The VRIF Trial: Hypoglycemia Reduction With Automated-Insulin Delivery System: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The VRIF Trial: Hypoglycemia Reduction With Automated-Insulin Delivery System (NCT03674281), a na na study. The primary outcome is Percentage of Time of Blood Glucose in Range 70-180 mg/dL.

2. Clinical Question

3. Trial Design

The VRIF Trial is registered as NCT03674281 and is sponsored by the University of Virginia. It is listed as a completed device study with a sequential model, no masking, and allocation recorded as not applicable. The registry lists enrollment of 50. The study start date is 2018-09-24, and the primary completion date is 2019-11-21. Phase is listed as NA.

The registered intervention names are sensor-augmented pump (SAP) and closed-loop control (CLC). The arm description states that participants used their own insulin pumps without automated insulin delivery, together with Dexcom G6 continuous glucose monitoring, for 4 weeks in the SAP period. Actigraphy watch monitoring and Ecological Momentary Assessments were planned in approximately the final 14 days of that period. The primary endpoint timeframe describes one month during each study period. The precise CLC regimen, order of study periods, and details of how participants moved between periods are not reported in the registry record.

The study objective includes testing an artificial pancreas system in children aged 6 to 10 years and adults aged 65 years or older with type 1 diabetes. It also describes assessing cognitive function and examining whether improved glycemic control is associated with cognitive function. The registry further characterizes the work as a pilot intended to generate preliminary data for future, larger and longer trials. Secondary outcomes are listed as Ecological Momentary Assessments and sleep patterns.

Because allocation is not applicable and the record explicitly cautions that the arms are not a like-for-like control comparison, the two periods should not be described as randomized parallel groups. The registry does not report the detailed allocation or sequence procedures, protocol eligibility criteria, or a sample-size justification.

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

Reasons for not completing this period:

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 Percentage of Time of Blood Glucose in Range 70-180 mg/dL, measured over One month during each study period (sensor augmented pump and closed loop control).

The primary endpoint is the percentage of time blood glucose is in the 70-180 mg/dL range, measured by continuous glucose monitoring. It is defined for one month during each study period, with the periods identified as sensor-augmented pump and closed-loop control. The registry specifies a mean as the summary parameter and standard deviation as the measure of dispersion, so the endpoint is treated as a continuous measure.

Time in a prespecified glucose range condenses repeated sensor readings into a measure of how much monitoring time falls within that range. In general, such a measure can describe glycemic patterns across a monitoring period, rather than relying on a single measurement. Its interpretation depends on the period observed, the sensor data available, and the analysis population; details about handling incomplete sensor records are not reported in the registry record.

The study summary also mentions stable control defined as more than 70% of the day in the 70-180 mg/dL range in connection with the cognitive-function objective. That threshold is distinct from the registered primary endpoint, which is the continuous percentage of time in range. The record does not report numerical endpoint values, differences between study periods, or uncertainty estimates. Therefore, the endpoint definition alone does not establish what values participants achieved or how the periods compared.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a percentage of monitoring time, summarized using a mean and standard deviation. In general, a mean gives the arithmetic average of observed values, while a standard deviation describes how dispersed those values are around the mean. These summaries are informative only when the analysis population and data handling are clear. The registry record does not report the actual means or standard deviations, nor does it specify how the endpoint was calculated for participants with incomplete sensor data.

What does a continuous endpoint tell us?

In general, a continuous outcome retains gradations in the measured quantity rather than reducing it to a yes-or-no category. For this trial, the endpoint is the percentage of time in the 70-180 mg/dL range during a month of monitoring. A continuous summary can distinguish values across the measurement scale, while the separate threshold mentioned in the study summary—more than 70% of the day in range—represents a categorization. The registry does not say that this threshold was the primary endpoint or define how it was used in a cognitive analysis.

Continuous glucose monitoring records repeated measurements over time. In general, repeated observations from a participant are related to one another, so they should not automatically be treated as independent participants. The registry identifies the participant-level outcome as a percentage of time in range, but does not report the detailed aggregation method, the sensor-data completeness rule, or the statistical model used for repeated measurements. It is therefore not possible to attribute a particular repeated-measures method to this study.

How should a comparison between study periods be understood?

The registry describes a sequential model and two named device periods. It does not describe random allocation and explicitly warns that the arms are not a like-for-like control comparison. In general, a within-participant comparison can account for characteristics that remain stable within an individual, but period order, changes over time, carryover, and differences in available observations can still affect interpretation. These are general design considerations, not claims about what occurred in this trial. The sequence details and any methods used to account for period effects are not reported in the registry record.

In general, an estimated difference summarizes the contrast chosen by an analysis, while a confidence interval expresses uncertainty around that estimate under the model’s assumptions. A p-value addresses compatibility of the data with a specified null hypothesis; it does not measure effect size or the probability that a treatment is effective. The registry record does not report an endpoint contrast, confidence interval, p-value, or the model and hypothesis used for an inferential comparison. No inferential conclusion can be reconstructed from the endpoint definition alone.

