Clinical Assessment of a Novel Microprobe Array Continuous Glucose Monitor for Type 1 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clinical Assessment of a Novel Microprobe Array Continuous Glucose Monitor for Type 1 Diabetes (NCT01908530), a non-randomized na study. The primary outcome is Number of Participant Developed the Skin Inflammation.

2. Clinical Question

The broader study purpose was to assess the safety and efficacy of the Imperial College Microprobe Array Continuous Glucose Sensor in healthy volunteers and subjects with type 1 diabetes. Secondary outcomes listed include pain score, participant skin penetration, detectable signal, correlation with venous blood glucose, and acceptability questionnaire. These outcomes indicate that the study also examined aspects of tolerability and sensor performance, but their results are not reported in the registry record.

3. Trial Design

The study is registered as completed, with Imperial College London listed as sponsor. Its start is recorded as 2013-11 and primary completion as 2017-06. The phase is listed as NA. Enrollment is recorded as 26, and the listed condition is Diabetes Mellitus.

The design is described as non-randomized, parallel, and unmasked. Participants were not randomly allocated according to the registry record, and no masking is listed. The arm descriptions distinguish phases of sensor application: healthy volunteers received the sensor for 6 hours in Phase 1 and for 24 hours in Phase 2, while Phase 3 and 4 involved participants with type 1 diabetes. The adverse-event group listing labels its final group Phase 3. The registry record does not explain this difference in labeling.

These phase descriptions do not constitute a like-for-like control comparison. The populations and, for the healthy-volunteer phases, exposure durations differ. Consequently, a contrast across phases could reflect those differences as well as any sensor-related experience. The record does not report a control intervention, random allocation procedure, or masking procedure.

The registry lists seven outcome measures and identifies the primary outcome as the number of participants who developed skin inflammation during 24 hours. It lists secondary outcomes concerning pain, skin penetration, signal detection, correlation with venous blood glucose, and acceptability. The registry record does not report participant flow, phase-specific enrollment beyond the adverse-event counts, or a protocol-level explanation of the study’s design choices.

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: Phase 4 did not go ahead

Overall Study

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 Number of Participant Developed the Skin Inflammation, measured over 24 hours.

The primary endpoint is the number of participants who developed skin inflammation, with a 24-hour timeframe. It is a count of participants with an event rather than a measure of inflammation severity or duration. The stated purpose is to assess safety of using the microprobe array continuous glucose sensor with respect to skin inflammation.

Skin inflammation is relevant to a sensor applied to the skin because a local reaction can describe tolerability of the device at its application site. As a binary outcome, each participant is classified according to whether the event occurred within the specified timeframe. A count can communicate how many participants experienced the event, but interpretation also depends on the number assessed and on consistent event definitions and ascertainment.

The registry record does not report the primary endpoint result, its analysis population, an endpoint denominator, or a definition of skin inflammation beyond the stated description. The adverse-event listing reports separate serious and other adverse-event counts, but it does not identify those entries as the primary endpoint result. Those safety entries therefore should not be substituted for the missing primary outcome result.

Without a reported endpoint result and supporting analysis, the record does not establish the observed frequency of skin inflammation under the primary outcome definition. It also does not report an estimate of uncertainty around that outcome. The endpoint’s safety purpose can be described, but its findings cannot be quantified from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical interpretation is constrained by what the registry record reports. It specifies a binary primary outcome as a count of participants with skin inflammation, but it does not give the primary outcome count or an analysis. The adverse-event counts are available as descriptive entries, but they are separate from the stated primary endpoint unless the registry explicitly connects them, which it does not.

What does a binary outcome represent?

For a binary outcome, each participant is classified as having experienced the event or not having experienced it within the outcome window. A count of participants with the event is a basic summary. To interpret that count fully, readers also need to know how many participants were assessed for the outcome, how missing observations were handled, and whether the event definition and assessment were applied consistently. Those details are not reported in the registry record for the primary endpoint.

A confidence interval is a range of values compatible with the observed data under a specified statistical method. It communicates precision as well as the direction and size of an estimate. A small observed count does not automatically imply a precise estimate; precision depends on the amount and structure of the data. No primary endpoint estimate or confidence interval is reported in the registry record, so its precision cannot be assessed.

Why does the design affect interpretation?

Random allocation, when used, can help balance measured and unmeasured participant characteristics across comparison groups. This study is listed as non-randomized, and its registry description says the arms are not a like-for-like control comparison. The phase groups also differ in their described populations or sensor application durations. As a result, comparisons among phases would be descriptive rather than a clean estimate of a treatment contrast. Differences could reflect the phase populations, duration, or other factors that the record does not describe.

