Model Driven Diabetes Care: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Model Driven Diabetes Care (NCT01774149), a randomized na study. The primary outcome is Change in the Frequency of Hyper- and Hypo-glycemic Events From Baseline to Week 8-12..

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. Its title is Model Driven Diabetes Care, and its official title is Data Driven Feedback as a Method to Improve Glycaemic Control in Type 1 Diabetes. The condition listed is type 1 diabetes. The sponsor was University Hospital of North Norway. The phase is recorded as NA.

The two groups used the Few Touch Application, with the timing of Diastat activation distinguishing the registered arms. In the Diastat arm, the module was turned on in week 4 post-enrollment. In the Delayed Diastat arm, participants used the regular version of the application, with Diastat turned on in week 12 post-enrollment. Thus, the registry describes a comparison of earlier and delayed access to the module, rather than a comparison between application use and no application use.

The registry records enrollment of 30 and gives a study start of 2013-03, primary completion of 2013-08, and status of COMPLETED. The registered primary outcome was assessed during baseline, defined as the first 4 weeks post-enrollment, and weeks 8-12 post-enrollment. The registry does not report how the random sequence was generated, whether allocation was concealed, or whether any masking was used for outcome assessment. It also does not report a detailed protocol, sample-size justification, or analysis plan.

The outcome population was narrower than the enrolled population: the record says 14 participants in each group were active users and had sufficient data for the primary outcome analysis. Safety reporting lists 15 participants at risk in each group. These are different reported counts for different purposes; the registry does not explain the difference. The record does not report why participants lacked sufficient primary-outcome data or whether any participants discontinued application use.

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

Analysis population for the primary outcome, as defined in the registry: 14 participants in each group were active users and had sufficient data to be analyzed for the outcome.

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 Change in the Frequency of Hyper- and Hypo-glycemic Events From Baseline to Week 8-12., measured over Up to 12 weeks post-enrollment.

The primary endpoint is the change in frequency of hyper- and hypoglycemic events from baseline to weeks 8-12, expressed in events. The registry defines qualifying events by self-measured blood glucose values below 4 mmol/L (72 mg/dL) or above 15 mmol/L (270 mg/dL). Baseline covers the first 4 weeks post-enrollment, and the later assessment period is weeks 8-12 post-enrollment.

This endpoint focuses on the frequency of readings that cross either threshold over the specified periods. It is not the same as the average blood glucose level, the proportion of readings outside the thresholds, or the number of participants who experienced an event. The record does not state whether repeated qualifying readings were counted individually, how readings were handled when several occurred close together, or how the change score’s sign was defined.

Counting threshold-based events can describe one aspect of glucose management that may be relevant to the study’s question about feedback on self-recorded data. The registry also lists change in HbA1c as a secondary outcome, but HbA1c is not the primary endpoint. It represents a separate outcome measure, and its result is not included in the registry entries for this analysis. The primary endpoint’s definition therefore sets the scope for interpreting the posted group estimates: they concern recorded threshold events over specified time periods, not every dimension of diabetes management or health.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry labels the primary outcome as a mean of a continuous measure and gives dispersion as a 95% confidence interval. The underlying quantity is a change in event frequency, expressed in events. Although event counts are discrete at the participant level, an average change across participants can be treated as a continuous summary for reporting. The registry does not provide further details about the calculation of the mean or the confidence interval.

What does a confidence interval describe?

In general, a confidence interval expresses the statistical uncertainty around an estimate under the method used to calculate it. A narrower interval indicates greater precision under that method, while a wider interval indicates less precision. A confidence interval is not a range expected to contain every participant’s value. Nor does a group-specific interval, on its own, answer whether two randomized groups differ. A direct comparison requires an estimate of the contrast and its uncertainty, which the registry does not report as estimable for this endpoint.

The reported intervals are -25.0 to -13.0 for the Delayed Diastat group and -15.5 to -0.5 for the Diastat group. These describe uncertainty around the separate group means, not uncertainty around a between-group difference. The registry record does not describe the interval calculation method, distributional assumptions, or whether any adjustment was applied. Those details should not be inferred from the label “95% Confidence Interval.”

Why is the group comparison unavailable?

The registry explicitly indicates that values are missing or not estimable for a two-arm comparison of the continuous endpoint. This matters because randomization creates the intended basis for comparing assigned groups, but a randomized design alone does not supply a numerical contrast. The posted summaries show each group separately; they do not give a comparison estimate, standard error, test statistic, or p-value. It is not appropriate to create a contrast by subtracting the group means, and the posted record does not support a recalculated comparison.

More generally, a comparison analysis must use information about the variation and relationship of participant-level outcomes, along with a specified analysis method. Group means and their separate confidence intervals do not necessarily contain all information needed to reproduce the intended analysis. The registry does not report participant-level data or a method that would allow the missing comparison to be reconstructed. The appropriate reading is therefore that the registry reports arm-level estimates but no estimable two-arm result.

