Second Degree Burns to lower distal extremeties | Injury, poisoning and procedural complications | 1 of 194 |
Most frequent non-serious adverse event terms
| Event term | Organ system | 530G Insulin Pump |
|---|---|---|
| Hyperglycaemia | Metabolism and nutrition disorders | 28 of 194 |
| Skin irritation | Skin and subcutaneous tissue disorders | 20 of 194 |
| Upper respiratory tract infection | Infections and infestations | 18 of 194 |
| Medical device site reaction | General disorders | 17 of 194 |
| Rash | Skin and subcutaneous tissue disorders | 10 of 194 |
| Medical device pain | General disorders | 8 of 194 |
| Pharyngitis streptococcal | Infections and infestations | 8 of 194 |
| Cough | Respiratory, thoracic and mediastinal disorders | 6 of 194 |
| Gastroenteritis | Infections and infestations | 6 of 194 |
| Headache | Nervous system disorders | 6 of 194 |
Serious events with the highest arm-level rate: dabetic ketoacidosis (1.0%), severe hyperglycemia (0.5%), gastroenteritis (0.5%), second degree burns to lower distal extremeties (0.5%).
The registry’s safety summary lists 194 at risk for the 530G Insulin Pump group. It reports 4 affected in the serious adverse-event field, 121 affected in the other-adverse-event field, and 0 deaths. These are registry counts, not rates calculated. The record does not describe a comparison group, so the figures do not provide a comparative estimate of safety against another treatment.
The listed serious terms are “Dabetic Ketoacidosis” at 1.0%, “Severe Hyperglycemia” at 0.5%, “Gastroenteritis” at 0.5%, and “Second Degree Burns to lower distal extremeties” at 0.5%. The terms are reproduced as recorded. The registry does not provide details in this summary about event timing, duration, recurrence, relationship to the device, or the number of events per affected participant. The listed term percentages should not be treated as a complete description of each event’s clinical course.
In general, serious and other adverse-event summaries are distinct categories, and event counts should not be interpreted as evidence of causation. The posted figures do not establish whether an event was caused by the device or how risk compares with an alternative. They should be read alongside the study’s single-group observational design and the limited event detail in the registry record.
12. What the Result Does, and Does Not, Mean
The registry establishes that this completed prospective, observational study assessed home use of the 530G insulin pump with Threshold Suspend in pediatric patients with type 1 diabetes. Its primary question concerned mean A1C change from baseline to one year and a stated non-inferiority margin of 0.4%. The endpoint description identifies the analysis group as subjects with both measurements, while the posted record does not give the numerical endpoint result.
Without the mean change, 95% confidence interval, or test result, the record cannot show whether the non-inferiority criterion was met. It also cannot show the estimated magnitude or precision of change. The study objective describes the intended assessment, not its outcome. No conclusion about the primary endpoint should be substituted for the missing result.
Even if the numerical change were available, the single-group observational design would limit causal interpretation. A baseline-to-follow-up change describes participants over time, but in the absence of a concurrent control group it cannot isolate the contribution of device use from other influences. The registry does not report a method that addresses this limitation.
The safety summary reports serious and other adverse-event counts and deaths for the study group. It does not provide a comparative safety estimate, and the limited event detail does not establish causation. The listed subgroup outcomes also lack definitions and reported results in the registry record.
In general, a registry record is most informative when its objective, prespecified methods, participant flow, outcome estimates, uncertainty, and missing-data handling can be considered together. In this case, design and endpoint details are available, but the key numerical efficacy result and several analysis details are not reported. The record therefore supports a description of what was studied and how the primary question was framed, but not a determination of the endpoint’s result or its causal meaning.
13. Why This Trial Matters Statistically
This study illustrates the distinction between an endpoint’s planned purpose and evidence that the endpoint met its criterion. The registry specifies a mean A1C change, a 0.4% non-inferiority margin, a one-sided significance level of 0.025, and a 95% confidence interval. Those details describe the intended inferential framework. Without the estimate, interval, and test result, the framework cannot be used to determine the study’s finding.
It also shows why design matters alongside statistical testing. A non-inferiority test addresses a defined margin, but a single-group observational study does not provide a concurrent randomized comparator. An estimate of change over time and a causal estimate of what would have happened under another condition are different quantities. Statistical precision cannot by itself resolve that design limitation.
The analysis population also deserves attention. The registry says the primary endpoint was based on 132 subjects with both baseline and end-of-study measurements, while 136 subjects completed the study. Without an account of the missing measurements or a missing-data analysis, readers cannot evaluate how representative the endpoint group may be or how robust the estimate would be to missingness.
Finally, safety counts and secondary-outcome titles should be interpreted in the context of the information actually recorded. Summary counts do not establish causation, and repeated subgroup titles without definitions or results do not permit independent assessment of subgroup findings. Careful reading therefore depends on separating what the registry documents from conclusions that would require unreported results or methods.
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 limitation for interpreting the main question is that the registry record does not report the mean A1C change, confidence interval limits, or non-inferiority test result. Although the endpoint and margin are described, the essential result needed to judge the stated criterion is absent. The record also does not provide the detailed hypothesis specification or test statistic.
The study is listed as single group, observational, and non-randomized, with no concurrent control group. Accordingly, a change from baseline to follow-up cannot be interpreted as a controlled estimate of the effect of the device. The registry does not describe an adjustment strategy that would separate device use from other influences on A1C.
The endpoint description reports 136 study completers and 132 subjects with both baseline and end-of-study A1C, but does not explain why measurements were unavailable or report a missing-data method or sensitivity analysis. The potential effect of missing paired measurements on the endpoint cannot be evaluated. Further details about analysis populations are also not reported in the registry record.
The three secondary-outcome entries share the same subgroup-stratified title. The subgroup definitions, results, and any multiplicity strategy are not reported. The safety summary gives counts and selected terms but lacks detail about timing, recurrence, and attribution. The registry also does not report the sample-size justification, interim analyses, or a full statistical analysis plan. These limits restrict assessment of both the primary inference and the broader evidential context.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Threshold Suspend in Pediatrics at Home — NCT02120794.
- Registry record last updated 2018-09-11; status last verified 2018-08.
- Results first posted 2018-08-15.
- 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
NCT02120794 is a completed, prospective, observational, multi-center study of home use of the 530G insulin pump with Threshold Suspend in pediatric patients with type 1 diabetes. The registry lists a single-group design, enrollment of 194, and no masking. The recorded start is 2014-06, and primary completion is 2017-06-16.
The primary endpoint is overall mean change in A1C from baseline to one year after screening. Its description specifies a 0.4% non-inferiority margin and a one-sided significance level of 0.025, and says that the endpoint was based on 132 subjects with both measurements. The registry record does not report the endpoint estimate, confidence interval limits, or test result. Safety fields report 4 serious adverse-event cases, 121 other adverse-event cases, and 0 deaths among 194 at risk. No concurrent control group is reported.