Masking can reduce the influence of expectations on outcome assessment or reporting. No masking is listed. That matters particularly for outcomes involving subjective assessment, such as pain score and acceptability. The record does not describe how these outcomes were measured or whether assessors were independent of the intervention.

What do the adverse-event counts show?

The registry lists serious and other adverse-event counts separately, with at-risk counts for each group. It records 0 serious events and 0 other events for each listed group, with deaths recorded as 0 in each group. These values are descriptive. They are not a comparative estimate, and the record gives no inferential analysis or precision measure for them. The at-risk counts in these entries should not be assumed to be the primary endpoint denominators, because the record does not specify that relationship.

When event counts are sparse, conventional large-sample approximations can be unreliable. Exact or other methods suited to sparse data may be considered in a fully specified analysis, but the registry record does not say which method was used. No method should be attributed to the study without a reported analysis plan or result.

What do the planning assumptions mean?

The registry record describes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect. Such calculations are prospective illustrations of how a future study might be planned under assumptions; they are not evidence that the completed study had a particular probability of detecting an effect. They are not observed power.

Interpreting those illustrations requires an observed effect estimate and a relevant comparison. The registry record does not report the primary effect or a like-for-like control comparison. Therefore, planning assumptions alone cannot establish what effect the study observed or how precisely it estimated that effect. No sample-size justification for the original design is reported in the registry record.

What analyses are available?

The record reports seven outcome measures but no analyses. It gives no statistical model, hypothesis test, adjustment strategy, confidence interval method, or analysis population for the primary outcome. It also does not report an interim analysis plan or say whether any interim analyses occurred. These details should not be inferred from the design label or from the existence of outcome listings. Without the corresponding results and methods, statistical interpretation must remain descriptive and limited to the recorded counts.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Microprobe Glucose Sensor Phase 1", "Microprobe Glucose Sensor Phase 2", "Microprobe Glucose Sensor Phase 3")
n      <- c(8, 8, 10)
events <- c(0, 0, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Microprobe Glucose Sensor Phase 1  8  0  0.0  0.0  36.9  0.0  32.4  0.3  33.6
ARM  Microprobe Glucose Sensor Phase 2  8  0  0.0  0.0  36.9  0.0  32.4  0.3  33.6
ARM  Microprobe Glucose Sensor Phase 3  10  0  0.0  0.0  30.8  0.0  27.8  0.2  28.5

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists seven outcome measures, including the primary skin-inflammation outcome and secondary measures relating to pain, skin penetration, signal detection, correlation with venous blood glucose, and acceptability. When a study examines multiple outcomes, the chance of obtaining at least one apparently unusual result can increase if each is tested separately. Statistical plans may address this through prespecified outcome hierarchy or adjustment methods. The registry record does not report a multiplicity strategy, and it reports no analyses. No claim can therefore be made about whether or how multiplicity was addressed.

Missing-data handling is also not reported in the registry record. The record does not state how participants with unavailable assessments were treated, whether outcomes were collected for all participants, or whether any imputation or sensitivity analysis was used. For a binary endpoint, missing observations can affect both the event count and the denominator used to interpret it. The adverse-event at-risk counts do not establish the endpoint-specific denominators.

The analysis populations are not reported in the registry record. There is no statement identifying an intention-to-treat, per-protocol, safety, or other analysis set for the primary endpoint. Nor are participant flow, exclusions, or reasons for missing data described. These omissions limit assessment of how closely the recorded outcomes represent all enrolled participants.

Because neither analyses nor missing-data methods are reported, the available record cannot establish whether outcome reporting was complete or whether analytic decisions could have influenced the results. These are reporting gaps, not evidence that data were missing or that a particular method was used.

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 registry’s adverse-event entries separate serious events from other events and list the number at risk for each group. For Microprobe Glucose Sensor Phase 1, the record lists 0 serious adverse events and 0 other adverse events among 8 at risk. For Phase 2, it lists 0 serious adverse events and 0 other adverse events among 8 at risk. For Phase 3, it lists 0 serious adverse events and 0 other adverse events among 10 at risk. Deaths are recorded as 0 in each group.

These are counts as recorded, not estimates of comparative risk. The groups are not described as like-for-like controls, and the record does not provide confidence intervals or an analysis of these event counts. A zero recorded count means that no event was entered in that category for the listed group; it does not establish that the true risk is zero or that an event could not occur under other circumstances.

The primary endpoint is specifically the count of participants who developed skin inflammation during 24 hours. The adverse-event listing does not explicitly identify its entries as that primary outcome, so the two should be kept distinct. The record does not report the primary outcome result.

