Feasibility Trial Using an Inpatient Insulin Dosing Calculator: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Feasibility Trial Using an Inpatient Insulin Dosing Calculator (NCT06206525), a non-randomized na study. The primary outcome is Clinical Significant Hypoglycemia.

2. Clinical Question

3. Trial Design

This is a completed, non-randomized, single-group feasibility trial. The intervention was an initial insulin dosing calculator for patients with type 2 diabetes mellitus on hospital admission. The registry lists an enrollment of 38 and identifies the study sponsor as the University of Washington. The recorded start date is 2024-03-22, the primary completion date is 2025-05-31, and the status is completed.

Participants received their initial insulin doses through the calculator. The registry states that all participants in the primary endpoint population received at least the first dose by calculator before clinicians adjusted their insulin doses. Care after initial dosing followed standard of care. The comparator was not a concurrent randomized control group: it was a historical UW observational cohort of patients who received initial doses by standard of care. The registry lists 140 people at risk in that cohort for its adverse-event entries.

Allocation is recorded as non-randomized, masking as none, and the model as single group. These labels matter when interpreting any contrast with historical patients. Random assignment can help balance measured and unmeasured differences, whereas a historical comparison does not provide that protection. Differences between the groups could reflect clinical practice, patient mix, or other factors rather than the calculator. The registry record does not report details that would establish how comparable the cohorts were.

The trial is labelled phase NA. Its stated purpose is feasibility: to examine whether calculator use is associated with avoiding excessive hypoglycemia and decreasing hyperglycemia. The registry record does not report a detailed protocol, recruitment flow, or reasons for any exclusions or non-participation.

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: The observational cohort arm was historical 140 participants, and therefore, were not included in the enrollment number.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All participants received at least the first dose by the calculator before clinicians adjusted their insulin doses

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 Clinical Significant Hypoglycemia, measured over During the first 48 hours after admission.

The primary endpoint is clinical significant hypoglycemia, recorded as the count of participants who developed a blood glucose concentration lower than 54 mg/dL during the first 48 hours after admission. It is a binary participant-level outcome: each participant either met the registry’s threshold during the stated timeframe or did not. The endpoint description identifies the analysis population as all participants who received at least the first dose by calculator before clinicians adjusted their insulin doses.

A threshold-based outcome can make a safety event clinically interpretable by focusing on whether a specified glucose level was crossed, rather than only comparing average glucose values. It also reflects the trial’s stated concern about excessive hypoglycemia during initial inpatient dosing. This is a general explanation of why a threshold endpoint may be used; the registry record does not provide a rationale for selecting this particular threshold or timeframe.

The primary endpoint’s title, definition, and timeframe are reported, but its observed count and any comparative estimate are not reported in the registry record. The listed secondary outcomes are mean blood glucose concentration in mg/dL, clinical significant hyperglycemia, and hypoglycemia lower than 70 mg/dL. Their outcome values are also not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study uses a non-randomized design with a calculator-treated group and a historical observational comparator. In general, randomization helps distribute known and unknown prognostic factors across groups. A historical comparator does not ensure that balance. As a result, any observed difference between the groups could be related to differences in participants, care processes, or the time periods represented, as well as to the initial dosing approach. The registry record does not report an adjustment strategy or enough information to assess whether such differences were addressed.

What does a binary endpoint represent?

The primary outcome is a participant count: whether a participant developed blood glucose concentration lower than 54 mg/dL during the first 48 hours after admission. This is a binary endpoint because each participant is classified according to whether the defined event occurred. In a complete report, counts can be accompanied by an estimate of the event frequency and an uncertainty interval. The registry record does not report the primary outcome count or an interval, so neither the event frequency nor its precision can be assessed from this record.

A confidence interval is a general statistical way to express uncertainty around an estimate. It gives a range of values compatible with the data under the method’s assumptions; it does not guarantee that the true value lies in that range. For a comparative study, an interval around a treatment contrast would help readers assess which effects remain compatible with the data. No confidence interval for the primary or secondary outcomes is reported in the registry record.

Why does the historical comparison need caution?

In a non-randomized comparison, differences in baseline risk can influence outcome contrasts. For example, if groups differ in clinical characteristics or in care practices, their outcomes may differ even if the dosing approach has no effect. This is general statistical reasoning, not a claim that the trial groups did differ in any particular way. The registry record does not report group characteristics, a matching procedure, covariate adjustment, or other details that would allow the extent of this concern to be evaluated.

The record identifies the model as single group while also describing an observational cohort comparator. These descriptions indicate that the calculator intervention was not assigned through random allocation and that the standard-care cohort was historical. They do not establish a like-for-like control comparison. The registry record does not report whether the outcome analyses account for the different sources of the two groups.

