Single Hormone Closed Loop Study With PDT Sensor: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Single Hormone Closed Loop Study With PDT Sensor (NCT03528174), a na na study. The primary outcome is Mean Absolute Relative Difference of Sensed Glucose Values.

2. Clinical Question

The question concerns agreement between a glucose-sensing device and a reference blood glucose measurement during an automated insulin-delivery study. It is not a comparison of the device or closed-loop system against a concurrent control group. The record describes a study in which insulin-only closed-loop management was used, and the primary endpoint was the mean absolute relative difference of sensed glucose values.

The protocol summary describes an all-day study. The closed-loop system was to start after the Dexcom G5 sensor's 2-hour start-up, with an IV catheter used for frequent blood withdrawal for 10 hours. These details establish the measurement context, but the registry record does not report the actual endpoint value or other numerical glucose results. It therefore does not answer how closely the PDT system's readings agreed with the reference measurements in the enrolled participants.

3. Trial Design

NCT03528174, titled “Single Hormone Closed Loop Study With PDT Sensor,” is registered as a completed, single-group study sponsored by Oregon Health and Science University. The registry lists enrollment of 10 and identifies type 1 diabetes mellitus as the condition. The listed study start is 2018-04-16, and primary completion is listed as 2018-07-01. Phase is recorded as NA.

The single group was to have glucose managed with the Artificial Pancreas Control system using insulin only. Insulin was to be infused through the Pacific Diabetes Technologies CGM Insulin Infusion system. The protocol summary also describes a Dexcom G5 transmitter, a Nexus 5 smart phone, and two Tandem tslim pumps connected to two Pacific Diabetes Technologies glucose-sensing cannulas. The closed-loop system was to start after a G5 2-hour sensor start-up, with frequent blood withdrawal through an IV catheter for 10 hours.

Masking is listed as NONE, allocation as NA, and the study model as SINGLE GROUP. There is no randomized allocation or separate control arm described in the registry record. Thus, the design is suited to describing measurements collected during the stated study setup, but it cannot provide a like-for-like between-arm comparison. The record lists a primary endpoint timeframe of up to 10 hours.

The registry does not report a detailed statistical analysis plan, a sample-size justification, or additional design details such as eligibility criteria. The completed status and listed completion dates describe the study record; they do not, by themselves, establish that the endpoint results were analyzed or reported.

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 Mean Absolute Relative Difference of Sensed Glucose Values, measured over up to 10 hours.

The primary endpoint is the mean absolute relative difference of sensed glucose values, expressed as a percentage of reference blood glucose. The registry describes the measure as based on all reference YSI venous blood glucose values and gives a timeframe of up to 10 hours.

In general, mean absolute relative difference (MARD) summarizes the size of discrepancies between sensor readings and reference values. Each discrepancy is considered in absolute terms, so readings above and below the reference both contribute as differences rather than canceling one another. The discrepancies are scaled relative to the reference glucose values and then averaged. This makes MARD a summary of relative measurement error rather than a measure of glucose control or insulin-delivery performance on its own.

Using YSI venous blood glucose as the reference gives the sensor readings a comparator for assessing measurement agreement. The registry record does not report the resulting MARD, its standard deviation, the number of paired sensor-reference measurements, or a confidence interval. It therefore identifies what was intended to be measured, but does not quantify the observed agreement or its precision. The endpoint also does not, by itself, describe the range of individual discrepancies or how those discrepancies varied over the study period.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure: mean absolute relative difference of sensed glucose values, with standard deviation as the dispersion measure. The outcome is based on reference YSI venous blood glucose values and has a timeframe of up to 10 hours. The record does not report the observed mean or standard deviation.

What does the primary measure represent?

General statistical explanation: MARD summarizes the average size of relative differences between a measurement device and a reference method. Taking absolute differences prevents positive and negative discrepancies from canceling out. Scaling by the reference value puts discrepancies on a relative rather than purely absolute scale. An average is compact and interpretable, but it can conceal variation: a similar mean may arise from readings with consistently small discrepancies or from a mixture of closer and more discrepant readings. The registry record does not provide a distribution of individual differences or another summary of agreement.

The standard deviation describes how dispersed individual measurements are around the mean in the analyzed data. It is not the same as a confidence interval. A confidence interval, in general, expresses uncertainty in an estimated quantity, while a standard deviation describes variation among observations. Neither a confidence interval nor a reported estimate of the standard deviation appears in the registry record.

What does the study design allow?

The study model is SINGLE GROUP, allocation is NA, and masking is NONE. The endpoint compares sensor measurements with a reference blood measurement, but the registry does not describe a separate participant group receiving a different system. A reference measurement can help evaluate agreement without creating a randomized treatment comparison. As a result, any observed endpoint would describe device-reference differences in this study setup rather than an effect estimated between randomized arms.