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

The registry record shows that the study was completed and that it was designed to assess a microprobe array continuous glucose sensor in healthy volunteers and participants with type 1 diabetes. It identifies skin inflammation within 24 hours as the primary safety outcome and lists secondary measures related to tolerability and sensor performance. It also records zero serious and zero other adverse events in each listed group, with zero deaths recorded.

Those details do not establish the observed primary endpoint result. The number of participants with skin inflammation is not reported in the registry record, and no statistical analyses are reported. The adverse-event counts cannot be treated as a substitute for that endpoint. Nor do they establish a precise estimate of safety, because the registry does not report uncertainty intervals, endpoint denominators, or details of event ascertainment.

The study’s non-randomized design and absence of masking also matter. Its phases do not form a like-for-like control comparison: the described healthy-volunteer phases differ in exposure duration, and later phases involve participants with type 1 diabetes. A difference between phases, if one were reported, would not automatically be attributable to the sensor alone.

The record does not report results for the secondary outcomes, including detectable signal, correlation with venous blood glucose, pain, skin penetration, or acceptability. It therefore cannot support a statistical conclusion about those outcomes. The distinction is between the study’s stated aims and the results that are available in the registry record: the aims are described, while numerical outcome findings and analyses are not.

Planning calculations, if presented alongside the record, should be understood as illustrations under stated assumptions, not as observed power or as evidence of the study’s findings. They cannot compensate for unreported outcome results.

14. Why This Trial Matters Statistically

This study raises statistical issues common to early evaluations of devices used across different populations and exposure conditions. Its phases involve healthy volunteers and participants with type 1 diabetes, and the healthy-volunteer phases have different stated application durations. Those design features make it important to distinguish safety description within a group from a controlled estimate of a device effect.

The primary endpoint is a participant-level event count, which can be useful for communicating whether a local safety event was observed. But the count is interpretable only alongside a clearly defined assessment population, event definition, follow-up window, and uncertainty estimate. The registry record states a 24-hour timeframe but does not report the primary outcome result or the associated analysis details. This makes transparent reporting especially important.

The posted adverse-event counts are all zero, but zero counts are not equivalent to proof of no risk. Statistical methods must account for uncertainty in sparse event data, and readers need adequate information about follow-up and assessment to judge what those counts can support. The record does not provide those details or inferential results.

The study also lists outcomes spanning safety, tolerability, and sensor performance. Multiple outcome domains can be valuable in device assessment, but a clear analysis plan helps readers distinguish primary conclusions from exploratory findings and understand how multiplicity was handled. That plan and the corresponding analyses are not reported in the registry record.

These considerations make the record useful for understanding the study’s aims and structure, while also showing why a completed status alone is not a substitute for transparent outcome reporting. The distinction between a stated question and a reported statistical answer is central to interpreting clinical research.

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.

Important details needed to interpret the findings are not reported in the registry record. Most notably, the primary outcome result for skin inflammation is absent, as are results for the listed secondary outcomes. No statistical analyses are reported, so there are no outcome estimates, confidence intervals, or tests to evaluate.

The design is non-randomized, unmasked, and described as lacking a like-for-like control comparison. The phase descriptions differ by population and, for healthy volunteers, duration of sensor application. These features limit causal interpretation of any comparison across phases. The record does not describe a control intervention or explain how potential differences between phases were addressed.

The record does not report the primary endpoint denominator, analysis population, event definition beyond the stated description, or method of outcome ascertainment. It also does not report missing-data handling, participant flow, multiplicity strategy, or the statistical plan. These omissions prevent assessment of endpoint completeness and analytic robustness.

Although the adverse-event entries record zero serious and zero other events in each group, the record does not provide uncertainty estimates or sufficient detail to interpret those entries as a comprehensive measure of device safety. The adverse-event entries also are not explicitly identified as the primary skin-inflammation endpoint. The discrepancy in phase labeling between the arm description and adverse-event group label is not explained in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Clinical Assessment of a Novel Microprobe Array Continuous Glucose Monitor for Type 1 Diabetes — NCT01908530.
• Registry record last updated 2019-11-20; status last verified 2019-10.
• Results first posted 2019-11-20.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes a completed, non-randomized, unmasked, parallel study sponsored by Imperial College London. Enrollment is listed as 26, with a start of 2013-11 and primary completion of 2017-06. The study assessed a Microprobe glucose sensor in healthy volunteers and participants with type 1 diabetes.

The primary outcome is the number of participants who developed skin inflammation during 24 hours. Secondary outcomes include pain score, skin penetration, detectable signal, correlation with venous blood glucose, and acceptability. The primary and secondary outcome results, statistical analyses, and analysis populations are not reported in the registry record.

The adverse-event entries record zero serious adverse events, zero other adverse events, and zero deaths for each listed group. These descriptive counts do not replace the missing primary endpoint result or establish a controlled comparison.