Why does distributional shape matter?

In general, a standard deviation is most straightforward to interpret when the distribution is reasonably well described by its mean and spread. A percentage is bounded, and observed values can cluster near a boundary or show unequal variability. Those features may matter when selecting a model or calculating uncertainty. The record specifies mean and standard deviation as the endpoint summaries, but does not report the observed distribution or a model assessment. No additional distributional assumption should be attributed to the investigators.

Over-dispersion is a general concern when observed variability is greater than a model assumes. If a model understates variability, its uncertainty estimates may be too narrow. The registry does not report a count model or an over-dispersion assessment for this continuous endpoint, so over-dispersion cannot be identified as a specific issue in the trial analysis. It is relevant as a general reminder that model fit and variability assumptions influence precision.

What do the planning assumptions mean?

The registry describes sample-size illustrations calculated from the observed effect, rather than the trial’s own design. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the illustrations consider detecting a stated fraction of the observed effect. For non-time-to-event outcomes, the convention is participants per arm. These illustrations are planning aids, not the number of participants actually analyzed and not observed power. The record does not report the original sample-size justification, the observed effect used in the illustrations, or numerical illustration results.

In general, a planning calculation asks how much information a future study might need to detect a specified effect under stated assumptions. Its answer depends on the effect size, variability, design, and analysis method. It does not show that a completed study had that probability of finding an effect, and it cannot replace a reported estimate with its uncertainty. The record does not report the assumptions used to choose enrollment of 50 for the trial itself.

Overall, the endpoint and some planning assumptions are described, but the numerical primary results and the analysis procedures needed to interpret them are not reported in the registry record. Any more specific account of the fitted model, adjustment strategy, treatment contrast, or inferential result would go beyond the registered information.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Older Adults:Sensor Augmented Pump (SAP)", "Older Adults: Closed-Loop Control (CLC)", "Young Children: SAP", "Young Children: CLC")
n   <- c(15, 15, 13, 13)
m   <- c(69.6, 79.6, 50.9, 69.2)
sp  <- c(14.2, 7.8, 12, 9.2)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Older Adults:Sensor Augmented Pump (SAP)  15  69.600  14.200  61.736  77.464
CARM  Older Adults: Closed-Loop Control (CLC)  15  79.600  7.800  75.281  83.919
CARM  Young Children: SAP  13  50.900  12.000  43.648  58.152
CARM  Young Children: CLC  13  69.200  9.200  63.640  74.760

11. Multiplicity, Missing Data and Analysis Populations

The study lists multiple aims: glucose time in range, cognitive function, and preliminary assessment in two age groups. It also lists Ecological Momentary Assessments and sleep patterns as secondary outcomes. When a study examines several outcomes, subgroups, time periods, or comparisons, general statistical practice considers the possibility that some results may appear unusual by chance. Whether and how to address this issue depends on which analyses were planned and how conclusions are framed.

The registry record does not report a multiplicity plan, such as a prespecified hierarchy of outcomes or an adjustment for multiple testing. It also does not report a detailed statistical analysis plan, so it is not possible to determine which outcomes or subgroup analyses were confirmatory. The absence of a reported plan should not be interpreted as evidence that no such plan existed outside the registry record.

Missing-data handling is also not reported in the registry record. The endpoint depends on continuous glucose-monitoring data over a month, but the record does not specify minimum sensor wear, rules for incomplete days, imputation, exclusions, or sensitivity analyses. The analysis populations are not described: there is no reported intention-to-treat, per-protocol, or other population definition, and no primary-outcome analysis denominator is given. Enrollment of 50 does not establish how many participants contributed usable endpoint data.

These omissions limit assessment of selection, missingness, and the extent to which the reported analysis—if available elsewhere—matches the registered aims. They also prevent a reader from determining how multiple outcomes and incomplete observations affected the precision or interpretation of the findings.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The adverse-event figures are reported separately for older adults and younger children during SAP and CLC. For older adults, the registry lists 15 at risk in each period; for younger children, it lists 13 at risk in each period. The serious adverse event count is 0 in every listed group, and the deaths count is 0 in every listed group. Other adverse events are also listed as 0 in each group.

These figures describe the recorded counts in the registry, not a complete characterization of safety. In general, a zero event count means no events of that category are recorded for the listed group; it does not establish that the underlying risk is zero. Interpretation depends on the observation period, event definitions, ascertainment, and completeness of reporting. The record does not report further details about those procedures or the duration of safety follow-up.

The group counts should not be combined into an overall comparison, especially because the study is sequential and the registry cautions that the arms are not a like-for-like control comparison. The adverse-event entries provide a count-based record, but do not support a causal comparison or a conclusion about relative safety.