The protocol summary describes an all-day closed-loop study using insulin only. It specifies a G5 sensor start-up before the closed-loop system began and frequent blood withdrawal through an IV catheter for 10 hours. The endpoint timeframe is listed as up to 10 hours. The registry does not report the actual timing or number of paired readings used in the analysis, so the detailed structure of the repeated measurements cannot be reconstructed.

How should the planning assumptions be read?

General statistical explanation: A sample-size calculation estimates how many observations may be needed under specified assumptions to detect a chosen effect with a chosen type-I error rate and power. Its result depends on the assumed effect, variability, design, and analysis. The registry's planning illustration states a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. The illustration is not the registered single-group design. The observed effect and the numerical planning outputs are not reported in the registry record.

Planning calculations are not observed power. Observed power is not a substitute for reporting an estimate and its uncertainty; a planned sample-size illustration also cannot show what the study found. No sample-size justification for the actual trial is reported in the registry record. The registry lists enrollment of 10, but enrollment alone does not show how many observations were available for each outcome or whether the planned analysis assumptions were met.

What can be concluded from the analysis fields?

The record lists 15 outcome measures and 0 analyses. This does not supply an estimate, uncertainty interval, or hypothesis-test result for the primary endpoint. It also does not establish whether analyses were performed outside the registry or whether the registry record is complete. The appropriate reading is limited to what is registered: the intended endpoint and its measurement basis are described, while numerical outcome analyses are not reported in the registry record.

General statistical explanation: For continuous outcomes, interpretation typically benefits from an estimate of the average, a measure of variability, and an uncertainty interval, along with a clear description of which observations entered the analysis. For repeated sensor readings, the analysis approach should also explain how measurements within the same participant are handled, since those readings are not necessarily independent. The registry record does not report these analysis details. They should not be inferred from the endpoint label or the brief protocol summary.

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("Single Hormone Closed Loop")
n   <- c(10)
m   <- c(13.5)
sp  <- c(4.6)
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  Single Hormone Closed Loop  10  13.500  4.600  10.209  16.791

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 15 outcome measures and 0 analyses. The primary outcome is mean absolute relative difference of sensed glucose values. Listed secondary outcomes include percent of time with sensed glucose between 70-180 mg/dl, less than 70 mg/dl, less than 54 mg/dl, and greater than 250 mg/dl; mean sensed glucose; number of carbohydrate treatments; mean amount of insulin delivered; and mean visual analog scale score at 15 minutes. Numerical results for these measures are not reported in the registry record.

When a study evaluates multiple outcomes or multiple tests, the chance of at least one apparently notable result can increase if many tests are considered without an appropriate plan. Statistical methods for addressing that issue depend on the intended questions and analysis plan. A prespecified hierarchy, adjustment approach, or other multiplicity strategy is not reported in the registry record. It is therefore not possible to determine how the listed outcomes were prioritized or how any multiple testing was handled.

The registry does not report how missing glucose readings, incomplete sensor-reference pairs, or other missing outcome data were handled. This does not establish that data were missing; it means the handling approach is not reported. The primary endpoint population is also not reported in the registry record. No analysis population definition or participant flow for the endpoint is described.

Interim analyses and a detailed statistical plan are not reported in the registry record. With no numerical outcome analyses or missing-data account, readers cannot assess whether the analyzed observations matched the intended measurement set or how robust any estimate might be to incomplete data.

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.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalization for emesis (10.0%).

The registry lists safety figures for the Single Hormone Closed Loop group. Serious adverse events are reported for 1 participant among 10 at risk. Other adverse events are reported for 0 among 10 at risk, and deaths are reported as 0. The top serious-event term is hospitalization for emesis, listed as 10.0%.

These figures describe the counts entered in the registry; they do not establish why an event occurred or whether it was related to a study device, insulin delivery, or another cause. The record does not report further event details, such as timing, duration, clinical course, or an assessment of relatedness. The serious-event term and the group-level count should therefore be read as safety reporting, not as evidence of causation.

The listed group is the study's single closed-loop group, not a comparison between treatment arms. With no separate control group, these entries do not support a comparative estimate of event risk. The registry's small enrolled group and the absence of additional event detail also limit how broadly the counts can be interpreted. The figures should be understood as the safety information recorded for this study, not as a complete characterization of safety in other settings or populations.

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

The record establishes that a single-group study was registered to assess mean absolute relative difference between sensed glucose values and reference YSI venous blood glucose values over up to 10 hours. It also describes the insulin-only closed-loop setup and lists enrollment of 10 participants. These details clarify the intended measurement and the conditions under which it was planned.