What do the planning assumptions mean?

The stated sample-size assumptions describe a planning illustration based on the observed effect, not the trial’s own design. The assumptions specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. For non-time-to-event outcomes, planning quantities are expressed as participants per arm for 80% or 90% power; time-to-event planning uses total events. These are planning conventions, not evidence about the probability that the completed study detected an effect. A sample-size calculation based on an observed effect is not observed power.

The registry record does not report a sample-size justification for the feasibility trial itself, nor does it report the underlying effect estimate or the calculated planning quantities in the registry record for this analysis. The planning assumptions therefore cannot be used to infer the trial’s intended or achieved statistical power. More generally, a planning calculation depends on choices about the effect worth detecting, variability or event frequency, allocation, and analysis method. It cannot replace reporting the observed outcomes and their uncertainty.

Why can over-dispersion matter?

Over-dispersion is a general concern when count data vary more than a statistical model assumes. If it is ignored, standard errors can be too small and confidence intervals too narrow. The primary endpoint in this trial is a participant count for a binary event, but the registry record does not report a model or dispersion assessment. It would therefore be inappropriate to assert that over-dispersion was present or that a particular method was used to address it.

The registry lists four outcome measures and four analyses, but detailed specifications of those analyses are not reported in the registry record. The named secondary outcomes are mean blood glucose concentration, clinical significant hyperglycemia, and hypoglycemia lower than 70 mg/dL. The record does not give the corresponding estimates, uncertainty intervals, or analysis procedures. Statistical interpretation is consequently limited to the stated design and outcome definitions, rather than conclusions about estimated effects.

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("Insulin Dosing Calculator", "Observational Cohort")
n      <- c(38, 140)
events <- c(0, 1)
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  Insulin Dosing Calculator  38  0  0.0  0.0  9.3  0.0  9.2  0.1  9.0
ARM  Observational Cohort  140  1  0.7  0.0  3.9  0.1  3.9  0.2  3.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and four analyses, but it does not report a multiplicity adjustment or explain how the analyses relate to one another. Multiplicity is a general issue that arises when several outcomes or statistical tests are considered: the chance of at least one apparently notable result can increase as more tests are conducted. A prespecified hierarchy, adjustment method, or clear designation of primary and secondary analyses can help readers understand how multiple results should be interpreted. The registry record does not report such details beyond naming the primary endpoint and listing secondary outcomes.

The primary endpoint population is described as all participants who received at least the first dose by calculator before clinicians adjusted their insulin doses. The registry record does not report how missing endpoint observations were handled, whether any participants had missing outcome data, or whether analyses used imputation or another missing-data method. It also does not report a detailed analysis population for the historical cohort or explain whether all enrolled participants were included in each outcome analysis.

The record does not report an intention-to-treat analysis, a per-protocol analysis, or sensitivity analyses. These terms describe general approaches to defining and testing analysis populations; their absence from the record should not be taken as proof that none were conducted. Rather, the registry record does not provide enough information to determine which populations were analyzed or how missing data and multiple comparisons were addressed.

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 list the calculator group with 38 participants at risk and zero affected by serious adverse events. The observational cohort has 140 participants at risk and zero affected by serious adverse events. Deaths are listed as zero in both groups, and no serious-event terms are listed. The entries also list zero affected by other adverse events in each group.

These entries describe the adverse-event fields in the registry; they are not the primary endpoint results. The primary endpoint concerns clinical significant hypoglycemia, defined as blood glucose concentration lower than 54 mg/dL during the first 48 hours after admission. The registry record does not report the observed count for that endpoint. A zero serious-adverse-event count therefore cannot be interpreted as a zero count for clinical significant hypoglycemia.

Zero observed events does not mean that the underlying risk is known to be zero. In general, when no events are recorded, uncertainty about the possible event frequency remains, and the amount of information depends in part on the group size and the event definition. The registry record does not provide confidence intervals for the adverse-event counts. Also, the groups are not randomized or like-for-like by design, so their counts should not be treated as a controlled safety comparison.

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

The registry record establishes that a completed feasibility trial evaluated an initial insulin dosing calculator for patients with type 2 diabetes mellitus on hospital admission. It identifies the primary safety endpoint, the observation timeframe, the intervention process, and a historical observational cohort. It also lists secondary glucose outcomes and adverse-event entries. These details clarify what the trial intended to examine.

The record does not report the primary endpoint result, secondary outcome values, comparative estimates, or confidence intervals. Consequently, it does not show whether clinical significant hypoglycemia occurred in the calculator group, whether hyperglycemia decreased, or how precise any difference might be. The zero serious adverse event entries are limited to those adverse-event fields and do not supply the missing endpoint results.