What is the role of the analysis population?

The primary outcome population is described as 14 active users in each group who had sufficient data to be analyzed. This identifies the reported outcome summaries as based on participants meeting those data-availability and use conditions. It does not establish that the analysis followed an intention-to-treat approach, nor does it describe how missing measurements were treated. The registry does not report whether all randomized participants were included in any analysis, how many lacked sufficient outcome data, or whether analysis populations were defined before results were examined.

In general, restricting an analysis to participants with available data can affect how well the estimate represents the randomized groups if availability is related to outcomes or treatment assignment. The direction and size of any such effect cannot be determined from the record. Information about reasons for missing data and the methods used to address them is not reported in the registry record.

How should planning estimates be understood?

In general, sample-size calculations are planning illustrations: they describe the number of participants that a hypothetical design would need under specified assumptions to target a chosen power for an effect. They are not evidence that a completed trial had that power, and they do not describe the observed precision of its estimates. The planning assumptions recorded for this trial specify a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as fractions of the observed effect. The registry does not report the trial’s own sample-size justification or provide the planning estimate values in the record available for this analysis.

Planning calculations based on an observed effect are particularly dependent on that effect estimate and on their assumptions. They should not be read as a retrospective statement of “observed power.” General statistical practice treats confidence intervals and the completeness of the comparison as more informative descriptions of the uncertainty in the posted outcome. In this trial, the registry provides arm-level confidence intervals but no estimable two-arm comparison, so planning assumptions cannot fill that gap.

The record does not report interim analyses, adjustment for multiple comparisons, a detailed statistical analysis plan, or a model used for the primary outcome. Because these elements are unspecified, no claims about them can be made. Statistical interpretation should remain tied to what is posted: randomized parallel groups, arm-level mean changes with confidence intervals, an outcome population defined by active use and sufficient data, and no estimable two-arm comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and identifies change in HbA1c as a secondary outcome. The primary outcome is change in the frequency of hyper- and hypoglycemic events. The record does not give results for the secondary outcome or describe the other outcome measures in the registry record for this analysis. It does not report a prespecified hierarchy among outcomes, a multiplicity adjustment, or a rule for interpreting several statistical tests. Whether and how multiplicity was handled is therefore not reported in the registry record.

The record reports zero analyses and states that a two-arm comparison for the continuous primary endpoint is missing or not estimable. It does not provide a p-value or comparative effect estimate. This is distinct from a finding of no difference: the registered information does not supply an estimable comparison from which such a finding could be made.

For the primary outcome, the registry says 14 participants in each group were active users and had sufficient data to be analyzed. It does not report how many participants were randomized to each arm, the number with missing primary-outcome data, the reasons data were unavailable, or how missing values were handled. There is no reported imputation method or sensitivity analysis. The record also does not state whether the analysis population was defined before outcome data were reviewed.

These omissions limit assessment of potential bias and the robustness of the result. In general, missing outcome data can affect an estimate if data availability is related to the outcome or group assignment, but the available information does not establish whether that occurred. The safety counts use a different at-risk population than the primary-outcome summaries; the registry does not explain that difference. Randomization procedures, allocation concealment, interim analyses, and the full statistical plan are not reported in the registry record.

11. Safety Results

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

Other (non-serious) adverse events

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

The registry reports zero serious adverse events among 15 participants at risk in the Delayed Diastat group and zero among 15 participants at risk in the Diastat group. It also reports zero other adverse events among 15 participants at risk in each group. No serious adverse-event terms are listed. Death counts are not reported in the registry record.

These figures describe events recorded in the registry for the stated at-risk groups. They do not establish that no adverse experiences occurred outside the categories or reporting period represented in the registry, and they do not establish that an intervention is free of risk. The record does not provide definitions of the adverse-event collection period, the method of surveillance, or further information about event ascertainment.

The safety population counts differ from the primary-outcome population counts: safety reporting lists 15 at risk in each group, whereas the primary outcome includes 14 active users with sufficient data in each group. The registry does not explain the difference. These counts should be kept tied to their respective outcomes rather than treated as interchangeable. The absence of reported events is a description of the posted safety figures, not a comparative safety analysis.

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

The posted primary-outcome results describe mean changes in threshold-event frequency within each randomized arm among active users with sufficient data. The Delayed Diastat group estimate is -18.0 events, with a 95% confidence interval from -25.0 to -13.0; the Diastat group estimate is -7.5 events, with a 95% confidence interval from -15.5 to -0.5. The registry does not define the sign convention for change, so the negative signs should not be translated into a direction of change without additional information.

The registry does not provide an estimable two-arm comparison. Therefore the group summaries cannot establish whether the earlier-versus-delayed timing of Diastat activation was associated with a difference in the primary outcome. The reported arm estimates are not a substitute for a comparative estimate, and no difference should be calculated from them. The absence of an estimable comparison also does not demonstrate that the groups were equivalent.