However, the registry record does not report the primary endpoint estimate, its standard deviation, or an uncertainty interval. It also does not report numerical results for the listed secondary outcomes. The record therefore does not quantify sensor-reference agreement, glucose-range measurements, insulin delivered, or other listed outcomes. In particular, the absence of an outcome estimate is not evidence that agreement was high or low, and it is not evidence that a clinically meaningful difference was or was not observed.

The design also limits the kinds of questions it can answer. It is registered as a single-group study with no separate control group and no randomized allocation. The reference blood measurements can serve as a basis for evaluating sensor readings, but they do not create a controlled comparison of the full closed-loop system against an alternative. Any conclusion about comparative performance or causal effects would require information not reported in this record.

The safety entries record serious adverse events for 1 participant, other adverse events for 0, and deaths for 0. These figures do not identify event cause or establish how the event profile compares with another system. More event detail is not reported in the registry record.

In short, the registry describes what the study set out to measure and records a small set of safety counts, but it does not provide the numerical outcome results needed to judge the observed measurement performance or its precision. The record supports a description of the study design, not a conclusion about the system's effectiveness or overall safety.

14. Why This Trial Matters Statistically

This study is statistically interesting because its primary question concerns measurement agreement in a device setup that combines glucose sensing and insulin infusion. Evaluating sensed glucose against reference YSI venous blood glucose focuses attention on the quality of the measurements that inform automated insulin delivery. The endpoint, mean absolute relative difference, is a continuous summary of relative discrepancies rather than a direct summary of treatment benefit.

The study also illustrates the distinction between a measurement reference and a randomized comparator. Reference blood samples can anchor an assessment of sensor readings, while a single-group design does not provide a concurrent alternative system for estimating a between-group contrast. Keeping those roles distinct prevents an accuracy endpoint from being misread as a controlled test of the entire system.

The record highlights why outcome estimates and uncertainty matter. Naming a primary endpoint and its measurement timeframe does not show the result. Without the mean, dispersion, and an uncertainty interval, readers cannot assess the observed size or precision of the discrepancy summary. Without the number and handling of paired readings, the analysis basis is also unclear.

Finally, the listed planning assumptions show why a sample-size illustration must be interpreted in light of the actual design and outcome information. The record describes assumptions framed around two arms, whereas the study is registered as single-group. Planning values cannot substitute for results or establish observed power. This distinction is important when translating statistical design language into claims about what a completed study demonstrates.

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.

A central limitation of the registry record is that it does not report the primary endpoint result or numerical results for the secondary outcomes. The mean absolute relative difference, its standard deviation, and a confidence interval are not reported in the registry record. Without them, the magnitude and precision of sensor-reference agreement cannot be assessed.

The registry describes a single-group study with no separate comparator and no randomized allocation. That design provides no concurrent control group for a comparative assessment of the full system. Although reference YSI venous measurements are named for the primary endpoint, the number and timing of paired readings and the handling of repeated measurements are not reported in the registry record.

Enrollment is listed as 10, but the number of participants contributing data to each outcome is not reported. Eligibility details, the primary analysis population, and handling of missing measurements are also not reported. These omissions prevent assessment of whether the analyzed data represent all enrolled participants or how incomplete observations might affect an estimate.

The study timeframe is described as up to 10 hours, with the protocol summary specifying an all-day study and frequent blood withdrawal for 10 hours after sensor start-up. This is a defined study setting; the registry does not establish how well measurements or any findings generalize beyond that setting. Further details about outcome analyses, multiplicity handling, and interim analyses are not reported in the registry record. The listed safety counts also lack event-level context and do not establish causation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Single Hormone Closed Loop Study With PDT Sensor — NCT03528174.
• Registry record last updated 2019-09-12; status last verified 2019-08.
• Results first posted 2019-09-12.
• 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

NCT03528174 is a completed, single-group study sponsored by Oregon Health and Science University, with enrollment of 10 participants and type 1 diabetes mellitus listed as the condition. Its primary endpoint is mean absolute relative difference of sensed glucose values, based on reference YSI venous blood glucose values over up to 10 hours. The study description specifies an insulin-only closed-loop system using the Artificial Pancreas Control system and the Pacific Diabetes Technologies CGM Insulin Infusion system.

The registry lists 15 outcome measures and 0 analyses. It records serious adverse events for 1 participant among 10 at risk, other adverse events for 0 among 10 at risk, and deaths as 0. The numerical primary and secondary outcome results, analysis population, missing-data methods, and detailed statistical plan are not reported in the registry record. Accordingly, the record describes the intended assessment but does not provide results with which to evaluate the observed glucose-measurement agreement.