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

The registry record describes a pilot study that measured the percentage of time blood glucose was in the 70-180 mg/dL range during SAP and CLC study periods. It also outlines aims related to cognitive function in children aged 6 to 10 years and adults aged 65 years or older. The record’s adverse-event entries list no serious events, other adverse events, or deaths in the displayed groups. These are descriptions of registered information, not a treatment recommendation.

What the record does not establish is the magnitude or precision of the primary endpoint results. It contains no numerical period means, standard deviations, estimated contrast, confidence interval, or p-value for the primary outcome. Cognitive-function findings and age-specific primary-outcome results are also not reported in the registry record. Without these results, the endpoint definition and study objectives cannot answer whether the periods differed or how large any difference was.

The sequential design and the registry’s warning that the arms are not a like-for-like control comparison further limit causal interpretation. The registry does not report the detailed sequence, the analysis method, or how period effects and missing sensor data were handled. Even if numerical results were available, those design and analysis details would be important to understanding what the comparison represents.

The study is characterized as a pilot intended to generate preliminary data for larger and longer trials. In general, a pilot can help inform later study planning, but its purpose and scale alone do not establish an effect. The planning illustrations described in the record are not observed power and do not substitute for outcome estimates and uncertainty. The defensible reading is that the registry documents the question, endpoint, and design features, while leaving comparative efficacy and cognitive findings unresolved.

14. Why This Trial Matters Statistically

This trial raises statistical questions that are important when evaluating device studies in populations identified as vulnerable. Its objective combines a continuous glucose-monitoring outcome with cognitive assessment and includes children and older adults. Those aims can require careful definitions of the analysis population, outcome timing, and subgroup interpretation. The registry identifies these broad aims, but does not report the results or the detailed analysis plan needed to evaluate them.

The sequential structure is also important. The registry explicitly cautions that the arms are not a like-for-like control comparison. This directs attention to the distinction between describing outcomes in two study periods and making a causal comparison. General concerns include period order, time trends, and incomplete monitoring; the record does not say how these were addressed. Clear reporting of sequence and analysis procedures would help readers judge the strength of any comparison.

The primary endpoint illustrates a broader measurement issue: a continuous time-in-range outcome and a threshold-based definition of stable control are related but not interchangeable. The registry uses the continuous percentage as its primary endpoint and mentions more than 70% of the day in range in its cognitive-function objective. Distinguishing the measurement scale from a threshold can prevent a continuous result from being misread as a categorical achievement.

Finally, the record’s pilot framing makes transparent reporting of uncertainty especially useful. Numerical estimates, confidence intervals, missing-data rules, and analysis populations would show what the observed data can support and how much remains uncertain. Without those details, the registry can identify a relevant research question and design, but cannot establish comparative results.

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 trial-specific details needed to interpret the primary endpoint are not reported in the registry record. Although the primary outcome timeframe is one month during each study period, the record does not give the exact analysis window, the CLC procedures, or the sequence and order of periods. It describes SAP use for 4 weeks, with actigraphy and Ecological Momentary Assessments planned in approximately the final 14 days, but does not explain whether those monitoring procedures were identical in the other period.

The registry lists enrollment of 50 but does not report the number of participants contributing primary-outcome data, the distribution across the stated age groups, or exclusions from an analysis. Eligibility criteria, demographic characteristics, baseline glucose measures, and the distribution of sensor data completeness are not reported in the registry record. These omissions preclude assessment of how representative the analyzed participants were of the intended populations.

Numerical primary-outcome results, cognitive results, and detailed adverse-event definitions are not reported in the registry record. Neither the statistical model nor the treatment contrast is described. Missing-data procedures, handling of repeated sensor measurements, multiplicity decisions, and analysis populations are also not reported. Although the registry characterizes the study as a pilot and gives planning assumptions, it does not report the original sample-size justification.

The registry’s warning that the arms are not a like-for-like control comparison is a central limitation for interpretation. Without the sequence details and analytic approach, period comparisons cannot be given a more specific causal interpretation. These reporting limitations do not show that the omitted procedures or analyses were absent; they mean they cannot be evaluated from the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The VRIF Trial: Hypoglycemia Reduction With Automated-Insulin Delivery System — NCT03674281.
• Registry record last updated 2021-01-12; status last verified 2020-12.
• Results first posted 2021-01-12.
• 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

NCT03674281, the VRIF Trial, is listed as a completed sequential device study sponsored by the University of Virginia, with enrollment of 50 and no masking. The registry identifies SAP and CLC study periods in children aged 6 to 10 years and adults aged 65 years or older with type 1 diabetes.

The primary endpoint is mean percentage of time with blood glucose in the 70-180 mg/dL range during one month in each period, measured by continuous glucose monitoring. The registry also lists Ecological Momentary Assessments and sleep patterns as secondary outcomes and describes cognitive-function aims. It cautions that the arms are not a like-for-like control comparison.

Primary endpoint results, cognitive findings, and detailed statistical and missing-data methods are not reported in the registry record. The adverse-event entries list 0 serious events, 0 other adverse events, and 0 deaths in each displayed group.