The endpoint captures specified self-measured glucose readings below 4 mmol/L (72 mg/dL) or above 15 mmol/L (270 mg/dL) during defined periods. It does not represent all clinical outcomes, the full pattern of glucose measurements, or HbA1c. Although change in HbA1c is listed as a secondary outcome, its result is not reported in the record for this analysis.

The analysis population comprised active users with sufficient outcome data, not necessarily every participant enrolled or randomized. The registry does not explain missing primary-outcome data or how it was handled. Consequently, the reported summaries should be understood as applying to the analyzed participants and should not automatically be generalized to all enrolled participants. The record also lacks details about participant characteristics, the randomization procedure, and the analysis method, which limits broader assessment.

In general, a trial’s posted estimates are interpreted alongside the definition of the outcome, the analyzed population, and the uncertainty of the estimates. Here, the confidence intervals convey uncertainty about group-specific means, but they do not answer the randomized comparison question. The results are descriptive registry findings and do not, on their own, establish a causal effect or support a clinical recommendation.

13. Why This Trial Matters Statistically

This trial raises a central statistical distinction: reporting an estimate in each randomized group is not the same as reporting the comparison created by randomization. The registry gives mean changes and confidence intervals for both arms, but indicates that a two-arm comparison for the continuous endpoint is missing or not estimable. A reader can describe the arm-specific summaries, but cannot infer a formal comparative result from them.

The distinction is important because the research question concerns the timing of Diastat activation. The primary endpoint is a change over time in threshold-based event frequency, and the two groups had different planned activation times. A direct estimate of the between-group contrast would address that randomized question more directly than separate within-group summaries. The record does not report such an estimate.

The analysis population also matters. The primary outcome includes active users with sufficient data, while safety counts are reported for a different at-risk population. These details make it important to identify which participants contribute to each result. The registry does not explain the difference between the populations or report the handling of missing outcome data, limiting assessment of whether the available summaries represent the randomized groups.

More generally, this record illustrates why a statistical report benefits from a clearly specified endpoint, analysis population, missing-data approach, comparison estimate, and uncertainty measure. The registry supplies some of these elements, including the endpoint definition and arm-level confidence intervals, but not the full analysis details. These gaps do not invalidate the descriptive numbers; they constrain what can be concluded from them.

The study is also a reminder that planning assumptions and observed trial results serve different purposes. A planning illustration based on an observed effect does not replace a reported treatment comparison or establish observed power. In this record, the key interpretive issue is not the label attached to a planning calculation but the absence of an estimable two-arm analysis for the primary outcome.

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.

Several trial-specific limitations follow from the registry record. The primary outcome population consists of active users with sufficient data, reported as 14 participants in each group. The record does not explain why other enrolled participants are not included in those summaries, how many lacked data, or whether the decision to analyze active users was prespecified. This limits assessment of selection and missing-data concerns.

The registry reports arm-level means and confidence intervals but no estimable two-arm comparison. It does not provide a p-value, comparative confidence interval, or analysis method for the primary endpoint. The confidence intervals describe separate group estimates and do not resolve the missing comparison. The sign convention for the change score is also not stated, restricting interpretation of the negative estimates.

The endpoint relies on self-measured blood glucose values meeting specified thresholds. The registry does not describe the completeness or timing of measurements, the handling of repeated readings, or procedures for checking recorded data. It also does not report participant characteristics that would help readers assess how closely the analyzed participants resemble other people with type 1 diabetes.

Details of random sequence generation, allocation concealment, masking of outcome assessment, and the statistical analysis plan are not reported. The trial is described as randomized and unmasked, but further procedural details cannot be assessed from the record. The registry lists a secondary HbA1c outcome without a result in the information analyzed; details for the other outcome measures are not reported. Finally, the safety counts and primary-outcome counts refer to different reported populations, and the reason for that difference is not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Model Driven Diabetes Care — NCT01774149.
• Registry record last updated 2015-09-16; status last verified 2015-08.
• Results first posted 2015-09-16.
• 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.

18. Record Summary

The registry describes a completed randomized, unmasked, parallel-group trial of the Few Touch Application with different timing of Diastat activation. Enrollment was 30, and the listed condition was type 1 diabetes. The primary outcome was change in the frequency of self-measured blood glucose events below 4 mmol/L (72 mg/dL) or above 15 mmol/L (270 mg/dL), from baseline to weeks 8-12.

For the primary outcome, 14 active users with sufficient data in each group contributed to the reported arm estimates. The Delayed Diastat mean change was -18.0 events (95% confidence interval -25.0 to -13.0); the Diastat mean change was -7.5 events (95% confidence interval -15.5 to -0.5). The registry does not report an estimable two-arm comparison. Safety records show zero serious and zero other adverse events among 15 at risk in each group; deaths are not reported in the registry record.

Change in HbA1c is listed as a secondary outcome. Missing-data methods, the detailed analysis plan, and the reason for the difference between outcome and safety populations are not reported.