The non-randomized design also limits causal interpretation. A contrast between participants receiving calculator-based initial doses and historical patients receiving standard-care doses could reflect differences between the cohorts or their care context. Without reported group characteristics and analytical details, the registry record does not allow those possibilities to be assessed.

These limitations do not negate the trial’s feasibility focus. A feasibility study can address whether an approach is suitable for further evaluation, but the registry record does not report specific feasibility metrics such as adherence to calculator recommendations, recruitment performance, or workflow measures. Such measures should not be assumed from the study label.

The result, as represented in the registry record, is therefore principally a description of the study question and design rather than a quantitative estimate of the calculator’s effects. It does not establish that the calculator improved glucose management, nor does it establish that it failed to do so. Those conclusions require outcome results and analysis details that are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial raises a practical statistical question: how should an initial dosing strategy be evaluated when the early outcome of interest includes both avoiding low glucose and reducing high glucose? The registry identifies clinical significant hypoglycemia as the primary endpoint and lists mean blood glucose concentration, clinical significant hyperglycemia, and hypoglycemia lower than 70 mg/dL as secondary outcomes. Considering outcomes on both sides of glucose management illustrates why a single summary measure may not capture every aspect of an intervention’s performance. This is a general point about outcome selection, not a conclusion about this trial’s results.

The trial also illustrates the importance of separating feasibility questions from comparative-effectiveness questions. A feasibility study can motivate further evaluation, while a non-randomized comparison with historical patients is vulnerable to differences between groups. Clear reporting of the population, outcome definitions, analysis methods, and uncertainty is especially important when the groups were not assigned at random.

Finally, the record shows why endpoint definitions and outcome values must be read separately. The primary endpoint threshold and timeframe are specified, but its event count is not reported. Likewise, zero serious adverse event entries do not answer the distinct question about the primary hypoglycemia endpoint. Statistical interpretation depends on having results for the particular outcome being discussed, together with enough information to judge their precision and the design’s limitations.

These are broader lessons in trial interpretation. Naming a primary endpoint does not provide its result, and a feasibility label does not substitute for reported feasibility outcomes. For this trial, the registry record makes the intended questions visible but does not report the quantitative information needed to answer them.

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.

The central limitation for interpreting this trial is the absence of reported outcome estimates. The registry record specifies the primary endpoint and its timeframe, but does not report the count of participants who developed blood glucose concentration lower than 54 mg/dL during the first 48 hours after admission. It also does not report results for the listed secondary outcomes. This prevents assessment of the observed glucose outcomes and their precision.

The comparator is a historical observational cohort, and allocation is non-randomized. The registry record does not provide cohort characteristics, a description of differences in care context, or adjustment methods that could clarify comparability. Any contrast between the groups is therefore subject to potential confounding, and the available information does not support attributing a difference to the calculator.

The primary population is described, but the record does not report missing-data handling, participant flow, or detailed analysis populations. It also does not report multiplicity procedures, outcome analysis methods, or confidence intervals. The sample-size assumptions describe a planning illustration, not the trial’s own design, and the registry record does not state a sample-size justification for the feasibility trial.

Finally, the record does not provide detailed feasibility measures such as calculator uptake or workflow performance. The reported zero serious adverse event entries do not resolve the missing primary endpoint result and are not a substitute for a full account of the glucose outcomes. These gaps constrain interpretation to the trial’s stated design and intended questions.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Feasibility Trial Using an Inpatient Insulin Dosing Calculator — NCT06206525.
• Registry record last updated 2026-05-11; status last verified 2026-04.
• Results first posted 2026-04-15.
• 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.

19. Record Summary

NCT06206525 is a completed feasibility trial titled “Feasibility Trial Using an Inpatient Insulin Dosing Calculator.” It enrolled 38 participants and evaluated calculator-based initial insulin dosing for patients with type 2 diabetes mellitus on hospital admission. The record identifies the primary endpoint as clinical significant hypoglycemia, defined as blood glucose concentration lower than 54 mg/dL during the first 48 hours after admission.

The design was non-randomized, unmasked, and described as single group, with a historical UW observational cohort receiving standard-care initial doses as comparator. The registry lists mean blood glucose concentration, clinical significant hyperglycemia, and hypoglycemia lower than 70 mg/dL as secondary outcomes. It does not report results for the primary or secondary outcomes. Its adverse-event entries list zero serious adverse events in both groups, with 38 and 140 participants at risk, respectively. The registry record does not report comparative effect estimates or confidence intervals.